Proefschrift de Vries

Patient perspectives in the benefit-risk evaluation of drugs

Uitnodiging

voor het bijwonen van de openbare verdediging van

mijn proefschrift

Patient perspectives in the benefit-risk

evaluation of drugs

op woensdag 22 april 2015om 16.15 uur in de aula

van het academiegebouw vande Rijksuniversiteit Groningen,

Broerstraat 5 te Groningen.

Aansluitend op de promotieis er een receptie in het

Academiegebouw.

Sieta de VriesHunze 12,

9204 BP Drachten06-14227380

[email protected]

Paranimfen

Marrit Groen ([email protected])

Freya Hornyák ([email protected])

Dianna de Vries ([email protected])

Sieta T. d

e Vries


evaluation of drugs

Sieta T. de Vries


The patient perspective in the process of drug evaluation and drug use is high on the agenda, which is demonstrated by an increased use of patient-reported outcome instruments to evaluate drugs and a shift towards patient-centred care in clinical practice. This thesis contains studies focusing on 1) the development and validation of a patient-reported outcome instrument to assess adverse drug events (ADEs), and 2) the role of patient characteristics and preferences on treatment decisions in clinical practice. The first part presents the development of a generic questionnaire to assess ADEs from the patient perspective. Although this questionnaire showed sufficient content and concurrent validity to detect ADEs at a general level, it was not sensitive enough to detect all ADEs perceived by patients. Suggestions are provided to improve the questionnaire for future use. In the second part, insight in decisions to start or intensify treatment with special attention for different patient age groups is provided. It was found that age influenced prescribing behaviour as well as the patient’s willingness to add a drug. For all patients, preventing death and ADEs were important considerations when choosing an additional drug. The influence of beliefs about benefits and risks on patients’ drug adherence, however, differed among types of drugs. These findings can be used to improve the assessment of ADEs from the patient perspective, to incorporate the patient perspective in treatment decisions and to develop better tailored interventions for improving drug adherence.


Sieta T. de Vries

The studies presented in this thesis were funded by the graduate school for Health Services Research (SHARE) of the University of Groningen and the Escher project (T6-202), a project of the Dutch Top Institute Pharma.

Printing of this thesis was partially supported by the University of Groningen, the SHARE graduate school and the University Medical Center Groningen.

ISBN: 978-90-367-7683-7(printed version)ISBN: 978-90-367-7682-0 (digital version)

Cover design, lay-out design and printed by: Gildeprint – Enschede © 2015, S.T. de VriesNo parts of this thesis may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording or any information storage and retrieval system, without permission of the author.


Proefschrift

ter verkrijging van de graad van doctor aan de Rijksuniversiteit Groningen

op gezag van de rector magnificus prof. dr. E. Sterken

en volgens besluit van het College voor Promoties.

De openbare verdediging zal plaatsvinden op

woensdag 22 april 2015 om 16.15 uur

door

Sietske Trijntje de Vries

geboren op 27 juli 1986 te Smallingerland

PromotoresProf. dr. P. DenigProf. dr. F.M. Haaijer-RuskampProf. dr. D. de Zeeuw

BeoordelingscommissieProf. dr. G. NijpelsProf. dr. E.P. van PuijenbroekProf. dr. B.H. Stricker

ParanimfenMarrit GroenFreya HornyákDianna de Vries

Table of contents General introduction 9

Part I. Development and validation of a patient-reported adverse drug event questionnaire

Chapter 1. Development and initial validation of a patient-reported 25 adverse drug event questionnaireSupplement I. User acceptance of a web-based version of a 49 patient-reported adverse drug event questionnaireChapter 2. Construct and concurrent validity of a patient-reported 57 adverse drug event questionnaire: a cross-sectional study Chapter 3. The validity of a patient-reported adverse drug event 75 questionnaire using different recall periodsSupplement II. Illustrations and possible solutions of problems in the use 89 of patient reports

Intermezzo. The assessment and management of adverse drug events by 95 patients and healthcare professionals in clinical practice: a case-report

Part II. The role of patient characteristics and preferences on treatment decisions in clinical practice

Chapter 4. Potential overtreatment and undertreatment of diabetes in 105 different patient age groups in primary care after the introduction of performance measuresChapter 5. The role of patient’s age on their preferences for choosing 121 additional blood pressure-lowering drugs: a discrete choice experiment in patients with diabetesChapter 6. Medication beliefs, treatment complexity, and non-adherence 141 to different drug classes in patients with type 2 diabetes

Summary and general discussion 155Nederlandse samenvatting 173Fryske gearfetting 181

Appendices

Appendix 1. Patient-reported adverse drug event questionnaire 191 Appendix 2. Supplemental tables chapter 1 217Appendix 3. Supplemental tables chapter 2 219 Appendix 4. Supplemental tables chapter 3 225Appendix 5. Supplemental tables chapter 4 226 Appendix 6. Supplemental tables chapter 5 229Appendix 7. Supplemental tables chapter 6 231Curriculum vitae 233Dankwoord 237SHARE publications 243

General introduction

R1R2R3R4R5R6R7R8R9

R10R11R12R13R14R15R16R17R18R19R20R21R22R23R24R25R26R27R28R29R30R31R32R33R34R35R36R37R38R39


10

Today, the patient perspective is high on the agenda in our society in general and particularly in the process of drug use. This process includes both the evaluation of drugs and their actual prescribing. The evaluation of a drug is based on balancing its efficacy and safety. Incorporating patient perspectives in the process of drug use has gained much attention over the past decade at both, regulatory and clinical practice level. At regulatory level, the use of patient-reported outcome instruments has increased [1,2]. In such instruments, the patient is the direct source of information without an interpretation of their responses by a healthcare professional [3-5]. At clinical practice level, the patient perspective is important to tailor a drug decision to characteristics and needs of the individual. This importance is illustrated by diabetes guidelines stating that patient-centred care is part of optimal diabetes management [6].

Drug evaluation in the regulatory processBefore a drug comes to the market, the effects of the drug have been evaluated in animals and clinical trials with healthy people and patients. However, in clinical trials the drug is followed for a relatively short period of time and a low number and selected sample of patients are included, with usually an underrepresentation of children, aged patients, women, ethnic minorities, and patients with comorbidity and polypharmacy [7-14]. Moreover, the quality of clinical trial evidence used as the basis for drug approvals may vary widely across indications [15]. Therefore, it would be better to use a life-cycle approach with a continually benefit-risk evaluation of the drug in both clinical trials and post-marketing studies.

Post-marketing studies should especially focus on the risks of the drug since clinical trials are often designed and powered to assess the benefits. This limits their ability to detect for instance less common risks [12,16]. In the literature, various terms and definitions have been used for the assessment of patients’ risks to drugs [17]. The term adverse event is generally used to indicate “[..] anything adverse that happened to a patient. It may happen as a consequence of a disease, a procedure, or an adverse drug reaction” [18]. An adverse drug reaction is defined by the World Health Organization as “a noxious and unintended response to a medicine that occurs at normal therapeutic doses used in humans for prophylaxis, diagnosis, or therapy of disease, or for the modification of physiologic function” [19]. The Food and Drug Administration (FDA) in the USA [20] and the European Medicines Agency (EMA) in Europe [21] register these adverse drug reactions in the safety profile of a drug. A third term is adverse drug event (ADE), defined by the World Health Organization as “any untoward medical occurrence that may present during treatment with a pharmaceutical product but which does not necessarily have a causal relationship with this treatment” [19]. Throughout this thesis, the term ADE is used without differentiating adverse drug reactions from ADEs. The term side effect is used as the lay-term for an ADE.

R1R2R3R4R5R6R7R8R9R10R11R12R13R14R15R16R17R18R19R20R21R22R23R24R25R26R27R28R29R30R31R32R33R34R35R36R37R38R39


11

Different methods can be used to evaluate the risks of the drug in the post-marketing phase. Spontaneous reporting of ADEs has worldwide been the leading method for decades [22]. Most governmental agencies around the world use this method to detect rare ADEs and the more serious ADEs [23]. The method is limited by, for instance, the often restricted information given about the ADE, and under-reporting of ADEs. These limitations restrict the possibility for a causality assessment and to quantify ADE rates [24,25]. Another method used in the post-marketing phase is event monitoring in which all ADEs, also the less serious ones, are assessed. In event monitoring, observational cohort studies are conducted assessing any adverse event experienced by the patient since the investigated drug has been prescribed [7,23,26,27]. Examples of such monitoring programs are the Intensive Monitoring Medicines Programme in New Zealand [26], the Prescription Event Monitoring in England [27], and the Lareb Intensive Monitoring programme in the Netherlands [28].

Patient perspective in the process of drug evaluationIn the past, drug evaluation in clinical trials and post-marketing studies was mainly based on reports of healthcare professionals. Over time, the use of patient-reported outcome instruments has increased. Patient-reported outcome instruments can be used to measure various types of outcomes such as physical functions, psychological well-being, treatment satisfaction, and ADEs [29,30]. The added value of incorporating the patient perspective in the evaluation of, for instance, the safety of drugs has generally been acknowledged [31-35]. ADEs reported by patients provides additional information to ADEs reported by healthcare professionals since the latter miss some potential ADEs. Examples are ADEs that are considered to be less relevant by healthcare professionals, that are less easily communicated by patients to the healthcare professional, that are not being reported by the patient due to a fear of the healthcare professional’s reaction, and that are of symptomatic nature [36-38].

With respect to clinical trials, the inclusion of patient perspectives in regulatory submissions for labelling drug claims began to appear in the mid-1990s [39]. In 2005, the EMA released a reflection paper to give some broad recommendations on the use of patient-reported outcome instruments in the evaluation of drugs [40]. In 2006, the FDA released a draft version of guidance to use patient-reported outcome instruments to support potential treatment claims in product labelling [41]. The final version of this guidance was released in 2009 [3]. Furthermore, there are several initiatives to involve patient representatives in the evaluation of drugs at a regulatory level [42,43]. These initiatives and the released documents reflect the increased attention to incorporate patient perspectives in the drug approval process [44]. A literature review of the period 2006-2010 showed that almost a quarter of newly approved drugs by the FDA included

R1R2R3R4R5R6R7R8R9



12

patient-reported outcome claims, with most of them granted for symptom reduction or increased functioning [45]. Although direct patient-reporting is considered essential in the assessment of such beneficial effects, a standard patient-reported outcome instrument to assess the safety of a drug is lacking [35]. In the USA, work is being conducted to create a patient-reported version of the Common Terminology Criteria for Adverse Events (CTCAE) which is the standard approach of ADE reporting by research staff in oncology trials [35]. The patient-reported version of the CTCAE cannot be generally used in clinical trials since the instrument is developed specifically for oncology trials.

In the post-marketing phase of drug evaluation, the increased attention to incorporate the patient perspective is illustrated by, for instance, the Dutch pharmacovigilance centre Lareb. From 2003 onwards, Lareb allows patients to directly submit their ADE reports to the spontaneous reporting system. The value of these patient reports was considered equal to the value of healthcare professional reports since 2004 [46]. In 2006, the Lareb Intensive Monitoring programme was introduced in which new users of a drug under investigation are asked to complete several questionnaires over time [28]. Studies conducted with Lareb Intensive Monitoring showed that this method can increase the knowledge about, for instance, quantification and the time course of ADEs in practice [47-49]. Patient-reported outcome instruments to assess ADEs have also been included in observational studies (e.g. [50,51]). However, a literature review showed that patient-reported outcome instruments in the assessment of ADEs are still underutilized [52]. Patient-reported outcome instruments to evaluate the safety of a drugPatient-reported outcome instruments to assess ADEs can be open-ended and checklist-based. Event monitoring programmes such as Lareb Intensive Monitoring [28] use an open-ended question to report experienced ADEs. Open-ended questions are less sensitive in identifying potential ADEs than checklists [53,54]. Checklists may lack specificity in the detection of true ADEs [53], but adding questions per ADE on its nature and causality may solve this problem.

Available checklists for detecting patient-reported ADEs mainly focus on specific ADEs (e.g. gastrointestinal ADEs [50]) or ADEs of a specific drug class (e.g. chemotherapy [55] and contraceptives [56]). The use of such specific instruments limits the ability to compare ADE profiles of different drugs [29,53]. In addition, the focus of such specific instruments is on expected ADEs. Generic instruments, on the other hand, allow for the detection of unexpected ADEs [29,57]. Disadvantages of generic instruments may be the inclusion of irrelevant items for some patients. Moreover, generic instruments may lack disease or drug specific items which may negatively influence the instrument’s sensitivity to change [29,57,58].



13

The reliability and validity of any instrument should be demonstrated before the instrument can be used [4]. This validation is especially important for patient-reported outcome instruments since scepticism about the reliability and validity of patient-reported ADEs is one of the reasons for being reluctant to use such data [59]. An overview of different validity aspects that influence the quality of a patient-reported outcome instrument is presented in Figure 1.

15

Figure 1. Overview and definitions of different aspects of quality assessment of patient-reported outcome (PRO) instruments (adapted from: [3,60-63]). Patient perspective in clinical practice

In clinical practice, patient-reported outcome instruments can be used to assess patient perspectives of care outcomes and need for treatment [64]. The importance to assess patient perspectives in clinical practice appears from several trends that stimulate patient-centred care. Examples of such trends are the individualisation of the society in which the autonomy of an individual is respected [65] and the increased knowledge about an individual’s genetic and molecular profile which improves personalized medicine [66]. Treatment guidelines are moving towards more tailored recommendations based on specific patient characteristics and preferences [67]. This shift towards patient-centred

Quality of a PRO-instrument

Reliability Stability of scores over time

when no change occurs in the concept of interest

Validity Degree to which the instrument

measures what it intends to measure

Responsiveness Ability to detect change over

time

Internal consistency Extent to which items in a

(sub)scale are intercorrelated

Reliability (e.g. test-retest)

Stability of scores over time when no change has

occurred

Content validity Extent to which items and

response options are relevant and

comprehensive for concept

Construct validity Degree to which scores are consistent with predefined

hypothesis

Criterion validity (e.g. concurrent validity)

Extent to which scores relate to a gold standard

Figure 1. Overview and definitions of different aspects of quality assessment of patient-reported outcome (PRO) instruments (adapted from: [3,60-63]).

Patient perspective in clinical practice In clinical practice, patient-reported outcome instruments can be used to assess patient perspectives of care outcomes and need for treatment [64]. The importance to assess patient perspectives in clinical practice appears from several trends that stimulate patient-centred care. Examples of such trends are the individualisation of the society in which the autonomy of an individual is respected [65] and the increased knowledge about an individual’s genetic and molecular profile which improves personalized medicine [66].

R1R2R3R4R5R6R7R8R9



14

Treatment guidelines are moving towards more tailored recommendations based on specific patient characteristics and preferences [67]. This shift towards patient-centred care has been shown in guidelines of, for instance, the prevention and treatment of diabetes [6,68,69]. The assessment of an individual’s preferences, needs and values is required in patient-centred care to individualise treatment decisions and goals [70,71].

When setting individualised goals, various aspects should be taken into account [6,68]. For diabetes treatment, both patient attitudes and clinical aspects are important to consider when setting treatment goals (Figure 2). In recent years, attention has been given to age-specific goals, since evidence of long-term benefit of tight glycaemic and blood pressure control in aged patients is lacking [72,73]. Therefore, guidelines advise to take a patient’s life-expectancy and preferences into account in setting more or less stringent treatment goals [69,72]. Currently, little is known about the influence of age on actual prescribing behaviour in clinical practice. Cross-sectional data from the Netherlands suggest that trends in drug treatment from 1998 to 2008 were similar for different age groups [74]. Survey studies on associations between age and drug treatment indicate that age may be related to a patient’s preferences for specific treatment options and the willingness to undergo a treatment [75-77].

Setting individualised goals may also be influenced by a patient’s motivation and adherence (Figure 2). A lack of motivation or adherence may induce less stringent treatment goals but it also necessitates targeted interventions to improve motivation and adherence. Non-adherence is common in clinical practice [78] and no intervention can be expected to be effective across all patients, conditions and settings [79]. Uncertainty about the best approach to improve medication adherence particularly exists in specific populations, such as patients with comorbidity [79]. Patients can be intentional and unintentional non-adherent to their drug treatment [80-82]. Intentional non-adherence is seen as a deliberate decision for not taking the drug as prescribed whereas unintentional non-adherence is a more passive behaviour [82,83]. More insight in the underlying processes of these types of non-adherence, especially in patients who need to take multiple drugs for different indications, may contribute to better tailored interventions for improving drug adherence [84].



15

17

More stringent goals Less stringent goals Patient attitudes and expected treatment efforts

e.g. Highly motivated, adherent

e.g. Less motivated, non-adherent

Risk of hypoglycaemia and other adverse events

Low

High

Disease duration

Newly diagnosed

Long-standing

Age/life-expectancy

Young/long

Aged/short

Comorbidities

Absent

Few/mild

Severe

Established vascular complications Absent

Few/mild

Severe

Figure 2. Influence of various aspects on setting goals for glycohemoglobin (HbA1c) (adapted from [85,86]).

Research aims and outline of the thesis

Patient-reported information about ADEs is relevant for regulatory authorities in the benefit-risk evaluation of a drug and for healthcare professionals and patients to make better informed decisions about preferred treatments [34,87]. The first part of this thesis focuses on the development and validation of a patient-reported ADE questionnaire intended for such drug evaluations. In the second part, the role of patient characteristics and preferences on treatment decisions in clinical practice is explored. In both parts of the thesis, the focus is on patients with type 2 diabetes. These patients are often prescribed multiple drugs which increases the risk of ADEs, may complicate treatment decisions and may decrease adherence and willingness to take drugs. In addition, they get these drugs until well advanced in age. Understanding the effect of age, treatment complexity, and beliefs on treatment preferences and decisions may contribute to a better patient-centred care.

Figure 2. Influence of various aspects on setting goals for glycohemoglobin (HbA1c) (adapted from [85,86]).

Research aims and outline of the thesisPatient-reported information about ADEs is relevant for regulatory authorities in the benefit-risk evaluation of a drug and for healthcare professionals and patients to make better informed decisions about preferred treatments [34,87]. The first part of this thesis focuses on the development and validation of a patient-reported ADE questionnaire intended for such drug evaluations. In the second part, the role of patient characteristics and preferences on treatment decisions in clinical practice is explored. In both parts of the thesis, the focus is on patients with type 2 diabetes. These patients are often prescribed multiple drugs which increases the risk of ADEs, may complicate treatment decisions and may decrease adherence and willingness to take drugs. In addition, they get these drugs until well advanced in age. Understanding the effect of age, treatment complexity, and beliefs on treatment preferences and decisions may contribute to a better patient-centred care.

Part I. Development and validation of a patient-reported ADE questionnaireThe aims of the first part are to:• develop a patient-reported ADE questionnaire;• assess the reliability and validity of this questionnaire.

The development and reliability testing of the questionnaire is described in chapter 1. The patient-reported ADE questionnaire is developed for research purposes, that is, to be used in clinical trials and post-marketing studies. More specifically, the aim is to

R1R2R3R4R5R6R7R8R9



16

quantify ADE rates and generate additional information about the ADEs as reported by patients. The questionnaire was paper-based but adapted to a web-based version. In supplement I, the user acceptance of the web-based version of the questionnaire is presented. The assessment of the construct and concurrent validity is reported in chapter 2. Additional concurrent validity assessment is presented in chapter 3 in which ADEs reported in the questionnaire are compared with ADEs reported in a daily diary. The influence of different recall periods in the questionnaire is also addressed in chapter 3. In supplement II, some biases in validated questionnaires and of patient-reporting in general that were encountered in the studies are presented.

In the intermezzo, the assessment and management of ADEs in clinical practice is illustrated from a patient’s perspective.

Part II. The role of patient characteristics and preferences on treatment decisions in clinical practiceThis is followed by three studies which provide insight in various patient influences on treatment decisions in clinical practice, focusing on: • the decisions to start or intensify treatment with special attention for different

patient age groups; • the influence of age and medication beliefs on patients’ drug preferences;• the role of medication beliefs and treatment complexity on patients’ non-adherence

to drugs.

In chapter 4, potential undertreatment and overtreatment for glucose-, and blood pressure-lowering treatment in different patient age groups over time is presented. In particular, it was assessed whether after the introduction of diabetes performance measures decreases in undertreatment corresponded with increases in overtreatment in different patient age groups. In chapter 5, it was evaluated whether age affects 1) the patients’ willingness to add a blood pressure-lowering drug and 2) the importance they attach to specific treatment characteristics. In addition, the influence of medication beliefs on the association between age and willingness to add a blood pressure-lowering drug is explored in this chapter. In chapter 6, the association between medication beliefs and treatment complexity on intentional and unintentional non-adherence is assessed for glucose-, blood pressure-, and lipid-lowering drugs. These associations were studied within one group of patients with type 2 diabetes to explore differences across therapeutic groups.



17

Finally, the main findings of these studies are summarized and the results are discussed in light of their implications for research and practice.

R1R2R3R4R5R6R7R8R9



18

References[1] Doward LC, Gnanasakthy A, Baker MG. Pa-

tient reported outcomes: looking beyond the label claim. Health Qual Life Outcomes 2010;8:89.

[2] Szende A, Leidy NK, Revicki D. Health-relat-ed quality of life and other patient-reported outcomes in the European centralized drug regulatory process: a review of guidance documents and performed authorizations of medicinal products 1995 to 2003. Value Health 2005;8(5):534-48.

[3] US Department of Health and Human Ser-vices, FDA, Center for Drug Evaluation and Research Center for Biologics Evalua-tion and Research Center for Devices and Radiological Health. Guidance for indus-try patient-reported outcome measures: use in medical product development to support labeling claims [online]. http://www.fda.gov/downloads/Drugs/Guidance ComplianceRegulatory Information/Guid-ances/UCM193282.pdf. Accessed Nov 10, 2014.

[4] Frost MH, Reeve BB, Liepa AM, Stauffer JW, Hays RD, Mayo/FDA Patient-Reported Out-comes Consensus Meeting Group. What is sufficient evidence for the reliability and validity of patient-reported outcome meas-ures? Value Health 2007;10(Suppl 2):S94-105.

[5] Patrick DL, Burke LB, Powers JH, Scott JA, Rock EP, Dawisha S, et al. Patient-reported outcomes to support medical product labe-ling claims: FDA perspective. Value Health 2007;10(Suppl 2):S125-37.

[6] American Diabetes Association. Standards of medical care in diabetes--2014. Diabetes Care 2014;37(Suppl 1):S14-80.

[7] Mann RD. Prescription-event monitor-ing--recent progress and future horizons. Br J Clin Pharmacol 1998;46(3):195-201.

[8] Heiat A, Gross CP, Krumholz HM. Rep-resentation of the elderly, women, and mi-norities in heart failure clinical trials. Arch Intern Med 2002;162(15):1682-8.

[9] Zarin DA, Young JL, West JC. Challenges to evidence-based medicine: a compar-ison of patients and treatments in ran-domized controlled trials with patients and treatments in a practice research net-work. Soc Psychiatry Psychiatr Epidemiol 2005;40(1):27-35.

[10] Stricker BH, Psaty BM. Detection, verifica-tion, and quantification of adverse drug re-actions. BMJ 2004;329(7456):44-7.

[11] Mosenifar Z. Population issues in clinical trials. Proc Am Thorac Soc 2007;4(2):185-7; discussion 187-8.

[12] Raine J, Wise L, Blackburn S, Eichler HG, Breckenridge A. European perspective on risk management and drug safety. Clin Pharmacol Ther 2011;89(5):650-4.

[13] Zulman DM, Sussman JB, Chen X, Cigolle CT, Blaum CS, Hayward RA. Examining the evidence: a systematic review of the inclu-sion and analysis of older adults in rand-omized controlled trials. J Gen Intern Med 2011;26(7):783-90.

[14] Boyd CM, Vollenweider D, Puhan MA. In-forming evidence-based decision-making for patients with comorbidity: availa-bility of necessary information in clini-cal trials for chronic diseases. PLoS One 2012;7(8):e41601.

[15] Downing NS, Aminawung JA, Shah ND, Krumholz HM, Ross JS. Clinical trial evi-dence supporting FDA approval of nov-el therapeutic agents, 2005-2012. JAMA 2014;311(4):368-77.

[16] Curtin F, Schulz P. Assessing the benefit: risk ratio of a drug--randomized and nat-uralistic evidence. Dialogues Clin Neurosci 2011;13(2):183-90.

[17] Pintor-Marmol A, Baena MI, Fajardo PC, Sabater-Hernandez D, Saez-Benito L, Gar-cia-Cardenas MV, et al. Terms used in pa-tient safety related to medication: a litera-ture review. Pharmacoepidemiol Drug Saf 2012;21(8):799-809.

[18] Rehan HS, Chopra D, Kakkar AK. Physician’s guide to pharmacovigilance: terminology and causality assessment. Eur J Intern Med 2009;20(1):3-8.

[19] World Health Organization Drug and ther-apeutics committee. Session 4. Assessing and managing medicine safety [online]. http://www.who.int/medicines/technical_briefing/tbs/04-Drug-Safety_final-08.ppt. Accessed Nov 10, 2014.

[20] Food and Drug Administration. Guideline for Industry. Clinical Safety Data Manage-ment: Definitions and Standards for Expe-dited Reporting (ICH-E2A) [online]. http://www.fda.gov/downloads/Drugs/Guid-ances/ucm073087.pdf. Accessed Nov 10, 2014.



19

[21] European Medicines Agency. Guideline on good pharmacovigilance practices (GVP) [online]. http://www.ema.europa.eu/docs /en_GB/document_library/Scientific_guideline/2013/05/ WC500143294.pdf. Accessed Nov 10, 2014.

[22] Segal ES, Valette C, Oster L, Bouley L, Edf-jall C, Herrmann P, et al. Risk management strategies in the postmarketing period: safety experience with the US and Euro-pean bosentan surveillance programmes. Drug Saf 2005;28(11):971-80.

[23] Fletcher AP. Spontaneous adverse drug reaction reporting vs event monitoring: a comparison. J R Soc Med 1991;84(6):341-4.

[24] Hazell L, Shakir SA. Under-reporting of ad-verse drug reactions: a systematic review. Drug Saf 2006;29(5):385-96.

[25] Ahmad SR. Adverse drug event monitoring at the Food and Drug Administration. J Gen Intern Med 2003;18(1):57-60.

[26] Coulter DM. The New Zealand Intensive Medicines Monitoring Programme. Phar-macoepidemiol Drug Saf 1998;7(2):79-90.

[27] Rawson NS, Pearce GL, Inman WH. Pre-scription-event monitoring: methodo-lo-gy and recent progress. J Clin Epidemiol 1990;43(5):509-22.

[28] Härmark L, van Grootheest AC. Pharma-covigilance: methods, recent developments and future perspectives. Eur J Clin Pharma-col 2008;64(8):743-52.

[29] Fitzpatrick R, Davey C, Buxton MJ, Jones DR. Evaluating patient-based outcome meas-ures for use in clinical trials. Health Technol Assess 1998;2(14):1-74.

[30] Deshpande PR, Rajan S, Sudeepthi BL, Ab-dul Nazir CP. Patient-reported outcomes: A new era in clinical research. Perspect Clin Res 2011;2(4):137-44.

[31] Blenkinsopp A, Wilkie P, Wang M, Rout-ledge PA. Patient reporting of suspected ad-verse drug reactions: a review of published literature and international experience. Br J Clin Pharmacol 2007;63(2):148-56.

[32] Black N, Jenkinson C. How can patients’ views of their care enhance quality im-provement? BMJ 2009;339(7714):202-5.

[33] Weingart SN, Gandhi TK, Seger AC, Seger DL, Borus J, Burdick E, et al. Patient-report-ed medication symptoms in primary care. Arch Intern Med 2005;165(2):234-40.

[34] Anker SD, Agewall S, Borggrefe M, Calvert M, Jaime Caro J, Cowie MR, et al. The im-

portance of patient-reported outcomes: a call for their comprehensive integration in cardiovascular clinical trials. Eur Heart J 2014;35(30):2001-9.

[35] Basch E, Reeve BB, Mitchell SA, Clauser SB, Minasian LM, Dueck AC, et al. Devel-opment of the National Cancer Institute’s patient-reported outcomes version of the common terminology criteria for adverse events (PRO-CTCAE). J Natl Cancer Inst 2014;106(9):10.

[36] van Grootheest K, de Graaf L, de Jong-van den Berg LT. Consumer adverse drug reac-tion reporting: a new step in pharmacovig-ilance? Drug Saf 2003;26(4):211-7.

[37] Britten N. Medication errors: the role of the patient. Br J Clin Pharmacol 2009;67(6):646-50.

[38] Hakobyan L, Haaijer-Ruskamp FM, de Zeeuw D, Dobre D, Denig P. Comparing adverse event rates of oral blood glu-cose-lowering drugs reported by patients and healthcare providers: a post-hoc analysis of observational studies pub-lished between 1999 and 2011. Drug Saf 2011;34(12):1191-202.

[39] Burke L. Exploring clinical outcome assess-ments in rare disease trials [online]. http://www.everylifefoundation.org/wp-content/uploads/images/workshopseries /11-Burke-Exploring-Clinical-Outcome- Assessments-in-Rare-Disease-Trials.pdf. Accessed Nov 10, 2014.

[40] European Medicines Agency. Commit-tee for medicinal products for human use (CHMP). Reflection paper on the regulatory guidance for the use of health-related qual-ity of life (HRQOL) measures in the evalua-tion of medicinal products. EMEA/CHMP/EWP/139391/2004 [online]. http://www.ispor.org/workpaper/EMEA-HRQL-Guid-ance.pdf. Accessed Nov 10, 2014.

[41] US Department of Health and Human Ser-vices FDA Center for Drug Evaluation and Research, Center for Biologics Evaluation and Research, Center for Devices and Ra-diological Health. Guidance for industry: patient-reported outcome measures: use in medical product development to support labeling claims: draft guidance. Health Qual Life Outcomes 2006;4:79.

[42] Food and Drug Administration (FDA). Structured approach to benefit-risk as-sessment in drug regulatory decision-mak-ing. Draft PDUFA V Implementation Plan

R1R2R3R4R5R6R7R8R9



20

- February 2013. Fiscal Years 2013-2017 [online]. http://www.fda.gov/downloads/ ForIndustry/UserFees/PrescriptionDrug UserFee/UCM329758.pdf. Accessed Nov 10, 2014.

[43] Bernabe RD, van Thiel GJ, van Delden J. Patient representatives’ contributions to the benefit-risk assessment tasks of EMA scientific committees. Br J Clin Pharmacol 2014 [Epub ahead of print].

[44] Bottomley A, Jones D, Claassens L. Pa-tient-reported outcomes: assessment and current perspectives of the guidelines of the Food and Drug Administration and the reflection paper of the European Medicines Agency. Eur J Cancer 2009;45(3):347-53.

[45] Gnanasakthy A, Mordin M, Clark M, DeMuro C, Fehnel S, Copley-Merriman C. A review of patient-reported outcome labels in the United States: 2006 to 2010. Value Health 2012;15(3):437-42.

[46] de Langen J, van Hunsel F, Passier A, de Jong-van den Berg L, van Grootheest K. Ad-verse drug reaction reporting by patients in the Netherlands: three years of experience. Drug Saf 2008;31(6):515-24.

[47] Härmark L, van Puijenbroek E, van Grooth-eest K. Intensive monitoring of dulox-etine: results of a web-based intensive monitoring study. Eur J Clin Pharmacol 2013;69(2):209-15.

[48] Härmark L, van Puijenbroek E, Straus S, van Grootheest K. Intensive monitoring of pregabalin: results from an observational, Web-based, prospective cohort study in the Netherlands using patients as a source of information. Drug Saf 2011;34(3):221-31.

[49] Härmark L, Puijenbroek E, Grootheest K. Longitudinal monitoring of the safety of drugs by using a web-based system: the case of pregabalin. Pharmacoepidemiol Drug Saf 2011;20(6):591-7.

[50] Bytzer P, Talley NJ, Jones MP, Horowitz M. Oral hypoglycaemic drugs and gastrointes-tinal symptoms in diabetes mellitus. Ali-ment Pharmacol Ther 2001;15(1):137-42.

[51] Vexiau P, Mavros P, Krishnarajah G, Lyu R, Yin D. Hypoglycaemia in patients with type 2 diabetes treated with a combination of metformin and sulphonylurea therapy in France. Diabetes Obes Metab 2008;10(Sup-pl 1):16-24.

[52] Hakobyan L, Haaijer-Ruskamp FM, de Zeeuw D, Dobre D, Denig P. A review of

methods used in assessing non-serious adverse drug events in observational studies among type 2 diabetes mellitus patients. Health Qual Life Outcomes 2011;9:83.

[53] Bent S, Padula A, Avins AL. Brief commu-nication: Better ways to question patients about adverse medical events: a rand-omized, controlled trial. Ann Intern Med 2006;144(4):257-61.

[54] Sheftell FD, Feleppa M, Tepper SJ, Rapoport AM, Ciannella L, Bigal ME. Assessment of adverse events associated with triptans--methods of assessment influence the re-sults. Headache 2004;44(10):978-82.

[55] Sitzia J, Dikken C, Hughes J. Psychometric evaluation of a questionnaire to document side-effects of chemotherapy. J Adv Nurs 1997;25(5):999-1007.

[56] Clerson P, Graesslin O, Gater A, Taylor F, Filonenko A, Schellschmidt I, et al. EVAP-IL-R Scale: Continuous Development and Validation of a Tool to Assess Patient-Re-ported Tolerability of Different Contracep-tive Methods in Longitudinal Studies. Clin Ther 2014;36(5):638-47.

[57] Fletcher A, Gore S, Jones D, Fitzpatrick R, Spiegelhalter D, Cox D. Quality of life meas-ures in health care. II: Design, analysis, and interpretation. BMJ 1992;305(6862):1145-8.

[58] McKenna SP. Measuring patient-report-ed outcomes: moving beyond misplaced common sense to hard science. BMC Med 2011;9:86.

[59] Banerjee AK, Okun S, Edwards IR, Wicks P, Smith MY, Mayall SJ, et al. Patient-Reported Outcome Measures in Safety Event Report-ing: PROSPER Consortium Guidance. Drug Saf 2013;36(12):1129-49.

[60] Mokkink LB, Terwee CB, Patrick DL, Alonso J, Stratford PW, Knol DL, et al. The COSMIN study reached international consensus on taxonomy, terminology, and definitions of measurement properties for health-related patient-reported outcomes. J Clin Epidemi-ol 2010;63(7):737-45.

[61] Mokkink LB, Terwee CB, Patrick DL, Alonso J, Stratford PW, Knol DL, et al. The COSMIN checklist for assessing the methodological quality of studies on measurement proper-ties of health status measurement instru-ments: an international Delphi study. Qual Life Res 2010;19(4):539-49.



21

[62] De Vet H, Terwee C, Mokkink L, Knol D. Measurement in Medicine: A practical guide. Cambridge: University Press; 2011.

[63] Terwee CB, Bot SD, de Boer MR, van der Windt DA, Knol DL, Dekker J, et al. Quality criteria were proposed for measurement properties of health status questionnaires. J Clin Epidemiol 2007;60(1):34-42.

[64] Dawson J, Doll H, Fitzpatrick R, Jenkinson C, Carr AJ. The routine use of patient reported outcome measures in healthcare settings. BMJ 2010;340:c186.

[65] Lindbladh E, Lyttkens CH, Hanson BS, Ostergren PO. Equity is out of fashion? An essay on autonomy and health policy in the individualized society. Soc Sci Med 1998;46(8):1017-25.

[66] Abrahams E, Ginsburg GS, Silver M. The Personalized Medicine Coalition: goals and strategies. Am J Pharmacogenomics 2005;5(6):345-55.

[67] Bitton A, Onega T, Tosteson AN, Haas JS. To-ward a better understanding of patient-re-ported outcomes in clinical practice. Am J Manag Care 2014;20(4):281-3.

[68] Rutten GEHM, De Grauw WJC, Nijpels G, Houweling ST, Van de Laar FA, Bilo HJ, et al. The NHG guideline Diabetes mellitus type 2. Huisarts Wet 2013;56(10):512-25.

[69] Verenso. Multidisciplinaire richtlijn diabe-tes. Verantwoorde diabeteszorg bij kwets-bare ouderen thuis en in verzorgings- of verpleeghuizen. Deel 1. [Multidisciplinary guideline diabetes. Responsible diabetes care in vulnerable elderly at home and in residential care or nursing homes. Part 1]. Utrecht, the Netherlands, Verenso, 2011.

[70] Glasgow RE, Peeples M, Skovlund SE. Where is the patient in diabetes perfor-mance measures? The case for including patient-centered and self-management measures. Diabetes Care 2008;31(5):1046-50.

[71] Kaldjian LC. Teaching practical wisdom in medicine through clinical judgement, goals of care, and ethical reasoning. J Med Ethics 2010;36(9):558-62.

[72] American Diabetes Association. Standards of medical care for patients with diabetes mellitus. Diabetes Care 2003;26(Suppl 1):S33-50.

[73] van Hateren KJ, Landman GW, Kleefstra N, Houweling ST, van der Meer K, Bilo HJ. Time for considering other blood pressure target values in elderly patients with type 2 diabe-

tes? Int J Clin Pract 2012;66(2):125-7. [74] van Hateren KJ, Drion I, Kleefstra N,

Groenier KH, Houweling ST, van der Meer K, et al. A prospective observational study of quality of diabetes care in a shared care setting: trends and age differences (ZODI-AC-19). BMJ Open 2012;2(4):e001387.

[75] Dibonaventura MD, Wagner JS, Girman CJ, Brodovicz K, Zhang Q, Qiu Y, et al. Multi-national Internet-based survey of patient preference for newer oral or injectable Type 2 diabetes medication. Patient Prefer Adherence 2010;4:397-406.

[76] Chin MH, Drum ML, Jin L, Shook ME, Huang ES, Meltzer DO. Variation in treatment pref-erences and care goals among older pa-tients with diabetes and their physicians. Med Care 2008;46(3):275-86.

[77] Fried TR, Van Ness PH, Byers AL, Towle VR, O’Leary JR, Dubin JA. Changes in prefer-ences for life-sustaining treatment among older persons with advanced illness. J Gen Intern Med 2007;22(4):495-501.

[78] Cramer JA. A systematic review of adher-ence with medications for diabetes. Diabe-tes Care 2004;27(5):1218-24.

[79] Ryan R, Santesso N, Lowe D, Hill S, Grim-shaw J, Prictor M, et al. Interventions to improve safe and effective medicines use by consumers: an overview of systemat-ic reviews. Cochrane Database Syst Rev 2014;4:CD007768.

[80] Clifford S, Barber N, Horne R. Under-standing different beliefs held by adh-erers, unintentional nonadherers, and intentional nonadherers: application of the Necessity-Concerns Framework. J Psychosom Res 2008;64(1):41-6.

[81] Schüz B, Marx C, Wurm S, Warner LM, Zie-gelmann JP, Schwarzer R, et al. Medication beliefs predict medication adherence in older adults with multiple illnesses. J Psy-chosom Res 2011;70(2):179-87.

[82] Wroe AL. Intentional and unintentional nonadherence: a study of decision making. J Behav Med 2002;25(4):355-72.

[83] Lehane E, McCarthy G. Intentional and un-intentional medication non-adherence: a comprehensive framework for clinical re-search and practice? A discussion paper. Int J Nurs Stud 2007;44(8):1468-77.

[84] Horne R, Chapman SC, Parham R, Freeman-tle N, Forbes A, Cooper V. Understanding patients’ adherence-related beliefs about medicines prescribed for long-term con-

R1R2R3R4R5R6R7R8R9



22

ditions: a meta-analytic review of the Ne-cessity-Concerns Framework. PLoS One 2013;8(12):e80633.

[85] Inzucchi SE, Bergenstal RM, Buse JB, Dia-mant M, Ferrannini E, Nauck M, et al. Man-agement of hyperglycemia in type 2 diabe-tes: a patient-centered approach: position statement of the American Diabetes Associ-ation (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care 2012;35(6):1364-79.

[86] Ismail-Beigi F, Moghissi E, Tiktin M, Hirsch IB, Inzucchi SE, Genuth S. Individualizing glycemic targets in type 2 diabetes melli-tus: implications of recent clinical trials. Ann Intern Med 2011;154(8):554-9.

[87] Soreide K, Soreide AH. Using patient-re-ported outcome measures for improved decision-making in patients with gastroin-testinal cancer - the last clinical frontier in surgical oncology? Front Oncol 2013;3:157.

Part I Development and validation of a patient-reported

adverse drug event questionnaire

Chapter 1

Development and initial validation of a

patient-reported adverse drug event questionnaire

Sieta T. de Vries1 Peter G.M. Mol1 Dick de Zeeuw1

Flora M. Haaijer-Ruskamp1 Petra Denig1

Drug Safety 2013;36(9):765-77.

1 Department of Clinical Pharmacy and Pharmacology, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands

R1R2R3R4R5R6R7R8R9


Chapter 1

26

Abstract

Background Direct patient reporting of adverse drug events (ADEs) is relevant for the evaluation of drug safety. To collect such data in clinical trials and post-marketing studies, a valid questionnaire is needed that can measure all possible ADEs experienced by patients.

Objective Our aim was to develop and test a generic questionnaire to identify ADEs and quantify their nature and causality as reported by patients.

Methods We created a draft list of common ADEs in lay-terms, which were classified in body categories and mapped to the Medical Dictionary for Regulatory Activities (MedDRA®) terminology. Questions about the nature and causality were derived from existing questionnaires and causality scales. Content validity was tested through cognitive debriefing, revising the questionnaire in an iterative process. Feasibility and reliability were assessed using a web-based version of the questionnaire. Patients received the questionnaire twice. Feasibility was assessed by the reported time needed for completion and ease of use. Reliability was calculated using Cohen’s kappa and proportion of positive agreement (PPA) on: 1) any ADE at patient level; 2) similar ADEs at MedDRA® System Organ Class level; and 3) the same ADE at ADE-specific level. Results In the development phase, 28 patients with type 2 diabetes or asthma/chronic obstructive pulmonary disease (COPD) participated. Questions and answer options were rephrased, lay-out was improved, and changes were made in the classification of ADEs. The final questionnaire consisted of 252 ADEs organized in 16 body categories, and included 14 questions per reported ADE. A total of 135 patients using a median of five different drugs completed the web-based questionnaire twice. The median completion time was 15 minutes for patients not reporting any ADE, and 30 minutes for patients reporting at least one ADE. Three quarters of the patients found the questionnaire easy to use. Test–retest reliability was acceptable at patient level (κ = 0.50, PPA = 0.64) and at MedDRA® System Organ Class level (κ = 0.52, PPA = 0.54), but was low at ADE-specific level (κ = 0.38, PPA = 0.38). Conclusions We developed a generic patient-reported ADE questionnaire and confirmed its content validity. The questionnaire was feasible and reliable for reporting any ADE and similar ADEs at MedDRA® System Organ Class level. Additional work is, however, needed to reliably quantify specific ADEs reported by patients.


Patient-reported adverse drug event questionnaire

27

1

IntroductionToday, patients are increasingly involved in information gathering and decision making at all levels of the healthcare system [1]. Patient self-reports of adverse drug events (ADEs) are an important additional source of information on the safety of drugs because they differ from healthcare professional reports [2–7]. Healthcare professionals often underestimate symptomatic ADEs experienced by patients [7,8]. The added value of patient reports is acknowledged by the Food and Drug Administration (FDA) as well as the European Medicines Agency [9,10]. The FDA advises the use of patient-reported outcome (PRO) questionnaires for the measurement of outcomes that are best known by patients [9] (e.g., pain [11]). In PRO questionnaires, the patient is the direct source of information without interpretation of the responses by a healthcare professional [9,12].

Patient-reported ADE questionnaires can be open-ended or checklist-based. Compared to open-ended questionnaires, checklist-based questionnaires are more sensitive in identifying potential ADEs [13,14]. However, these methods may lack specificity in the detection of true ADEs [13]. Adding questions per ADE on its nature and causality might solve this problem. To assess unknown ADEs of (new) drugs and comparing ADE profiles of different drugs, a generic PRO questionnaire is needed that can measure all possible ADEs [13,15]. Most available patient-reported ADE questionnaires focus on specific ADEs, such as gastrointestinal ADEs [16] or ADEs specific for a drug class, such as inhaled corticosteroids [17] or chemotherapy [18]. Previously, a generic questionnaire was developed that contained approximately 600 symptoms classified by body category [19]. More recently, a questionnaire with 84 ADEs classified in 19 body categories was developed [3]. Although both questionnaires have been piloted, no explicit validation has been reported. Furthermore, both questionnaires lack questions supporting causality assessment and questions about the nature of the ADE such as those regarding seriousness, severity, frequency, and time course, which are relevant attributes in the evaluation of the ADE [20,21].

The aim of our study was to develop and test a generic questionnaire for identifying ADEs and assessing their nature (e.g., frequency, severity) and causality as reported by patients. We tested the content validity and feasibility of the questionnaire as well as the reliability for reporting ADEs.

MethodsThe study consisted of three parts: 1) development of a draft ADE questionnaire, 2) content validation and revision of the questionnaire in an iterative process, and 3) feasibility and reliability testing of the revised questionnaire.

R1R2R3R4R5R6R7R8R9


Chapter 1

28

Questionnaire development The questionnaire consists of four sections with questions about: 1) general patient characteristics; 2) drug use in the past 4 weeks, diseases for which these drugs were used, whether the patient had other diseases; 3) ADEs experienced in the past 4 weeks using structured checklists; and 4) for each ADE a question to describe the ADE in the patient’s own words with additional questions about its nature and causality. We expected that a period of 4 weeks would be sufficient for capturing a wide range of ADEs for which patients would be able to recall the relevant details. In the development phase, ADEs were selected, named, coded, and categorized into a body category, and questions were constructed to assess the nature and causality of the ADEs. ADE selection and naming in lay-termsWe aimed to include a wide range of common symptomatic ADEs. We identified possible ADEs from the Common Terminology Criteria for Adverse Events version 4.0 [22], and existing symptom and ADE checklists [3,13,18,23–29]. Patient-reported data about ADEs from the Lareb Intensive Monitoring System of the Netherlands pharmacovigilance centre Lareb [30] were used to translate ADEs into lay-terms. We excluded ADEs based on laboratory results (e.g., hyperkalaemia) and those related to specific devices (e.g., uncomfortable pressure of the mask). The first selection included 252 possible ADEs with an open-ended option for reporting “other” experienced ADEs. Coding of ADEs Two researchers (STdV and PD) independently coded each lay-term ADE to a Lowest Level Term of the Medical Dictionary for Regulatory Activities (MedDRA®) terminology version 13.0, making use of codings suggested by pharmacovigilance experts from Lareb. MedDRA® is the international medical terminology developed under the auspices of the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH). Agreement between the codings existed in 74% of the ADEs. Dissimilarities were resolved by discussion, and translation of the Dutch lay-terms into English by a professional translator was used to reach agreement on all MedDRA® terms. Two ADEs, “Bone fracture or fractures” and “Stroke”, were classified at a higher hierarchical ADE group definition because of their nonspecific nature. One ADE (dry teeth) showed overlap in the MedDRA® terminology with another included ADE (dry mouth), and they were therefore combined. Categorization of ADEs To increase the efficiency of completing the questionnaire, the ADEs were classified in body categories. By first checking body categories in which patients experienced ADEs,



29

1

they were directed to short checklists of specific ADEs within that body category. These lists with specific ADEs also include the option to report other ADEs. The body categories in the initial questionnaire were based on the classification used in the MedDRA® and in existing questionnaires [3,19].

Assessing nature and causality of ADEs Relevant known attributes of ADEs were duration, frequency, severity, and seriousness of the ADE; its impact on activities; and the patient’s benefit–risk assessment of the drug [24,30–32]. Existing questionnaires were screened for questions covering these topics [26,27,33–35]. Questions regarding causality were included, based on medical [36], and patient-reported considerations [37].

Content validation The draft questionnaire was subjected to cognitive debriefing interviewing to eliminate ambiguity in questions and answer options. Cognitive debriefing is a qualitative interview method in which the patient’s understanding and interpretation of items and answer options of the questionnaire are assessed [38,39]. A separate classification task was used to assess the appropriateness of the body categories.

Study population Patients included in the study were 18 years or older; diagnosed with type 2 diabetes, asthma, and/or chronic obstructive pulmonary disease (COPD); using drugs for these conditions; and able to speak, read, and write the Dutch language. Patients with these diagnoses were included to cover a population with a broad age range in which many different types of drugs are commonly used, both daily and as needed. Eligible patients were recruited by three general practitioners and two dieticians in the northern part of the Netherlands in 2011–2012.

Study procedure After signing informed consent, patients completed the questionnaire during which they were observed by a researcher (STdV) to detect any problems with completing the questionnaire. Immediately thereafter, a semi-structured interview was conducted using a topic list based on the “question-and-answer” model [38,39]. A subset of patients was asked to do a classification task, for which all ADEs were randomly split into five lists. Patients were instructed to classify each ADE of one list into a body category. Each ADE was classified by at least four patients.

R1R2R3R4R5R6R7R8R9


Chapter 1

30

AnalysesThe audio-recorded interviews were transcribed verbatim, and transcripts were screened by two researchers (STdV and PD) to identify problems in understanding the questions and answer options. The questionnaire was adapted in an iterative process by which changes were made addressing detected problems until no new problems were identified regarding understanding the questions and answer options (Figure 1.1) [38]. Regarding the classification task, we considered an ADE classification as problematic when more than two patients classified the ADE in body categories different from our original classification, or when two patients were consistent in choosing a different category. These problematic ADEs were subsequently judged by four additional patients and a pharmacovigilance expert. Based on their judgements, revisions were made. This revised questionnaire was then translated from Dutch to English by a professional translator. The English version was screened for differences with the original Dutch version through informal back translation by the researchers, and final changes were made. A web-based version of the content-valid questionnaire was then constructed using the Unipark Enterprise Feedback Suite 8.0 version 1.1 (http://www.unipark.de).

35

version through informal back translation by the researchers, and final changes were made. A web-based version of the content-valid questionnaire was then constructed using the Unipark Enterprise Feedback Suite 8.0 version 1.1 (http://www.unipark.de).

Figure 1.1. Iterative process in adapting the developed questionnaire to a content validated questionnaire

Feasibility and reliability testing The web-based version was used to assess the feasibility of completing the questionnaire, its ability to measure the ADEs in a consistent manner (test–retest reliability), and to assess the impact of using body categories on feasibility and ADE reporting. Study population Included patients were aged 18 years or older, had been dispensed an oral glucose-lowering drug, had an e-mail address, and were able to access the internet. These patients were recruited via pharmacists in the northern part of the Netherlands in 2012.

cognitive debriefing interviews after

completion transcription of

interviews

independent screening of transcripts to identify

relevant problems

discussion between researchers about detected problems

adapting questionnaire after

2-3 patients

patients fill in questionnaire

pilot questionnaire

content validated version of the questionnaire after no more relevant issues

Figure 1.1. Iterative process in adapting the developed questionnaire to a content validated questionnaire



31

1

Feasibility and reliability testing The web-based version was used to assess the feasibility of completing the questionnaire, its ability to measure the ADEs in a consistent manner (test–retest reliability), and to assess the impact of using body categories on feasibility and ADE reporting. Study population Included patients were aged 18 years or older, had been dispensed an oral glucose-lowering drug, had an e-mail address, and were able to access the internet. These patients were recruited via pharmacists in the northern part of the Netherlands in 2012. Study design and procedure In a test–retest design study, consenting patients received an e-mail message with the URL (uniform resource locator) to open the web-based version. A personal login code was used to prevent multiple completions of patients [40]. After completion of the ADE part, questions were asked regarding feasibility, including self-reported time to complete the questionnaire and ease of use on a five-point Likert scale. In addition, the total time between opening and closing of the digital questionnaire was logged (registered time), as well as the proportion of patients completing the questionnaires, and the number of ADEs reported in the “other” category. One week after completion, patients received an e-mail for the second questionnaire for the reliability analysis. Patients were randomly assigned to three groups using simple randomization [41] to receive: A) the same questionnaire twice (the “test–retest group”); B) a questionnaire with the body category structure at the first measurement (T1) and without these categories at the second measurement (T2) (the “group with body categories at T1”); or C) reversing the order used in B (the “group with body categories at T2”). One reminder was sent to the patients who did not complete the first questionnaire within a month. Patients who did not complete the second questionnaire were sent a reminder twice. We aimed to include about 50 patients per group, which has been reported as a reasonable number for reliability studies [42].

AnalysesDifferences in gender and age between responders and non-responders were assessed using Pearson χ2-test and Mann–Whitney U test respectively. Descriptive statistics were used for the feasibility parameters, including self-reported completion time, ease of use, proportion of patients completing the questionnaires, and number of ADEs reported in the “other” category. ADEs that were reported as “other” were evaluated and, if possible, classified by the researchers within the provided ADE lists. To assess the number of chronic diseases, we classified each self-reported disease in 1 of 12 chronic diseases, excluding conditions of normal ageing (e.g., loss of hearing).

R1R2R3R4R5R6R7R8R9


Chapter 1

32

We measured the agreement between ADE reporting at T1 and T2 at three levels: any ADE at “patient level”, similar ADEs at primary System Organ Class “MedDRA®

level”, and the same ADE at the lowest description “ADE specific level”. Cohen’s kappa coefficient and proportion of positive agreement were calculated as measures of agreement. Especially at the lowest level, where specific ADEs will be checked by few patients, the kappa statistic is negatively affected by the skewed distribution and proportion of positive agreement has been proposed as an alternative [43]. The proportion of positive agreement was calculated by the formula 2a/[N + (a – d)], in which N is the total number of observations, a is the number of patients reporting ADE at T1 and T2, and d is the number of patients not reporting ADE at T1 and T2 [44]. Kappa and proportion of positive agreement values of >0.5 were considered to be acceptable [45]. We conducted additional analyses aggregating experienced ADEs using the patients’ own description of the ADEs. Based on these descriptions, two researchers (STdV and PD) clustered ADEs that were checked as separate ADEs but described by the patients as being one problem. Although one might expect that this clustering is similar to the aggregation at MedDRA® level, it is possible that patients use terms from different MedDRA® classes to describe one problem. For instance, goose bumps, shivering, and cold limbs can be seen as one problem by the patient but are coded in different primary MedDRA® System Organ Classes. Misclassification can also occur when patients check similar but not the same symptomatic ADEs at T1 and T2. Finally, we calculated how often patients checked a symptom only as a symptom at one time point but as a possible ADE at another time. The effect of including body categories was tested by comparing feasibility parameters and the number of reported ADEs between the questionnaire with body categorization and without at baseline, using Pearson χ2-tests and Mann–Whitney U tests. Additionally, the agreement values of the group with the body categories at T1 and the group with the body categories at T2 were compared using the normal curve deviate statistic (Z value) [46]. Sensitivity analyses were conducted to investigate whether the number of days between completing the first and second questionnaire influenced the agreement values. All analyses were conducted using IBM SPSS Statistics version 20 (Armonk, New York, USA). P-values of <0.05 were considered to be statistically significant.

Results

Questionnaire development The initial version of the questionnaire contained 252 ADEs categorized in 21 body categories, and 11 questions regarding the nature and causality assessment for every ADE identified.



33

1

Content validation Twenty-eight patients, 54% of them women, participated (Table 1.1). Ages ranged from 22 to 90 years, with a median of 61 years. Almost all patients used more than one drug.

Table 1.1. Patient characteristics of content validationTotal number of participants 28 (15 females)Median age in years (range) 61 (22-90)Education Lowa 14 Middleb 10 Highc 4Diagnosis Type 2 diabetes 16 Asthma/COPD 10 Type 2 diabetes and asthma/COPD 2Multiple versus single drug users Multiple drug users 26 Single drug users 2Median number of self-reported prescription drugs (range) 5 (1-14)Median number of self-reported chronic diseases, including asthma/COPD, diabetes and cardiovascular diseases (range)

3 (1-6)

a No education; elementary school; junior secondary vocational educationb Junior general secondary education; senior secondary vocational educationc Senior general secondary education; higher professional education; university education COPD: Chronic obstructive pulmonary disease

Content validation, cognitive debriefing interviewsBased on the cognitive debriefing interviews, the questionnaire was revised 14 times. This included a revision of the general structure of the questionnaire, and a major revision by asking for ADEs as well as symptoms. The final revision was tested in five patients and no major problems in the interpretation of questions and answer options were detected. Problems detected in the questionnaire are presented according to the domains of the question-and-answer model, with examples given in Table 1.2. Wording of the body categories and ADEs was generally clear for the patients (Table 1.2: “Comprehension”). Several ambiguous interpretations, reading difficulties, and vague statements were reported by patients regarding specific question and answer options, which were subsequently changed. Eight patients reported that the recall period of 4 weeks for the experienced ADEs was short (Table 1.2: “Retrieval”). Because this did not reflect the content validation, no changes regarding the recall period were made during the study period.

R1R2R3R4R5R6R7R8R9


Chapter 1

34

The initial questionnaire asked to indicate “experienced ADEs”. However, it became clear that patients, when confronted with a checklist of possible symptomatic ADEs, incorrectly started to check symptoms that they actually did not see as ADEs (Table 1.2: “Judgement”). Asking to check both experienced symptoms and ADEs solved this problem. The answer option “do not know” was added because some patients were not sure whether the experienced symptom was related to a drug they used. Almost half of the patients either skipped the body categories to go directly to the specific checklists (navigation) or had difficulties in deciding which body category their symptom might be classified into. Other patients who used the body categories found them helpful and easy to use. As a result, we kept the body category structure as a supportive step in the questionnaire, but patients no longer needed to check body categories before going to the specific checklists.

Answer options that did not fit with the judgements of the patients were detected and adapted, and answer options were added (Table 1.2: “Response”). The answer options of the question “how often did you experience this side effect during the past 4 weeks (on how many or which days)?” were changed multiple times. Problems remained especially for intermittently occurring ADEs, and this question was therefore adapted into an open-ended question (Table 1.3).

Two questions were added to the initial questionnaire because they yielded additional information regarding causality (Table 1.3). One question was added to cover an additional attribute, namely actions taken (Table 1.3).

One patient reported difficulties with the sequence of the questions per ADE (Table 1.2: “Respondent burden”). This was improved by clustering the topics of the questions. One patient had some problems with the size of the letters in the questionnaire (font size 11, Arial), but none of the other patients reported such reading difficulties. Problems regarding navigation in the questionnaire, especially due to lay-out issues were detected and resolved. After seven interviews, the questionnaire was split into two distinct parts, separating the specific questions about the ADEs from the first part of the questionnaire. Two patients mentioned that they felt many questions per ADE were included but that this was not a problem for them. Comments on the length and number of answer options of a causality question led to shortening these phrases (Table 1.3).



35

1

Table 1.2. Examples of issues mentioned by patients per topic during the cognitive debriefingDomain of the question-and-answer model

Topic in topics list

Examples of issues mentioned by patients

Comprehension Text “Difficult question. I do not understand it entirely. A bit difficult question. I am reading it 7 times and still do not know what they mean.” (female, 58 years)

Adverse drug events

“Most of this is common language, no medical terms, and otherwise it is explained.” (female, 44 years)

Retrieval Recall “In my opinion, the period of 4 weeks is quite short.” (male, 61 years)

Judgment Symptom or adverse drug event

- “I thought, I experience all kind of things. But if you read further, it is about medication, then you say, no that thing has nothing to do with it. But I experience that symptom but it has nothing to do with medication.” (male, 80 years)- “I find it difficult to say which are side effects. I do experience symptoms but are they symptoms or side effects. No idea. And I think I still reported it [the symptom] because I do not know and because perhaps you may think when it is reported by everyone, it can be a side effect.” (female, 61 years)

Body categories

- “I think of only some things with that [body category] and then later I had to go back, no, this fits with that one [body category].” (female, 71 years)- “Do you have problems with your eyes? Yes. Sometimes I have a blurred vision, I cannot tolerate sunlight very well, so in that case you check eyes. Bladder, I use that tolbutamide from which I have to pee a lot, so the bladder. The skin, I have quite a dry skin lately. Often, my back or my hands are itching, so then you check skin. So, you just go by this [list]”(female, 53 years)

Response Answer options

Regarding ‘how often’ ADE is experienced:“Almost every day and that for a period of 14 days.” (female, 58 years)

Lack of… “I would include whether the side effect is treated or whether it disappeared spontaneously.” (female, 44 years)

Respondent burden

Structure “It can be confusing, at one time you are asked for the drug. And the next time not. Then again about side effects, and then again about drugs.” (male, 58 years)

Lay-out - “For me it [the size of the letters] is a little bit small.” (male, 90 years)- “At a certain point I found it [the navigation in the questionnaire] a bit chaotic.” (female, 44 years)

Relevance Regarding the number of answer options: “Somewhat less. Maybe half of it can go. I think, everything a little bit more concise.” (female, 53 years)

R1R2R3R4R5R6R7R8R9


Chapter 1

36

Table 1.3. Comparison of questions regarding nature and causality of ADEs between initial and last revisionAttribute assessed

Initial question Question in last revision Answer options

Duration/ timeline

Since when have you experienced this side effect*? Try to be as specific as possible (for example 1 March 2009 or June 2010 or 2006).

When did you first experience this side effect of your medication?

Adapted to checklist-based

Duration/ timeline

– New: Has this side effect gone away by now or improved?

–

Frequency How often did you experience this side effect during the past 4 weeks?

How often did you experience this side effect during the past 4 weeks (on how many or which days)?

Adapted to open-ended

Severity How much did this side effect bother you in the past 4 weeks (how bad or intense was it)?

On the days that you experienced this side effect, how much did it bother you (how bad or intense was it)?

No changes

Severity + impact on activities

How much influence did this side effect have on your daily functioning in the past 4 weeks?

On the days that you experienced this side effect, how much influence did it have on your daily functioning?

No changes

Actions taken – New: What action did you take in relation to this side effect during the past 4 weeks?

–

Seriousness Did this side effect result in serious medical situations for yourself during the past 4 weeks?

Unchanged 1 answer option excluded

Importance to patient + benefit-risk assessment

How satisfied are you with the drug/drugs described in question IV-8 when you consider both this particular side effect and the effect of this drug/these drugs?

How satisfied are you with the drug (or drugs) described in question 38 when you consider both this particular side effect and the effect of the drug or drugs?

No changes

Causality Which drug(s) do you think caused this side effect?

Which drug or drugs do you think caused this side effect?

No changes

Why do you think this side effect is caused by this drug/these drugs (several answers possible)?

Why do you think this symptom was caused by your medication (you may give more than one answer)?

Length of answer options has been shortened

How sure are you that this side effect is caused by this drug?

How sure are you that this side effect is caused by this drug or these drugs?

No changes

Do you think there are possible other factors for your experiencing this side effect (other than your medication)?

Do you think there are other reasons for your experiencing this side effect (other than your medication)?

No changes

Have you experienced this symptom in the past in combination with other medication?

Have you experienced this side effect in the past in combination with other medication?

No changes

Causality (also timeline)

– New: How long had you been using this drug or these drugs before this side effect started occurring?

–

* The term side effect is used in the questionnaire as lay-term for adverse drug event



37

1

Classification task Based on the classification task, where the patients had to assign ADEs to body categories, 51 problematic ADEs (20%) were detected. As a consequence, we made the following adaptations: shifting the ADE to a more fitting body category (5 ADEs), renaming the ADE (2 ADEs), a combination of shifting and renaming of the ADE (2 ADEs), renaming a body category (8 ADEs), combining body categories (16 ADEs), and creating a new body category (6 ADEs). For 12 ADEs, no changes were made. Final revision Based on the English translation, one ADE was detected that was considered ambiguous in the original Dutch version. To solve this, two ADE descriptions instead of one were introduced (“blood with feces” and “blood in feces”). The comparison of the English version with the Dutch version resulted in a few minor changes regarding the wording in both versions. Finally, after combining the ADEs with an overlapping MedDRA® term (dry teeth/mouth), the final questionnaire contained 252 ADEs categorized in 16 body categories with 14 questions per ADE regarding its nature and causality (Appendix 1, and Table 1.3). Feasibility and reliabilityIn total, 187 patients gave informed consent in response to an invitation that was mailed to 958 patients. These 187 patients were slightly younger (65 vs 67 years, Z = -2.653, P < 0.01) than patients not responding. There was no significant difference regarding gender (39.6 vs 44.7% women, χ2 = 1.638, P = 0.20). Of the consenting patients, 152 started with the study by opening the questionnaire, and 137 completed both questionnaires (73.3%). Four times, a patient reported an ADE in the “other” box, which could all be classified to one of the listed ADEs by the researchers. One patient reported in the comments that the reported ADE was probably not due to a drug but due to surgery. This ADE was excluded from further analysis. One patient was excluded from the test–retest analysis for reporting to have experienced the “same symptoms” at T2 as at T1, instead of checking the symptoms again. Another was excluded because of this patient’s comment that several symptoms had been wrongly checked. Further analyses were thus based on 135 patients, 45 in each group. The median age of this population was 65 years; on average, they used five prescription drugs (Table 1.4). The median number of days between completing the first and second questionnaires was 8 days (SD: 4). At T1, 25.2% (N = 34) of the 135 patients reported one or more ADEs, and 27.4% (N = 37) at T2. In total, 173 ADEs were reported at T1, and 146 ADEs at T2. The most common type of ADEs were gastrointestinal disorders (Table 1.5). Less than 1% of the

R1R2R3R4R5R6R7R8R9


Chapter 1

38

questions about the nature and causality of the ADE were not completed (0.4% missing at T1, and 0.2% at T2). For most ADEs (124 at T1 and 96 at T2), patients checked only one reason for suspecting the ADE. The most common reason was that they did not experience the symptom before they took the drug. In three quarters of the cases, the patients indicated which drug they thought caused the symptom, and in most of these cases they were quite sure about the relationship between the drug and the ADE (Table 1.5). Finally, there were 51 cases where a symptom was reported only as a symptom at one point but as a possible ADE at another time (22 times as symptom at T1 but ADE at T2, and 29 times as ADE at T1 but symptom at T2).

Table 1.4. Patient characteristics, number of ADEs reported per group (P-values for differences among the three groups)

Total Test-retest group (both with body categories)

Group with the body categories at T1

Group with the body categories at T2

P-value

Number of participants 135 45 45 45Females (%) 49 (36.3) 16 (35.6) 14 (31.1) 19 (42.2) .544Median age in years (range)

65 (41-86) 64 (44-86) 67 (47-82) 63 (41-83) .210

Education (%) Low educateda

Middle educatedb

High educatedc

Other

38 (28.1)50 (37.0)40 (29.6)7 (5.2)

15 (33.3) 13 (28.9)14 (31.1)3 (6.7)

11 (24.4)17 (37.8)15 (33.3)2 (4.4)

12 (26.7)20 (44.4)11 (24.4)2 (4.4)

.796

Median number of self- reported prescription drugs (range)

5 (2-18) 5 (2-14) 5 (2-13) 6 (3-18) .095

Median number of self- reported chronic diseases, including DM and CVD (range)

3 (1-10) 3 (2-8) 3 (1-5) 3 (2-10) .367

Number of patients reporting an ADE at T1 & T2 (%)

T1: 34 (25.2)T2: 37 (27.4)

T1: 12 (26.7)T2: 13 (28.9)

T1: 11 (24.4)T2: 11 (24.4)

T1: 11 (24.4)T2:13 (28.9)

T1: .961T2: .862

Number of ADEs reported (range)

T1: 173 (1-19)T2: 146 (1-11)

T1: 64 (1-15)T2: 51 (1-10)

T1: 35 (1-10)T2: 34 (1-9)

T1: 74 (1-19)T2: 61 (1-11)

T1: .339T2: .394

T1 = First measurement; T2 = Second measurement after one week periodDM = Diabetes mellitus; CVD = Cardiovascular diseasesa No education; elementary school; junior secondary vocational educationb Junior general secondary education; senior secondary vocational educationc Senior general secondary education; higher professional education; university education



39

1

Table 1.5. Nature and causality reported at ADE level for the three groups at first (T1) and second (T2) measurement

T1 (N=173)

T2(N=146)

MedDRA System Organ Class (%): Cardiac disorders Ear and labyrinth disorders Eye disorders Gastrointestinal disorders General disorders and administration site conditions Infections and infestations Injury, poisoning and procedural complications Investigations Metabolism and nutrition disorders Musculoskeletal and connective tissue disorders Nervous system disorders Psychiatric disorders Renal and urinary tract disorders Reproductive system and breast disorders Respiratory, thoracic and mediastinal disorders Skin and subcutaneous tissue disorders Vascular disorders

1 (0.6)1 (0.6)17 (9.8)53 (30.6)11 (6.4)2 (1.2) 0 (0.0)4 (2.3)5 (2.9)4 (2.3)31 (17.9)9 (5.2)2 (1.2)8 (4.6)6 (3.5)14 (8.1)5 (2.9)

0 (0.0)1 (0.7)6 (4.1)37 (25.3)10 (6.8)1 (0.7)1 (0.7)9 (6.2)6 (4.1)5 (3.4)17 (11.6)22 (15.1)2 (1.4)8 (5.5)7 (4.8)10 (6.8)4 (2.7)

Number of times a reason for a relationship among drug and ADE was reported: - I did not experience this symptom before I started taking the drug - The symptom started soon after I started taking the drug - I experienced this symptom less often before I started taking the drug - The symptom was less serious before I started taking the drug - The symptom went away when I stopped taking the drug and came back when I started taking it again - The symptom went away when I stopped taking the drug - The symptom started or grew worse when the drug dosage was increased - The symptom decreased or went away when the drug dosage was decreased - A healthcare professional (for example a doctor or pharmacist) confirmed this - The symptom is described in the patient leaflet - Other

97492082

5142203411

87331782

191222212

Number of times it was not known by the patient which drug caused the ADE (%)

75 (43.4) 61 (41.8)

Patients’ certainty about the relationship among the reported ADE and reported drug (%)*: Very sure Quite sure Not very sure Very unsure

30 (17.3)48 (27.7)14 (8.1)2 (1.2)

21 (12.1)50 (28.9)7 (4.0)4 (2.3)

* Percentages do not sum to 100% due to missings or not indicating a causal drug

R1R2R3R4R5R6R7R8R9


Chapter 1

40

Self-reported time for questionnaire completion was in general lower than the registered time (Table 1.6). On average, the median self-reported time was 15 minutes for patients not reporting any ADE (with three patients reporting >30 minutes), and 30 minutes for those reporting one or more ADEs (with four patients reporting >60 minutes). Differences observed in completion time between the questionnaire with and without body categorization were not significant (Table 1.6). Most of the patients agreed that the questionnaire was easy to use (74.4% for the questionnaire with body categories; 75.6% for the questionnaire without body categories), which did not significantly differ between the two versions of the questionnaire (χ2 = 0.028, P = 0.986). Overall, this percentage was lower for patients reporting one or more ADEs than for patients not reporting any ADE (52.9 vs 82.2%, χ2 = 12.791, P = 0.002). Table 1.6. Time in minutes needed to complete the questionnaire with and without body categories for reporting no ADE and ≥1 ADEs

With body categories (Group with body categories at T1 + Test-retest group, N=90)

Without body categories (Group with body categories at T2, N=45)

P-value

No ADE reported: Median self-reported time needed to complete questionnaire (range; SD)

15 (3-40; 7.7) 13 (2-60; 10.7) 0.377

Median registered time needed to complete questionnaire (range; SD)

17 (5-48; 10.1) 16 (7-82; 16.2) 0.720

One or more ADEs reported: Median self-reported time needed to complete questionnaire (range; SD)

23 (15-240; 63.6) 40 (20-120; 29.0) 0.166

Median registered time needed to complete questionnaire (range; SD)

54 (22-96; 24.1) 71 (32-147; 36.5) 0.115

SD = Standard deviation; T1 = First measurement; T2 = Second measurement after one week period

The agreement of reported ADEs regarding the test–retest reliability was acceptable at patient level and at MedDRA® level (κ >0.5, proportion of positive agreement >0.5). At ADE specific level, the agreement was lower (κ = 0.38, proportion of positive agreement = 0.38, Table 1.7). By aggregating separately checked but related ADEs according to the patient’s own description, the 64 ADEs reported at T1 were reclassified as 34 distinct ADEs, and the 51 ADEs at T2 as 31 distinct ADEs. There was agreement for 16 of these ADEs and the proportion of positive agreement was 0.49. Agreement between the two measurements was slightly higher for patients who completed the questionnaire including body categories at first measurement in comparison to those who first completed the questionnaire without this categorization. However, kappa values did not significantly differ between the group with the body



41

1

categories at T1 and the group with the body categories at T2 (Table 1.7). The two-by-two tables of the agreement analyses are presented in Appendix 2 (supplemental table 1 and 2). The number of reported ADEs was similar between the questionnaire with and without body categories (Z = -0.049, P = 0.961). Sensitivity analyses including only those patients who completed the second questionnaire within 10 days did not lead to significant differences in agreement measures (Appendix 2; supplemental table 3 and 4).

Table 1.7. Kappa values and proportion of positive agreement for test-retest reliability and body categories at patient level, MedDRA® level and ADE specific level

Test-retest group Group with body categories at T1

Group with body categories at T2

Differences in kappa values between group with body categories at T1 and T2 Z-value

Kappa values(95% CI)

PPA Kappa values(95% CI)

PPA Kappa values (95% CI)

PPA

Patient level(N=45)

0.502 (0.21-0.79)

0.64 0.639 (0.37-0.91)

0.73 0.433 (0.12-0.74)

0.58 0.387

MedDRA level(N=810)

0.521 (0.35-0.69)

0.54 0.395 (0.19-0.60)

0.42 0.264 (0.12-0.40)

0.30 0.330

ADE specific level(N=11340)

0.380 (0.24-0.52)

0.38 0.259 (0.06-0.46)

0.26 0.158 (0.003-0.31)

0.16 0.301

T1 = First measurement; T2 = Second measurement after one week period; PPA = Proportion Positive Agreement calculated by the formula 2a/[N+(a-d)], in which N=total number of observations, a=patients reporting ADE at T1 and T2, and d=patients not reporting ADE at T1 and T2. CI = Confidence interval.

DiscussionWe developed and tested a generic questionnaire for patient reporting of ADEs. The questionnaire adds to the available questionnaires in that it is both generic and checklist-based and includes specific questions about causality, severity, duration, seriousness, and frequency of each experienced ADE. The questionnaire is intended for use in post-marketing studies and clinical trials.

Through cognitive debriefing interviews, significant problems were detected in several domains of the question-and-answer model that needed to be resolved. After initial adaptations, some problems reoccurred, underlining the relevance of an iterative process. The input of patients was found to be vital for the development and content validation. It became clear that directly asking for ADEs can lead to over-reporting because some patients accidently checked symptoms as well as ADEs when confronted with a list of symptomatic ADEs. While going through the lists, patients sometimes forgot that they should only check symptoms perceived as being ADEs. This happened even

R1R2R3R4R5R6R7R8R9


Chapter 1

42

while patients were able to distinguish ADEs from symptoms, as has been established before [37,47]. Some of the available checklist-based ADE questionnaires use terms such as symptoms, problems, and ADEs interchangeably (e.g., see [18,27]). We recommend clear differentiation between symptoms that could be related to the underlying disease and ADEs, as is done in other checklists [23], ensuring that respondents maintain the distinction while completing the questionnaire. This mechanism may explain in part why more ADEs are reported in checklists than in open-ended questionnaires [13]. Several patients reported that a recall period of 4 weeks was quite short, for instance, to capture ADEs that fluctuate over time, as has been identified before [48]. On the other hand, the period should not be too long when the aim is to collect information on symptomatic ADEs that can be mild in nature. The optimal recall period may depend on the nature of the ADE [48]. Although a recall period of 4 weeks is quite common, and even shorter recall periods have been used in ADE questionnaires [17], the reliability of various recall periods needs to be tested in further studies. Reducing respondent burden is relevant for the feasibility of using the questionnaire. We identified problems in navigating the questionnaire and these were solved by formatting the questionnaire along principles of cognitive design [49]. Around half of the patients found the body category structure helpful, but we detected some difficulties with our initial ADE classification based on the MedDRA® System Organ Classes. We thus adapted this to a more patient-based classification system. The feasibility test showed, however, that the categorization structure only marginally decreased the time to complete the questionnaire for patients reporting at least one ADE. Only four ADEs were reported as “other”, indicating that most patients were able to identify their experienced ADE within the provided lists. For most of the patients reporting at least one ADE, the time needed to complete the questionnaire was <60 minutes. In our opinion, this time is acceptable for a questionnaire intended for research purposes, in which questions about general characteristics and drug use were included. It should, however, be noted that only a quarter of the patients reported at least one ADE. The majority of the patients agreed that the questionnaire was easy to use, but this number was lower for those reporting an ADE than those reporting no ADEs. Of the patients who opened the questionnaire, around 10% were lost to follow-up. Although the test–retest reliability of the patient-reported ADE questionnaire was considered acceptable at patient level and at MedDRA® level, it was below the threshold of 0.6–0.8 recommended for reliability coefficients [50]. For ADE reporting, however, a skewed distribution is observed where many patients report no ADEs on both measurements, which decreases the kappa values used for the reliability assessment [51,52]. Formulas to adjust for such effects have been proposed, for example, the prevalence-adjusted bias-adjusted kappa [53], but their inappropriateness has also been demonstrated [51]. We



43

1

therefore calculated the proportions of positive agreement as an alternative agreement measure, which showed similar results. Future studies assessing the reliability of ADE reporting are advised to recruit a more balanced group of patients experiencing and not experiencing ADEs [51]. Based on a combined approach, that is, looking at kappa values, alternative agreement measures, and additional analysis of ADEs at patient level, we conclude that our questionnaire was not sufficiently reliable at the ADE-specific level. This result implies that the distinct symptoms reported by patients as ADEs using these checklists should not be used blindly to quantify rates at the lowest ADE-specific level. Part of the lack of reliability might be solved by improving the questionnaire, but some lack of reliability at the lowest ADE level could be inherent to patient reporting. One can expect that uncertainty by patients about a symptom being an ADE may lead to inconsistent answers. The finding that some patients checked a symptom as an ADE on one measurement but not on the other indicates such uncertainty. Furthermore, in around half of the cases the patients did not mention a potential drug that they believed was causing that specific ADE or were not very sure about the causal relationship. On the other hand, some of the inconsistency was caused by using a checklist that does not require differentiation between related and disparate ADEs. Patients often checked multiple related ADEs, but not exactly the same ADEs on the two measurements. When aggregated at MedDRA® level or using the patient’s own descriptions, patients were therefore found to be more consistent. This problem could be a consequence of direct patient reporting; that is, reporting without involvement of a healthcare professional who can interpret and cluster specific symptoms to a more general ADE description. However, a more intelligent questionnaire flow or an interactive questionnaire, might solve this problem. For instance, using an interactive questionnaire requiring patients to cluster related symptoms that are considered as one problem before they move to answer more detailed questions. Such a questionnaire should incorporate a more flexible linkage to the MedDRA® System Organ Class by not only focusing on the primary MedDRA® class. This prevents symptoms with different primary MedDRA® classes used to describe one ADE being classified in different MedDRA® classes. Notwithstanding these possible improvements to the questionnaire, some patients clearly checked totally different ADEs at the two measurements. We chose a period of 1 week between the measurements to exclude memory effects, but this period may have been too long to exclude true changes in the experience of ADEs in the previous 4 weeks, especially for ADEs that might change from day to day [12].

R1R2R3R4R5R6R7R8R9


Chapter 1

44

The comparison between the questionnaire with and without the body category structure showed no significant differences in the number of reported ADEs or in agreement measures. From this we conclude that including a body category system did not influence the reliability of the ADE reporting. Because the cognitive debriefing showed that the body categories were helpful and increased the feasibility for some patients, we still recommend the use of such a categorization as a supportive element. To our knowledge, this is the first study to validate a generic patient-reported questionnaire intended for systematic data collection of ADEs. We conducted a broad search for symptomatic ADEs, which we translated in lay-terms and linked to MedDRA® terms. The use of these standard terms makes it possible to compare ADE data across different studies, which is important in the evaluation of drug safety [54]. We included a heterogeneous population with respect to age and education level in the content-validation study. Patients were selected for having type 2 diabetes, asthma, or COPD, but many of them used multiple drugs, and also used drugs for other diseases. We expect that the questionnaire is suitable for adult patients on a steady drug regimen who are able to read and write. We cannot, however, guarantee that all ADE terms are content valid. In addition, we tested the Dutch version of the questionnaire. The use of the questionnaire in other languages requires additional testing [55]. We expect that the reliability for ADE reporting of the web-based version is comparable to the paper-based version. The navigation through the questionnaire and the time needed to complete, however, may differ between the web-based and paper-based versions [56]. We tested the questionnaire in an observational, post-marketing setting. We expect that the questionnaire is also applicable in clinical trials in which patients are initial drug users, but this should be confirmed in future studies. Further validation studies are needed (e.g., establishing the probability of a causal relationship between the reported ADEs and the drugs using an external reference) because content validation is an essential but only first step in providing evidence of full validity [12,57,58]. ConclusionsParticipants in post-marketing studies and clinical trials can use multiple drugs that may interact and cause unexpected ADEs. Using a generic questionnaire in which all experienced ADEs can be reported by patients is therefore important. In terms of content validity, our patient-reported ADE questionnaire can be used for assessing the nature and causality of symptomatic ADEs as experienced by patients undergoing chronic drug therapy. The questionnaire is feasible for research purposes, and reliable to identify numbers of patients experiencing ADEs in general and at MedDRA® System Organ Class level. To quantify specific patient-reported ADEs, improvements to the structure of the questionnaire are required.



45

1

Acknowledgements The authors thank Lareb for providing a dataset with ADEs as worded and reported by patients in the Lareb Intensive Monitoring Project, Dr. L. Härmark for judging problematic ADEs into a body category, and Dr. F.L.P. van Sonderen for providing support in the construction of the questionnaire and agreement analyses. MedDRA® is a registered trademark of the International Federation of Pharmaceutical Manufacturers and Associations (IFPMA).

R1R2R3R4R5R6R7R8R9


Chapter 1

46

References[1] Meadows KA. Patient-reported outcome

measures: an overview. Br J Community Nurs 2011;16(3):146–51.

[2] Zhu J, Stuver SO, Epstein AM, Schneider EC, Weissman JS, Weingart SN. Can we rely on patients’ reports of adverse events? Med Care 2011;49(2):948–55.

[3] Jarernsiripornkul N, Krska J, Capps PA, Richards RM, Lee A. Patient reporting of potential adverse drug reactions: a meth-odological study. Br J Clin Pharmacol 2002;53(3):318–25.

[4] Weingart SN, Gandhi TK, Seger AC, Seger DL, Borus J, Burdick E, et al. Patient-report-ed medication symptoms in primary care. Arch Intern Med 2005;165(2):234–40.

[5] Basch E, Jia X, Heller G, Barz A, Sit L, Frus-cione M, et al. Adverse symptom event re-porting by patients vs clinicians: relation-ships with clinical outcomes. J Natl Cancer Inst 2009;101(23):1624–32.

[6] Wetzels R, Wolters R, van Weel C, Wensing M. Mix of methods is needed to identify ad-verse events in general practice: a prospec-tive observational study. BMC Fam Pract 2008;9:35.

[7] Blenkinsopp A, Wilkie P, Wang M, Rout-ledge PA. Patient reporting of suspected ad-verse drug reactions: a review of published literature and international experience. Br J Clin Pharmacol. 2007;63(2):148–56.

[8] Hakobyan L, Haaijer-Ruskamp FM, de Zeeuw D, Dobre D, Denig P. A review of methods used in assessing non-serious ad-verse drug events in observational studies among type 2 diabetes mellitus patients. Health Qual Life Outcomes 2011;9:83.

[9] US Department of Health and Human Ser-vices, FDA, Center for Drug Evaluation and Research Center for Biologics Evaluation and Research Center for Devices and Ra-diological Health. Guidance for industry patient-reported outcome measures: use in medical product development to support labeling claims [online]. http://www.fda.gov/downloads/Drugs/GuidanceCompli-ance RegulatoryInformation/Guidances/UCM193282.pdf. Accessed 14 Nov 2012.

[10] European Medicines Agency. Fourth report on the progress of the interaction with patients’ and consumers’ organizations (2010) and results/analysis of the degree

of satisfaction of patients and consumers involved in EMA activities during 2010. EMA/ 632696/2011 [online]. http://www.ema.europa.eu/docs/en_GB/document_li-brary/Report/2011/10/WC500116866.pdf. Accessed 14 Nov 2012.

[11] European Medicines Agency. Note for guid-ance on clinical investigation of medici-nal products for treatment of nociceptive pain [online]. http://www.ema.europa.eu/docs/en_GB/document_library/Scientif-ic_guideline/2009/09/WC500003525.pdf. Accessed 14 Nov 2012.

[12] Frost MH, Reeve BB, Liepa AM, Stauffer JW, Hays RD; Mayo/FDA Patient-reported out-comes consensus meeting group. What is sufficient evidence for the reliability and validity of patient-reported outcome meas-ures? Value Health 2007;10(Suppl 2):S94–105.

[13] Bent S, Padula A, Avins AL. Brief commu-nication: better ways to question patients about adverse medical events: a rand-omized, controlled trial. Ann Intern Med 2006;144(4):257–61.

[14] Sheftell FD, Feleppa M, Tepper SJ, Rapoport AM, Cianella L, Bigal ME. Assessment of adverse events associated with triptans—methods of assessment influence the re-sults. Headache 2004;44(10):978–82.

[15] Ruiz MA, Pardo A, Rejas J, Soto J, Villasante F, Aranguren JL. Development and valida-tion of the ‘‘Treatment Satisfaction with Medicines Questionnaire’’ (SATMED-Q). Value Health 2008;11(5):913–26.

[16] Bytzer P, Talley NJ, Jones MP, Horowitz M. Oral hypoglycaemic drugs and gastrointes-tinal symptoms in diabetes mellitus. Ali-ment Pharmacol Ther 2001;15(1):137–42.

[17] Foster JM, van Sonderen E, Lee AJ, San-derman R, Dijkstra A, Postma DS, et al. A self-rating scale for patient-perceived side effects of inhaled corticosteroids. Respir Res 2006;7:131.

[18] Sitzia J, Dikken C, Hughes J. Psychometric evaluation of a questionnaire to document side-effects of chemotherapy. J Adv Nurs 1997;25(5):999–1007.

[19] Corso DM, Pucino F, DeLeo JM, Calis KA, Gal-lelli JF. Development of a questionnaire for detecting potential adverse drug reactions. Ann Pharmacother 1992;26(7–8):890–6.



47

1

[20] International Conference on Harmonisa-tion of Technical Requirements for Reg-istration of Pharmaceuticals for Human Use. ICH harmonised tripartite guideline pharmacovigilance planning E2E [online]. http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Effi- cacy/E2E/Step4/E2E_Guideline.pdf. Acces- sed 14 Nov 2012.

[21] Härmark L, Puijenbroek E, Grootheest K. Longitudinal monitoring of the safety of drugs by using a web-based system: the case of pregabalin. Pharmacoepidemiol Drug Saf 2011;20(6):591–7.

[22] US Department of Health and Human Ser-vices, National Institutes of Health National Cancer Institute. Common Terminology Cri-teria for Adverse Events (CTCAE) version 4 [online]. http://evs.nci.nih.gov/ftp1/CT-CAE/CTCAE_4.03_2010-06-14_Quick Ref-erence_8.5x11.pdf. Accessed 14 Nov 2012.

[23] Uher R, Farmer A, Henigsberg N, Rietschel M, Mors O, Maier W, et al. Adverse reac-tions to antidepressants. Br J Psychiatry. 2009;195(3):202–10.

[24] Welch V, Singh G, Strand V, Fries J, Boers M, Ramey D, et al. Patient based method of assessing adverse events in clinical trials in rheumatology: the revised Stanford Toxic-ity Index. J Rheumatol 2001;28(5):1188–91.

[25] Basch E, Iasonos A, Barz A, Culkin A, Kris MG, Artz D, et al. Long-term toxicity mon-itoring via electronic patient-reported out-comes in patients receiving chemotherapy. J Clin Oncol 2007;25(34):5374–80.

[26] Broström A, Strömberg A, Mårtensson J, Ulander M, Harder L, Svanborg E. Associ-ation of Type D personality to perceived side effects and adherence in CPAP-treat-ed patients with OSAS. J Sleep Res 2007;16(4):439–47.

[27] Day JC, Wood G, Dewey M, Bentall RP. A self-rating scale for measuring neurolep-tic side-effects. Validation in a group of schizophrenic patients. Br J Psychiatry 1995;166(5):650–3.

[28] Brown V, Sitzia J, Richardson A, Hughes J, Hannon H, Oakley C. The development of the Chemotherapy Symptom Assessment Scale (C-SAS): a scale for the routine clinical assessment of the symptom experiences of patients receiving cytotoxic chemotherapy. Int J Nurs Stud 2001;38(5):497–510.

[29] The Ohio State University Medical Center. Checklist for reporting side effects [online]. https://patienteducation.osumc.edu/Pag-es/search.aspx?k=checklist?for?reporting. Accessed 14 Nov 2012.

[30] Härmark L, van Grootheest K. Web-based intensive monitoring: from passive to ac-tive drug surveillance. Expert Opin Drug Saf 2012;11(1):45–51.

[31] DeWitt JE, Sorofman BA. A model for un-derstanding patient attribution of ad-verse drug reaction symptoms. Drug Inf J 1999;33:907–20.

[32] De Smedt RH, Jaarsma T, Ranchor AV, van der Meer K, Groenier KH, Haaijer-Ruskamp FM, et al. Coping with adverse drug events in patients with heart failure: exploring the role of medication beliefs and perceptions. Psychol Health 2012;27(5):570–87.

[33] Härmark L, van Puijenbroek E, Straus S, van Grootheest K. Intensive monitoring of pregabalin: results from an observational, web-based, prospective cohort study in the Netherlands using patients as a source of information. Drug Saf 2011;34(3):221–31.

[34] Grootenhuis PA, Snoek FJ, Heine RJ, Bou-ter LM. Development of a type 2 diabetes symptom checklist: a measure of symptom severity. Diabet Med 1994;11(3):253–61.

[35] Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30(6):473–83.

[36] Naranjo CA, Busto U, Sellers EM, et al. A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther 1981;30(2):239–45.

[37] Krska J, Anderson C, Murphy E, Avery AJ. How patient reporters identify adverse drug reactions: a qualitative study of re-porting via the UK Yellow Card Scheme. Drug Saf 2011;34(5):429–36.

[38] Ploughman M, Austin M, Stefanelli M, God-win M. Applying cognitive debriefing to pre-test patient-reported outcomes in old-er people with multiple sclerosis. Qual Life Res 2010;19(4):483–7.

[39] Collins D. Pretesting survey instruments: an overview of cognitive methods. Qual Life Res 2003;12(3):229–38.

[40] Schleyer TK, Forrest JL. Methods for the design and administration of web-based surveys. J Am Med Inform Assoc 2000;7(4):416–25.

R1R2R3R4R5R6R7R8R9


Chapter 1

48

[41] Urbaniak GC, Plous S. Research Randomizer (version 3.0) [online computer software]. http://www.randomizer.org/. Accessed 14 Nov 2012.

[42] De Vet H, Terwee C, Mokkink L, Knol D. Measurement in medicine: a practical guide. Cambridge: University Press; 2011.

[43] Cicchetti DV, Feinstein AR. High agreement but low kappa: II. Resolving the paradoxes. J Clin Epidemiol 1990;43(6):551–8.

[44] St Sauver JL, Hagen PT, Cha SS, Bagniewski SM, Mandrekar JN, Curoe AM, et al. Agree-ment between patient reports of cardiovas-cular disease and patient medical records. Mayo Clin Proc 2005;80(2):203–10.

[45] Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33(1):159–74.

[46] Foody GM. Thematic map comparison: evaluating the statistical significance of differences in classification accura-cy. Photogramm Eng Remote Sensing 2004;70(5):627–33.

[47] De Smedt RH, Denig P, Haaijer-Ruskamp FM, Jaarsma T. Perceived medication ad-verse effects and coping strategies report-ed by chronic heart failure patients. Int J Clin Pract. 2009;63(2):233–42.

[48] Stull DE, Leidy NK, Parasuraman B, Chas-sany O. Optimal recall periods for pa-tient-reported outcomes: challenges and potential solutions. Curr Med Res Opin 2009;25(4):929–42.

[49] Mullin PA, Lohr KN, Bresnahan BW, McNul-ty P. Applying cognitive design principles to formatting HRQOL instruments. Qual Life Res 2000;9(1):13–27.

[50] Kottner J, Audigé L, Brorson S, Donner A, Gajewski BJ, Hróbjartsson A, et al. Guide-lines for Reporting Reliability and Agree-ment Studies (GRRAS) were proposed. J Clin Epidemiol 2011;64(1):96–106.

[51] Hoehler FK. Bias and prevalence ef-fects on kappa viewed in terms of sen-

sitivity and specificity. J Clin Epidemiol 2000;53(5):499–503.

[52] Sim J, Wright CC. The kappa statistic in reliability studies: use, interpretation, and sample size requirements. Phys Ther 2005;85(3):257–68.

[53] Byrt T, Bishop J, Carlin JB. Bias, prev-alence and kappa. J Clin Epidemiol 1993;46(5):423–9.

[54] US Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER), Center for Biologics Evaluation and Research (CBER). Guidance for indus-try Premarketing risk assessment [online]. http://www.fda.gov/downloads/Regula-toryInformation/Guidances/ucm126958.pdf. Accessed 14 Nov 2012.

[55] Coyne KS, Tubaro A, Brubaker L, Baven-dam T. Development and validation of patient-reported outcomes measures for overactive bladder: a review of concepts. Urology 2006;68(2 Suppl):9–16.

[56] Dillman DA, Gertseva A, Mahon-Haft T. Achieving usability in establishment sur-veys through the application of visual de-sign principles. J Off Stat 2005;21(2):183–214.

[57] Brod M, Tesler LE, Christensen TL. Qualita-tive research and content validity: develop-ing best practices based on science and ex-perience. Qual Life Res 2009;18(9):1263–78.

[58] Rothman M, Burke L, Erickson P, Leidy NK, Patrick DL, Petrie CD. Use of existing patient-reported outcome (PRO) instru-ments and their modification: the ISPOR good research practices for evaluating and documenting content validity for the use of existing instruments and their modifi-cation PRO task force report. Value Health 2009;12(8):1075–83.

Supplement I

User acceptance of a web-based version of a

patient-reported adverse drug event questionnaire

R1R2R3R4R5R6R7R8R9


Supplement I

50

IntroductionThe use of web-based questionnaires in research has several advantages compared to paper-based questionnaires, such as direct storage of the data, objective measurement of the response time, response checks, and automatic presentation of additional questions based on the given answers [1-5]. However, the use of web-based questionnaires may also introduce specific problems, which are in part related to selective participation or completion of such questionnaires [2,5,6]. Problems have been observed for aged patients [7-9], patients with a lower education level [8], and patients with a low level of computer and/or internet literacy [2,3,10]. Such patients are also expected to have problems with the specific features of web-based questionnaires [11], which influences the reliability and validity of the instrument. This includes, for instance, problems with visual aspects or respondent required actions, such as the use of a mouse and a scroll down option on a questionnaire page [2,6,12].

Although computer use and internet access have increased over time [5], also in the aged population [13], attention to the design of a web-based questionnaire and its general applicability is important [2,6]. Muylle et al. [14] have developed and validated a user satisfaction model as a theoretical and practical base for testing the user acceptance of websites. The model includes the dimensions 1) lay-out (perceived look), 2) ease of use (user friendliness), 3) entry guidance (quality of the start page), 4) structure (perception of the order in which different parts are linked up), 5) hyperlink connotation (easiness of using hyperlinks), 6) speed (whether it works slow or fast), 7) language customization (degree to which the language is customized to the user), and 8) information (degree to which the content is covered) [14].

Previously, we developed a paper-based questionnaire for the assessment of adverse drug events (ADEs) reported by patients (Appendix 1) [15]. Because of the advantages of web-based questionnaires, we developed a web-based version of the ADE questionnaire. The user acceptance of the web-based version was assessed in a pilot study using the website user satisfaction model as theoretical base.


Web-based version of a patient-reported adverse drug event questionnaire

51

I

Methods

ParticipantsWe included a pragmatic sample of 10 people who were being prescribed at least one chronic drug and who were aged 18 years or older.

Measures and procedureThe web-based version of the questionnaire was developed using Unipark Enterprise Feedback Suite 8.0 version 1.1 (www.unipark.de) and respondent-friendly design principles for web surveys [4]. The questionnaire contains general questions, checklists with possible ADEs, and additional questions per reported ADE about its nature [15]. The web-based version contains additional questions about the drug that patients relate to their experienced ADE (e.g. about the dose of the drug).

Participants of this pilot study received an e-mail with the URL (Uniform Resource Locator) to the website with the ADE questionnaire. Instructions about how to open the website and entering the questionnaire using a personal code were included in the e-mail. The personal entering code was used to prevent multiple completions of one patient [3]. The first page in the questionnaire contained information about how responses to the questions should be given.

After completing the questionnaire, telephone interviews were conducted by the researcher using a structured questionnaire with closed- and open-ended questions. These questions were based on the dimensions of the website user satisfaction model [14]. The dimension information was not assessed in this pilot study since the content validity of the questionnaire has been demonstrated for the paper-based version of the questionnaire (Chapter 1) [15]. The questions during the interviews were based on existing questionnaires and previously used questions related to website satisfaction [4,16-19], and were adapted to be used for questionnaire satisfaction (Table I.1). Closed-ended questions could be answered on a five-point Likert scale ranging from 1) totally disagree to 5) totally agree. Completion time was measured subjectively by asking the patients about the time needed to complete the questionnaire, and objectively by using the time registered in the Unipark program.

R1R2R3R4R5R6R7R8R9


Supplement I

52

Table I.1. Examples of questions used for dimensions of the website user satisfaction model [14].Dimension Examples of questionsLay-out; 3 items I can read the text in the questionnaire sufficiently Ease of use; 2 items In my opinion, the questionnaire is easy to completeEntry guidance; 4 items The first page in the questionnaire provides sufficient

information about completing the questionnaire Structure; 2 items In my opinion, a good structure is used in the questionnaire Hyperlink connotation; 2 items For me it was clear that I had to click the underlined sentence

in the e-mail to open the questionnaireSpeed; 2 items Opening the questionnaire took me a lot of timeLanguage customization; 1 item The questions are asked in an easy way

AnalysesFor each patient, the mean of the closed-ended questions was calculated for each dimension. With these means, the overall population mean was calculated per dimension. The descriptive analyses were conducted using IBM SPSS Statistics version 20 (Armonk, New York, USA). Answers given to the open-ended questions were screened by the researcher to identify specific problems in the user acceptance of the questionnaire.

ResultsThe 10 participants in this pilot study had a median age of 54 years, and included 4 females. Their self-assessed computer skills are reported in Table I.2. Most patients reported to use a computer daily (7 patients), or once in two days (3 patients), with a median of 12 hours per week (range: 3-35 hours). The patients used drugs for diseases such as type 2 diabetes, asthma, heart disease, and psychiatric problems. Four patients reported in total 15 ADEs in the questionnaire.

User acceptanceOverall, scores on the closed-ended questions were high (Figure I.1), indicating that the patients were positive about the questionnaire. The speed dimension was moderately scored, since a patient had problems opening the questionnaire due to the personal log-in code. During the pilot, this problem was resolved by including additional information in the e-mail.



53

I

Table I.2. Patient characteristics and number of reported adverse drug events (ADEs)Total number of participants 10 (4 females)Age in years <50 4 50-65 3 >65 3Education Lowa 3 Middleb 4 Highc 3Number of patients reporting an ADE 4Number of ADEs reported (range) 15 (1-6)Self-reported assessment of one’s computer skills Poor/reasonable 3 Moderate 3 Good/very good 4a No education; elementary school; junior secondary vocational educationb Junior general secondary education; senior secondary vocational educationc Senior general secondary education; higher professional education; university education

Some other problems were detected by the open-ended questions during the interview. One patient sometimes forgot to scroll down on a page in the questionnaire (lay-out). However, continuing the questionnaire was only possible when the button below the page was pressed. A difficulty regarding the ease of use was returning to a previous page in the questions per ADE (2 patients). In the Unipark software, this could not be adapted. Some patients reported that it was not possible to report the whole name of the drugs they had used. Therefore, the length of the text that could be added should be changed.

Another remark, made by two patients, was that they would have liked to print their answers given on the questionnaire. Therefore, information about how to print the answers should be added to the instruction text. Other remarks, applicable for the questionnaire in generalOne patient mentioned being confused about what to report on the item in the questionnaire in which a description of the ADE should be given. The patient had no additional information besides the ADE that was checked in the symptom list and therefore simply reported the ADE. Another remark was that it is sometimes difficult to indicate whether or not a symptom is an ADE.

R1R2R3R4R5R6R7R8R9


Supplement I

54

Lay-outEase of useEntry guidanceStructureHyperlink connotationsSpeedLanguage customization

Negative to positive1 2 3 4 5

Figure I.1. Patients’ evaluation of the questionnaire on the basis of the different dimensions

Completion timeSelf-reported completion time ranged from 5 to 60 minutes with a median of 25. Objective completion time as registered by the program ranged from 10 to 62 minutes with a median of 31.

DiscussionThis pilot study showed that the web-based version of the patient-reported ADE questionnaire was positively evaluated by the users. The completion time was on average 25-30 minutes, which is in our opinion acceptable for a questionnaire intended for research purposes. Few problems were detected, which were resolved by or are expected to be resolved by adapting or adding information to the questionnaire.

However, there are still some issues that should be considered. First of all, no extensive content validity assessment of the web-based version has been performed. The content validity is expected to be sufficient since it has been demonstrated for the paper-based version [15]. Previously, it has been shown that the mode of questionnaire presentation does not affect the interpretation of the questions [20]. In addition, the language customization dimension was positively evaluated, supporting the hypothesis that the content validity of the web-based version is sufficient. One patient was a little confused about an item in the questionnaire, indicating that a critical evaluation of a questionnaire is an ongoing process.

On some pages of the questionnaire, patients have to scroll down to see all the answer options. A previous study showed that this may induce frustration by the respondents [21]. Although, one patient in the current study sometimes forgot the scroll down to see all the answer options, no frustration was reported. Not seeing all the answer options



55

I

may introduce bias to the given answers. This bias is not expected to have occurred in the current study, since patients were forced to scroll down to press a button for continuing the questionnaire.

The used software limited the flexibility of the construction of the questionnaire. In future studies, the use of more flexible software is important.

This pilot study has the limitation of a small sample size. However, our patient sample was heterogeneous with respect to age and computer experience. Another limitation is that we did not systematically assess whether the questionnaire could be opened using every internet browser and service provider [12]. The browsers and providers used by the patients in this pilot study did not reveal any problems.

To conclude, the web-based version of the patient-reported ADE questionnaire has sufficient user acceptance to allow its use in further studies assessing the reliability and validity of the instrument.

R1R2R3R4R5R6R7R8R9


Supplement I

56

References[1] Eysenbach G, Wyatt J. Using the Internet for

surveys and health research. J Med Internet Res 2002;4(2):E13.

[2] Wyatt JC. When to use web-based surveys. J Am Med Inform Assoc 2000;7(4):426-29.

[3] Schleyer TK, Forrest JL. Methods for the design and administration of web-based surveys. J Am Med Inform Assoc 2000;7(4):416-25.

[4] Dillman DA, Tortora RD, Bowker D. Princi-ples for constructing web surveys [online]. http://134.121.51.35/dillman/papers /1998/PrinciplesforConstructingWebSur-veys.pdf. Accessed 25 Feb 2014.

[5] van Gelder MMHJ, Bretveld RW, Roeleveld N. Web-based Questionnaires: The Fu-ture in Epidemiology? Am J Epidemiol 2010;172(11):1292-8.

[6] Manfreda KL, Batagelj Z, Vehovar V. Design of Web Survey Questionnaires: Three Basic Experiments. J Comput Mediat Commun 2002;7(3).

[7] Härmark L, Alberts S, van Puijenbroek E, Denig P, van Grootheest K. Representative-ness of diabetes patients participating in a web-based adverse drug reaction monitor-ing system. Pharmacoepidemiol Drug Saf 2013;22(3):250-255.

[8] Bech M, Kristensen MB. Differential re-sponse rates in postal and Web-based sur-veys among older respondents. Surv Res Methods 2009;3(1):1–6.

[9] Klovning A, Sandvik H, Hunskaar S. Web-based survey attracted age-biased sample with more severe illness than paper-based survey. J Clin Epidemiol 2009;62(10):1068-1074.

[10] Basch E, Artz D, Dulko D, Scher K, Sabbatini P, Hensley M, et al. Patient online self-re-porting of toxicity symptoms during chemo-therapy. J Clin Oncol 2005;23(15):3552-3561.

[11] Taylor MJ, Stables R, Matata B, Lisboa PJ, Laws A, Almond P. Website design: Tech-nical, social and medical issues for self-re-porting by elderly patients. Health Infor-matics J 2013 [Epub ahead of print].

[12] Couper MP. Review: Web Surveys: A Review of Issues and Approaches. The Public Opin-ion Quarterly 2000;64(4):464-494.

[13] Sleijpen, G. Centraal Bureau voor de Statis-tiek (Central Bureau of Statistics). Een derde van de 75-plussers gebruikt internet [One third of the people over 75 years uses internet] [online]. http://www.cbs.nl/nl- NL/menu/themas/vrije-tijd-cultuur/pub- licaties/artikelen/ archief/2013/2013-383 4-wm.htm. Accessed 25 Feb 2014.

[14] Muylle S, Moenaert R, Despontin M. The conceptualization and empirical validation of web site user satisfaction. Inf Manage 2004;41(5):543-560.

[15] de Vries ST, Mol PG, de Zeeuw D, Haai-jer-Ruskamp FM, Denig P. Development and Initial Validation of a Patient-Reported Adverse Drug Event Questionnaire. Drug Saf 2013;36(9):765–77.

[16] Pitkow JE, Recker MM. Using the Web as a survey tool: results from the second WWW user survey. Comput Netw ISDN Syst. 1995;27(6):809-822.

[17] Bunz U. The Computer-Email-Web (CEW) Fluency Scale-Development and Validation. Int J Hum Comput Interact 2004;17(4):479-506.

[18] Giesen D, Meertens V, Vis-Visschers R, Beukenhorst D. Vragenlijstontwikkeling Centraal Bureau voor de Statistiek. [Ques-tionnaire development Central Bureau of Statistics] [online] http://www.cbs.nl/NR/rdonlyres/F8FB2360-C9A3-4379-8314-9C13C2938FCE/0/2010 x3705pub.pdf. Accessed 25 Feb 2014.

[19] Elling S, Lentz L, de Jong M. Website Eval-uation Questionnaire: Development of a Research-Based Tool for Evaluating In-formational Websites. Electronic Govern-ment Lecture Notes in Computer Science 2007;4656:293-304.

[20] Leung DYP, Kember D. Comparability of Data Gathered from Evaluation Question-naires on Paper and Through the Internet. Res High Educ 2005;46(5):571-591.

[21] Dillman DA, Bowker D. The web question-naire challenge to survey methodologists [online]. http://www.websm.org/uploadi/editor/Dillman.pdf. Accessed 25 Feb 2014.

Chapter 2

Construct and concurrent validity of a patient-reported

adverse drug event questionnaire:

a cross-sectional study

Sieta T. de Vries1

Flora M. Haaijer-Ruskamp1 Dick de Zeeuw1

Petra Denig1

Health and Quality of Life Outcomes 2014;12:103.


R1R2R3R4R5R6R7R8R9


Chapter 2

58

Abstract

BackgroundDirect patient-reported information about adverse drug events (ADEs) is important since it adds to healthcare professional-reported information about the safety of drugs. Previously, we developed an instrument to assess patient-reported ADEs in research settings. The aim of this study is to assess the construct and concurrent validity of the questionnaire.

MethodsPatients on at least an oral glucose-lowering drug completed the ADE questionnaire, the World Health Organization Quality of Life-BREF, and the Treatment Satisfaction Questionnaire for Medication (TSQM). The ADE questionnaire assesses ADEs for any drug that the patient uses. Construct validity was assessed by testing whether patients reporting an ADE had a lower general quality of life and physical health than those not reporting an ADE, using Mann–Whitney U tests and t-tests (significance level <0.05). For concurrent validity, we tested whether ADEs that patients associate with particular drugs in the ADE questionnaire are documented in the Summary of Product Characteristics (SPC) of those drugs, and whether patients who report an ADE with the use of metformin on the TSQM, mention metformin as a drug associated with an ADE on the ADE questionnaire. Agreement of 70% with the SPC was considered satisfactory. Sensitivity and positive predictive value (PPV) were calculated for the comparison with the TSQM, where 70% was used as the cut-off level for sufficient concurrent validity.

ResultsWe included 135 patients (mean age 64 years, 35% women). Patients who reported an ADE (N = 37) had a lower general quality of life and physical health than those not reporting an ADE (P < 0.05). For 78 of the 146 reported ADEs (53%), patients mentioned at least 1 particular drug associated with the ADE. After clustering related ADEs, this resulted in 56 patient-reported ADE-drug associations. Of these, 41 (73%) were in agreement with information in the SPC. Finally, the questionnaire had a sensitivity of 38% and PPV of 79% for assessing ADEs associated with metformin.

ConclusionsThe construct validity of the patient-reported ADE questionnaire was sufficient for reporting any versus no ADE, but the concurrent validity was only partly demonstrated. Therefore, the questionnaire needs to be adapted before it can be used.


Validity of a patient-reported adverse drug event questionnaire

59

2

IntroductionThe safety of a drug is monitored and assessed in clinical trials and observational studies [1,2]. Currently, the attribution of adverse events to a drug and the assessment of the severity of adverse drug events (ADEs) in research settings is primarily conducted by healthcare professionals [3]. It has, however, been shown that healthcare professionals downgrade the severity of ADEs experienced by patients [4]. Additionally, it has been shown that healthcare professionals underestimate symptomatic, subjective ADEs [4-7]. In a literature review, it was for instance shown that ADE rates of constipation with the use of the glucose-lowering drug metformin ranged from 0.6-1.0% when reported by healthcare professionals, and was 21% when reported by patients [6]. Therefore, regulatory authorities acknowledge the added value of patient-reported outcome instruments [8,9] in which the patient is the direct source of information [8,10]. This acknowledgement is especially the case for many symptomatic ADEs for which there is no objective test. Assessment of such ADEs is important, since they influence a patient’s quality of life (QOL) [7]. Previous studies showed that an increase in total scores of the number, frequency, and severity of experienced ADEs is associated with a decrease in QOL [11,12]. In addition, patients who report an ADE have a lower general health perception than patients who do not report an ADE [13].

Although some patient-reported instruments to assess ADEs exist (e.g. [14-16]), a generic instrument not limited to a specific ADE, or drug, and including questions about the nature (e.g. frequency, severity) and causality is not available. Therefore, we previously developed such an instrument (Appendix 1) [17]. This patient-reported ADE questionnaire is generic, checklist-based, includes questions about the nature and causality of the ADE, and is intended for research purposes in clinical trials and observational studies. The content validity of the instrument has been established and was adequate (Chapter 1) [17]. Further validation is needed, in particular given reported concerns about the validity of patient-reported ADEs (e.g. incorrect attributions of symptoms to drugs) [5].

The aim of the current study is to assess the construct and concurrent validity of the patient-reported ADE questionnaire. For the construct validity, the association between patient-reported ADEs and QOL is tested. With respect to the concurrent validity, the focus is on 1) concurrence between reported ADE-drug associations and known ADEs of those drugs, and on 2) agreement between ADE-drug reporting in the generic ADE questionnaire and a treatment/drug-specific questionnaire with a differently phrased question. The results of this study will help to establish whether or not the patient-reported ADE questionnaire is sufficiently valid to investigate experienced ADEs in clinical trials and observational studies, and how to further improve the questionnaire. More information about patient-reported ADEs will increase the knowledge about the

R1R2R3R4R5R6R7R8R9


Chapter 2

60

safety of drugs, which in the end can be used in the decision to approve or disapprove the drug to the market and in clinical practice.

MethodsThis study had a cross-sectional design and is part of a larger study about the development and validation of the patient-reported ADE questionnaire. For pragmatic reasons, we focused on patients with type 2 diabetes, who may be expected to use a variety of drugs with associated ADEs. Patients prescribed oral glucose-lowering drugs were included in this study since that is a valid proxy for type 2 diabetes. Previously, we have reported on the development process of the ADE questionnaire and assessment of its content validity and reliability (Chapter 1) [17]. We now present the follow-up study, assessing the construct and concurrent validity of this questionnaire. For this part, additional data were collected during the second measurement of the previous study, assessing the patients’ QOL and experiences with metformin, a drug which was expected to be used by most of the included patients (see below). The study was carried out in accordance with the Code of Ethics of the World Medication Association (Declaration of Helsinki) for experiments involving humans. The Medical Ethics Committee of the University Medical Center Groningen (METc UMCG) in the Netherlands determined that ethical approval was not needed for this study (reference number M12.112446).

Participants, procedure and data collectionPatients who were aged 18 years or older and had been dispensed at least an oral glucose-lowering drug were recruited in 2012 via pharmacists in the northern part of the Netherlands, including two pharmacies in a village and two in a town. The pharmacists selected patients who fulfilled the inclusion criteria from their database and sent an information letter with consent form to these patients. One of the researchers (STdV) contacted pharmacists until the number of consenting patients was around 150. Those patients who had an e-mail address, were able to access the internet, and gave informed consent completed the ADE questionnaire twice with a one week period in between. Patient characteristics reported in the current study were based on patient-reported information collected at the second measurement. The pharmacists provided prescription data covering a 1 year period for all patients. They did this at one point in time, after most patients had completed the second measurement. Since chronic medication is prescribed for 3 months in the Netherlands, the most recent prescriptions between 3 months before and up to the date of questionnaire completion were used to describe the medication prescribed to the patients.



61

2

Measures

Patient-reported ADEs, ADE-drug associations, and causality scoreThe patient-reported ADE questionnaire is a generic questionnaire and contains a checklist with 252 symptoms, categorized in 16 body categories, which patients can check as being a potential ADE for any of the drugs they use (Appendix 1) [17]. Fourteen additional questions about its nature are asked for each potential ADE, including questions about causality and causal drugs (e.g. “Of which drug or drugs do you think this side effect is the result?”). Branching was used for these additional questions, which means that they were only presented if a symptom was checked as being a potential ADE. Each included symptom in the checklist is linked to a Lowest Level Term of the Medical Dictionary for Regulatory Activities (MedDRA®) terminology version 13.0. For the current study, a web-based version of the questionnaire was used which was constructed using the Unipark Enterprise Feedback Suite 8.0 version 1.1 [18]. Patients could mention one or more than one drug to be associated with one ADE in the questionnaire, for example, increased stool frequency caused by metformin and tolbutamide. Also, patients could mention the same drug or drugs for multiple ADEs, for example decreased weight and abdominal discomfort caused by liraglutide. Furthermore, patients could check multiple, related ADEs describing one overall ADE, for example, diarrhea and fecal incontinence, which they associated with the same drug or drugs. These related ADEs were clustered by two researchers (STdV and PD) independently [17]. This resulted in four possible ADE-drug associations: single ADE-single drug, single ADE-multiple drugs, clustered ADE-single drug, clustered ADE-multiple drugs.

Quality of lifeQOL was assessed using the Dutch version of the World Health Organization Quality of Life-BREF (WHOQOL-BREF) questionnaire [19]. The WHOQOL-BREF measures a patient’s QOL and contains 26 items, including one item assessing general health, and one item assessing general well-being. These two general items were summed resulting in a general QOL score. The other items comprise four domains of QOL, namely physical health, psychological health, social relationships, and environment. An association between the ADE questionnaire and the general QOL score and physical health domain were used as indicating construct validity of the ADE questionnaire. The WHOQOL-BREF is a reliable and valid questionnaire to measure QOL [20]. In our study, the internal consistency ranged from 0.651 (domain social relationships) to 0.836 (domain psychological health) and was 0.806 for the physical health domain.

R1R2R3R4R5R6R7R8R9


Chapter 2

62

Known ADEsAlthough there is no gold standard for which ADEs are ‘true’ ADEs, the ADEs that are documented in the Summary of Product Characteristics (SPC) can be considered as known ADEs of a drug. Two researchers (STdV and PD) independently determined whether all the patient-reported ADE-drug associations (of both the single and clustered ADEs) could be confirmed with the information in the SPC as indicating concurrent validity. Dissimilarities between the two researchers were resolved by discussion. The SPCs of the reported drugs were retrieved from the website of the Dutch Medicines Evaluation Board [21].

The comparison of ADE-drug associations with the information in the SPC was additionally assessed by taking into account a patient-reported causality assessment. We developed a patient-reported causality scoring system using the additional questions per potential ADE in the questionnaire. This scoring system was based on the items of the Naranjo causality classification [22], and patient-reported aspects of causality assessment [23], and includes the following items of the patient-reported ADE questionnaire: the question why the patient links the symptom to drug use with the option to check more than one of the answer categories assessing causality, the question which drug the patient associates with the ADE, how certain the patient is about this ADE-drug association, and the yes-no question assessing whether or not the patient can think of other reasons for experiencing the ADE. Scores could range from −2 to +10 (Appendix 3; supplemental table 1).

Treatment/drug-specific ADEsThe Treatment Satisfaction Questionnaire for Medication (TSQM) [24] was applied for the use of metformin. Metformin is the first-choice initial glucose-lowering drug in the treatment of patients with diabetes [25,26], and likely to be used by most of the patients included in our study. Although the TSQM aims to assess treatment satisfaction, the questionnaire contains four items about side effects. One of these items asks for whether or not any ADE was experienced as a result of taking the drug in question (“Do you experience any side effects of the use of metformin?”). The concurrent validity of the ADE questionnaire was examined by using the answers given to this question as the gold standard and comparing them with the reporting of metformin as a particular drug associated with an ADE in the ADE questionnaire. Of note, no recall period is specified for this question in the TSQM. Patients who reported that they had used metformin in the previous 4 weeks had to complete the TSQM.



63

2

AnalysesThe QOL of patients who reported one or more ADEs and those who did not report an ADE in the patient-reported ADE questionnaire was expressed with mean and median scores and differences were tested using Mann–Whitney U tests (for non-normally distributed variables, observed for general QOL in the study population, where kurtosis was >1) and t-tests (physical health, psychological health, social relationships, and environment domains). Significance levels were used to indicate sufficient validity, and effect-size of the difference in QOL was calculated to indicate clinical relevance. We calculated effect-size as the (mean QOL of those reporting ≥1 ADEs – mean QOL of those not reporting an ADE) / standard deviation of the QOL scores of those not reporting an ADE. Moderate (0.5 – 0.79) or large (>0.8) effect-sizes were interpreted as clinically relevant [27].

Differences in patient characteristics between those who mentioned at least one particular drug for any of their ADEs and those who did not mention a particular drug for any of their ADEs were tested. In addition, differences in ADE characteristics between the ADEs for which patients mentioned at least one particular drug and those ADEs for which patients did not mention a particular drug were tested. T-tests, Fisher-Freeman-Halton tests, and Pearson χ2-tests were used for these analyses.

The percentage of agreement was used in the comparison between the patient-reported ADE-drug association and the information in the SPC of the drug. The percentage of agreement was calculated at the ADE level. This calculation means that in case multiple drugs were reported for one ADE, the confirmation that the ADE was acknowledged in the SPC of at least one of the drugs was considered as agreement in the analyses. We did not expect full agreement between the patient-reported ADE-drug associations and the information in the SPC, since it has been demonstrated that some ADEs are lacking in the SPC [28,29] and that patient reports may provide additional ADEs to those noted in the SPC [5,30]. Therefore, we used a cut-off level of at least 70% agreement between the patient-reported ADE-drug associations and known ADEs as indicating sufficient validity. A sensitivity analysis was performed in which the patient-reported causality score was taken into account. In this analysis, only the cases indicating a single ADE-single drug association could be included, since the causality score was measured at this level. We assessed the agreement for all such cases, and for the cases with a causality score higher than or equal to 1) the median of the causality scores, and 2) the third quartile.

For the comparison of the patient-reported ADE-metformin association in the ADE questionnaire and the TSQM, a positive outcome was defined as the detection of an ADE with the use of metformin. The number of true positives, true negatives, false positives, and false negatives are presented. These numbers were used to calculate the sensitivity

R1R2R3R4R5R6R7R8R9


Chapter 2

64

[31] and positive predictive value [32]. A cut-off level of 70% was used as indicating sufficient sensitivity and/or positive predictive value. Most of the patients on a drug are expected not to experience an ADE which will result in a high number of true negatives. Therefore, specificity and negative predictive value were not calculated.

The analyses were conducted using IBM SPSS Statistics version 20 (Armonk, New York, USA) and P-values of <0.05 were considered statistically significant. Exact confidence intervals based on binomial probabilities were calculated for the validity measures [33], using Stata version 12 (Stata Corp., College Station, TX).

ResultsThe number of patients included in this study was 135. The mean age of these patients was 64 years (SD: 9), and most of them were male (65%) (Table 2.1). Metformin was the most commonly prescribed oral glucose-lowering drug (94%) and the median number of total systemic drugs prescribed was 7 (interquartile range: 5–9). Thirty-seven patients (27%) reported at least one ADE, with a total of 146 ADEs (median: 3, range: 1–11). In 58% of the cases, the ADE had started more than 12 months ago. Most of the reported ADEs had not yet improved or disappeared (82%).

Construct validityPatients who reported one or more ADEs on the questionnaire had a lower self-reported QOL than patients who did not report an ADE (Table 2.2). This difference was statistically significant for physical health (P < 0.01), and the general QOL score (P < 0.05), and turned out to be clinically relevant (d = 0.53 for the general QOL score and d = 0.50 for physical health). Although not statistically significant, patients who reported more than one ADE (N = 26) had a lower general QOL score and physical health than those who reported one ADE (N = 11) (median of 6.5 versus 8.0 on the general QOL score, P = 0.032; mean of 13.6 versus 14.9 on physical health, P = 0.138).



65

2

Table 2.1. Patient characteristics and nature of the reported ADEsCharacteristic N (%)Patients 135Women 47 (35)Mean age in years (SD) 64 (9)Education Lowa 34 (25) Middleb 53 (39) Highc 39 (29) Other 9 (7)Number of prescribed oral glucose-lowering drugsd

1 oral glucose-lowering drug 91 (68) 2 oral glucose-lowering drugs 38 (29) 3 oral glucose-lowering drugs 4 (3)Classes of prescribed oral glucose-lowering drugsd

Biguanides 125 (94) Dipeptidyl peptidase 4 (DPP-4) inhibitors 4 (3) Glucagon-like peptide-1 (GLP-1) 5 (4) Sulfonamides 42 (32) Thiazolidinediones 3 (2)Median number of systemic drugs prescribed (IQR) 7 (5–9)Patients additionally on insulin 24 (18)Patients reporting an ADE 37 (27)Number of ADEs reported (range) 146 (1–11)First time experiencing the ADE Today 6 (4) Yesterday 1 (1) 2-7 days ago 5 (3) Between 1 week and 1 month ago 14 (10) Between 1 and 6 months ago 14 (10) Between 6 and 12 months ago 21 (14) More than 12 months ago 85 (58)ADE gone away or improved Not yet 120 (82) Clearly improved 14 (10) ADE was treated and has improved 4 (3) ADE went away by itself 0 (0) ADE went away after quitting medication 1 (1) ADE went away after treatment 1 (1) Other 6 (4)How much bothersome Not at all 11 (8) Only a bit 22 (15) Somewhat 71 (49) Quite a lot 31 (21) Very much 11 (8)Influence daily functioning None 47 (32) Only a bit 24 (17) Somewhat 54 (37) Quite a lot 18 (12) Very much 2 (1)ADEs = Adverse drug events; SD = Standard deviation; IQR = Interquartile range a No education; elementary school; junior secondary vocational education b Junior general secondary education; senior secondary vocational education c Senior general secondary education; higher professional education; university education d Medication use is based on 133 patients since drug information of 2 patients was not available

R1R2R3R4R5R6R7R8R9


Chapter 2

66

Table 2.2. QOL and differences between patients who report and those who do not report an ADETotal Patients not

reporting an ADE

Patients reporting ≥1 ADEs

P-value Effect size d

Number of patients 135 98 37Mean QOL (SD) Median general QOL score (IQR) 8.0 (6.0-8.0) 8.0 (7.0-8.0) 7.0 (6.0-8.0) 0.021† −0.53 Physical health 14.9 (2.5) 15.2 (2.4) 14.0 (2.6) 0.009* −0.50 Psychological health 14.7 (2.4) 15.0 (2.4) 14.1 (2.5) 0.064* −0.38 Social relationships 14.5 (2.6) 14.6 (2.7) 14.4 (2.6) 0.666* −0.07 Environment 16.0 (2.2) 16.2 (2.2) 15.6 (2.3) 0.134* −0.27QOL = Quality of life; ADE = Adverse drug event; SD = Standard deviation; IQR = Interquartile range. † Mann–Whitney U test; * T-test.

Concurrent validity

Comparison of patient-reported ADE-drug association with known ADEs For 78 reported ADEs (53%), patients mentioned at least one particular drug for the ADE. These patients (N = 25) did not significantly differ in age, gender, and education level from the patients who did not mention a particular drug for any of their ADEs (N = 12) (Appendix 3; supplemental table 2). For the ADEs that were associated with one or more particular drugs (N = 78), it was more often reported that the ADE occurred (soon) after the start or dosage increase of a drug and that a healthcare professional confirmed the symptom being an ADE than for the ADEs not associated with a particular drug (N = 68) (Appendix 3; Supplemental table 3). ADEs that were already perceived before the start of a drug but increased in frequency after the start of a drug were less likely to be associated with a particular drug. For 15% of both ADEs, those that were associated with a particular drug and those that were not, the patients reported that the symptom was described in the patient leaflet as being an ADE for that particular drug.

For almost half of the ADEs that were associated with one or more particular drugs (38 cases; 49%), multiple related ADEs describing one overall ADE were observed resulting in 16 clustered ADE-drug associations (Figure 2.1). All of the single and clustered ADE-drug associations (N = 56) were compared with the known ADEs for those drugs as documented in the SPC (Appendix 3; supplemental table 4). In 41 of the cases (73%), the ADE-drug association was in agreement with known ADEs for at least one of the reported drugs, indicating sufficient concurrent validity.



67

2

78

38 related ADEs resulting in 16 clustered ADEs

9 clustered ADE-single drug

7 clustered ADE-multiple drugs

40 single ADEs

29 single ADE-single drug

11 single ADE- multiple drugs

68 potential ADEs with no particular drugs

reported (47%) 78 potential ADEs for which ≥1 particular

drugs were reported

146 symptoms checked as potential ADEs

37 patients reporting ≥1 ADEs

135 patients

ADE-drug association was in agreement with known ADEs for at least one of the reported drugs, indicating sufficient concurrent validity.

Figure 2.1. Flow-chart of reported adverse drug events (ADEs) and ADE-drug associations

There were 29 single ADE-single drug associations. Of these ADEs, 22 (76%) were in agreement with the known ADEs. The patient-reported causality score for these single ADE-single drug associations ranged from 1 to 3 with a median of 2 (interquartile range: 1.0 – 2.5). The single ADE-single drug associations with a causality assessment score ≥2 (N = 21) were in 76% of the cases (N = 16) in agreement with the known ADEs. This agreement increased to 100% when only causality scores of 3 (N = 7) were taken into account. Comparison of ADE-metformin association between ADE questionnaire and treatment/drug-specific questionnaire Of the 135 patients, 125 reported that they had used metformin in the previous 4 weeks.

Figure 2.1. Flow-chart of reported adverse drug events (ADEs) and ADE-drug associations

There were 29 single ADE-single drug associations. Of these ADEs, 22 (76%) were in agreement with the known ADEs. The patient-reported causality score for these single ADE-single drug associations ranged from 1 to 3 with a median of 2 (interquartile range: 1.0 – 2.5). The single ADE-single drug associations with a causality assessment score ≥2 (N = 21) were in 76% of the cases (N = 16) in agreement with the known ADEs. This agreement increased to 100% when only causality scores of 3 (N = 7) were taken into account. Comparison of ADE-metformin association between ADE questionnaire and treatment/drug-specific questionnaireOf the 135 patients, 125 reported that they had used metformin in the previous 4 weeks. The comparison of reporting an ADE with metformin use between the ADE questionnaire and the TSQM revealed 11 true positives, 93 true negatives, 3 false positives, and 18 false negatives. These numbers resulted in a sufficient, high positive predictive value (79%; 95% confidence interval 49-95%), but an insufficient, low sensitivity (38%; 95% confidence interval 21-58%).

R1R2R3R4R5R6R7R8R9


Chapter 2

68

DiscussionConstruct validity of the ADE questionnaire was demonstrated by an association between ADEs and QOL, where patients who reported an ADE on the questionnaire had a lower general QOL and physical health than patients who did not report an ADE. The 73% agreement between the ADE-drug associations as reported by the patients and the known ADEs for those drugs as documented in the SPC indicates concurrent validity. However, additional concurrent validity assessment revealed that the ADE questionnaire has sufficient positive predictive value but insufficient sensitivity when comparing the linkage of an ADE to metformin in the ADE questionnaire with the treatment/drug-specific TSQM questionnaire.

For the construct validity, patients who reported an ADE had statistically significant lower scores than those not reporting an ADE on physical health and on the general QOL score. The lowest QOL scores were observed for patients reporting more than 1 ADE. The effect sizes indicate that the QOL differences were clinically relevant. The underlying mechanism for the association between ADEs and QOL is not fully understood yet. Although, an increase in the experience of ADEs may be associated with a decrease in QOL [11,12], it has also been shown that baseline QOL scores were lower for patients who report an ADE at follow-up than for patients who do not report an ADE at follow-up [34]. This finding may imply that patients with a low QOL have a higher risk of experiencing ADEs [35], or that patients with negative health perceptions or certain personality traits are more disposed to experiencing and/or reporting ADEs [13,36]. The underlying mechanism for the association between ADEs and QOL does not influence its usefulness to assess construct validity, since the principle of construct validity is based on the existence of an adequate association between scores on instruments [37].

Agreement of at least one drug that patients mentioned for their ADE on the patient-reported ADE questionnaire with known ADEs as documented in the SPC supports its concurrent validity. Full agreement was shown when only cases with a high patient-reported causality score were included, indicating that a patient-reported causality assessment increases the validity of ADE reporting by patients. Previously, it was shown that patients cannot perceive all ADEs and also do not always make the connection between a symptom and their drug use [7]. In addition, we found that only 53% of the patients who reported an ADE mentioned a particular drug for the ADE. This low number may especially occur in a patient population, such as included in this study. Patients with diabetes using oral glucose-regulating drugs often use multiple drugs for various related and unrelated diseases, which complicates the assignment of a drug to the symptom [38]. We found that mentioning a particular drug was more likely when a healthcare professional confirmed the symptom being an ADE. This finding indicates that patient-reported ADE instruments cannot fully replace healthcare professional



69

2

reports. Special attention of healthcare professionals may be necessary for symptoms that are already present but increase in frequency after starting a drug, and for ADEs that do not occur soon after the start of a drug. Our study results support that for reliable knowledge about ADEs, information from both healthcare professionals and patients is needed [39].

For 15% of the ADEs, the patients reported that the ADE was described in the patient leaflet. Surprisingly, this was also observed for ADEs for which patients did not mention a particular drug. It may be that the ADE is reported in the patient leaflet of multiple drugs and the patient is unsure about the exact drug that may cause the ADE. Additional preferably qualitative research is needed to better understand these reports.

Comparing the ADE questionnaire with the TSQM only partly supported the concurrent validity. The positive predictive value indicates that an ADE with the use of metformin detected by the ADE questionnaire is likely to be an ADE. The low sensitivity, on the other hand, implies that not all experienced ADEs associated with metformin use are detected by the ADE questionnaire. There are, however, notable differences between the ADE questionnaire and the TSQM that may negatively influence the validity measures. First, the patient-reported ADE questionnaire includes a recall period of 4 weeks whereas the TSQM does not include a recall period. Second, in the patient-reported ADE questionnaire it is asked which drug is associated with the ADE, whereas in the TSQM it is asked whether or not an ADE is experienced with the use of the metformin.

More than half of the ADEs reported in this study started more than 12 months ago. This timing of the ADE may have influenced our validity assessment. For construct validity, the experience of an ADE for a long period of time may lead to a stronger association with QOL due to the longer period of burden but could also lead to a weaker association due to adaptation of getting used to the symptoms. The timing of the ADE seemed to influence whether or not a particular drug was mentioned for the ADE (Appendix 3; supplemental table 3). This finding indicates that particularly the more recent ADEs were included in the concurrent validity assessment, since only patients who report a specific drug were included in this assessment. It is not clear whether this could have impact on the associations with the SPC and the TSQM conducted for the concurrent validity.

An important limitation of this study is the low number of patients reporting an ADE. The construct validity of the ADE questionnaire was only assessed by testing the association between patient-reported ADEs and QOL. Future studies should assess the construct validity using additional constructs, also at specific ADE level and taking the severity of the ADE into account. The severity of an ADE has shown to influence a patient’s QOL [35]. Participating patients in this study were significantly younger than

R1R2R3R4R5R6R7R8R9


Chapter 2

70

the non-responders [17], and were more often prescribed metformin and less often a sulfonamide or thiazolidinedione than other, on average elder, patients with diabetes in the Dutch primary care [40]. These differences may be due to the use of a web-based version of the questionnaire, because it has been shown that respondents to web-based questionnaires are younger and have a shorter diabetes duration [41,42]. It should also be noted that the validity of the paper-based version of the questionnaire may differ from the web-based version due to differences between these methods in, for instance, branching of questions, visual aspects, or respondent required actions such as the use of a mouse in the web-based version [43-45]. Another limitation was that about half of the reported ADEs that were associated by the patients with particular drugs were assessed by the researchers as being related ADEs describing one overall ADE. In this study, these ADEs were clustered by the researchers. We recommend that future research with checklist-based patient-reported ADE questionnaires should include the option for patients to cluster related ADEs. This option will likely increase the validity of the instrument and the practicability for the researchers. We included patients using drugs for type 2 diabetes in this study, who particularly reported ADEs belonging to the gastrointestinal disorders SOC of the MedDRA® [17]. Although these patients also often used drugs for other diseases, the use of the questionnaire in other patient populations and in the general population in which other ADEs are also common requires additional validation [10]. The validity of the questionnaire is expected to differ especially for more serious ADEs (e.g. rhabdomyolysis with statin use) and for ADEs that can be distinguished more clearly from the symptoms related to the disease (e.g. alopecia with the use of chemotherapy in patients with cancer). Finally, it should be noted that this validation was conducted in an observational study, where most patients were current users and familiar with their drug treatment. Therefore, additional studies are needed assessing the validity of the instrument in a clinical trial, where patients are often new users and blinded to the drug treatment. However, the insufficient concurrent validity indicates that the first next step is to make adaptations to the questionnaire.

ConclusionsOur results indicate that the patient-reported ADE questionnaire has construct validity of reporting any ADE versus no ADE, since the reporting of an ADE was associated with a lower QOL. Additional assessment is needed to test the construct validity for individual ADEs. The concurrent validity was only partly demonstrated by 1) sufficient agreement between the ADE-drug associations and the known ADEs when the researchers clustered related ADEs into one ADE, 2) sufficient positive predictive value to detect ‘real’ ADEs associated with metformin use, but 3) insufficient sensitivity to detect all ADEs that are associated with metformin use. Therefore, adaptations to the ADE questionnaire, such



71

2

as allowing patients to cluster multiple related ADEs as one overall ADE, need to be made and tested in future validation studies.

AcknowledgementsMedDRA® is a registered trademark of the International Federation of Pharmaceutical Manufacturers and Associations (IFPMA). We would like to thank the WHO for providing the WHOQOL-BREF questionnaire and Quintiles for providing the TSQM.

R1R2R3R4R5R6R7R8R9


Chapter 2

72

References[1] Suvarna V. Phase IV of drug development.

Perspect Clin Res 2010;1(2):57–60.[2] Berlin JA, Glasser SC, Ellenberg SS. Ad-

verse event detection in drug develop-ment: recommendations and obliga-tions beyond phase 3. Am J Public Health 2008;98(8):1366–71.

[3] Walshe K. Adverse events in health care: issues in measurement. Qual Health Care 2000;9(1):47–52.

[4] Basch E, Iasonos A, McDonough T, Barz A, Culkin A, Kris MG, et al. Patient versus clinician symptom reporting using the na-tional cancer institute common terminolo-gy criteria for adverse events: results of a questionnaire-based study. Lancet Oncol 2006;7(11):903–9.

[5] Blenkinsopp A, Wilkie P, Wang M, Rout-ledge PA. Patient reporting of suspected ad-verse drug reactions: a review of published literature and international experience. Br J Clin Pharmacol 2007;63(2):148–56.

[6] Hakobyan L, Haaijer-Ruskamp FM, de Zeeuw D, Dobre D, Denig P. Comparing adverse event rates of oral blood glu-cose-lowering drugs reported by patients and healthcare providers: a post-hoc analysis of observational studies pub-lished between 1999 and 2011. Drug Saf 2011;34(12):1191-202.

[7] Gawert L, Hierse F, Zink A, Strangfeld A. How well do patient reports reflect adverse drug reactions reported by rheumatol-ogists? Agreement of physician- and pa-tient-reported adverse events in patients with rheumatoid arthritis observed in the German biologics register. Rheumatology (Oxford) 2011;50(1):152–60.

[8] US Department of Health and Human Ser-vices, FDA, Center for Drug Evaluation and Research Center for Biologics Evalua-tion and Research Center for Devices and Radiological Health. Guidance for indus-try patient-reported outcome measures: use in medical product development to support labeling claims. http://www.fda.gov/ downloads/Drugs/GuidanceCompli-anceRegulatory Information/Guidances/UCM193282.pdf.

[9] European Medicines Agency. Fourth report on the progress of the interaction with patients’ and consumers’ organizations

(2010) and results/analysis of the degree of satisfaction of patients and consumers involved in EMA activities during 2010. EMA/632696/2011. http://www.ema.eu-ropa.eu/docs/en_GB/document_library/ Report/ 2011/10/WC500116866.pdf.

[10] Frost MH, Reeve BB, Liepa AM, Stauffer JW, Hays RD, Mayo/FDA Patient-Reported Out-comes Consensus Meeting Group. What is sufficient evidence for the reliability and validity of patient-reported outcome meas-ures? Value Health 2007;10(Suppl 2):94–105.

[11] Azevedo AM, Alonso NB, Vidal-Dourado M, Noffs MH, Pascalicchio TF, Caboclo LO, et al. Validity and reliability of the Portu-guese-Brazilian version of the quality of life in epilepsy inventory-89. Epilepsy Behav 2009;14(3):465–71.

[12] Carreno M, Donaire A, Falip M, Maestro I, Fernandez S, Nagel AG, et al. Validation of the Spanish version of the Liverpool ad-verse events profile in patients with epilep-sy. Epilepsy Behav 2009;15(2):154–9.

[13] De Smedt RH, Denig P, Haaijer-Ruskamp FM, Jaarsma T. Perceived medication ad-verse effects and coping strategies report-ed by chronic heart failure patients. Int J Clin Pract 2009;63(2):233–42.

[14] Foster JM, van Sonderen E, Lee AJ, San-derman R, Dijkstra A, Postma DS, et al. A self-rating scale for patient-perceived side effects of inhaled corticosteroids. Respir Res 2006;7:131.

[15] Jarernsiripornkul N, Krska J, Capps PA, Richards RM, Lee A. Patient reporting of potential adverse drug reactions: a meth-odological study. Br J Clin Pharmacol 2002;53(3):318–25.

[16] Corso DM, Pucino F, DeLeo JM, Calis KA, Gal-lelli JF. Development of a questionnaire for detecting potential adverse drug reactions. Ann Pharmacother 1992;26(7-8):890–6.

[17] De Vries ST, Mol PG, de Zeeuw D, Haai-jer-Ruskamp FM, Denig P. Development and initial validation of a patient-reported adverse drug event questionnaire. Drug Saf 2013;36:765–77.

[18] Unipark Enterprise Feedback Suite 8.0 ver-sion 1.1. http://www.unipark.de.

[19] De Vries J, Van Heck GL. WHOQOL-Bref. Til-burg: Universiteit van Tilburg; 1996.



73

2

[20] Skevington SM, Lotfy M, O’Connell KA, WHOQOL Group. The world health organ-ization’s WHOQOL-BREF quality of life assessment: psychometric properties and results of the international field trial. A report from the WHOQOL group. Qual Life Res 2004;13(2):299–310.

[21] Dutch Medicines Evaluation Board. http://www.cbg-meb.nl/cbg/nl.

[22] Naranjo CA, Busto U, Sellers EM, San-dor P, Ruiz I, Roberts EA, et al. A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther 1981;30(2):239–45.

[23] Krska J, Anderson C, Murphy E, Avery AJ. How patient reporters identify adverse drug reactions: a qualitative study of re-porting via the UK yellow card scheme. Drug Saf 2011;34(5):429–36.

[24] Atkinson MJ, Sinha A, Hass SL, Colman SS, Kumar RN, Brod M, et al. Validation of a general measure of treatment satisfaction, the treatment satisfaction questionnaire for medication (TSQM), using a national panel study of chronic disease. Health Qual Life Outcomes 2004;2:12.

[25] American Diabetes Association: Standards of medical care in diabetes–2013. Diabetes Care 2013;36(Suppl 1):11–66.

[26] Rutten GEHM, De Grauw WJC, Nijpels G, Houweling ST, Van de Laar FA, Bilo HJ, et al. The NHG guideline diabetes mellitus type 2. Huisarts Wet 2013;56:512–25.

[27] Wyrwich KW, Bullinger M, Aaronson N, Hays RD, Patrick DL, Symonds T, Clinical Significance Consensus Meeting Group. Estimating clinically significant differenc-es in quality of life outcomes. Qual Life Res 2005;14(2):285–95.

[28] Aursnes I, Gjertsen MK. Common adverse events associated with an SSRI: meta-anal-ysis of early paroxetine data. Pharmacoepi-demiol Drug Saf 2008;17(7):707–13.

[29] Bergk V, Haefeli WE, Gasse C, Brenner H, Martin-Facklam M. Information deficits in the summary of product characteristics preclude an optimal management of drug interactions: a comparison with evidence from the literature. Eur J Clin Pharmacol 2005;61(5-6):327–35.

[30] Aagaard L, Hansen EH. Consumers’ reports of suspected adverse drug reactions volun-teered to a consumer magazine. Br J Clin Pharmacol 2010;69(3):317–8.

[31] Altman DG, Bland JM. Diagnostic tests: 1: sensitivity and specificity. BMJ 1994; 308(6943):1552.

[32] Altman DG, Bland JM. Diagnostic tests 2: pre-dictive values. BMJ 1994;309(6947):102.

[33] Deeks JJ, Altman DG. Sensitivity and speci-ficity and their confidence intervals cannot exceed 100%. BMJ 1999;318(7177):193–4.

[34] Ernst EJ, Ernst ME, Hoehns JD, Bergus GR. Women’s quality of life is decreased by acute cystitis and antibiotic adverse effects associated with treatment. Health Qual Life Outcomes 2005;3:45.

[35] Guo N, Marra F, Fitzgerald JM, Elwood RK, Marra CA. Impact of adverse drug reaction and predictivity of quality of life status in tuberculosis. Eur Respir J 2010;36(1):206–8.

[36] Imayama I, Plotnikoff RC, Courneya KS, Johnson JA. Determinants of quality of life in type 2 diabetes population: the inclusion of personality. Qual Life Res 2011;20(4):551–8.

[37] De Vet H, Terwee C, Mokkink L, Knol D. Measurement in Medicine: A practical guide. Cambridge: University Press; 2011.

[38] Chaipichit N, Krska J, Pratipanawatr T, Uchaipichat V, Jarernsiripornkul N. A qualitative study to explore how patients identify and assess symptoms as adverse drug reactions. Eur J Clin Pharmacol 2014;70(5):607–15.

[39] Basch E, Jia X, Heller G, Barz A, Sit L, Frus-cione M, et al. Adverse symptom event re-porting by patients vs clinicians: relation-ships with clinical outcomes. J Natl Cancer Inst 2009;101(23):1624–32.

[40] Voorham J, Haaijer-Ruskamp FM, Wolffen-buttel BH, Stolk RP, Denig P, Groningen Initiative to Analyze Type 2 Diabetes Treat-ment Group. Medication adherence affects treatment modifications in patients with type 2 diabetes. Clin Ther 2011;33(1):121–34.

[41] Härmark L, Alberts S, van Puijenbroek E, Denig P, van Grootheest K. Representative-ness of diabetes patients participating in a web-based adverse drug reaction monitor-ing system. Pharmacoepidemiol Drug Saf 2013;22(3):250–5.

[42] Schiotz M, Bøgelund M, Willaing I. Chal-lenges using online surveys in a Danish population of people with type 2 diabetes. Chronic Illn 2012;8(1):56–63.

R1R2R3R4R5R6R7R8R9


Chapter 2

74

[43] Wyatt JC. When to use web-based surveys. J Am Med Inform Assoc 2000;7(4):426–9.

[44] Manfreda KL, Batagelj Z, Vehovar V. Design of Web Survey Questionnaires: Three Basic

Experiments. J Computer-Mediated Com-mun 2002;7.

[45] Couper MP. Web surveys: a review of is-sues and approaches. Public Opin Q 2000, 64(4):464–494.

Chapter 3

The validity of a patient-reported adverse drug event

questionnaire using different recall periods

Sieta T. de Vries1 Flora M. Haaijer-Ruskamp1

Dick de Zeeuw1 Petra Denig1

Quality of Life Research 2014;23(9):2439-45.


R1R2R3R4R5R6R7R8R9


Chapter 3

76

Abstract

ObjectiveTo assess the validity of a patient-reported adverse drug events (ADEs) questionnaire with a 3-month or 4-week recall period.

Methods Patients receiving at least one oral glucose-lowering drug were asked to report potential ADEs they experienced related to any drug in a daily diary for a 3-month period. Thereafter, they completed the ADE questionnaire with either a 3-month or 4-week recall period. The validity was assessed by comparing ADEs reported in each version with those reported in the diary at class level and at specific ADE level. At class level, a comparison was made using 1) primary system organ classes (SOCs) of the Medical Dictionary for Regulatory Activities and 2) other related SOCs. Sensitivity and positive predictive value (PPV) were calculated.

Results Each version of the questionnaire was completed by 39 patients. In the 3-month group, 21 patients reported 70 ADEs in the diary. In the 4-week group, six patients reported seven ADEs in the last 4 weeks of the diary. Sensitivity to assess ADEs at primary SOC was low for both recall groups (33%). PPV was 51 and 10% for, respectively, the 3-month and 4-week group. Taking other related SOCs into account slightly increased the sensitivity for the 3-month group (38%). Sensitivity of reporting the same ADE was 41 and 43% for, respectively, the 3-month and 4-week group.

Conclusions Regardless of the recall period and level of comparison, the validity for assessing ADEs was low with the patient-reported ADE questionnaire. Further refinement is needed to improve the validity.


Validity of a patient-reported adverse drug event questionnaire using different recall periods

77

3

IntroductionIn clinical trials, a daily diary is often used to record adverse events experienced by patients [1,2]. The diary method brings the reporting closer to the occurrence of adverse events than retrospective questionnaires [3] and can be seen as a gold standard in the assessment of symptoms due to the quality and richness of the collected information [4,5]. Daily diaries have also been used in observational studies assessing adverse events [6]. However, keeping a daily diary is burdensome. The use of a retrospective questionnaire with a longer recall period might be an alternative. The recall period in a questionnaire is the time period for which the patient has to consider the answer of the question [7].

A patient’s recall is influenced by factors such as forgetting an event or its correct date, its accessibility in mind (influenced by for instance its recency, frequency, and salience), and one’s mood [5,8–10]. In addition, a patient’s evaluation of a specific event or health state may change over time due to a response shift, which is defined as the use of a different reference category [11,12]. Therefore, the recall period in a questionnaire is seen as a limitation of patient reporting [13]. An inappropriate recall period may introduce measurement error [14]. The food and drug administration recommends to pay attention to this issue when constructing patient-reported outcome (PRO) instruments [15].

An optimal recall period depends on various issues, and debate is ongoing about what recall period is suitable for which questionnaire [14,16]. Some studies have assessed the impact of different recall periods on reporting symptoms in questionnaires (e.g., [17–19]). The general conclusion is that there is an inverse association between the length of recall period and accuracy of recall used in questionnaires [14].

Currently, information about the optimal recall period to assess adverse drug events (ADEs) is lacking. Commonly used recall periods in PRO instruments are between 1 day up to 4 weeks [14]. However, from our study about the content validation of a patient-reported ADE questionnaire, it became clear that several patients found a recall period of 4 weeks relatively short (Chapter 1) [20]. A longer recall period can be preferred since it is not always immediately clear for patients whether or not a symptom is an ADE, and ADEs that occur irregularly or after some time may not be captured in a 4-week period. In the current study, we examined the validity of a retrospective questionnaire using a 4-week or a 3-month recall period for the assessment of ADEs. The primary objective was to assess the validity of reported ADEs at aggregated class level and at individual ADE level. The secondary objective was to explore whether the validity of the questionnaire might be dependent on either the class of ADE, or characteristics at patient level.

R1R2R3R4R5R6R7R8R9


Chapter 3

78

MethodsThe study had a longitudinal design, where patients first completed a daily diary for a period of 3 months followed by a previously developed retrospective questionnaire [20]. The reporting of ADEs in the questionnaire was compared with a daily diary, which was used as the gold standard. Although the patient-reported questionnaire can be used to assess adverse drug reactions (ADRs), we use the term ADE instead of ADR for two reasons. First, patients may be uncertain about a causal relation between a symptom and a drug [20]. Second, patients may perceive unintended responses due to medication errors or overdoses. This implies that our questionnaire is not restricted to assess ADRs as defined by the World Health Organization as ‘‘a noxious and unintended response to a medicine that occurs at normal therapeutic doses used in humans for prophylaxis, diagnosis, or therapy of disease, or for the modification of physiologic function’’ [21]. The study was carried out in accordance with the Code of Ethics of the World Medication Association (Declaration of Helsinki) for experiments involving humans. The Medical Ethics Committee of the University Medical Center Groningen in the Netherlands determined that ethical approval was not needed for this study.

ParticipantsInclusion criteria for the patients were age 18 years or older, being dispensed an oral glucose-lowering drug, availability of an e-mail address, access to the internet, and ability to read and write Dutch language. These patients were recruited in 2012 and 2013 via pharmacies in the northern part of the Netherlands. In around 30 pharmacies, a randomly selected sample of 15 patients aged 18 years or older, being dispensed an oral glucose-lowering drug, were contacted by telephone and sent an information letter when they fulfilled the inclusion criteria and were interested in participation. In another 4 pharmacies, an information letter was sent to all patients aged ≥18 years and being dispensed an oral glucose-lowering drug (three pharmacies) or glucose-lowering drug (one pharmacy). The patients who returned a completed consent form and who fulfilled the inclusion criteria were included in the study. After completing the study, patients were compensated with a voucher of €10 for participation.

Reported ADEsAs primary outcome, we compared reported ADEs at the primary System Organ Class level of the Medical Dictionary for Regulatory Activities (MedDRA®) terminology version 13.0. Each symptom included in the patient-reported questionnaire was assigned to a Lowest Level Term of the MedDRA® [20], which is linked to at least one System Organ Class. This is the level at which all adverse reactions should be tabulated according to the European guideline of Summary of Product Characteristics [22]. Although each



79

3

symptom has at least a primary System Organ Class, the symptom may also be linked to a secondary and even tertiary System Organ Class. Therefore, we additionally assessed the validity by including secondary and tertiary System Organ Class levels of the MedDRA®, if applicable, which takes into account possible misclassifications. As a third step, the validity was assessed for reporting the same ADE at the lowest level.

Material and procedureA paper-based diary was sent by mail to the participants, to be filled in daily for a period of 3 months. The diary was developed for this study and consisted of an open-ended question asking for symptoms experienced. An additional closed-ended question asked whether or not the patient attributed the symptom(s) to any drug they used, not restricted to their oral glucose-lowering drug. Telephone reminders were given to patients who did not return their diary within a month after it should have been completed.

After returning the completed diary, the patient received an e-mail message with the URL (uniform resource locator) to open the web-based version of the patient-reported ADE questionnaire, which was constructed using the Unipark Enterprise Feedback Suite 8.0 version 1.1 (http://www.unipark.de). The e-mail message included a personal login code to prevent multiple completions of the questionnaire by a patient [23]. The patient-reported ADE questionnaire is a generic questionnaire which includes general questions about patient characteristics and drug use, and a list with symptoms in lay-terms which can be checked by the patient as a symptom unrelated to any drug or as a potential ADE (Chapter 1) [20]. Additional questions about the nature of the ADE and the drugs a patient relates to the ADE are asked for each potential ADE. Two versions of the patient-reported ADE questionnaire were used, one with a recall period of 3 months (e.g., ‘‘Which symptoms involving your ‘eyes and/or eyelids’ did you experience during the past 3 months’’) and one with a recall period of 4 weeks (e.g., ‘‘Which symptoms involving your ‘eyes and/or eyelids’ did you experience during the past 4 weeks’’). Patients were randomized using blocked randomization [24] to one of the two groups that differed in the recall period of the questionnaire. We aimed to include 100 patients (50 per group), which has been suggested as a reasonable number for reliability studies [25]. Although the current study does not assess the reliability, we used this number as a reference since no data about different recall periods in assessing ADEs were available for calculating the required sample size.

R1R2R3R4R5R6R7R8R9


Chapter 3

80

AnalysesDifferences in patient characteristics between those who completed the questionnaire with a recall period of 4 weeks and those who completed the questionnaire with a recall period of 3 months were compared using the Pearson χ²-test, Fisher–Freeman–Halton test, and t-test, depending on the type of variable.

For the comparison of the questionnaire with the diary, the ADEs were used as unit of analysis and the number of true-positive, true-negative, false-positive, and false-negative ADEs are presented where relevant. The validity of reporting ADEs at primary System Organ Class level of the MedDRA® was assessed by calculating the sensitivity [26] and positive predictive value [27]. Specificity and negative predictive value were not calculated since these values are expected to be high and non-informative due to the high number of true negatives. A positive outcome was defined as the detection of an ADE at this primary class level. Exact confidence intervals (CI) based on binomial probabilities were calculated for these validity measures [28]. The questionnaire with a recall period of 3 months was compared with the full 3-month diary, whereas the questionnaire with a recall period of 4 weeks was compared with the last 4 weeks reported in the diary. In addition, the questionnaire with a recall period of 4 weeks was compared with the full 3-month period in the diary to assess the validity in ADE reporting within this wide time frame to allow for incorrect recall of the date of occurrence. Sensitivity analyses were performed to assess whether delayed completion of the diary or the questionnaire affected the results by excluding 1) those patients with >14 days between the last date reported in the diary and receiving the completed diary by the researchers (delayed diary completers), 2) those patients who completed the questionnaire >14 days after the diary was received by the researchers (delayed questionnaire completers), and 3) both the delayed diary and delayed questionnaire completers. Differences between the two recall groups in days of delay were compared using Mann–Whitney U tests.

The sensitivity of both versions of the questionnaire was additionally calculated at 1) MedDRA® additional class level for taking not only the primary but also the secondary or tertiary System Organ Classes of the MedDRA® into account if applicable and 2) specific ADE level for reporting the same ADEs among the questionnaire and the diary. Two researchers (PD and STdV) independently classified the reported ADEs in the diary to a System Organ Class of the MedDRA® and checked whether or not the specific ADEs reported in the diary were the same ADEs as reported in the retrospective questionnaire. Discrepancies in the judgments of the researchers were resolved by discussion. All participants were included in the analyses comparing the questionnaire with the diary.

To explore whether the validity of the questionnaire was dependent on the class of ADE, or on characteristics at patient level, the reports of both recall groups were combined. The sensitivity per primary System Organ Class of the MedDRA® was



81

3

assessed for those classes in which at least five ADEs were reported in the diary and/or the questionnaire. The age, gender, and education level of the patients were compared between those patients with no agreement (no corresponding ADEs), partial agreement (some but not all corresponding ADEs), and full agreement (all corresponding ADEs) between the ADEs reported in the diary and the questionnaire.

The analyses were conducted using IBM SPSS Statistics version 20 (Armonk, New York, USA), and P-values <0.05 were considered statistically significant. Confidence intervals were calculated using Stata version 12 (Stata Corp., College Station, TX).

ResultsOf the 113 patients who returned an informed consent form, 78 patients (69%) completed the study. These patients did not significantly differ in age, gender, and education level from the patients who did not complete the study (data not shown). No differences between the completers of the 2 recall groups were found in age and education level, but more males were included in the 3-month recall group than in the 4-week recall group (P < 0.05; Table 3.1). In total, 27 of the 78 participants reported 77 individual ADEs in the diary. Of these ADEs, 61 were linked to a System Organ Class of the MedDRA® (multiple ADEs reported by one participant within the same System Organ Class were counted as one).

Table 3.1. Patient characteristics per recall group4 weeks(N = 39)

3 months(N = 39)

P-value

Mean age (SD) 63 (10.0) 67 (7.2) 0.069•

Females (%) 20 (51.3) 11 (28.2) 0.037*

Education (%) 0.594‡

Lower educationa 8 (20.5) 8 (20.5) Middle educationb 17 (43.6) 21 (53.8) Higher educationc 12 (30.8) 7 (17.9) Other 2 (5.1) 3 (7.7)a No education; elementary school; junior secondary vocational educationb Junior general secondary education; senior secondary vocational educationc Senior general secondary education; higher professional education; university education• T-test; * Pearson χ²-test; ‡ Fisher-Freeman-Halton test

R1R2R3R4R5R6R7R8R9


Chapter 3

82

Validity of reporting ADEs at primary class levelThe sensitivity and positive predictive value were low for both recall periods (Table 3.2). Sensitivity analyses by excluding delayed diary and/or questionnaire completers revealed similar validity levels (Appendix 4; supplemental table 1). The comparison of the 4-week recall questionnaire with the full 3-month diary revealed a similar sensitivity (32%; 95% confidence interval 14–55%) and a slightly increased positive predictive value (from 10%; 95% confidence interval 1–30%, to 33%; 95% confidence interval 15–57%).

Table 3.2. Validity of the retrospective questionnaire with a recall period of 4 weeks or 3 months compared with the daily diary in reporting adverse drug events at MedDRA® primary class level (N=702)*

TP FP TN FN Se (95% CI) PPV (95% CI)4-week recall; last 4 weeks of diary 2 19 677 4 33% (4-78) 10% (1-30)3-month recall; full 3-month diary 18 17 630 37 33% (21-47) 51% (34-69)* The N at MedDRA® level is the number of patients per recall group (39) times 18 different MedDRA® System Organ Classes which are covered by the ADEs in the questionnaire. TP = True positive; FP = False positive; TN = True negative; FN = False negative; Se = Sensitivity; PPV = Positive Predictive Value; CI = Confidence interval.

Validity of reporting ADEs at additional class levelThe sensitivity of the 4-week recall group remained the same when taking also secondary and tertiary System Organ Classes of the MedDRA® into account. For the 3-month recall group, a slightly increased sensitivity was shown (from 33%; 95% confidence interval 21–47%, to 38%; 95% confidence interval 25–52%).

Specific ADE levelIn the 3-month recall group, 21 patients (54%) reported in total 70 ADEs in the diary. The sensitivity of the questionnaire in reporting the same ADE was 41% (95% confidence interval 30–54%; number of true positives 29; number of false negatives 41). In the 4-week recall group, 6 patients (15%) reported in total 7 ADEs in the last 4 weeks of the diary, and the sensitivity of the questionnaire was 43% (95% confidence interval 10–82%; number of true positives 3; number of false negatives 4).

Differences per class of ADE and in characteristics at patient level Sensitivity levels ranged from 0 to 50% per System Organ Class of the MedDRA®, but confidence intervals were overlapping (Table 3.3). Of the 27 patients who reported one or more ADEs in the diary, 6 (22%) patients had full agreement by reporting all of these ADEs also in the questionnaire, 11 (41%) had partial agreement, and 10 (37%) had no agreement. Patients with no agreement were somewhat younger than patients with full



83

3

or partial agreement [mean age in years 64 (SD: 6) vs. 66 (SD: 10) and 67 (SD: 8)], and more often female (60 vs. 33 and 36%). The education level of the patients appeared to be similar among the three groups.

Table 3.3. Validity of the questionnaire* per MedDRA® System Organ Class level for those classes in which ≥5 adverse drug events are reported in the diary or the questionnaire.MedDRA® System Organ Class level TP FP TN FN Se (95% CI**)Gastrointestinal disorders 7 2 62 7 50% (23-77)General disorders and administration site conditions 2 3 69 4 33% (4-78)Metabolism and nutrition disorders 0 4 72 2 0% (0-84)Musculoskeletal and connective tissue disorders 3 3 66 6 33% (7-70)Nervous system disorders 4 6 62 6 40% (12-74)Psychiatric disorders 0 2 73 3 0% (0-71) Respiratory, thoracic and mediastinal disorders 0 5 71 2 0% (0-84)Skin and subcutaneous tissue disorders 3 3 66 6 33% (7-70)* Results of the questionnaire with a 4-week and 3-month recall period are combined** one-sided, 97.5% confidence interval when TP equals 0 TP = True positive; FP = False positive; TN = True negative; FN = False negative; Se = SensitivityPPV = Positive Predictive Value; CI = Confidence interval

DiscussionRegardless of the recall period, the patient-reported ADE questionnaire had a low sensitivity to identify patients who experienced an ADE at organ class level and at specific ADE level. In addition, the questionnaire had low positive predictive value. There may be differences among classes of ADEs but additional studies are needed to confirm this finding. In addition, further studies are needed to assess whether characteristics at patient level, such as age and gender, influence the validity of patient-reported ADE questionnaires.

The positive predictive value of the questionnaire was especially low for the 4-week recall period. Patients in this recall group more often reported an ADE at MedDRA® level in the questionnaire than in the diary. This higher reporting could be due to reporting additional ADEs in the questionnaire, or to forward telescoping, that is, ADEs were reported as being more recent than they actually occurred [5,29,30]. Forward telescoping probably occurred at least five times in the 4-week recall group, since five additional true positives were found when the full 3-month diary was taken into account. However, it should be noted that patients may have been primed to the 3-month period when answering the questionnaire because of our study design, since they completed the diary for this time period.

There are several factors that influence the validity of the questionnaire when comparing it with a diary. First of all, patients complete a diary with the knowledge they

R1R2R3R4R5R6R7R8R9


Chapter 3

84

have at that moment, whereas the questionnaire is completed with the knowledge they gained over time about their symptoms. This additional knowledge may change their opinion about, for instance, a symptom being an ADE. Second, an open-ended question was used in the diary, whereas a symptom checklist was used in the questionnaire. Previously, it was shown that more patients report an ADE and that the number of reported ADEs is higher on a checklist than on an open-ended question [31]. Using the open-ended question in the diary as the gold standard, the use of a closed instead of open-ended question in the questionnaire may lead to higher false-positive rates. Furthermore, some patients appeared to have delayed the completion of the diary or the questionnaire, which can be expected to result in lower validity. The sensitivity analyses, however, showed also low validity levels when such patients were excluded.

Previously, we found low test–retest reliability of assessing ADEs at specific level using the patient-reported ADE questionnaire, which may be due to problems in the questionnaire as well as a patient’s uncertainty about a symptom being an ADE (Chapter 1) [20]. This uncertainty was also demonstrated in the current study, in which patients related a specific drug to the ADE in less than half of the cases (data not shown). In addition, reported symptoms were sometimes indicated as an ADE and sometimes as ‘I do not know’ in the diaries. The uncertainty may particularly occur in patients with multiple comorbidity and comedication, which is common in the patient population included in this study [32]. Therefore, the low validity observed in our study may in part be due to the complexity of acknowledging ADEs in this specific patient population. The performance of the questionnaire might be better in patients who only have one disease or use one drug. Qualitative studies are needed to assess to what extent patients in general, and patients with type 2 diabetes more specifically, are able to report all (possible) ADEs, and to gain more knowledge about discrepancies in reported ADEs between a diary and a questionnaire.

We observed slightly lower sensitivities at organ class levels as compared to specific ADE level. This suggests that the direct linkage of symptoms in the checklist to MedDRA® terms may be inadequate. We observed that the System Organ Class of the checked symptom may differ from the System Organ Class that would be linked to the additional information given by the patients about the symptom (e.g., the System Organ Class of a checked symptom ‘‘tingling or prickling sensation’’ differs from the System Organ Class of the additional description provided by the patient being ‘‘muscle pain’’).

Only small differences were found in the validity of reporting specific ADEs between a questionnaire with a recall period of 4 weeks and 3 months. This finding of similar validity among the recall periods differs from a recent study in which higher accuracy of reporting headache frequency was found when a recall period of 30 days was used compared to 90 days [17]. This inconsistency indicates that conclusions about recall



85

3

periods cannot easily be transferred from one questionnaire to another, as has been stated before [14,16].

Strengths and limitationsTo our knowledge, this is the first study assessing the validity of different recall periods for assessing ADEs in a patient-reported questionnaire. Some limitations need to be acknowledged. The first and major limitation is the small sample size included in this study in combination with the low number of patients reporting an ADE, especially in the 4-week recall group. This limitation resulted in wide confidence intervals. A post hoc analyses showed that given the current data, a sample size of 43 for the 3-month recall group and 378 for the 4-week recall group would be necessary to achieve an accuracy of 5% for the observed sensitivity of 33% at primary System Organ Class level of the MedDRA® [33]. This finding indicates that for a more precise indication of the validity of a recall period of 4 weeks in an ADE questionnaire, a (preselected) sample of patients with a higher expected ADE rate would be preferable. Secondly, we included a selective sample of patients that responded to the letters sent via pharmacists. These were patients consenting to keep a diary for 3 months, using an oral glucose-lowering drug, and with internet access. A previous study with a web-based version of the patient-reported ADE questionnaire showed that the responders were younger than the non-responders (Chapter 1) [20]. Thirdly, more males were included in the 3-month recall group than in the 4-week recall group, indicating that the randomization was not completely successful. Fourthly, there are some limitations with the use of a daily diary as a gold standard in the reporting of ADEs. It has been noted that daily diaries also require recall and may be influenced by the same factors that apply to retrospective questionnaires [3]. In addition, we are not sure whether patients completed the diary each day with the risk of loss of validity [17]. On the other hand, keeping a daily diary before completing a questionnaire may positively affect the recall in the questionnaire. We expect these factors to be similar for both recall groups. Furthermore, patients may become tired of keeping a diary, which can lead to less validity in the last period and therefore lower validity in the 4-week recall group. Although the number of patients reporting an ADE was relatively stable over time (data not shown), we cannot exclude this possibility.

Practice implicationsThe patient-reported ADE questionnaire is a generic questionnaire which is intended to measure all ADEs experienced by patients. However, the questionnaire is not sufficiently sensitive to detect all experienced ADEs. In addition, the questionnaire has low positive predictive value. Therefore, adaptations to the patient-reported ADE questionnaire

R1R2R3R4R5R6R7R8R9


Chapter 3

86

are needed before it can be generally used. The direct linkage of checked symptoms to MedDRA® terms may introduce misclassifications. In addition, patients check multiple symptoms describing one ADE, as has been shown previously [20]. Therefore, starting with an open-ended question in which patients give a description of their ADEs which is then linked to a MedDRA® term may be preferred. In additional research, the validity of such an adaptation should be tested. Further research is also needed to gain more insight into whether there are differences in accuracy among classes of ADEs. For observational studies assessing ADEs in patients using chronic medication, a recall period of 3 months may be preferable compared to a 4-week recall period. A 3-month recall period has the advantage of covering a longer time period facilitating the identification of more ADEs. Our study suggests that the validity of reporting specific ADEs is hardly affected using a longer recall period, but further validation may be needed when the questionnaire is adapted. Shorter recall periods, however, may be needed for clinical trials and studies that try to assess ADEs experienced at different stages of treatment [14,17].

ConclusionsThis study showed that a retrospective patient-reported ADE questionnaire is insufficiently valid for assessing ADEs, regardless of the recall period and the level of comparison. The use of a 3-month recall period may be preferred over a 4-week recall period since it covers a longer time period. However, further refinement of the questionnaire is needed to improve its validity.

Acknowledgements MedDRA® is a registered trademark of the International Federation of Pharmaceutical Manufacturers and Associations (IFPMA).



87

3

References[1] Stone AA, Shiffman S, Schwartz JE, Broder-

ick JE, Hufford MR. Patient compliance with paper and electronic diaries. Control Clin Trials 2003;24(2):182–99.

[2] Shiffman S. Electronic diaries: Impact on drug development. (2004). http:// www. google.nl/url?sa=t&rct=j&q=&esrc=s&- frm=1&source=web&cd=2&ved=0CDo QFjAB& url=http%3A %2F%2Fwww.fda. gov%2Fohrms%2Fdockets%2Fac%2F04 %2Fslides%2F2004-4078OPH1_01_Shiff- man.ppt&ei=doRfUqz2CsbZswam54HoDg &usg=AFQjCNG2FAPyq7Q1qPEzbt-4gy-BCsO2CEA. Accessed March 28, 2014.

[3] Gendreau M, Hufford MR, Stone AA. Meas-uring clinical pain in chronic widespread pain: Selected methodological issues. Best Pract Res Clin Rheumatol 2003;17(4):575–92.

[4] Monk TH, Buysse DJ, Kennedy KS, Pods JM, DeGrazia JM, Miewald JM. Measuring sleep habits without using a diary: The sleep tim-ing questionnaire. Sleep 2003;26(2):208–12.

[5] Verbrugge LM. Health diaries. Med Care 1980;18(1):73–95.

[6] Klocke KR, Stauch K, Landen H. Effect of add-on acarbose to insulin therapy in rou-tine clinical practice. Clin Drug Investig 2003;23(10):621–7.

[7] DeWalt DA, Rothrock N, Yount S, Stone AA, PROMIS Cooperative Group. Evaluation of item candidates: The PROMIS qualitative item review. Med Care 2007;45(5 Suppl 1):S12–21.

[8] Hufford MR, Shiffman S. Methodological is-sues affecting the value of patient-reported outcomes data. Expert Rev Pharmacoecon Outcomes Res 2002;2(2):119–28.

[9] Crosby RD, Kolotkin RL, Williams GR. De-fining clinically meaningful change in health-related quality of life. J Clin Epide-miol 2003;56(5): 395–407.

[10] Bohner G, Wänke M. Attitudes and attitude change. (2002). East Sussex: Psychology Press Ltd.

[11] Breetvelt IS, Van Dam FS. Underreport-ing by cancer patients: The case of re-sponse-shift. Soc Sci Med 1991;32(9):981–7.

[12] Litwin MS, McGuigan KA. Accuracy of recall in health-related quality-of-life assessment among men treated for prostate cancer. J Clin Oncol 1999;17(9):2882–8.

[13] King A, Daniels J, Lim J, Cochrane DD, Tay-lor A, Ansermino JM. Time to listen: A re-view of methods to solicit patient reports of adverse events. Qual Saf Health Care 2010;19(2):148–57.

[14] Stull DE, Leidy NK, Parasuraman B, Chas-sany O. Optimal recall periods for pa-tient-reported outcomes: Challenges and potential solutions. Cur Med Res Opin 2009;25(4):929–42.

[15] U.S. Department of Health and Human Services, FDA, Center for Drug Evaluation and Research Center for Biologics Evalua-tion and Research Center for Devices and Radiological Health. Guidance for indus-try patient-reported outcome measures: Use in Medical Product Development to Support Labeling Claims. http://www.fda.gov/downloads/Drugs /GuidanceCompli-anceRegulatory Information/Guidances/UCM193282.pdf. Accessed March 28, 2014.

[16] Norquist JM, Girman C, Fehnel S, De-Muro-Mercon C, Santanello N. Choice of recall period for patient-reported outcome (PRO) measures: Criteria for consideration. Qual Life Res 2012;21(6):1013–20.

[17] Heyer GL, Perkins SQ, Rose SC, Aylward SC, Lee, JM. Comparing patient and parent recall of 90-day and 30-day migraine disa-bility using elements of the PedMIDAS and an Internet headache diary. Cephalalgia 2014;34(4):298-306.

[18] Broderick JE, Schwartz JE, Vikingstad G, Pribbernow M, Grossman S, Stone AA. The accuracy of pain and fatigue items across different reporting periods. Pain 2008;139(1):146–57.

[19] Broderick JE, Schneider S, Schwartz JE, Stone AA. Interference with activities due to pain and fatigue: Accuracy of ratings across different reporting periods. Qual Life Res 2010;19(8):1163–70.

[20] de Vries ST, Mol PG, de Zeeuw D, Haai-jer-Ruskamp FM, Denig P. Development and initial validation of a patient-reported adverse drug event questionnaire. Drug Saf 2013;36(9):765–77.

R1R2R3R4R5R6R7R8R9


Chapter 3

88

[21] World Health Organization Drug and ther-apeutics committee. Session 4. Assessing and managing medicine safety [online]. http://www.who.int/medicines/technical_briefing/tbs/04-Drug-Safety_final-08.ppt. Accessed March 28, 2014.

[22] European commission. Enterprise and industry directorate-general. A guideline on summary of product characteristics (SmPC) September 2009. http://ec.europa.eu/health/files/eudralex/vol-2/c/smpc_guideline_rev2_en.pdf. Accessed February 3, 2014.

[23] Schleyer TK, Forrest JL. Methods for the design and administration of web-based surveys. J Am Med Inform Assoc 2000;7(4):416–25.

[24] Randomizer recall. Randomly assignment of subjects to groups were performed us-ing the Web site: GraphPad, QuickCalcs, Random number calculators. http://www.graphpad.com/quickcalcs/.Accessed Mar-ch 28, 2014.

[25] De Vet H, Terwee C, Mokkink L, Knol D. Measurement in medicine: A practical guide. (2011). Cambridge: Cambridge Uni-versity Press.

[26] Altman DG, Bland JM. Diagnostic tests. 1: Sensitivity and specificity. BMJ 1994;308(6943):1552.

[27] Altman DG, Bland JM. Diagnostic tests 2: Pre-dictive values. BMJ 1994;309(6947):102.

[28] Deeks JJ, Altman DG. Sensitivity and speci-ficity and their confidence intervals cannot exceed 100%. BMJ 1999;318(7177):193–4.

[29] Prohaska V, Brown NR, Belli RF. Forward telescoping: The question matters. Memory 1998;6(4):455–65.

[30] Sale H, Hedman L, Isberg A. Accuracy of pa-tients’ recall of temporomandibular joint pain and dysfunction after experiencing whiplash trauma: A prospective study. J Am Dent Assoc 2010;141(7):879–86.

[31] Bent S, Padula A, Avins AL. Brief commu-nication: Better ways to question patients about adverse medical events: a rand-omized, controlled trial. Ann Intern Med 2006;144(4):257–61.

[32] Suh DC, Choi IS, Plauschinat C, Kwon J, Bar-on M. Impact of comorbid conditions and race/ethnicity on glycemic control among the US population with type 2 diabetes, 1988–1994 to 1999–2004. J Diabetes Com-plications 2010;24(6):382–91.

[33] Jones SR, Carley S, Harrison M. An introduc-tion to power and sample size estimation. Emerg Med J 2003;20(5):453–8.

Supplement II

Illustrations and possible solutions of

problems in the use of patient reports

R1R2R3R4R5R6R7R8R9


Supplement II

90

IntroductionDirect patient reporting is a commonly used method to assess health-related quality of life or a patient’s health status, including the experience of adverse drug events (ADEs) [1]. Direct patient reporting is useful since it adds to the knowledge about the quality of care [2,3].

Recently, initiatives have been taken to use direct patient reporting as part of performance measurement in clinical practice [1,4]. However, several types of biases have been acknowledged in patient reporting which should be taken into account when developing and interpreting these measures. A literature review revealed 48 types of biases [5], such as ambiguous questions, social desirability, and primacy and recency effects [5,6]. In addition, patients can misread or misinterpret questions [7].

Previously, we developed a patient-reported ADE questionnaire and tested its reliability and validity [8,9]. The content validity of the developed questionnaire was assessed in a study with a qualitative design (Chapter 1). Further validation and reliability testing was performed in a cross-sectional study in which patients using oral glucose-lowering drugs completed the questionnaire twice, with a one week period in between (Chapter 1 and Chapter 2). When conducting these validation studies, we included several other questions partly from validated questionnaires, such as the Dutch version of the RAND/Short Form-36 [10,11] and the Dutch version of the Treatment Satisfaction Questionnaire for Medication (TSQM) [12]. The TSQM was applied for the use of metformin. During the validation studies, we encountered some of the known biases of patient reporting and less obvious problems partly in validated questionnaires. In this short communication, the encountered problems are presented and possible solutions are given.

Encountered problems

InconsistencyAn inconsistency was found in the reporting of drug use. Patients were asked about the drugs they had used in the previous 4 weeks. In addition, we asked them whether or not they had used the drug metformin nearing the end of the questionnaire. Two patients who reported that they did not use metformin on the latter question, did report the drug on the former.

Answering not applicable questions or not answering applicable questions It is common in questionnaires that patients are guided to skip some questions depending on the answers given. In the TSQM, patients who report to have experienced an ADE have to complete three additional questions about the influence of the ADE


Illustrations and possible solutions of problems in the use of patient reports

91

II

on physical health, mental function, and their mood or emotions. Of the patients who reported to have experienced an ADE with the use of metformin, 17%, 45%, and 41% reported that the additional question of respectively physical health, mental function, and mood or emotions, was not applicable for them. On the other hand, of the 97 patients who reported that they did not experience an ADE of metformin, 65 (68%) did complete the additional questions of physical health and mental function and 64 patients (67%) completed the mood or emotions item.

Different interpretationAn item that was interpreted differently than intended was the item ‘I’m as healthy as anybody I know’ of the RAND/Short Form-36. A patient reported ‘mostly false’ because in his/her opinion he/she was more healthy then the people he/she knows. However, the scoring of this question indicates that a mostly false answer would indicate lower health.

Potential biases and possible solutionsAt least five biases may be related to the encountered problems, that is unconsciously checking wrong answers (making mistakes), confusion about what to answer, reading/interpretation problems, ambiguous questions, and problems with the reporting of drug names. Multiple changes may be necessary to solve the biases. Examples of possible changes are given in Table II.1. The illustrated problems underline the need to critically evaluate patient-reported questionnaires, as is the case for every instrument. This critical evaluation is an ongoing process that needs to be continued after the validity of the instrument has been demonstrated. In addition, the use of an existing instrument in a different patient population or in a different language always requires pilot testing using cognitive debriefing methods.

R1R2R3R4R5R6R7R8R9


Supplement II

92

Table II.1. Potential biases for the encountered problems and possible solutionsPotential bias Encountered problem Possible change for solutionUnconsciously checking wrong answer

- Inconsistent reporting of metformin

- Answering not applicable questions or not answering applicable questions in TSQM

- Consistency checks in digital version of a questionnaire

- Formatting issues such as sufficient space between answer options, sufficient font size of letters

Confusion about what to answer when questions do not apply

Answering not applicable questions or not answering applicable questions in TSQM

Digital questionnaire in which patients are directly guided to questions that apply to them

Reading/interpretation problems of question

Answering not applicable questions or not answering applicable questions in TSQM

Not completely solvable, but chance will be reduced by pilot testing of questionnaire in other population using cognitive debriefing

Ambiguous question Different interpretation in RAND/Short Form-36

Not completely solvable, but chance will be reduced by pilot testing of questionnaire in other population using cognitive debriefing

Problems with drug name Inconsistent reporting of metformin

Open-ended question to report drugs. For additional questions about a specific drug, patients should be guided based on answers on open-ended question


Illustrations and possible solutions of problems in the use of patient reports

93

II

References[1] Black N, Jenkinson C. How can patients’

views of their care enhance quality im-provement? BMJ 2009;339(7714):202-5.

[2] Department of Health: Guidance on the routine collection of Patient-Reported Out-come Measures (PROMs). London: Depart-ment of Health; 2008.


[4] Basch E, Torda P, Adams K. Standards for patient-reported outcome-based perfor-mance measures. JAMA 2013;310(2):139-40.

[5] Choi BC, Pak AW. A catalog of biases in questionnaires. Prev Chronic Dis 2005; 2(1):A13.

[6] Bowling A. Mode of questionnaire admin-istration can have serious effects on data quality. J Public Health 2005;27(3):281-91.

[7] Bohner G, Wänke M. (2002). Attitudes and attitude change. East Sussex: Psychology Press Ltd.

[8] de Vries ST, Mol PG, de Zeeuw D, Haai-jer-Ruskamp FM, Denig P. Development and

initial validation of a patient-reported ad-verse drug event questionnaire. Drug Saf 2013;36(9):765-77.

[9] de Vries ST, Haaijer-Ruskamp FM, de Zeeuw D, Denig P. Construct and concurrent validi-ty of a patient-reported adverse drug event questionnaire: a cross-sectional study. Health Qual Life Outcomes 2014;12(1):103.

[10] Van der zee K, Sanderman R. (1993). Het meten van de algemene gezondheidstoe-stand met de RAND-36. Een handleiding. Noordelijk Centrum voor Gezondheids-vraagstukken: Rijksuniversiteit Groningen.

[11] Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30(6):473-83.

[12] Atkinson MJ, Kumar R., Cappelleri JC, Hass SL. Hierarchical construct validity of the treatment satisfaction questionnaire for medication (TSQM version II) among outpatient pharmacy consumers. Value Health 2005;8(Suppl 1):S9-24. Those seek-ing information regarding or permission to use the TSQM are directed to Quintiles, Inc. at www.quintiles.com/TSQM or [email protected].

Intermezzo

The assessment and management of adverse drug events

by patients and healthcare professionals

in clinical practice: a case-report

R1R2R3R4R5R6R7R8R9



Assessment and management of adverse drug events in clinical practice

97

IntroductionPrevious studies showed the added value of incorporating the patient’s perspective in the evaluation of the safety of drugs [1-4]. It is important to incorporate both the patient’s and the healthcare professional’s perspective since they are complementary in providing clinically useful information [3,5]. Case-reports about adverse drug events (ADEs) usually focus on the healthcare professional’s perspective. In this case-report, the assessment and management of ADEs in clinical practice is assessed from a patient’s perspective accompanied with the available clinical information. The case-report demonstrates the assessment and management of ADEs in clinical practice and supports the importance of incorporating the perspective of both, the patient and the healthcare professional.

Case-reportA woman, born in 1966, diagnosed with type 2 diabetes in 2003, experienced three drug-related problems over a period of four years. In each situation, the ADE, the drug causing the ADE, and the patient’s and healthcare professional’s perspective and management of the ADE were different.

In the first situation, a known ADE occurred shortly after the start of diclofenac (Figure A). The patient experienced symptoms and the healthcare professional recognized them as an allergic reaction to the use of diclofenac. The healthcare professional stopped the prescription of diclofenac upon which the symptoms disappeared. The patient was satisfied with the healthcare professional recognizing the symptoms as an allergic reaction and stopping the medication.

In the second situation, symptoms such as sweating and in particular loss of consciousness occurred after a switch from Coversyl plus (perindopril 5 mg, indapamide 1,25 mg) to perindopril/indapamide (perindopril 4 mg, indapamide 1,25 mg) (Figure B). The patient asked whether or not the symptoms could be due to this switch. The healthcare professionals questioned the association between the symptoms and the drug, and performed investigations to assess the cause of the symptoms. After several months of investigations, only the switch to the former drug reduced the symptoms. The management of this situation increased the burden of the patient and her family, and resulted in a decreased quality of life and treatment satisfaction.

In the third situation, symptoms (such as increased pain) which were not known as an ADE were experienced after a switch from prednison to prednisolone (Figure C). The patient asked whether the symptoms could be caused by prednisolon. The healthcare professional was not sure but switched back to prescribing prednison. This helped to reduce the symptoms and resulted in satisfaction of the patient since the healthcare professional had taken her seriously.

R1R2R3R4R5R6R7R8R9


Intermezzo

98

ConclusionsThis case-report is probably not unique but describes a patient’s perspective on the assessment and management of ADEs in clinical practice. It illustrates that ADE assessment can be complex and that there are different ways to deal with the perspective of the patient. It also indicates that over time, perceptions of both the patient and the healthcare professional may be influenced by past experiences. Incorporating the perspective of both the patient and the healthcare professional may improve the assessment and management of ADEs in clinical practice and may increase patient satisfaction.



99

112

Epico

ndyli

tis la

terali

s. Pr

escri

ption

of

diclof

enac

Na p

ch 50

mg

MSR 3

D1T.

Viole

nt al

lergic

reac

tion d

iclofe

nac:

eryth

ema e

xsud

ativu

m, un

well,

dizzy

, wr

etche

d. ST

OP di

clofen

ac. P

resc

riptio

n of

triam

cinolo

n crè

me 2D

; pre

dniso

lon

20mg

1D1T

; clem

astin

e 1mg

1D1T

.

I had

an ap

point

ment

with

my G

P be

caus

e I ha

d a ve

ry th

ick ar

m wh

ich

looke

d like

the s

kin of

a fis

h fille

d with

sc

ales a

nd th

ick flu

id bli

sters.

I felt

sick

. Th

e GP t

old m

e tha

t it w

as an

aller

gic

reac

tion.

Neve

r dicl

ofena

c aga

in. I w

as

pres

cribe

d pre

dniso

n, a c

reme

, and

pa

race

tamol.

I had

an ap

point

ment

wi

th GP

beca

use o

f an

inflam

matio

n in m

y elb

ow.

Two d

ays l

ater, m

y arm

wa

s nor

mal a

gain

and I

did

n’t fe

lt sick

anym

ore.

Da

y 1

Day 2

Da

y 3

Day 4

Da

y 5

Day 6

Da

y 7

Day 8

Da

y 9

Day 1

0 Da

y 11

Figu

re A

. Situ

ation

1.

=

Healt

hcar

e pro

fessio

nal, c

linica

l data

= Pa

tient

GP im

media

tely

said

this

was a

n all

ergic

reac

tion.

R1R2R3R4R5R6R7R8R9


Intermezzo

100

ECG:

no is

chem

ia. M

aybe

hype

rven

tilati

on an

d he

artb

urn.

Extra

omep

razo

le re

comm

ende

d. No

w a l

ittle

bette

r. Stil

l che

st pr

essu

re an

d stu

ffy. A

lso st

arted

coug

hing.

Burn

ing pa

in in

esop

hagu

s. Ting

ling f

eelin

g in l

eft ha

nd. P

ain

radia

tes to

left a

rm an

d leg

. Te

mp 37

.5, RR

175/

85, p

ulse 7

2ra,

O2 sa

t 99%

, Co

r/pu

lm -,

Trop

onin

T <0.0

30ug

/l.

Sync

ope a

t the

prac

tice.

Hosp

ital a

dmiss

ion. D

rugs

: me

tform

in 3d

d 100

0mg,

omep

razo

le 1d

d 40m

g, No

vora

pid 16

-16-

32 un

its, L

antu

s 44 u

nits, T

hyra

x 1dd

12

5mcg

, per

indop

ril 1d

d 5mg

, simv

astat

in 1d

d 20m

g, pa

race

tamol

4dd 1

g. EC

G: si

nus r

hyth

m, H

b 8,

Sedim

entat

ion 27

, leuk

ocyte

s 11.8

, lymp

hocy

tes 4.

0, th

romb

ocut

es 35

5, D-

dimer

<0.15

, arti

rial b

loodg

as ph

7.4

3, pC

O2 5.

0, bic

arbo

nate

24, B

exce

ss 0.8

, PO2

9.0,

O2

satu

ratio

n 95%

, gluc

ose 7

.1, so

dium

141,

potas

sium

4.2,

creati

ne 68

, calci

um 2.

35, a

lbumi

n 40,

norm

al liv

eren

zyme

s, TSH

1.8,

24-h

urine

on ca

techo

lamine

s no

rmal.

CT: N

o ind

icatio

n of p

ulmon

ary e

mboli

sm, n

o pn

eumo

nal d

eviat

ion. N

o ind

icatio

n of b

asal

path

ology

. CT

brain

: no d

eviat

ion. 2

4-h b

lood p

ressu

re no

clea

r nigh

tly

dip. M

ean S

BP: 1

57, D

BP: 8

2. Me

an H

R 96/

min.

Durin

g nig

ht SB

P: 14

7, DB

P: 79

, HR:

98. N

o arrh

ythmi

a. EC

G du

ring c

ollap

s: no

indic

ation

of ar

rhyth

mia o

r con

ducti

on

disor

der. E

EG: n

o dev

iation

s. Ind

icatio

n of b

elle

indiff

eren

ce. C

onsu

lt psy

chiat

rist: a

t this

mom

ent, n

o int

erve

ntion

need

ed. P

atien

t thin

ks sy

mptom

s are

due t

o oth

er Co

versy

l dru

g.

Figu

re B

. Situ

ation

2.

Pulm

onolo

gist: n

o clea

r dev

iation

s. So

mewh

at un

clear

para

trach

eal r

e. Sp

irome

tric f

lowvo

lume c

urve

no

rmal.

No c

hang

e on F

EV1 l

evel.

No

rmal

brea

th so

unds

. Pur

e hea

rt ton

es. H

ighly

susp

ect a

nd ty

picall

y for

hy

perv

entil

ation

with

out u

nder

lying

pa

tholo

gy.

I fell b

ackw

ards

whe

n I st

ood u

p fro

m th

e cou

ch. I

was

weak

and d

idn’t r

eact

on an

ything

. Amb

ulanc

e cam

e an

d the

y lift

ed m

e up.

I ope

ned m

y eye

s and

was

ap

proa

chab

le bu

t had

viole

nt he

adac

he. M

y hus

band

tol

d the

m ho

w we

ak an

d sick

I was

and t

hat m

y bloo

d pr

essu

re w

as to

o high

in th

e las

t wee

ks. I

was n

ot ad

mitte

d to t

he ho

spita

l. I as

ked t

he pr

ofessi

onals

wh

ether

or no

t it co

uld be

due t

o per

indop

ril. T

hey

didn’t

think

so.

I was

very

stuff

y and

had c

hest

pain.

I wa

s swe

ating

and a

t the

same

time

shak

ing fr

om th

e cold

and p

ain in

left

uppe

r arm

. The

ambu

lance

came

. The

y th

ough

t it w

as hy

perv

entil

ation

.

No tw

itchin

g, ur

ine lo

ss or

tong

ue bi

te.

Befor

ehan

d hea

dach

e. Gl

uc 5.

5, sa

t 99%

, pols

80

/min,

rr 18

0/10

0, no

rmal

pupil

lary

reac

tion,

norm

al fol

low m

ovem

ents

of th

e eye

s on

requ

est. D

oes n

ot re

act o

n pain

. Dec

reas

ed

cons

cious

ness

eci, p

roba

bly co

nver

sion.

Conc

lusion

ECG v

ertic

al ax

is, no

furth

er

abno

rmali

ties.

Cove

rsyl p

lus

ARG 5

/1,25

mg

1D1T

pe

rind/

indap

4/

1,25m

g

I ord

ered

Cove

rsyl-p

lus at

th

e pha

rmac

y. I r

eceiv

ed

perin

dopr

il. Th

ey to

ld me

th

at th

is wa

s che

aper

.

With

diffi

culty

I arri

ved i

n th

e hos

pital

for th

e ap

point

ment

with

the

inter

nist. T

hey c

ouldn

’t wa

ke m

e up.

I was

wea

k. Th

ey de

cided

to ad

mit m

e to

the i

nten

sive c

are.

The

next

day,

I cou

ld go

home

be

caus

e we h

ad to

wait

for

the t

est r

esult

s.

In th

e wait

ing ro

om of

my

GP, I

lost m

y co

nscio

usne

ss. An

ap

point

ment

was

mad

e for

an EE

G and

to vi

sit

the n

euro

logist

.

I lost

my co

nscio

usne

ss.

My hu

sban

d cou

ld br

ing

me to

the h

ospit

al wh

ere

I was

admi

tted.

Furth

er

tests

were

done

.

The 3

th da

y in t

he ho

spita

l: I fe

lt a lit

tle st

rong

er. N

urse

s wer

e sur

prise

d, ho

w co

uld th

is ha

ppen

? I as

ked w

hich b

rand

of bl

ood p

ressu

re-lo

werin

g me

dicine

I use

d her

e: Co

versy

l plus

. Late

r, I lo

st my

cons

cious

ness

again

, bu

t this

time i

t was

muc

h sho

rter. P

revio

usly,

I ask

ed al

l the s

pecia

lists

wheth

er th

is co

uld be

relat

ed to

perin

dopr

il. Th

ey al

l said

that

wasn

’t the

ca

se. B

ut fo

r me i

t was

clea

r now

! The

4th d

ay: A

gain,

I felt

a lit

tle be

tter

and t

he te

st re

sults

wer

e goo

d. Th

e psy

chiat

rist c

onclu

ded t

hat it

was

n’t

a psy

cholo

gical

prob

lem. T

he nu

rse as

ked t

he ph

ysici

an to

pres

cribe

Co

versy

l plus

to se

e wha

t will

happ

en. R

elucta

ntly,

he di

d. In

the

phar

macy

, I go

t the

Cove

rsyl w

ith th

e lab

el ‘m

edica

lly ne

cessa

ry’.

It wa

s com

pletel

y im

prov

ed, e

xcep

t my

cond

ition

. No o

ne kn

ows

what

has h

appe

ned:

medic

ally,

ther

e wer

e no

prob

lems. I

do no

t beli

eve

it! W

hy th

e med

ically

ne

cessa

ry re

port

on th

e lab

el? In

the l

eafle

t of

perin

dopr

il, I a

lso re

ad th

at in

very

rare

case

s, fain

ting

could

occu

r!

Af

ter th

e pre

scrip

tion o

f Co

versy

l plus

, I los

t my

cons

cious

ness

only

once

and

only

for a

few m

inutes

.

Durin

g a re

gular

visit

, the

only

thing

my i

nter

nist s

aid

was ‘y

ou w

ere v

ery s

ick, I

was s

hock

ed w

hen I

saw

you’.

She d

id no

t give

any

excu

ses a

nd di

dn’t s

ay th

at I

was r

ight.

Card

iolog

ist: A

typica

l thor

acic

pain

by a

high r

isk pr

ofile

of va

scula

r dise

ase.

Exce

llent

ex

ercis

e tole

ranc

e with

out s

igns

of co

rona

ry in

suffi

cienc

y, wi

th

symp

toms f

itting

with

hy

perv

entil

ation

. ECG

is no

rmal.

Cove

rsyl p

lus AR

G 5/1

,25mg

1D

1T M

edica

lly ne

cessa

ry

Au

g Se

pt

Oct

Nov

Dec

Jan

Feb

Mar

Apr

May

June

The p

rofes

siona

ls pe

rsiste

d: Th

ey w

ere

right

and I

was

wro

ng.

With

all it

s co

nseq

uenc

es. If

they

ha

d list

ened

, my f

amily

an

d I w

ouldn

’t hav

e ha

d this

mise

ry an

d ten

sion.

It wo

uld ha

ve

save

d the

healt

h ins

uran

ce a

lot of

mo

ney.

At th

e end

, I go

t my

old d

rug.

They

ne

ver s

aid th

at th

ey

were

wro

ng. L

et alo

ne

that

they

apolo

gized

.



101

114

Pray

ers s

ign is

comp

atible

with

dia

betic

arth

ropa

thy,

with

some

ten

domy

algy.

LJM co

vers

symp

toms b

ut it

is in

early

stag

e. Gi

ven t

he al

lergy

of di

clofen

ac an

d co

morb

idity

of es

pecia

lly

hype

rtens

ion, N

SAID

is le

ss de

sirab

le. Pr

escri

ption

of

pred

nison

5mg 1

D1T b

eside

s pa

race

tamol

and e

xerci

ses f

or

hand

s.

I was

told

that

I hav

e LJM

. pr

ednis

on an

d par

aceta

mol

were

pres

cribe

d. I r

eacte

d goo

d on

this:

less

pain

and m

y han

ds

were

less

crook

ed.

Patie

nt ha

s join

t pr

oblem

s. Ph

ysiot

hera

pist

expe

cts

rheu

matis

m.

Patie

nt ha

s stif

fness

in ha

nds. F

amily

histo

ry of

rh

euma

tism.

No s

wellin

g, re

dnes

s of s

kin, h

eat o

f joi

nts, o

r pre

ssure

pain.

Sp

ecial

ist:

arth

ritis/

rheu

matis

m un

likely

. No o

sseou

s ab

norm

alitie

s of th

e han

ds.

BSE:

31* R

Ftes

t:<10

anti-

CCP:<

7.0.

Pres

cript

ion of

pr

ednis

olon 5

mg

1D1T

.

The p

harm

acist

gave

me

pred

nisolo

n bec

ause

this

bran

d was

chea

per.

Pray

ers s

ign no

w ne

gativ

e. BS

E 30,

CRP 3

, HB

6.7,

MCV 8

2, cre

at 70

, glu

cose

6.1,

ALAT

20,

HbA 1

c 56.

Conc

lusion

: cra

mping

, an

emia.

Pres

cript

ion of

fer

rofum

araa

t 200

mg

3dd,

pred

nison

5mg d

d an

d par

aceta

mol 1

000

mg 3d

d. Re

duce

use o

f pr

ednis

on.

The p

ain in

creas

ed. M

y wh

ole bo

dy st

iffen

ed. I

aske

d my r

heum

atolog

ist

wheth

er th

is co

uld be

due

to th

e cha

nge o

f the d

rug.

He di

dn’t t

hink s

o, bu

t we

could

try.

He pr

escri

bed

pred

nison

with

the

indica

tion t

hat it

was

me

dicall

y nec

essa

ry. T

his

helpe

d.

Stiffn

ess i

ncre

ases

. Pa

tient

has r

ead

artic

le ab

out li

mited

joi

nt m

obilit

y (LJM

) in

diabe

tes. R

eferra

l to

rheu

matol

ogist

.

De

c Jan

Fe

b Ma

r Ap

r Ma

y Jun

e Jul

y Au

g Se

pt

Oct

Figu

re C

. Situ

ation

3.

The

rheu

matol

ogist

did

n’t th

ink it

was d

ue to

the

drug

, but

it is

my

body

and i

f I th

ough

t the

drug

wa

s the

caus

e we

could

try.

R1R2R3R4R5R6R7R8R9


Intermezzo

102

References[1] Weingart SN, Gandhi TK, Seger AC, Seger

DL, Borus J, Burdick E, et al. Patient-report-ed medication symptoms in primary care. Arch Intern Med 2005;165(2):234-40.

[2] Wetzels R, Wolters R, van Weel C, Wensing M. Mix of methods is needed to identify ad-verse events in general practice: a prospec-tive observational study. BMC Fam Pract 2008;9:35.

[3] Basch E, Jia X, Heller G, Barz A, Sit L, Frus-cione M, et al. Adverse symptom event re-porting by patients vs clinicians: relation

ships with clinical outcomes. J Natl Cancer Inst 2009;101(23):1624-32.


[5] de Vries ST, Haaijer-Ruskamp FM, de Zeeuw D, Denig P. Construct and concurrent validi-ty of a patient-reported adverse drug event questionnaire: a cross-sectional study. Health Qual Life Outcomes 2014;12:103.

Part II The role of patient characteristics and preferences

on treatment decisions in clinical practice

Chapter 4

Potential overtreatment and undertreatment of

diabetes in different patient age groups in primary

care after the introduction of performance measures

Sieta T. de Vries1

Jaco Voorham1 Flora M. Haaijer-Ruskamp1

Petra Denig1

Diabetes Care 2014;37(5):1312-20.


R1R2R3R4R5R6R7R8R9


Chapter 4

106

Abstract

ObjectiveTo assess whether after the introduction of diabetes performance measures decreases in undertreatment correspond with increases in overtreatment for blood pressure (BP) and glycaemic control in different patient age groups.

MethodsWe conducted a cohort study using data from the Groningen Initiative to Analyze Type 2 Diabetes Treatment database. General practices were included when data were available from 1 year before to at least 1 year after the introduction of diabetes performance measures. Included patients had a confirmed diagnosis of type 2 diabetes. Potential overtreatment was defined as prescribing maximum treatment or a treatment intensification to patients with a sustained low-risk factor level. Potential undertreatment was defined as a lack of treatment intensification in patients with a sustained high-risk factor level. Percentages of over- and undertreated patients at baseline were compared with those in subsequent years, and stratified analyses were performed for different patient age groups.

Results For BP, undertreatment significantly decreased from 61 to 57% in the first year after the introduction of performance measures. In patients >75 years of age, undertreatment decreased from 65 to ∼61%. Overtreatment was relatively stable (∼16%). For glycaemic control, undertreatment significantly increased from 49 to 53%, and overtreatment remained relatively stable (∼7%).

ConclusionsThe improvement of BP undertreatment after introduction of the performance measures did not correspond with an increase in overtreatment. The performance measures appeared to have little impact on improving glucose-regulating treatment. The trends did not differ among patient age groups.


Potential over- and undertreatment of diabetes

107

4

IntroductionThe quality of diabetes care, and in particular potential undertreatment of cardiometabolic risk factors, has received much attention in the past decade. Several improvements have been observed in the process as well as the outcomes of the care for patients with type 2 diabetes [1–3]. These improvements have been stimulated by quality assurance and pay-for-performance programs, which incorporate performance measurements focusing on achieving risk factor targets [4–6].

In clinical guidelines for diabetes management (DM), the general target for glycohemoglobin (HbA1c) concentration is set at <7% (53 mmol/mol) and for systolic blood pressure (SBP) at <140 mmHg or even <130 mmHg for some patients [7,8]. However, the debate about the target levels being too strict has intensified, in particular with regard to the aged patients [9–12]. In addition, concerns have been raised that the introduction of performance measures may stimulate potential overtreatment since providers are rewarded with financial incentives for achieving strict targets [9,13]. Recently, Kerr et al. [14] reported that among veterans with diabetes, potential overtreatment for BP was approaching that of undertreatment. These findings may not be unique for BP treatment, and similar trends may be expected for glucose-regulating treatment.

One may expect that the influence of performance measures will differ in different settings as well as among general practices (GPs) [15]. Performance rates and the extent of potential overtreatment may vary among practitioners and facilities [14,16], and changing this performance is a complex process that is influenced by multiple factors [17]. Kerr et al. [14] found that facilities with low levels of undertreatment were more likely to have higher levels of overtreatment.

Our aim was to assess whether, after the introduction of performance measures in the Netherlands, decreases in potential undertreatment for BP and glycaemic control correspond with increases in potential overtreatment in patients with type 2 diabetes. In addition, we assessed whether under- and overtreatment differ for different patient age groups.

Methods

Study designWe conducted an observational, dynamic cohort study from 2007–2011 of patients with type 2 diabetes in GPs in the province of Groningen in the Netherlands. The GPs are all member of a diabetes care group (DCG). Such DCGs, comparable with accountable care organizations in the USA, were formed after the introduction of bundled payment in 2007 in the Netherlands [18]. The DCGs are responsible for the organization and

R1R2R3R4R5R6R7R8R9


Chapter 4

108

provision of diabetes care in accordance with the Dutch Diabetes Federation Health Care Standard [19]. Diabetes performance measures were instituted in this region from 2008 onwards as part of this program. GPs received yearly feedback comparing their own practice performance with performance measures of the whole region and with benchmarks set by the DCG. There were no personal incentives or penalties linked to this benchmarking. GPs entered the DM program at different time points (Figure 4.1) [20–24].

121

Methods

Study design

We conducted an observational, dynamic cohort study from 2007–2011 of patients with type 2 diabetes in GPs in the province of Groningen in the Netherlands. The GPs are all member of a diabetes care group (DCG). Such DCGs, comparable with accountable care organizations in the USA, were formed after the introduction of bundled payment in 2007 in the Netherlands [18]. The DCGs are responsible for the organization and provision of diabetes care in accordance with the Dutch Diabetes Federation Health Care Standard [19]. Diabetes performance measures were instituted in this region from 2008 onwards as part of this program. GPs received yearly feedback comparing their own practice performance with performance measures of the whole region and with benchmarks set by the DCG. There were no personal incentives or penalties linked to this benchmarking. GPs entered the DM program at different time points (Figure 4.1) [20–24].

2007 2008 2009 2010 2011 Figure 4.1. Time flow of study with key publications and national guidelines relevant for the treatment of (aged) patients with high blood pressure or glycohemoglobin levels.

ACCORD and ADVANCE [20]: debate about strict target levels for glycaemic control HYVET study [21]: Antihypertensive treatment in aged patients reduces risk of death from stroke, death from any cause, and heart failure

Verenso guidelines [24]: Dutch guideline about less strict target levels for aged patients

ACCORD study [22]: strict target levels of systolic blood pressure did not reduce rate of cardio-vascular events Meta-analysis [23]: antihypertensive treatment in aged patients reduces stroke and heart failure with no effect on total mortality

32 GPs entered in 2008: Baseline = 2007 Follow-up year 1 = 2009 Follow-up year 2 = 2010 Follow-up year 3 = 2011

65 GPs entered in 2009: Baseline = 2008 Follow-up year 1 = 2010 Follow-up year 2 = 2011

36 GPs entered in 2010: Baseline = 2009 Follow-up year 1 = 2011

Figure 4.1. Time flow of study with key publications and national guidelines relevant for the treatment of (aged) patients with high blood pressure or glycohemoglobin levels.

Study population and data collectionData were collected from the Groningen Initiative to Analyze Type 2 Diabetes Treatment database including almost all regional patients with type 2 diabetes (<1% opted out). The cohort of patients was based on the GPs of which data were available for 1 year before up to at least 1 year after entry in the DM program.

Based on the GPs’ entry date in the DM program, the following cohort years were created: year before entry (baseline year), year of entry, 1 year after entry, and, if available, 2 and 3 years after entry. For practices entering in the second half of a calendar year, the next year was used as year of entry.

Per cohort year, patients were included who had a confirmed diagnosis of type 2 diabetes before January 1. Routinely collected data of the patients, including full prescription data, laboratory test results, and physical examinations, were extracted from electronic medical records using validated procedures [25]. In the Netherlands, no approval from an ethics committee is needed for studies using data from anonymous medical records.



109

4

Outcome measuresPrimary outcome measures were potential over- and undertreatment. The definitions of potential overtreatment were based on those suggested by Kerr et al. [14] and validated by assessing the association between overtreatment and experiencing an adverse drug event. The definitions of potential undertreatment were derived from practice guidelines [7,14,26–28]. We defined separate definitions for BP- and glucose-lowering treatment. The first measurement of SBP or HbA1c in a year was taken as index date. Treatment status and changes in treatment were assessed relative to this index measurement.

Potential overtreatment was defined as:SBP <130 mmHg and receiving ≥3 BP-lowering drugs or an increase in dose within 120 days after the index date or a start of a new BP-lowering drug class within 120 days after the index date and without next SBP measurement ≥130 mmHg within 120 days after the index date; HbA1c <6.5% (48 mmol/mol) and receiving ≥3 glucose-lowering drugs or insulin or an increase in dose within 120 days after the index date or a start of a new glucose-lowering drug class within 120 days after the index date and without next HbA1c measurement ≥6.5% (48 mmol/mol) within 120 days after the index date.

Potential undertreatment was defined as:SBP ≥140 mmHg and not on ≥3 BP-lowering drugs without any increase in dose within 120 days after the index date, start of a new BP-lowering drug class within 120 days after the index date, or switch to another BP-lowering drug class within 120 days after the index date and without next SBP measurement <140 mmHg in the period up to 120 days after the index date;HbA1c ≥7% (53 mmol/mol) and (not on ≥3 glucose-lowering drugs or insulin without any increase in dose within 120 days after the index date, start of a new glucose-lowering drug class within 120 days after the index date, or switch to another glucose-lowering drug class within 120 days after the index date and without next HbA1c measurement <7% (53 mmol/mol) in the period up to 120 days after the index date.

We used a period of 120 days to assess changes in treatment after the index date to capture clinical actions that were postponed to the next regular visit, which is commonly after 3 months in the Netherlands [28,29]. The therapeutic groups of BP- and glucose-lowering treatment included seven and eight drug classes, respectively (Table 4.1). The start of a drug was defined as a new drug prescription for a drug that had not been prescribed in 270 days before the start of the first prescription after the index date. A stop was defined as no repeat prescription within 270 days after the start date of

R1R2R3R4R5R6R7R8R9


Chapter 4

110

the last prescription. Stops before 7 days after index date were not considered as stops related to the measurement at index date. When a drug class was started within 7 days after the stop date of another drug class, the stop was considered a switch.

The following combined changes were considered as no change when assessing over- or undertreatment: a dose decrease combined with a dose increase, addition of a drug class combined with a dose decrease, and stop of a drug combined with a dose increase.

Table 4.1. Characteristics of general practices and patients in the year of entry to the disease management programCharacteristic NGeneral practices 133Median number of patients with type 2 diabetes per general practice (IQR) 117 (91-162)Total number of patients with type 2 diabetes 14,876Female patients (%) 7,674 (51.6) Mean age of patients in years at systolic blood pressure measurement (SD) 66.8 (12.2)Median diabetes duration at systolic blood pressure measurement (IQR) 5 (2-9)Number of patients without prescription of blood pressure-lowering drug in 6 months up to index date

3,077 (23.8)

Drug classes of blood pressure-lowering drugs* ACE-inhibitors (%) 5,117 (39.6) Angiotensin-II-antagonists (%) 2,673 (20.7) Drugs acting on the renin–angiotensin system (%) 4 (0.0) Diuretics (%) 5,854 (45.3) β-blockers (%) 5,133 (39.7) Calcium-channel blockers (%) 2,679 (20.7) Centrally-acting antihypertensives (%) 235 (1.8)Number of patients without prescription of glucose-lowering drug in 6 months up to index date

2,525 (18.6)

Drug classes of glucose-lowering drugs* Insulin (%) 1,940 (14.3) Biguanides (%) 8,693 (64.2) Sulfonamides (%) 5,356 (39.5) Alpha glucosidase inhibitors (%) 14 (0.1) Thiazolidinediones (%) 588 (4.3) Dipeptidyl peptidase 4 inhibitors (%) 92 (0.7) Repaglinide (%) 5 (0.0) Exenatide or liraglutide (%) 6 (0.0)SD = standard deviation; IQR = interquartile range* Number of patients who have been prescribed the drug class in the 6 months up to index date



111

4

AnalysisCharacteristics of the patient population were assessed for the year of entry to the DM program. Age and diabetes duration of the patients were calculated on the index date. For patients without an index date, the average index date of the other patients was used to assess their age and diabetes duration. The validation of the definition of potential overtreatment was performed using χ²-statistics.

Percentages of potential over- and undertreatment to BP- and glucose-lowering treatment were assessed in all patients and in eligible patients only. Eligible patients for overtreatment are patients with low-risk factor levels without an apparent need for intensified treatment, whereas eligible patients for undertreatment include those with high-risk factor levels, who are not on maximum treatment, and in whom additional treatment is usually indicated. Maximum treatment was defined as a prescription of three or more drug classes or insulin (only for the glucose-lowering drugs) in 6 months before and up to the index date.

Percentages of over- and undertreatment in the baseline year were compared with those in the subsequent years using z-approximation for differences between proportions. Subsequently, stratified analyses were conducted for the age groups <60, 60–75, and >75 years [3]. Sensitivity analyses were conducted using more relaxed definitions for overtreatment ━ that is, including only patients with levels of SBP <120 mmHg and HbA1c <6% (42 mmol/mol) as eligible for overtreatment.

The influence of the introduction of diabetes performance measures at GP level was assessed by comparing the percentages of over- and undertreated patients at baseline with 1 year after entry. GPs were divided into three groups, namely those with a ≥5% increase, ≥5% decrease, or stable percentage of patients with over- or undertreatment between the two measurements. We used χ²-statistics to test for associations between changes in over- and undertreatment at the GP level.

The analyses were conducted using Stata version 12 (Stata Corp., College Station, TX), and P-values <0.05 were considered statistically significant.

ResultsIn total, 133 GPs entered the DM program: 32 in the cohort of 2008, 65 in 2009, and 36 in 2010. This resulted in 133 GPs with follow-up data of at least 1 year, 97 GPs with follow-up data of 2 years, and 32 GPs with follow-up data of 3 years after entering the program (Figure 4.1). The patient population at year of entry consisted of 14,876 patients with a mean age of 67 years and 52% females (Table 4.1). Of the BP-lowering drugs, the diuretics were the most commonly prescribed drug class (45%), followed by the β-blockers (40%) and the ACE inhibitors (40%). Metformin (64%) and sulfonylurea derivatives (40%) were the most commonly prescribed glucose-lowering

R1R2R3R4R5R6R7R8R9


Chapter 4

112

drugs. Potential overtreatment was associated with more possible adverse drug events related to the specific drug classes of BP and glucose-lowering treatment (Appendix 5; supplemental table 1). For all patients with SBP measurements, potential overtreatment was observed in 3.2–3.8% in the study period, whereas potential undertreatment was seen in 18.6–25.4%. For all patients with HbA1c measurements, potential overtreatment was seen in 2.0–2.4% and undertreatment in 14.8–16.5%.

BP-lowering treatmentPotential overtreatment among eligible patients with an SBP <130 mmHg was 15.9% at baseline and remained relatively stable in the years after entry to the DM program (P > 0.05) (Table 4.3). This pattern was similar for the different patient age groups (Figure 4.2). Potential overtreatment mainly involved patients receiving maximum treatment of BP-lowering drugs (~14% of eligible patients), which was generally more common in patients >75 years of age (data not shown). Intensification of treatment occurred in ~3% of the eligible patients (Table 4.3) and was comparable among the patient age groups (data not shown). Similar non-significant patterns of overtreatment were found in patients with an SBP <120 mmHg, being the more relaxed definition of overtreatment (Appendix 5; supplemental table 2).

Potential undertreatment of eligible patients with an SBP ≥140 mmHg was extensive but decreased from 60.7% in the baseline year to 56.9–50.7% in the years after entry (Table 4.3). The percentages were significantly (P < 0.05) different for all 3 years after entry in the DM program in comparison with the baseline year and were largely due to the improvements in patients aged ≤75 years (Figure 4.2). In patients >75 years of age, undertreatment decreased from 64.7% to ~61% in the years after entry. In general, potential undertreatment of BP treatment was more common in aged patients.



113

4

Table 4.3. Numbers of potential overtreatment and undertreatment for patients with SBP or HbA1c measurements

Baseline Entry DM-program

Follow-up year 1

Follow-up year 2

Follow-up year 3

Blood pressure-lowering overtreatmentWith SBP measurement (% of all patients) 11,517 (84.4) 12,927 (86.9) 14,579 (89.8) 11,702 (91.2) 4,573 (94.2) Mean age in years (SD) 67.1 (12.0) 67.0 (12.0) 67.0 (11.9) 67.0 (12.0) 67.1 (11.5) Percent females 52.9 51.9 51.6 51.9 50.8 Median DM duration (range IQR) 4 (6) 5 (7) 5 (7) 5 (7) 5 (7) Mean SBP (SD) 144.0 (20.4) 144.7 (20.6) 144.2 (20.0) 142.7 (19.4) 142.6 (19.2) N SBP <130 mmHg 2,451 2,618 2,969 2,542 1,000 N SBP <130 mmHg with Potential overtreatment (% of eligible patients)

389 (15.9) 413 (15.8) 472 (15.9) 443 (17.4) 162 (16.2)

N classes >=3† 331 (13.5) 366 (14.0) 408 (13.7) 381 (15.0) 136 (13.6) N intensified† 82 (3.3) 69 (2.6) 87 (2.9) 81 (3.2) 36 (3.6)Glucose-lowering overtreatmentWith HbA1c measurement (% of all patients) 12,117 (88.8) 13,548 (91.1) 14,999 (92.4) 11,884 (92.6) 4,499 (92.7) Mean age in years (SD) 66.9 (12.0) 66.9 (12.0) 66.9 (12.0) 66.9 (12.0) 67.0 (11.6) Percent females 52.8 51.8 51.6 51.7 50.6 Median DM duration (range IQR) 4 (6) 5 (7) 5 (7) 5 (7) 5 (7) Mean HbA1c (SD) 6.9 (1.0) 7.0 (1.0) 7.0 (1.0) 7.0 (1.0) 7.0 (0.9) N HbA1c <6.5% (48 mmol/mol) 3,980 4,150 4,510 3,518 1,209 N HbA1c <6.5% (48 mmol/mol) with potential overtreatment (% of eligible patients)

296 (7.4) 310 (7.5) 341 (7.6) 239 (6.8) 103 (8.5)

N classes >=3† 46 (1.2) 28 (0.7)* 25 (0.6)* 20 (0.6)* 4 (0.3)* N insulin use† 178 (4.5) 181 (4.4) 193 (4.3) 126 (3.6) 58 (4.8) N intensified† 77 (1.9) 102 (2.5) 130 (2.9)* 100 (2.8)* 45 (3.7)*Blood pressure-lowering undertreatmentWith SBP measurement and not on max treatment (% of all patients)

8,387 (61.5) 9,250 (62.2) 10,373 (63.9) 8,246 (64.3) 3,178 (65.5)

Mean age in years (SD) 65.9 (12.3) 65.8 (12.4) 65.6 (12.2) 65.6 (12.3) 65.6 (11.7) Percent females 51.6 50.8 50.4 50.9 49.8 Median DM duration (range IQR) 4 (6) 4 (6) 4 (6) 5 (6) 5 (7) Mean SBP (SD) 142.7 (19.6) 143.5 (19.7) 142.9 (19.1) 141.3 (18.5) 141.0 (18.2) N SBP ≥140 mmHg not on max treatment

4,826 5,449 5,924 4,430 1,676

N SBP ≥140 mmHg not on max treatment with potential undertreatment (% of eligible patients)

2,931 (60.7) 3,209 (58.9) 3,370 (56.9)* 2,466 (55.7)* 850 (50.7)*

Glucose-lowering undertreatmentWith HbA1c measurement and not on max treatment (% of all patients)

10,219 (74.9) 11,413 (76.7) 12,593 (77.6) 9,969 (77.7) 3,712 (76.5)

Mean age in years (SD) 66.7 (12.0) 66.7 (12.0) 66.7 (11.9) 66.7 (11.9) 66.8 (11.5) Percent females 51.9 50.8 51.0 51.2 50.6 Median DM duration (range IQR) 4 (5) 4 (6) 4 (6) 4 (6) 4 (6) Mean HbA1c (SD) 6.8 (1.0) 6.9 (0.9) 6.8 (0.9) 6.8 (0.9) 6.9 (0.9) N HbA1c ≥7% (53 mmol/mol) not on max treatment

3,652 4,187 4,405 3,479 1,368

N HbA1c ≥7% (53 mmol/mol) not on max treatment with potential undertreatment (% of eligible patients)

1,796 (49.2) 2,145 (51.2) 2,335 (53.0)* 1,962 (56.4)* 725 (53.0)*

Baseline = Year before entry to the disease management program; Entry DM-program = Entry to disease management program; Follow-up year 1 = 1 year after entry; Follow-up year 2 = 2 years after entry; Follow-up year 3 = 3 years after entry; SBP = Systolic blood pressure; HbA1c = glycohemoglobin; SD = standard deviation; IQR = interquartile range. † Percentages do not sum to the percentages of patients with potential overtreatment because patients can be included in multiple categories of overtreatment. * Percentages of overtreatment or undertreatment in years of participation in the disease management program that significantly differ from the baseline year (P<0.05).

R1R2R3R4R5R6R7R8R9


Chapter 4

114

128

† Percentages do not sum to the percentages of patients with potential overtreatment because patients can be included in multiple categories of overtreatment. * Percentages of overtreatment or undertreatment in years of participation in the disease management program that significantly differ from the baseline year (P<0.05).

Baseline = Year before entry to the disease management program; Entry DM-program = Entry to disease management program; Follow-up year 1 = 1 year after entry; Follow-up year 2 = 2 years after entry; Follow-up year 3 = 3 years after entry. * Percentages of overtreatment or undertreatment in years of participation in the disease management program that significantly (P < 0.05) differ from the baseline year (reference year). Figure 4.2. Trends in percentages of over- and undertreated patients based on eligible patients, and P- values of the comparison of baseline year with subsequent years.

Baseline = Year before entry to the disease management program; Entry DM-program = Entry to disease management program; Follow-up year 1 = 1 year after entry; Follow-up year 2 = 2 years after entry; Follow-up year 3 = 3 years after entry. * Percentages of overtreatment or undertreatment in years of participation in the disease management program that significantly (P < 0.05) differ from the baseline year (reference year).

Figure 4.2. Trends in percentages of over- and undertreated patients based on eligible patients, and P-values of the comparison of baseline year with subsequent years.



115

4

Glucose-lowering treatmentPotential overtreatment among eligible patients with an HbA1c <6.5% (48 mmol/mol) was observed in 7.4% of the patients at baseline (Table 4.3). This percentage did not significantly change in the years after entry in the DM program. In patients <60 years of age, overtreatment was 6.7% at baseline, which increased to 7.5–10.3% in the years after entry (Figure 4.2). The percentage of patients with an HbA1c <6.5% (48 mmol/mol) receiving intensification of glucose-lowering treatment was significantly higher in later years (1.9% in the baseline year and ~3% in the years after entry). However, the percentage of patients receiving maximum treatment decreased significantly in later years (1.2% in the baseline year and 0.3–0.6% in the years after entry; Table 4.3). The increase in intensification over time was particularly seen in patients <60 years of age, whereas the decrease in receiving maximum treatment was seen in all patient age groups. Insulin use was generally more common in patients >75 years of age (data not shown). The pattern of overtreatment was similar for patients with the more relaxed definition of <6% (42 mmol/mol) for HbA1c (Appendix 5; supplemental table 2).

Potential undertreatment of patients with an HbA1c ≥7% (53 mmol/mol) increased from 49.2% at baseline to 53.0% in the first year after entry (Table 4.3). The percentages were significantly higher in all 3 years after entry in the DM program compared with the baseline year. This increase was largest in the second year after entry, a pattern that was observed in all three age groups (Figure 4.2). Overall, potential undertreatment for glycaemic control was more common in aged patients.

GP levelA decrease in undertreatment for SBP was seen for 44% (N = 57) of the GPs, while for only 29% (N = 38) an increase in overtreatment for SBP was seen. Of the 57 GPs that showed a decrease in undertreatment, 26 also showed a decrease in overtreatment (46%), whereas only 20 (35%) showed an increase in overtreatment. This association between improvements in under- and overtreatment at the GP level for SBP was statistically significant (P = 0.02; Appendix 5; supplemental table 3). For HbA1c, 25% of the GPs had a decrease in undertreatment, and the same number had an increase in overtreatment. No significant associations were seen between changes in under- and overtreatment at the GP level for HbA1c (P = 0.13; Appendix 5; supplemental table 3). DiscussionDuring the entire period from 2007–2011, potential overtreatment was much less common than potential undertreatment for both BP and glycaemic control. Following the introduction of diabetes performance measures, in the period 2008–2010, there was a significant decrease in potential undertreatment and a relatively stable level

R1R2R3R4R5R6R7R8R9


Chapter 4

116

of potential overtreatment for BP. For HbA1c, we observed a relatively stable level of overtreatment and an unexpected increase in potential undertreatment after the introduction of performance measures. These results hardly differed among the patient age groups, although levels of potential undertreatment for both BP and HbA1c were generally higher in aged patients.

There can be good reasons to deviate from guideline recommendations for individual patients. Regarding overtreatment, there are currently no minimum levels for BP or HbA1c, and measures of overtreatment have been criticized [30]. We used the same cut-off levels of overtreatment for all patient age groups, since the prevailing guidelines during the study period did not distinguish different target levels across age groups. Using different levels for defining potential overtreatment, we found similar results. We observed a decrease in the percentages of patients on maximum treatment. However, we also observed a small increase over time in treatment intensification rates in patients already having low HbA1c levels. This finding was particularly seen in younger patients, which implies that GPs need to be more cautious with intensifying treatment in this specific patient group. Potential overtreatment was more common in BP-lowering treatment than in glucose-lowering treatment. However, the need for additional treatment with BP-lowering drugs may be appropriate in patients needing these drugs for (cardiovascular) comorbidities [14]. Since aged patients more often have comorbidities than younger patients [31], this may also explain why being on maximum treatment was more common in aged patients.

We can only speculate why potential undertreatment for HbA1c increased in the years after the introduction of performance measures. This finding could be a temporary effect caused by changes in the underlying patient population. The performance measures were introduced as part of a DM program that also included a new payment system. Financial incentives related to this program may have led to unintended shifts of patients. Concerns have been expressed about increasing numbers of patients with preliminary stages of diabetes and patients being moved from specialist to primary care for financial reasons [32]. In contrast, our finding that this increase was especially seen in aged patients suggests that this is not the most likely explanation. An alternative explanation would be the intensifying call for using less strict target levels of <7.5 (58 mmol/mol) to <8.5% (69 mmol/mol) for HbA1c and <150 mmHg to <160 mmHg for SBP for aged patients in recent years [24,33]. Our definition of undertreatment was based on the 2006 guidelines that promoted treatment for strict target levels in general. Although these guidelines were changed after our study period in 2013 [34], it is likely that norms about less intensive treatment in aged patients were already starting to percolate in practice during the study period [29]. Given the current debate and introduction of guidelines that recommend less strict targets in aged patients, future studies need to apply age-specific definitions of overtreatment as well as undertreatment.



117

4

In contrast, undertreatment of both BP- and glucose-lowering treatment was more common in aged patients than in younger patients throughout the whole study period. GPs appear to be more restrictive in prescribing drugs in aged patients. A study about the prescription of β-blockers in patients with coronary artery disease found a similar result [35]. Several patient-related reasons for potential undertreatment have been proposed [36,37], some of which are likely to differ among age groups. Aged patients have, for instance, more often comorbidities [31] and an increased risk of adverse drug events that may restrict the therapeutic options, and their treatment preferences and needs may also differ from younger patients [38]. Future studies are needed to investigate the reason behind the difference in undertreatment among age groups.

Our study does not support the concerns about increasing overtreatment after the introduction of performance measures. Previously, Kerr et al. [14] found an association between low levels of undertreatment and high levels of overtreatment within veterans affairs facilities. In our study, most improvements (i.e., reductions) in undertreatment were observed for BP treatment. We found that decreases in undertreatment were significantly associated with decreases in overtreatment, which refutes the hypothesis that GPs felt pressured to prescribe more treatment in general after the introduction of performance measures. This dissimilarity between our findings and those of Kerr et al. [14] may be due to differences in the studied patient population or to slightly different definitions of over- and undertreatment, but are more likely due to differences in the way the performance measures have been implemented (e.g., different financial incentives) and in the organization of the healthcare system. The system that was intended to reduce undertreatment may have been less enforced in our country, which is then expected to result in less aggressive treatment in general. Indeed, we observed less overtreatment but also more undertreatment in our patient population in comparison with the population in the study of Kerr et al. [14]. The level of undertreatment in the years after the implementation of performance measures, being ~20% of all patients with a BP measurement and 16% of all patients with an HbA1c measurement, was much higher than the level of 6% for BP as seen in the study of Kerr et al. [14].

Strengths of our study comprise the large unrestricted cohort of patients with diabetes and the detailed longitudinal information on risk factors and drug prescribing. This allowed us to assess changes in prescribed treatment relative to risk factor levels. During the study period, risk factors were assessed in 84–94% of the patients. Data were collected from electronic medical records, and all included GPs prescribe electronically using the electronic medical records system. In the Netherlands, each patient is registered with a single GP who is the gatekeeper and obliged to keep adequate medical records, including out-of-hours prescriptions made by other practitioners.

The study is limited by its observational design and the dynamic cohort captured. Due to the rolling cohorts, year 1 covered the period 2009–2011, whereas year 3 was

R1R2R3R4R5R6R7R8R9


Chapter 4

118

restricted to 2011 (Figure 4.1). Therefore, changes observed in year 3 can be due to changing norms over time as well as sustained or delayed effects of the program. The findings on which we base our conclusions, however, were observed already in the first year after the introduction of performance measures and consistent for all year cohorts. Our outcome measures have been derived from guideline recommendations and have only in part shown associations with clinical outcomes (Appendix 5; supplemental table 1) [39,40]. We included the first measurement of SBP and HbA1c in a year and assessed whether the levels of a follow-up measurement in the 120 days after the index date returned to control. It is possible that GPs base the treatment changes on a longer period. A previous study showed, however, that an extended period of 180 days does not significantly lead to a higher number of changes after elevated levels [28]. Finally, we only had information about changes in drug treatment. Therefore, actions related to nondrug treatment, including lifestyle and medication adherence, were not accounted for when assessing potential undertreatment.

In summary, the introduction of performance measures reduced undertreatment for BP, which did not correspond with an increase in overtreatment. It seemed that the performance measures had little impact on improving glucose-regulating treatment. There were no clear differences in trends among different patient age groups. During the whole period, undertreatment was higher in aged patients than in younger patients, possibly reflecting concerns about the need for intensive medication treatment in aged patients.



119

4

References[1] Mudaliar U, Kim WC, Kirk K, Rouse C,

Narayan KM, Ali M. Are recommended standards for diabetes care met in Central and South America? A systematic review. Diabetes Res Clin Pract 2013;100:306–29.

[2] Si D, Bailie R, Wang Z, Weeramanthri T. Comparison of diabetes management in five countries for general and indigenous populations: an internet-based review. BMC Health Serv Res 2010;10:169–87.

[3] van Hateren KJ, Drion I, Kleefstra N, Groenier KH, Houweling ST, van der Meer K, et al. A prospective observational study of quality of diabetes care in a shared care setting: trends and age differences (ZODI-AC-19). BMJ Open 2012;2:e001387.

[4] British Medical Association and National Health Service Employers. Quality and Out-comes Framework Guidance for GMS Con-tract 2013/14. London, UK, British Medical Association, National Health Service Con-federation, 2013.

[5] National Quality Forum. National Voluntary Consensus Standards for Ambulatory Care Using Clinically Enriched Data: A Consen-sus Report. Washington, D.C., U.S. National Quality Forum, 2010.

[6] O’Connor PJ, Bodkin NL, Fradkin J, Glas-gow RE, Greenfield S, Gregg E, et al. Di-abetes performance measures: current status and future directions. Diabetes Care 2011;34:1651–9.

[7] Bouma M, Rutten GE, de Grauw WJ, Wiers-ma T, Goudswaard AN; Nederlands Huisart-sen Genootschap. [Summary of the practice guideline ‘Diabetes mellitus type 2’ (second revision) from the Dutch College of Gener-al Practitioners]. Ned Tijdschr Geneeskd 2006;150:2251–6.

[8] Executive summary: Standards of medi-cal care in diabetesd2013. Diabetes Care 2013;36(Suppl. 1):S4–10.

[9] Lehman R, Krumholz HM. Tight control of blood glucose in long standing type 2 dia-betes. BMJ 2009;338:b800.

[10] Pogach L, Aron D. The other side of quality improvement in diabetes for seniors: a pro-posal for an overtreatment glycemic meas-ure. Arch Intern Med 2012;172:1510–2.

[11] Havas S. The ACCORD Trial and control of blood glucose level in type 2 diabetes mel-litus: time to challenge conventional wis-dom. Arch Intern Med 2009;169:150–4.

[12] Jindal A, Whaley-Connell A, Sowers JR. Type 2 diabetes in older people; the importance of blood pressure control. Curr Cardiovasc Risk Rep 2013;7:233–7.

[13] Choe HM, Bernstein SJ, Standiford CJ, Hay-ward RA. New diabetes HEDIS blood pres-sure quality measure: potential for over-treatment. Am J Manag Care 2010;16:19–24.

[14] Kerr EA, Lucatorto MA, Holleman R, Ho-gan MM, Klamerus ML, Hofer TP; VA Di-abetes Quality Enhancement Research Initiative (QUERI) Workgroup on Clini-cal Action Measures. Monitoring perfor-mance for blood pressure management among patients with diabetes mellitus: too much of a good thing? Arch Intern Med 2012;172:938–45.

[15] Cabana MD, Rand CS, Powe NR, Wu AW, Wilson MH, Abboud PA, et al. Why don’t physicians follow clinical practice guide-lines? A framework for improvement. JAMA 1999;282:1458–65.

[16] Ryan AM, Doran T. The effect of improving processes of care on patient outcomes: evidence from the United Kingdom’s qual-ity and outcomes framework. Med Care 2012;50:191–9.

[17] Grant A, Sullivan F, Dowell J. An ethno-graphic exploration of influences on pre-scribing in general practice: why is there variation in prescribing practices? Imple-ment Sci 2013;8:72.

[18] Campmans-Kuijpers MJ, Lemmens LC, Baan CA, Gorter KJ, Groothuis J, van Vuure KH, et al. Defining and improving quality manage-ment in Dutch diabetes care groups and outpatient clinics: design of the study. BMC Health Serv Res 2013;13:129.

[19] De Grauw WJC. NDF Zorgstandaard. Trans-parantie en Kwaliteit van Diabeteszorg voor Mensen met Diabetes Type 2 [Trans-parency and quality of diabetes care for people with type 2 diabetes]. Amersfoort, the Netherlands, Nederlandse Diabetes Federatie, 2007.

[20] Dluhy RG, McMahon GT. Intensive glycemic control in the ACCORD and ADVANCE trials. N Engl J Med 2008;358:2630–3.

[21] Beckett NS, Peters R, Fletcher AE, Staessen JA, Liu L, Dumitrascu D, et al.; HYVET Study Group. Treatment of hypertension in pa-tients 80 years of age or older. N Engl J Med 2008;358:1887–98.

R1R2R3R4R5R6R7R8R9


Chapter 4

120

[22] Cushman WC, Evans GW, Byington RP, Goff DC Jr, Grimm RH Jr, Cutler JA, et al.; ACCORD Study Group. Effects of intensive blood-pressure control in type 2 diabetes mellitus. N Engl J Med 2010;362: 1575–85.

[23] Bejan-Angoulvant T, Saadatian-Elahi M, Wright JM, Schron EB, Lindholm LH, Fagard R, et al. Treatment of hypertension in pa-tients 80 years and older: the lower the better? A meta-analysis of randomized con-trolled trials. J Hypertens 2010;28: 1366–72.

[24] Verenso. Multidisciplinaire Richtlijn Diabe-tes. Verantwoorde Diabeteszorg bij Kwets-bare Ouderen Thuis en in Verzorgings of Verpleeghuizen. Deel 1. [Multidisciplinary Guideline Diabetes. Responsible Diabetes Care in Vulnerable Elderly at Home and in Residential Care or Nursing Homes. Part 1]. Utrecht, the Netherlands, Verenso, 2011.

[25] Voorham J, Denig P. Computerized extrac-tion of information on the quality of diabe-tes care from free text in electronic patient records of general practitioners. J Am Med Inform Assoc 2007;14:349–54.

[26] Martirosyan L, Braspenning J, Denig P, de Grauw WJ, Bouma M, Storms F, et al. Pre-scribing quality indicators of type 2 dia-betes mellitus ambulatory care. Qual Saf Health Care 2008;17:318–23.

[27] Voorham J, Denig P, Wolffenbuttel BH, Haa-ijer-Ruskamp FM. Cross-sectional versus sequential quality indicators of risk factor management in patients with type 2 diabe-tes. Med Care 2008;46:133–41.

[28] Sidorenkov G, Haaijer-Ruskamp FM, de Zeeuw D, Denig P. A longitudinal study ex-amining adherence to guidelines in diabe-tes care according to different definitions of adequacy and timeliness. PLoS ONE 2011;6:e24278.

[29] Houweling ST, Kleefstra N, Verhoeven S, van Ballegooie E, Bilo HJG. Protocollaire Diabeteszorg. Mogelijkheden voor Taakde-legatie Editie 2009/2010. Apeldoorn, the Netherlands, Langerhans School of Diabe-tes, 2008.

[30] Handberg E. How do guidelines impact measures of performance? Can they keep up? Arch Intern Med 2012;172:945–6.

[31] Britt HC, Harrison CM, Miller GC, Knox SA. Prevalence and patterns of multimorbidity in Australia. Med J Aust 2008;189:72–7.

[32] Struijs JN, van Til JT, Baan CA. Experiment-ing with a Bundled Payment System for Diabetes Care in the Netherlands. The First Tangible Effects. Bilthoven, the Nether-lands, National Institute for Public Health and the Environment, 2010.

[33] Sue Kirkman M, Briscoe VJ, Clark N, Florez H, Haas LB, Halter JB, et al.; Consensus De-velopment Conference on Diabetes and Older Adults. Diabetes in older adults: a consensus report. J Am Geriatr Soc 2012;60:2342–56.

[34] Rutten GEHM, De Grauw WJC, Nijpels G, Houweling ST, van de Laar FA, Bilo HJ, et al. The NHG guideline Diabetes mellitus type 2. Huisarts Wet 2013;56:512–25.

[35] Vitale C, Spoletini I, Volterrani M, Iellamo F, Fini M. Pattern of use of β-blockers in older patients with stable coronary artery disease: an observational, cross-sectional, multicentre survey. Drugs Aging 2011;28: 703–11.

[36] Steinman MA, Patil S, Kamat P, Peterson C, Knight SJ. A taxonomy of reasons for not prescribing guideline-recommended medi-cations for patients with heart failure. Am J Geriatr Pharmacother 2010;8:583–94.

[37] AB E, Denig P, van Vliet T, Dekker JH. Rea-sons of general practitioners for not pre-scribing lipid-lowering medication to pa-tients with diabetes: a qualitative study. BMC Fam Pract 2009;10:24.

[38] Chilton F, Collett RA. Treatment choices, preferences and decision-making by pa-tients with rheumatoid arthritis. Musculo-skelet Care 2008;6:1–14.

[39] Sidorenkov G, Voorham J, de Zeeuw D, Haa-ijer-Ruskamp FM, Denig P. Treatment qual-ity indicators predict short-term outcomes in patients with diabetes: a prospective co-hort study using the GIANTT database. BMJ Qual Saf 2013;22: 339–47.

[40] Sidorenkov G, Voorham J, Haaijer-Ruskamp FM, de Zeeuw D, Denig P. Association be-tween performance measures and glycemic control among patients with diabetes in a community-wide primary care cohort. Med Care 2013;51:172–9.

Chapter 5

The role of patients’ age on their preferences for

choosing additional blood pressure-lowering drugs:

a discrete choice experiment in patients with diabetes

Sieta T. de Vries1

Dianna F.M. de Vries1,2

Thijs Dekker3

Flora M. Haaijer-Ruskamp1

Dick de Zeeuw1

Adelita V. Ranchor4 Petra Denig1

1 Department of Clinical Pharmacy and Pharmacology, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands.

2 Department of Pharmacy, Unit of PharmacoEpidemiology & PharmacoEconomics, University of Groningen, Groningen, The Netherlands

3 Institute for Transport Studies, University of Leeds, Leeds, United Kingdom4 Department of Health Psychology, University of Groningen,

University Medical Center Groningen, Groningen, The Netherlands

R1R2R3R4R5R6R7R8R9


Chapter 5

122

Abstract

BackgroundToday, the incorporation of patient preferences in treatment decisions is high on the agenda. Patient drug preferences may differ between aged and non-aged patients in particular regarding preventive medication.

ObjectiveThe aim of this study is to assess whether patients’ willingness to add a blood pressure-lowering drug and the importance they attach to specific treatment characteristics differ among age groups in patients with type 2 diabetes. In addition, the influence of medication beliefs on the association between age and willingness to add a blood pressure-lowering drug was explored.

MethodsPatients aged ≥18 years and being prescribed at least an oral glucose-lowering and a blood pressure-lowering drug in the past 6 months, were asked to complete a questionnaire containing general questions, the Beliefs about Medicines Questionnaire, and a discrete choice experiment in which they had to imagine that their blood pressure was uncontrolled. In the discrete choice experiment, patients were presented choice sets containing two hypothetical blood pressure-lowering drugs and a no additional drug alternative which differed in their characteristics or so-called attributes. The included attributes were the treatment’s effect on blood pressure, risk of death within the next 5 years, limitations due to a heart attack, limitations due to a stroke, experiencing adverse drug events (ADEs), and the treatment’s intake moment. Differences in willingness to add a drug between patients <75 years (non-aged) and ≥75 years (aged) and the effect of different medication beliefs were tested using Pearson χ2-tests. Multinomial logit models were used to assess the importance attached to the attributes by non-aged and aged patients.

ResultsOf the 210 eligible consenting patients, 161 completed the questionnaire and 151 (72%) could be included in the analyses. These patients were on average 68 years old and 42% of them were female. Aged patients were less willing to add a blood pressure-lowering drug than non-aged patients (67% versus 84% respectively; P = 0.017). Non-aged patients with high necessity beliefs were more willing to add a drug than non-aged patients with low necessity beliefs (90% versus 74% respectively; P = 0.040). The most important attributes in both age groups were the treatment’s effect on the risk of death within the next 5 years and on experiencing ADEs. The effect on limitations due to stroke was only significant in the non-aged group. The treatment’s effect on blood pressure was more important in the non-aged group compared to the aged group (P = 0.043).

ConclusionsAged patients were less willing to add a blood pressure-lowering drug than non-aged patients. The willingness of non-aged patients seemed to be influenced by necessity beliefs. The effects on the risk of death and on experiencing ADEs were important attributes for both age groups in choosing a drug. Only the effect on the blood pressure was different among the age groups. Aged patients attached less importance to this drug effect than non-aged patients.


Preferences for additional blood pressure-lowering drugs

123

5

Introduction There is growing interest to tailor drug treatment to the individual patient’s clinical and personal needs. In general, the incorporation of patient preferences in treatment decisions is high on the agenda in our society. Particularly in drug treatment it is important to take patient preferences into account since such preferences are related to a patient’s willingness to take a drug. The willingness to take a drug improves adherence and leads to a more effective treatment [1-3].

A patient’s willingness to take a drug is influenced by factors related to the drug, the physician, the disease, and patient’s own characteristics such as medication beliefs [4-7]. Drug-related factors include expected effects on, for example, life extension or factors related to quality of life, such as good health, ease of administration (e.g. frequency of taking the drug) and burden of adverse drug events (ADEs) [1]. Patients often value life extension as one of the most important drug effects [8-11]. On the other hand, many aged or frail patients are likely to value quality of life over life extension [12-14]. Therefore, one may expect that preferences for specific drugs are influenced by a patient’s age or life-expectancy, as has been shown previously [15,16]. In addition, aged patients or patients with a limited life-expectancy may be less willing to add a treatment [12,15].

The role of patient’s age or life-expectancy on drug choices may be particularly important for preventive treatment, such as the use of blood pressure-lowering drugs in patients with type 2 diabetes to reduce their risk for cardiovascular morbidity and mortality [17]. With respect to balancing life extension with quality of life, the need for blood pressure-lowering drugs may be less in aged patients since 1) evidence of long-term benefit in aged patients is lacking [18], 2) having a high blood pressure level is usually not perceived as burdensome [19], and 3) adverse events related to these drugs are common in the aged [20,21]. In general, patients with type 2 diabetes appear to have lower necessity beliefs for blood pressure-lowering drugs than, for instance, for glucose-lowering drugs [19,22]. Currently, little is known about preferences for choosing blood pressure-lowering treatment among aged and non-aged patients.

A useful and commonly used method to assess patient preferences for treatment is the discrete choice experiment [23,24]. In such experiments, a hypothetical situation is presented to the patient with treatment alternatives described by their characteristics, or so-called attributes [25].The relative importance of the attributes can be inferred from the choices made [26]. This method is useful because people have to make trade-offs between positive and negative consequences of a choice, similar to decisions that they have to make in practice.

The aim of this study is to assess whether patients’ willingness to add a blood pressure-lowering drug and the importance they attach to specific treatment characteristics differ among age groups in patients with type 2 diabetes. In addition, we

R1R2R3R4R5R6R7R8R9


Chapter 5

124

explored whether medication beliefs have an influence on the association between age and willingness to add a blood pressure-lowering drug.

Methods

Study design and participantsThis study has a cross-sectional design in which we compared patients <75 years with ≥75 years of age. Patients were eligible to participate when they were aged ≥18 years and had been prescribed at least an oral glucose-lowering and a blood pressure-lowering drug in the past 6 months. We aimed to include 150 patients in our study, since it has been shown that the precision of discrete choice experiments rapidly decreases at sample sizes less than 150 [26]. Pharmacists in the northern part of the Netherlands sent invitation letters to eligible patients identified from their electronic records. Patients who gave informed consent received a questionnaire composed of general questions including Cantril’s ladder to assess a patient’s quality of life [27], a validated questionnaire assessing medication beliefs, and subsequently the discrete choice experiment to evaluate their treatment preferences. Patients were called when they did not return the questionnaire within two months or in case of missing data in the general or beliefs questions. In case of missing data in the discrete choice experiment, the questionnaire was returned to the patient for further completion. The Medical Ethics Committee of the University Medical Center Groningen (METc UMCG) in the Netherlands determined that ethical approval was not needed for this study (reference number M14.150721).

Outcome variableThe outcome variable in this study was the choice patients make in a hypothetical situation with respect to adding a blood pressure-lowering drug.

Discrete choice experimentAttributes in the discrete choice experiment were based on a two-step literature review. In the first step, the literature was assessed for factors which may influence a patient’s willingness to take blood pressure-lowering drugs or drugs for primary cardiovascular disease prevention in general. This review revealed that lowering the blood pressure, achieving risk reduction of complications (myocardial infarction, stroke), reducing ADEs and improving quality of life were relevant factors [4-7]. During the second step, previous studies with discrete choice experiments to assess patient preferences for any drug were screened. This screening revealed two additional attributes, that is, costs of treatment and number of tablets needed per day [28-35]. We decided not to include costs in our experiment since expenditure for preventive drugs, such as blood pressure-lowering drugs, are covered by the health insurance in the Netherlands.



125

5

Levels of efficacy were established using clinical trial data of blood pressure-lowering treatment effects and cardiovascular risk reduction between tight and less tight blood pressure control, and using the UKPDS risk engine for assessing differences among different ages [36]. We estimated risks of complications or death within the next 5 years. Levels for ADEs were based on the prevalence of known ADEs for blood pressure-lowering drugs, such as cough and headache, as reported in the national drug compendium for healthcare professionals. Levels for the intake moments were based on possible schemes mentioned in the literature [37,38]. The list of attributes and levels was discussed and finalized by interviews with ten experts (2 nurse practitioners, 3 general practitioners, 2 specialists, and 3 pharmacists) (Table 5.1).

Table 5.1. Overview of the attributes and levels used in the discrete choice experimentAttributes Levels Coding of

variablesBlood pressure level1 Remains 160*

Decrease from 160 to 140Decrease from 160 to 150‡

160140150

Risk of death by a heart attack or stroke in the next 5 years1

13 of the 100 die and 87 don’t*

9 of the 100 die and 91 don’t11 of the 100 die and 89 don’t‡

0.130.090.11

Risk of limitations due to a heart attack, such as fatigue and difficulty walking in the next 5 years1

7 of the 100 get limitations and 93 don’t*

5 of the 100 get limitations and 95 don’t6 of the 100 get limitations and 94 don’t‡

0.070.050.06

Risk of limitations due to a stroke, such a speech problems and forgetfulness in the next 5 years1

7 of the 100 get limitations and 93 don’t*

5 of the 100 get limitations and 95 don’t 6 of the 100 get limitations and 94 don’t ‡

0.070.050.06

Risk of side effect3, such as cough and headache1

No side effects*

5 of the 100 get side effects and 95 don’t10 of the 100 get side effects and 90 don’t‡

00.050.10

Intake moment2 1 tablet in the morning*

1 tablet in the morning and 1 in the evening‡

1 combination tablet2 tablets in the morning†

-

1 Continuous variable; 2 Categorical variable; 3 Side effect was used as lay-term for adverse drug events;* Level used for ‘no additional drug’ option (never used for the additional drug options); ‡ Levels used for the non-preferable drug in the dominant choice set; † Reference category in the categorical attribute

A d-efficient design [26] of 30 choice sets, divided in three blocks, was generated using Ngene (version 1.1.1). No restrictions of level combinations were included. Patients were randomly assigned to one of the blocks containing ten choice sets. This blocking

R1R2R3R4R5R6R7R8R9


Chapter 5

126

was used to reduce the cognitive burden of the patients. Patients had to imagine that they were using one blood pressure-lowering drug and that their blood pressure was uncontrolled (160 mmHg). In the choice sets, patients could indicate how they would want to continue their treatment, choosing from hypothetical treatment options presented as ‘no additional drug’, ‘additional drug A’, or ‘additional drug B’. The profile of ‘no additional drug’ described the situation as presented in the case, whereas the profiles of ‘additional drug A’ and ‘additional drug B’ were generated with the d-efficient design. Patients were asked to complete the ten choice sets plus a choice set in which one drug was preferable on all attributes compared to the other drug. This dominant choice set was added to identify the responders who may not understand the task. A pilot study was conducted in which ten patients completed the full questionnaire. Based on this pilot study, some changes in wordings were made and a separate instruction form for the choice sets was included. An example of a final choice set is presented in Figure 5.1.

Patients’ age and life-expectancyIn the questionnaire, patients were asked their age. We used a cut-off level of ≥75 years to define aged patients. This cut-off level has been used in guidelines [39] and observational studies [40] looking at age differences for preventive treatment. To test its value in relation to perceived life-expectancy, we additionally asked patients an open-ended question: “In 2012, men became on average 75 years old and women on average 80 years old [41]. How old do you think you will become?”. With this we determined that age was a reasonable proxy for self-reported life-expectancy (Appendix 6; supplemental table 1).

Medication beliefsA patient’s medication beliefs were assessed using the validated Beliefs about Medicines Questionnaire (BMQ) specific [42]. The BMQ contains ten items, five of them assessing necessity beliefs (e.g. ‘My health in the future will depend on my blood pressure-lowering drugs’) and five assessing concern beliefs (e.g. ‘Having to take blood pressure-lowering drug worries me’). Agreement with each item is indicated on a five-point Likert scale ranging from totally disagree to totally agree. Per subscale, scores on the items were summed resulting in a range from 5 (totally disagree) to 25 (totally agree). Internal consistency assessed with Cronbach’s α was 0.86 for the necessity and 0.72 for the concerns subscale. Patients were divided in having high necessity or concern beliefs (score higher than or equal to the median) and having low necessity or concern beliefs (score lower than the median).



127

5

144

Figure 5.1. Example of a choice set presented in the questionnaire

Figure 5.1. Example of a choice set presented in the questionnaire

R1R2R3R4R5R6R7R8R9


Chapter 5

128

Additional data collection Pharmacists provided age and gender for all eligible patients, as well as overviews of prescribed drugs in the last six months for those patients who gave informed consent. The Anatomical Therapeutic Chemical (ATC) classification system of the World Health Organization was used to classify the prescribed drugs.

Statistical analysesDescriptive statistics are presented for patient characteristics. Differences between responders and non-responders in age and gender were assessed using the T-test and Pearson χ2-test respectively. Differences in characteristics between non-aged and aged patients were tested using Pearson χ2-tests for categorical variables and Mann-Whitney U tests for non-normally distributed, continuous variables. The Fisher freeman-halton test was used for categorical variables in which one or more cells contained less than five patients.

Differences in the number of patients who chose at least once an additional drug on the choice sets (willing to add) versus never (unwilling to add) were compared between the age groups using the Pearson χ2-test. In addition, this association was assessed for the age groups stratified by having low and high necessity and concern beliefs. The choices were further analysed using multinomial logit models (asclogit function in Stata) to assess 1) the willingness to add a blood pressure-lowering drug when controlling for all attributes, and 2) the relative importance of the attributes. Three models were assessed, that is, one including non-aged patients, one including aged patients, and one with all patients in which the interaction terms between the age groups and the attributes were included.

We followed the random utility model by assuming that patients choose the alternative in each choice set which maximizes their utility. The estimated model was the following:U = V + ε = β0additional drug + β1blood pressure + β2death + β3limitations heart attack + β4limitations stroke + β5ADEs + β6one in morning one in evening + β7combination tablet + εwith U indicating the utility that a patient assigns to a treatment which is the sum of a systematic, explainable component V and a random, unexplainable component ε. The explainable component is a function of the attributes of the alternatives. The constant β0 indicates the relative weight patients place on choosing an additional blood pressure-lowering drug versus no additional blood pressure-lowering drug when controlling for the attributes. The β1 to β7 coefficients indicate the relative importance of each of the attributes. Coefficients reflect continuous variables of the attributes, except β6 and β7 which reflect the dummy-coding of respectively the level one drug in the morning and



129

5

one in the evening (coded as 1) versus two drugs in the morning (coded as 0), and the level combination tablet (coded as 1) versus two drugs in the morning (coded as 0). The sign of the beta-coefficients indicates whether the effect on the utility is positive or negative. Attributes with beta-coefficients with a two-sided P-value <0.05 were considered as being important for the treatment choice. Patients who chose the non-preferable drug in the dominant choice set were excluded from the analyses. Sensitivity analyses were conducted in which these patients were not excluded. Additional sensitivity analyses were conducted to assess the multinomial logit model for patients <65 years and patients aged ≥80 years. The analyses were conducted using Stata version 13 (Stata Corp., College Station, TX).

Results Three pharmacies sent information letters to 933 patients, resulting in 210 eligible consenting patients. Of these, 161 completed the questionnaires and 151 were included in the analyses (Figure 5.2). The included patients were on average 68 + 7 yrs and most of them were males (58%). There was no significant age difference between responders and non-responders (68 and 70 years, P=0.265), but less females participated (42% versus 52% females, P=0.037). Metformin was the most commonly prescribed glucose-lowering drug in both aged and non-aged patients (Table 5.2). Agents acting on the renin-angiotensin system (RAS-inhibitors) were the most commonly prescribed blood pressure-lowering drug class in non-aged patients, whereas the β-blockers were the most commonly prescribed class in aged patients. Most of the patients reported that their blood pressure was <160 mmHg, that they used one or two drugs to lower their blood pressure, and that they preferred to leave decisions about their drug to the general practitioner (Table 5.2).

R1R2R3R4R5R6R7R8R9


Chapter 5

130

148

Figure 5.2. Patient inclusion flow-chart

Information letter sent to eligible patients (n=933)

Excluded (n=723) ♦ Did not provide informed consent (n=717) ♦ Not meeting inclusion criteria (n=2) ♦ Other reasons (n=4)

Questionnaires with complete discrete choice experiments were received (n=161; 77%)

Analysed (n=151)

Questionnaire sent to the patients (n=210; 23%)

Contacted

Responders

Lost to follow-up (n=49) ♦ Did not return the questionnaire (n=39) ♦ Did not return a complete discrete choice experiment (n=10) (mean age 72 + 7 yrs and 60% males)

Excluded from analyses (n=10) ♦ Failed the dominant choice set (n=10) (mean age 70 + 10 yrs, 60% males, and 60% were of lower education)

Completers

Analyses

Figure 5.2. Patient inclusion flow-chart

Table 5.2. Patient characteristics per age groupCharacteristic All <75 years ≥75 years P-valueIncluded patients 151 106 45Mean age (SD) 68 (9.2) 64 (7.1) 79 (3.6)Females (%) 64 (42.4) 40 (37.7) 24 (53.3) 0.0761

Median BMI (IQR) 28 (26-32) 29 (27-33) 26 (24-29) 0.0002

Education (%) 0.4723

Lower educationa 86 (57.0) 59 (55.7) 27 (60.0) Middle educationb 44 (29.1) 34 (32.1) 10 (22.2) Higher educationc 17 (11.3) 11 (10.4) 6 (13.3) Other 4 (2.7) 2 (1.9) 2 (4.4)Smoking 0.0111

Current smokers 22 (14.6) 16 (15.1) 6 (13.3) Past smokers 75 (49.7) 60 (56.6) 15 (33.3) Non smokers 54 (35.8) 30 (28.3) 24 (53.3)Median quality of life (IQR)◦ 3 (3-5) 3 (3-4) 4 (3-5) 0.3512

Classes of prescribed blood pressure-lowering drugs (ATC code)*

Centrally acting antihypertensives (C02) 2 (1.3) 0 (0.0) 2 (4.4) 0.0893

Diuretics (C03) 49 (32.7) 31 (29.5) 18 (40.0) 0.2101

β-Blockers (C07) 87 (58.0) 56 (53.3) 31 (68.9) 0.0771

Calcium channel blockers (C08) 33 (22.0) 20 (19.1) 13 (28.9) 0.1821

Agents acting on the renin-angiotensin system (C09) 104 (69.3) 74 (70.5) 30 (66.7) 0.6431

Combination tablet‡ 27 (18.0) 16 (15.2) 11 (24.4) 0.1791

Classes of prescribed glucose-lowering drug (ATC code)*

Insulin (A10A) 33 (22.0) 27 (25.7) 6 (13.3) 0.0931



131

5

Biguanides (metformin) (A10BA) 135 (90.0) 96 (91.4) 39 (86.7) 0.3731

Sulfonamides (A10BB) 58 (38.7) 39 (37.1) 19 (42.2) 0.5581

Thiazolidinediones (A10BG) 1 (0.7) 1 (0.95) 0 (0.0) 1.0003

Dipeptidyl peptidase 4 inhibitors (A10BH) 13 (8.7) 9 (8.6) 4 (8.9) 1.0003

Combination Metformin and Sulfonamide (A10BD02) 1 (0.7) 1 (1.0) 0 (0.0) 1.0003

Liraglutide (A10BX07) 2 (1.3) 2 (1.9) 0 (0.0) 1.0003

Use of lipid-lowering drugs (%)* 0.0193

No lipid-lowering drug 27 (18.0) 13 (12.4) 14 (31.1) 1 lipid-lowering drug 117 (78.0) 88 (83.8) 29 (64.4) 2 lipid-lowering drugs 6 (4.0) 4 (3.8) 2 (4.4)Drug burden expressed as median number of chronic treatments from 8 anatomical chapters (IQR)ø

3 (3-4) 3 (3-4) 4 (3-4) 0.0252

High blood pressureHow serious do you think that having a high blood pressure is in general? (%)

0.5843

Very serious 20 (13.4) 13 (12.4) 7 (15.9) Reasonable serious 93 (62.4) 69 (65.7) 24 (54.6) A little serious 28 (18.8) 18 (17.1) 10 (22.7) Not serious 8 (5.4) 5 (4.8) 3 (6.8)How high was your systolic blood pressure during the last measurement conducted by your general practitioner or nurse practitioner? (%)

0.1621

<120 mmHg 10 (6.6) 7 (6.6) 3 (6.7) 120-139 mmHg 66 (43.7) 52 (49.1) 14 (31.1) 140-159 mmHg 47 (31.1) 32 (30.2) 15 (33.3) ≥160 mmHg 16 (10.6) 9 (8.5) 7 (15.6) I do not know 12 (8.0) 6 (5.7) 6 (13.3)Number of patients who report ever having experienced a symptom of high blood pressure (%)

36 (23.8) 25 (23.6) 11 (24.4) 0.9101

Blood pressure-lowering drugsNumber of drugs that the patients report to use for high blood pressure

0.1021

None 6 (4.0) 6 (5.7) 0 (0.0) One 79 (52.3) 55 (51.9) 24 (53.3) Two 38 (25.2) 29 (27.4) 9 (20.0) More than two 19 (12.6) 9 (8.5) 10 (22.2) I do not know 9 (6.0) 7 (6.6) 2 (4.4)Have you ever experienced a side effect of a blood pressure-lowering drug

0.5111

No 110 (72.9) 78 (73.6) 32 (71.1) Yes 24 (15.9) 18 (17.0) 6 (13.3) I do not know 17 (11.3) 10 (9.4) 7 (15.6)Beliefs about blood pressure-lowering drugs Median specific-necessity (IQR) 15 (13-18) 15 (13-18) 15 (13-19) 0.8912

Median specific-concerns (IQR) 12 (10-15) 13 (10-15) 12 (10-14) 0.0602

I prefer to leave decisions about my drugs to my general practitioner†

142 (94.0) 100 (94.3) 42 (93.3) 0.8111

a No education; elementary school; junior secondary vocational educationb Junior general secondary education; senior secondary vocational educationc Senior general secondary education; higher professional education; university educationSD = Standard deviation; IQR = Interquartile range; ATC = Anatomical Therapeutic Chemical◦ Measured with Cantril’s ladder [27] with a range of 1 (best possible life) – 10 (worst possible life)* N=150 (medication overview of one patient was not extracted at the time of data collection since the patient had not given informed consent yet)‡ ATC codes: C03EA01, C07BB02, C09BA03, C09BA04, C09BA06, C09BB04, C09DA01, C09DA03, C09DA04, C09DA06, C09DB02ø Drug burden was counted at the anatomical ATC level for the chapters: A, B, C, H, L, M, N, R (maximum of 8).† Statement adapted from [43]. Scored on a 6-point Likert scale and divided by (partially, totally) agree and (partially, totally) disagree. Number is presented for those who agree.1 Pearson χ2-test; 2 Mann-Whitney U test; 3 Fisher freeman-halton test

R1R2R3R4R5R6R7R8R9


Chapter 5

132

Influence of age on willingness to add a blood pressure-lowering drugOf the aged patients, 67% chose an additional drug in at least one of the ten presented choice sets. This percentage was significantly higher in non-aged patients (84%; P = 0.017). The same is reflected in the drug choice model where the constant is more negative in the aged than in the non-aged (Table 5.3). Having low or high concern beliefs did not influence whether or not patients chose at least once an additional blood pressure-lowering drug (Figure 5.3). Necessity beliefs, on the other hand, influenced the preferences of the non-aged patients. Non-aged patients with high necessity beliefs more often chose at least once an additional drug than non-aged patients with low necessity beliefs (90% versus 74% respectively; P = 0.040) (Figure 5.4).

151

Figure 5.3. Influence of concern beliefs on choosing an additional drug per age groups

Figure 5.4. Influence of necessity beliefs on choosing an additional drug or not per age group Influence of age on importance attached to drug attributes

Drug attributes that significantly influenced the choices for an additional blood pressure-lowering drug in both age groups were the risk of death within the next 5 years, the risk of experiencing ADEs, and the effect on the blood pressure (Table 5.3). For non-aged patients, the risk of limitations due to a stroke within the next 5 years also significantly contributed to the choice of an additional drug.


151


Figure 5.4. Influence of necessity beliefs on choosing an additional drug or not per age group Influence of age on importance attached to drug attributes

Drug attributes that significantly influenced the choices for an additional blood pressure-lowering drug in both age groups were the risk of death within the next 5 years, the risk of experiencing ADEs, and the effect on the blood pressure (Table 5.3). For non-aged patients, the risk of limitations due to a stroke within the next 5 years also significantly contributed to the choice of an additional drug.

Figure 5.4. Influence of necessity beliefs on choosing an additional drug or not per age group



133

5

Influence of age on importance attached to drug attributesDrug attributes that significantly influenced the choices for an additional blood pressure-lowering drug in both age groups were the risk of death within the next 5 years, the risk of experiencing ADEs, and the effect on the blood pressure (Table 5.3). For non-aged patients, the risk of limitations due to a stroke within the next 5 years also significantly contributed to the choice of an additional drug.

A sensitivity analysis in which the patients who failed the dominant choice set were included, revealed similar results (Appendix 6; supplemental table 2). In addition, sensitivity analyses including younger (<65 years) or older patients (≥80 years) showed the same direction for all coefficients but some coefficients became insignificant (Appendix 6; supplemental table 3).

Only the impact of the blood pressure-lowering effect turned out to be significantly different among the age groups (Table 5.3, interaction, P = 0.043). This finding indicates that the effect of an additional drug on the blood pressure level was seen as more important by non-aged patients than aged patients.

Table 5.3. Preferences of patients aged <75 years (non-aged) and ≥75 years (aged)Constant and attributes

<75 yearsa ≥75 yearsb P-value of the interaction between age groups and preferencesc

Coefficient (95% CI)

P-value Coefficient (95% CI)

P-value

Constant (additional drug)

-1.05 (-1.60 – -0.50)

0.000 -1.65(-2.52 – -0.77)

0.000 0.257

Blood pressure -0.09 (-0.11 – -0.08)

0.000 -0.06(-0.09 – -0.03)

0.000 0.043

Death within the next 5 years

-21.79(-29.59 – -13.99)

0.000 -24.43(-37.32 – -11.54)

0.000 0.731

Limitations heart attack

-9.13(-24.61 – 6.36)

0.248 -11.31(-36.83 – 14.22)

0.385 0.886

Limitations stroke

-30.22 (-45.83 – -14.61)

0.000 -15.71(-41.42 – 10.00)

0.231 0.344

Adverse drug events

-15.59 (-18.86 – -12.31)

0.000 -10.80(-16.02 – -5.58)

0.000 0.128

Additional tablet in the evening

0.13 (-0.08 – 0.34)

0.216 -0.10(-0.44 – 0.24)

0.578 0.264

Combination tablet

0.10(-0.11 – 0.31)

0.361 0.21(-0.12 – 0.54)

0.206 0.566

a Number of observations 3,180 (106 patients * 10 choice sets * 3 alternatives per choice set)b Number of observations 1,350 (45 patients * 10 choice sets * 3 alternatives per choice set)c Number of observations 4,530 (151 patients * 10 choice sets * 3 alternatives per choice set)

R1R2R3R4R5R6R7R8R9


Chapter 5

134

DiscussionThis study shows that aged patients with type 2 diabetes were less willing to add a blood pressure-lowering drug than non-aged patients when they had to imagine that their blood pressure was too high. Concern beliefs about the drugs did not seem to influence the willingness to add a drug. Among the non-aged patients, however, those with high necessity beliefs were more willing to add a blood pressure-lowering drug than those with low necessity beliefs. The effect of a drug on the blood pressure was more important for non-aged patients than aged patients. The effects on the risk of death within the next 5 years and experiencing ADEs were important drug characteristics for choosing a drug in both age groups. For non-aged patients, also the risk of limitations due to a stroke was important. Previously, it was found that patients with a limited life-expectancy have a decreased willingness to add treatment because they may value quality of life over life extension [12,15]. On the other hand, it was also found that a large proportion of such patients remain willing to undergo burdensome treatment for a small risk reduction of death [12]. Our study confirms the finding that aged patients are less willing to add a drug than non-aged patients. The finding that the drug effect on reducing the risk of death within the next 5 years was of similar importance for aged and non-aged patients may be surprising, but fits with the finding that many aged patients remain willing to undergo burdensome treatment for a risk reduction of death [12]. This might be explained by the fact that the aged group still perceived they had sufficient life-expectance. Our sensitivity analysis suggests that the drug effect on reducing the risk of death within the next 5 years may become less important in patients of 80 years or older. The number of patients in this group, however, was too small to draw firm conclusions. Previous studies have not directly compared the importance attached to various treatment outcomes between aged and non-aged patients. Our study revealed that most characteristics were similarly valued by aged patients in comparison to non-aged patients. Both groups attached a similar importance, for example, to the risk of ADEs. The only significant difference in importance of drug effects between aged and non-aged patients was the effect on the blood pressure. This could imply that aged patients are less willing to add a blood pressure-lowering drug because they believe that decreasing the blood pressure is of less importance. Whether this is influenced by the current advise of guidelines, and thus practitioners, to take a patient’s age or life-expectancy into account in setting blood pressure targets [44,45], remains to be determined. In both age groups, the choice for a blood pressure-lowering drug was not significantly influenced by the risk reductions in limitations in daily life due to a heart attack. This finding may be due to a more subjective interpretation of such an attribute compared to an attribute such as death [46]. However, the preferences of the non-aged patients



135

5

were influenced by a similar attribute, that is, the risk of limitations due to a stroke. Therefore, it seems that reducing potential limitations due to a heart attack from 7% to 5% was not decisive for the patients’ treatment choices, whereas a similar reduction for limitations due to stroke was. These findings are comparable with a previous study showing that patients are less willing to risk cognitive disability than physical disability [12]. For the aged patients, however, reducing potential limitations due to a stroke from 7% to 5% was not critical. Possibly, the presented limitations in our study may have been perceived by aged patients as an expected part of aging [47]. Presenting more severe problems, such as becoming dependent on others, might have revealed different results since maintaining independent is important for aged patients [48,49]. In our study, patients’ preferences for a blood pressure-lowering drug were not significantly influenced by the intake moment of the drug. A previous discrete choice experiment in patients with type 2 diabetes showed that the intake moment influences patient preferences, but that this is more important for patients who are taking less than five drugs a day compared to patients with five or more drugs per day [50]. This may explain the non-significant influence of the intake moment in our study since many patients were prescribed not only glucose-lowering and blood pressure-lowering drugs but also lipid-lowering drugs and drugs for other chronic diseases. This study was conducted in the north of the Netherlands, which includes a mostly caucasian population. Selection bias may have occurred since only 23% of the contacted patients gave informed consent and the percentage of females who responded was lower compared to the non-responders. Regarding age, however, there were no differences between responders and non-responders. The aged group included only 45 patients, which may have reduced the efficiency of the model. In the aged group, more blood pressure-lowering drugs and especially β-blockers were used than in the non-aged group. It is not clear whether this higher drug burden may have influenced the willingness to add a drug. There are also some strengths and limitations to a discrete choice experiment. A major strength is that it comes close to the trade-offs and choices that have to be made in real life, and can thus provide better estimates of the relative importance of different treatment characteristics than when asking patients to rate the importance of each characteristic separately. The evaluation of several treatment alternatives and attributes at one time reveals a rich source of data [26,51]. Moreover, the ‘no additional drug’ option was included which represents actual choices in practice. However, hypothetical situations have to be assessed which do not necessarily represent actual behaviour [52,53]. Some patients may have difficulty in making hypothetical choices. We excluded 10 patients from the analyses who failed the dominant choice set. There is discussion in the literature about the fairness of excluding such responders [54], but the sensitivity analysis in which these patients were included revealed similar

R1R2R3R4R5R6R7R8R9


Chapter 5

136

results. Another limitation may be that for most of the attributes only a reduction in the frequency of a risk was presented. Other information may be relevant for patients. For instance, patients may want information about the severity and duration of the ADEs. In conclusion, it is important to acknowledge that aged patients may be less willing to add a blood pressure-lowering drug than non-aged patients. The willingness to add a drug does not seem to be influenced by a patient’s concern beliefs but non-aged patients with high necessity beliefs were found to have a higher willingness to add a drug than those with low necessity beliefs. These findings underline the importance of discussing patients’ preferences even when they prefer to leave the final treatment decision to their general practitioner. When choosing a drug treatment, aged patients attach as much importance to reduce their risk of death within the next 5 years and of experiencing ADEs as non-aged patients. Therefore, treatment decisions in clinical practice should focus on quality of life as well as life extension in both age groups in which the individual patient’s preferences and willingness to add a drug should be taken into account.

AcknowledgementsThe study was funded by the Research Institute SHARE, University Medical Center Groningen, Groningen, The Netherlands.



137

5

References[1] Aristides M, Weston AR, FitzGerald P, Le

Reun C, Maniadakis N. Patient preference and willingness-to-pay for Humalog Mix25 relative to Humulin 30/70: a multicountry application of a discrete choice experiment. Value Health 2004;7(4):442-54.

[2] Vermeire E, Hearnshaw H, Van Royen P, Denekens J. Patient adherence to treat-ment: three decades of research. A com-prehensive review. J Clin Pharm Ther 2001;26(5):331-42.

[3] Mitchell AJ, Selmes T. Why don’t patients take their medicine? Reasons and solu-tions in psychiatry. Advances in Psychiatric Treatment 2007;13(5):336-46.

[4] Gascon JJ, Sanchez-Ortuno M, Llor B, Skid-more D, Saturno PJ, Treatment Compliance in Hypertension Study Group. Why hyper-tensive patients do not comply with the treatment: results from a qualitative study. Fam Pract 2004;21(2):125-30.

[5] Benson J, Britten N. Patients’ decisions about whether or not to take antihy-pertensive drugs: qualitative study. BMJ 2002;325(7369):873.

[6] Svensson S, Kjellgren KI, Ahlner J, Saljo R. Reasons for adherence with antihyperten-sive medication. Int J Cardiol 2000;76(2-3):157-63.

[7] Fried TR, Tinetti ME, Towle V, O’Leary JR, Iannone L. Effects of benefits and harms on older persons’ willingness to take medica-tion for primary cardiovascular prevention. Arch Intern Med 2011;171(10):923-8.

[8] Porzsolt F, Clouth J, Deutschmann M, Hipp-ler HJ. Preferences of diabetes patients and physicians: a feasibility study to identify the key indicators for appraisal of health care values. Health Qual Life Outcomes 2010;8:125.

[9] Hauber AB, Mohamed AF, Johnson FR, Falvey H. Treatment preferences and medi-cation adherence of people with Type 2 di-abetes using oral glucose-lowering agents. Diabet Med 2009;26(4):416-24.

[10] Polster M, Zanutto E, McDonald S, Conner C, Hammer M. A comparison of preferenc-es for two GLP-1 products--liraglutide and exenatide--for the treatment of type 2 dia-betes. J Med Econ 2010;13(4):655-61.

[11] Hiligsmann M, van Durme C, Geusens P, Dellaert BG, Dirksen CD, van der Weijden

T, et al. Nominal group technique to select attributes for discrete choice experiments: an example for drug treatment choice in osteoporosis. Patient Prefer Adherence 2013;7:133-9.


[13] Tsevat J, Dawson NV, Wu AW, Lynn J, Soukup JR, Cook EF, et al. Health values of hospital-ized patients 80 years or older. HELP Inves-tigators. Hospitalized Elderly Longitudinal Project. JAMA 1998;279(5):371-5.

[14] Sherbourne CD, Keeler E, Unutzer J, Lenert L, Wells KB. Relationship between age and patients’ current health state preferences. Gerontologist 1999;39(3):271-8.


[16] Dibonaventura MD, Wagner JS, Girman CJ, Brodovicz K, Zhang Q, Qiu Y, et al. Multi-nationalInternet-based survey of patient preference for newer oral or injectable Type 2 diabetesmedication. Patient Prefer Adherence 2010;4:397-406.

[17] Zanchetti A, Ruilope LM. Antihypertensive treatment in patients with type-2 diabe-tes mellitus: what guidance from recent controlled randomized trials? J Hypertens 2002;20(11):2099-110.

[18] van Hateren KJ, Landman GW, Kleefstra N, Houweling ST, van der Meer K, Bilo HJ. Time for considering other blood pressure target values in elderly patients with type 2 diabe-tes? Int J Clin Pract 2012;66(2):125-7.

[19] Stack RJ, Bundy C, Elliott RA, New JP, Gibson JM, Noyce PR. Patient perceptions of treat-ment and illness when prescribed multiple medicines for co-morbid type 2 diabetes. Diabetes Metab Syndr Obes 2011;4:127-35.

[20] Chan DC, Chen JH, Kuo HK, We CJ, Lu IS, Chiu LS, et al. Drug-related problems (DRPs) identified from geriatric medica-tion safety review clinics. Arch Gerontol Geriatr 2012;54(1):168-74.

R1R2R3R4R5R6R7R8R9


Chapter 5

138

[21] Leendertse AJ, Egberts AC, Stoker LJ, van den Bemt PM, HARM Study Group. Fre-quency of and risk factors for preventa-ble medication-related hospital admis-sions in the Netherlands. Arch Intern Med 2008;168(17):1890-6.

[22] de Vries ST, Keers JC, Visser R, de Zeeuw D, Haaijer-Ruskamp FM, Voorham J, et al. Medication beliefs, treatment complexity, and non-adherence to different drug class-es in patients with type 2 diabetes. J Psy-chosom Res 2014 2;76(2):134-8.

[23] Lancsar E, Louviere J. Conducting discrete choice experiments to inform healthcare decision making: a user’s guide. Pharma-coeconomics 2008;26(8):661-77.

[24] de Bekker-Grob EW, Ryan M, Gerard K. Discrete choice experiments in health eco-nomics: a review of the literature. Health Econ 2012;21(2):145-72.

[25] Ryan M. Discrete choice experiments in health care. BMJ 2004;328(7436):360-1.

[26] Reed Johnson F, Lancsar E, Marshall D, Kilambi V, Muhlbacher A, Regier DA, et al. Constructing experimental designs for discrete-choice experiments: report of the ISPOR Conjoint Analysis Experimental De-sign Good Research Practices Task Force. Value Health 2013;16(1):3-13.

[27] Cantril H. The pattern of human concerns. New Brunswick, NJ: Rutgers University Press; 1965.

[28] Kellett N, West F, Finlay AY. Conjoint analysis: a novel, rigorous tool for de-termining patient preferences for topi-cal antibiotic treatment for acne. A ran-domised controlled trial. Br J Dermatol 2006;154(3):524-32.

[29] Herbild L, Bech M, Gyrd-Hansen D. Estimat-ing the Danish populations’ preferences for pharmacogenetic testing using a discrete choice experiment. The case of treating de-pression. Value Health 2009;12(4):560-7.

[30] Darba J, Restovic G, Kaskens L, Balbona MA, Carbonell A, Cavero P, et al. Patient preferences for osteoporosis in Spain: a discrete choice experiment. Osteoporos Int 2011;22(6):1947-54.

[31] Shafey M, Lupichuk SM, Do T, Owen C, Ste-wart DA. Preferences of patients and phy-sicians concerning treatment options for relapsed follicular lymphoma: a discrete choice experiment. Bone Marrow Trans-plant 2011;46(7):962-9.

[32] Brown TM, Pashos CL, Joshi AV, Lee WC. The perspective of patients with haemo-philia with inhibitors and their care givers: preferences for treatment characteristics. Haemophilia 2011;17(3):476-82.

[33] Scalone L, Watson V, Ryan M, Kotsopoulos N, Patel R. Evaluation of patients’ prefer-ences for genital herpes treatment. Sex Transm Dis 2011;38(9):802-7.

[34] Kauf TL, Mohamed AF, Hauber AB, Fetzer D, Ahmad A. Patients’ willingness to accept the risks and benefits of new treatments for chronic hepatitis C virus infection. Patient 2012;5(4):265-78.

[35] Dwight Johnson M, Apesoa-Varano C, Hay J, Unutzer J, Hinton L. Depression treat-ment preferences of older white and Mex-ican origin men. Gen Hosp Psychiatry 2013;35(1):59-65.

[36] Tight blood pressure control and risk of macrovascular and microvascular compli-cations in type 2 diabetes: UKPDS 38. UK Prospective Diabetes Study Group. BMJ 1998;317(7160):703-13.

[37] Carter BL, Chrischilles EA, Rosenthal G, Gryzlak BM, Eisenstein EL, Vander Weg MW. Efficacy and safety of nighttime dos-ing of antihypertensives: review of the literature and design of a pragmatic clin-ical trial. J Clin Hypertens (Greenwich) 2014;16(2):115-21.

[38] Lewanczuk R, Tobe SW. More medications, fewer pills: combination medications for the treatment of hypertension. Can J Cardi-ol 2007;23(7):573-6.

[39] Proeftuin farmacie Groningen. K86, K87, Hypertensie [online]. http://proeftuinfar-maciegroningen.tmade.nl/qmfiles/Hyper-tensie.pdf. Accessed Okt 22, 2014.

[40] van Hateren KJ, Drion I, Kleefstra N, Groenier KH, Houweling ST, van der Meer K, et al. A prospective observational study of quality of diabetes care in a shared care setting: trends and age differences (ZODI-AC-19). BMJ Open 2012;2(4):e001387.

[41] Centraal Bureau voor de Statistiek [Cen-tral Bureau for Statistics]. Sterfte; kerncij-fers naar diverse kenmerken [Mortality; highlights to various characteristics]. Den Haag/Heerlen 19-9-2014.

[42] Horne R, Weinman J, Hankins M. The beliefs about medicines questionnaire: The devel-opment and evaluation of a new method for assessing the cognitive representa-



139

5

tion of medication, Psychology & Health 1999;14(1):1-24.

[43] Levinson W, Kao A, Kuby A, Thisted RA. Not all patients want to participate in decision making. A national study of public prefer-ences. J Gen Intern Med 2005;20(6):531-5.

[44] NICE National Institute for Health and Care Excellence. Hypertension. Clinical manage-ment of primary hypertension in adults [online]. http://www.nice.org.uk/guidance /cg127/resources/guidance-hypertension -pdf. Accessed Nov 7, 2014.

[45] Verenso. Multidisciplinaire Richtlijn Diabe-tes. Verantwoorde Diabeteszorg bij Kwets-bare Ouderen Thuis en in Verzorgings of Verpleeghuizen. Deel 1. [Multidisciplinary Guideline Diabetes. Responsible Diabetes Care in Vulnerable Elderly at Home and in Residential Care or Nursing Homes. Part 1]. Utrecht, the Netherlands, Verenso, 2011.

[46] Steel N. Thresholds for taking antihyper-tensive drugs in different professional and lay groups: questionnaire survey. BMJ 2000;320(7247):1446-7.

[47] Sarkisian CA, Hays RD, Mangione CM. Do older adults expect to age successfully? The association between expectations regard-ing aging and beliefs regarding healthcare seeking among older adults. J Am Geriatr Soc 2002;50(11):1837-43.

[48] Fried TR, Tinetti M, Agostini J, Iannone L, Towle V. Health outcome prioritization to elicit preferences of older persons with

multiple health conditions. Patient Educ Couns 2011;83(2):278-82.

[49] Fried TR, Tinetti ME, Iannone L, O’Leary JR, Towle V, Van Ness PH. Health out-come prioritization as a tool for decision making among older persons with multi-ple chronic conditions. Arch Intern Med 2011;171(20):1854-6.

[50] Hauber AB, Han S, Yang JC, Gantz I, Tunceli K, Gonzalez JM, et al. Effect of pill burden on dosing preferences, willingness to pay, and likely adherence among patients with type 2 diabetes. Patient Prefer Adherence 2013;7:937-49.

[51] Viney R, Lancsar E, Louviere J. Discrete choice experiments to measure consum-er preferences for health and healthcare. Expert Rev Pharmacoecon Outcomes Res 2002;2(4):319-26.

[52] Laba TL. Using Discrete Choice Experiment to elicit patient preferences for osteoporo-sis drug treatments: where to from here? Arthritis Res Ther 2014;16(2):106.

[53] Hauber AB, Arellano J, Qian Y, Gonzalez JM, Posner JD, Mohamed AF, et al. Patient preferences for treatments to delay bone metastases. Prostate 2014;74(15):1488-97.

[54] Ryan M, Watson V, Entwistle V. Rational-ising the ‘irrational’: a think aloud study of discrete choice experiment responses. Health Econ 2009;18(3):321-36.

Chapter 6

Medication beliefs, treatment complexity, and

non-adherence to different drug classes

in patients with type 2 diabetes

Sieta T. de Vries1 Joost C. Keers2,3 Rosalie Visser2,4 Dick de Zeeuw1

Flora M. Haaijer-Ruskamp1 Jaco Voorham1

Petra Denig1

Journal of Psychosomatic Research 2014;76(2):134-8.


2 Department of Endocrinology, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands

3 Van Swieten Research Institute, Martini Hospital, Groningen, The Netherlands 4 University of Groningen, University Medical Center Groningen,

The Lifelines Cohort Study, Groningen, The Netherlands

R1R2R3R4R5R6R7R8R9


Chapter 6

142

Abstract

ObjectiveTo assess the relationship of patients’ medication beliefs and treatment complexity with unintentional and intentional non-adherence for three therapeutic groups commonly used by patients with type 2 diabetes.

MethodsSurvey data about adherence (Medication Adherence Report Scale) and beliefs about medicines (Beliefs about Medicines Questionnaire) were combined with prescription data from the Groningen Initiative to ANalyse Type 2 diabetes Treatment (GIANTT) database. Patients were classified as being adherent, mainly unintentional non-adherent, or partly intentional non-adherent per therapeutic group (glucose-, blood pressure-, and lipid-lowering drugs). Treatment complexity was measured using the Medication Regimen Complexity Index, which includes the dosage form, dosing frequency and additional directions of taking the drug. Analyses were performed using Kruskal–Wallis and Mann–Whitney U tests. ResultsOf 257 contacted patients, 133 (52%) returned the questionnaire. The patients had a mean age of 66 years and 50% were females. Necessity beliefs were not significantly different between the adherers, mainly unintentional non-adherers, and partly intentional non-adherers (differences smaller than 5 points on a scale from 5 to 25). For blood pressure-lowering drugs, patients reporting intentional non-adherence had higher concern beliefs than adherers (8 point difference, P = 0.01). Treatment complexity scores were lower for adherers but similar for mainly unintentional and partly intentional non-adherers to glucose- and blood pressure-lowering drugs. ConclusionsTreatment complexity was related to non-adherence in general. Beliefs about necessity were not strongly associated with non-adherence, while patients’ concern beliefs may be associated with intentional non-adherence. However, the role of these determinants differs per therapeutic group.


Beliefs, treatment complexity and non-adherence

143

6

IntroductionAlthough a drug should be taken as prescribed to achieve its intended effectiveness, adherence to drug therapy is a well-known problem in clinical practice [1–3]. Patients can be intentional as well as unintentional non-adherent to their drug treatment [4–6]. Intentional non-adherence is seen as a conscious decision for not taking the drug as prescribed after balancing the pros and cons, whereas unintentional non-adherence is a more passive behaviour which is more strongly associated with demographics [3,6]. Non-adherence is influenced by many factors, including patient and treatment characteristics [3,7]. Of the modifiable factors, beliefs about a drug and treatment convenience or complexity are important predictors of non-adherence [7–10]. Within the belief domain it is relevant to distinguish between concern and necessity beliefs [11]. Concern beliefs are about the adverse consequences of taking a drug, whereas necessity beliefs are about the positive effects of a drug on someone’s health [12]. Little is known about the influence of these different beliefs on unintentional versus intentional non-adherence. Two studies showed that concern and necessity beliefs were associated with intentional non-adherence, whereas only one study found that concern beliefs were associated with unintentional non-adherence [4,13]. In these studies, however, people using different therapeutic groups were combined. Another study showed that the association between beliefs and types of non-adherence can differ across therapeutic groups [7].

Focusing on different therapeutic groups also has implications for the treatment complexity. Treatment complexity includes the number of drugs that have to be taken, the route of drug administration, dosing frequency, and additional directions of taking the drug [14]. Higher treatment complexity is associated with lower rates of optimal adherence [10]. Previous studies showed for instance higher adherence to a once-daily than a twice-daily regime [15,16] and a study using a composite score of drug administration, dosing frequency and additional directions found that patients with low complexity scores were more often adherent than patients with high complexity scores [17]. At present, it is not known how this association varies for unintentional or intentional non-adherence.

Patients with type 2 diabetes are often treated with drugs from multiple therapeutic groups, including glucose-, blood pressure-, and lipid-lowering drugs [18]. Previously, it was shown that patients with type 2 diabetes reported more often unintentional non-adherence to the glucose-lowering drugs than to the blood pressure- and lipid-lowering drugs, and that intentional non-adherence did not differ among the therapeutic groups [19]. The aim of the current study is to assess the role of different kinds of beliefs (necessity and concern) and treatment complexity on unintentional and intentional non-adherence, and whether this differs for glucose-, blood pressure-, and lipid-lowering drugs in patients with type 2 diabetes.

R1R2R3R4R5R6R7R8R9


Chapter 6

144

MethodsIn this study, cross-sectional survey data were collected in 2007, which were combined with prescription data collected in the Groningen Initiative to ANalyse Type 2 diabetes Treatment (GIANTT)-project [20]. The GIANTT-project is a regional initiative of healthcare professionals and researchers focusing on the primary care of patients with type 2 diabetes in the province of Groningen in the Netherlands. The study was carried out in accordance with the Code of Ethics of the World Medication Association (Declaration of Helsinki) for experiments involving humans. Ethical approval was not needed for this study, as determined by the Medical Ethics Committee of the University Medical Center Groningen in the Netherlands.

ParticipantsOf the 32 general practitioners (GPs) included in the GIANTT-project in 2006, 19 GPs (59%) agreed to recruit patients for this study. For these 19 GPs, we selected a total of 345 patients with type 2 diabetes from the GIANTT-database who had been prescribed an oral glucose-lowering drug in 2005. To recruit a balanced group of adherent and non-adherent participants, half of the patients were selected based on a medication possession ratio (MPR) <80% of their oral glucose-lowering drug indicating possible low adherence [21]. Of the 345 selected patients, 69 (20%) were excluded from the study by their GP because of: psychosocial problems (14), language issues (13), cognitive limitations (10), patient died (8), patient moved (6), GP expects unwillingness (4), serious comorbidity (3), admission to hospital or nursing home (3), or other reasons (8). The remaining 257 patients were contacted by mail, and those who gave informed consent received a survey composed of general questions and validated questionnaires assessing beliefs and adherence which were applied to glucose-, blood pressure-, and lipid-lowering drugs. Patients were asked to report the name of their glucose-lowering drugs, and when applicable, their blood pressure- and lipid-lowering drugs. In addition, patients were asked to indicate whether they self-measured their glucose levels, their blood pressure, and whether their lipid-levels had been measured without a GP order.

BeliefsPatients’ beliefs about the three therapeutic groups were assessed using the Beliefs about Medicines Questionnaire (BMQ) specific [12]. The BMQ contains 5 items about necessity beliefs (e.g. ‘My health at present depends on my glucose-lowering drugs’), and 5 items about concern beliefs (e.g. ‘I sometimes worry about becoming too dependent on the glucose-lowering drugs’). Participants indicate their agreement with each item on a five-point Likert scale, ranging from totally disagree to totally agree. Scores on items per subscale were summed, ranging from 5 (totally agree) to 25 (totally



145

6

disagree) for the necessity and concern subscales. Internal consistency was assessed using Cronbach’s α. For those patients included in the analyses of glucose-lowering drugs, the internal consistency was 0.721 and 0.777 for the necessity and the concern subscales, respectively. These values were 0.834 and 0.823, and 0.779 and 0.828 for the patients included in the analyses of respectively blood pressure-, and lipid-lowering drugs. Besides the assessment of concern and necessity beliefs, the necessity–concern differential was measured. This differential gives an indication of which beliefs the patient weighs more heavily [4].

Treatment complexityIn the questionnaire, patients reported which glucose-, blood pressure-, and lipid-lowering drugs they had used in the previous 3 months. Dosing information on the drugs was derived from the GIANTT-database. A treatment complexity score was computed for each reported drug using the Medication Regimen Complexity Index [14]. For each patient using more than one drug within a therapeutic group, the scores of the Medication Regimen Complexity Index were combined resulting in one complexity score per therapeutic group. The Medication Regimen Complexity Index takes into account the dosage form, dosing frequency and additional directions of taking the drug. For the dosage form, the scores 1 and 3 were used for tablets and injections, respectively. The dosing frequency was registered in the GIANTT-database. The following additional directions were included in the complexity score based on additional dosing information in the GIANTT-database: break or crush a tablet, intake of multiple units at one time, variable dosing, and alternating dosing. Two researchers (STdV and PD) independently computed the complexity score for each patient. The researchers agreed on 96%, 100% and 99% of the scores for the patients using glucose-, blood pressure-, and lipid-lowering drugs, respectively. All disagreements were solved by discussion between the researchers.

AdherenceAdherence was assessed using the Medication Adherence Report Scale (MARS) [22]. The MARS contains one item that reflects unintentional non-adherence (‘I forget to take my glucose-lowering drugs’) and four items that largely reflect different forms of intentional non-adherence (e.g. ‘I alter the dose of my glucose-lowering drugs’) [4,22]. Participants indicate how often each statement applied to them in the last 3 months on a five-point Likert scale ranging from always to never. Cronbach’s α values were 0.715, 0.595 and 0.699 for intentional non-adherence to respectively glucose-, blood pressure- and lipid-lowering drugs. The intentional non-adherence items were summed. Non-adherence was defined as a score of lower than the maximum of 5 for unintentional and

R1R2R3R4R5R6R7R8R9


Chapter 6

146

lower than the maximum of 20 for intentional non-adherence, indicating any degree of non-adherence.

Statistical analysesAnalyses were conducted per therapeutic group, including data from those patients who reported the name of their drug in the correct therapeutic group, who completed all MARS-questions, who had no more than one missing value at the belief questions, and for whom drug dosing information was available in the database. For the patients with one missing value for the belief questions, the value was imputed using the median value of the other patients for that item. Since the median values are used to test for differences between groups (see below), this method does not affect the median value of the whole sample for that item.

Per therapeutic group, patients were divided into being fully adherent, mainly unintentional non-adherent, or in part intentional non-adherent. This last group includes patients who report some form of intentional non-adherence but may also report to be unintentional non-adherent. Differences in patient characteristics between adherers, mainly unintentional non-adherers and partly intentional non-adherers were tested using Pearson χ²-tests, one-way analyses of variance, and Kruskal–Wallis tests, depending on the distribution of the variables. Associations between beliefs and (un)intentional non-adherence, and treatment complexity and (un)intentional non-adherence were tested using Kruskal–Wallis tests. A P-value < 0.05 was considered statistically significant. Mann–Whitney U tests with Bonferroni adjustment to correct for multiple testing (a P-value ≤ 0.01 was considered statistically significant) were used for subsequent testing for differences between two specific groups. All analyses were conducted using IBM SPSS Statistics version 20 (Armonk, New York, USA).

ResultsOf the 257 contacted patients, 133 (52%) returned the questionnaire. Half of these patients were female, and the mean age was 66 years (Table 6.1). Patients included in the analyses with the glucose-lowering drugs had longer diabetes duration than those that were not included. No other differences were found between patients included in the analyses per therapeutic group and those that were not included in the analyses (Appendix 7; supplemental table 1). Of the patients reporting non-adherence to their drugs, almost all patients reported to be at least unintentional non-adherent (Figure 6.1). Intentional non-adherers to glucose- or blood pressure-lowering drugs more often reported to alter their dose or take less than instructed than the intentional non-adherers to lipid-lowering drugs. Patients who were intentional non-adherent to their lipid-lowering drugs more often reported that they stopped taking the drug (Figure 6.1).



147

6

We observed no significant differences between fully adherent, mainly unintentional non-adherent and partly intentional non-adherent patients in age, gender, education level, and diabetes duration. However, intentional non-adherers to glucose-lowering drugs more often self-measured their glucose levels than adherent patients (P < 0.01) (Appendix 7; supplemental table 2). In addition, they had a higher body mass index than the unintentional non-adherers to these drugs (P < 0.01).

Table 6.1. Patient characteristics per therapeutic groupTotal

(N=133)

Glucose-lowering drugs (N=85)

Blood pressure-lowering drugs (N=67)

Lipid-lowering drugs (N=85)

Females (%) 66 (49.6) 38 (44.7) 34 (50.7) 39 (45.9)Mean age in years (SD) 66.3 (9.6) 65.8 (9.5) 65.9 (10.1) 65.7 (9.9)Education (%) Low education 70 (53.0) 41 (48.8) 32 (48.5) 42 (50.0) Middle education 29 (22.0) 23 (27.4) 17 (25.8) 22 (26.2) High education 25 (18.9) 15 (17.9) 13 (19.7) 14 (16.7) Other 8 (6.1) 5 (6.0) 4 (6.1) 6 (7.1)Mean BMI (SD) 29.3 (4.3) 29.0 (4.0) 29.7 (4.7) 29.2 (4.3)Median diabetes duration (IQR)

7 (4.0-10.0)

7 (5.0-11.0)

7 (5.0-11.0)

7 (3.0-10.0)

Measurement outside GPs office Yes (%) 22 (27.2) 18 (26.9) 8 (9.4)GPs = General practitioners; BMI = Body mass index; SD = Standard deviation; IQR = Interquartile range

Beliefs and non-adherenceFor all three therapeutic groups, no significant differences in necessity beliefs were found between the adherers and unintentional and intentional non-adherers (Table 6.2). For glucose-lowering drugs, the median necessity scores showed only 1 point difference, whereas this was 1.5 point for blood pressure-lowering drugs and 4 points for lipid-lowering drugs. In general, higher necessity beliefs were reported for the glucose-lowering drugs than for the blood pressure- and lipid-lowering drugs.Intentional non-adherers to glucose- and blood pressure-lowering drugs had more concerns about these drugs than the adherers and unintentional non-adherers, which was only statistically significant for the blood pressure-lowering drugs (P < 0.05). The median concern scores showed differences of 2.5 points for glucose-lowering drugs, 8 points for blood pressure-lowering drugs, and 0 points for lipid-lowering drugs. The significant difference for the blood pressure-lowering drugs was mainly due to the difference between the adherers and intentional non-adherers (P = 0.01).

R1R2R3R4R5R6R7R8R9


Chapter 6

148

The adherent and the unintentional non-adherent patients weighed the necessity of their drug use heavier than their concerns about these drugs, as shown by the positive necessity–concern-differential (Table 6.2). This finding applied for the three therapeutic groups. For the intentional non-adherers to blood pressure- and lipid-lowering drugs, however, concerns weighed more heavily than necessity.

Treatment complexity and non-adherenceTreatment complexity scores were lower for adherers than non-adherers to glucose- (P < 0.05) and blood pressure-lowering drugs (P < 0.05) (Table 6.2). For all three therapeutic groups, no differences in complexity scores were seen between the unintentional and intentional non-adherers (Table 6.2).

169

points for blood pressure-lowering drugs, and 0 points for lipid-lowering drugs. The significant difference for the blood pressure-lowering drugs was mainly due to the difference between the adherers and intentional non-adherers (P = 0.01).

The adherent and the unintentional non-adherent patients weighed the necessity of their drug use heavier than their concerns about these drugs, as shown by the positive necessity–concern-differential (Table 6.2). This finding applied for the three therapeutic groups. For the intentional non-adherers to blood pressure- and lipid-lowering drugs, however, concerns weighed more heavily than necessity. Treatment complexity and non-adherence

Treatment complexity scores were lower for adherers than non-adherers to glucose- (P < 0.05) and blood pressure-lowering drugs (P < 0.05) (Table 6.2). For all three therapeutic groups, no differences in complexity scores were seen between the unintentional and intentional non-adherers (Table 6.2).

Figure 6.1. Percentage of patients reporting different types of non-adherence per therapeutic group

Figure 6.1. Percentage of patients reporting different types of non-adherence per therapeutic group



149

6

Table 6.2. Beliefs and treatment complexity for glucose-, blood pressure-, and lipid- lowering drugs according to adherence

All Adherent Unintentionally non-adherent

Intentionally non-adherent

P-value*

Glucose-lowering drugs N = 85 N = 53 N = 22 N = 10Differential median (IQR) 8.0

(4.0-13.0)8.0 (3.0-11.0)

8.5 (5.8-15.0)

8.0 (0.5-13.8)

0.370

Median (IQR) of necessity beliefs (scale 5 – 25)

20.0 (17.0–22.0)

20.0 (16.5-22.0)

21.0 (18.8-23.0)

20.0(15.8-24.0)

0.152

Median (IQR) of concern beliefs (scale 5 – 25)

11.0 (9.0–14.0)

11.0 (9.0-14.0)

10.5 (8.8-13.3)

13.0(7.8-16.5)

0.516

Median (IQR) of treatment complexity scores

6 (3.0 – 7.5)

4 (3.0 – 7.0)

7 (5.5 – 7.3)

7.5 (2.8 – 13.0)

0.126²

Blood pressure-lowering drugs

N = 67 N = 53 N = 10 N = 4

Differential median (IQR) 5.0 (2.0-9.0)

5.0 (2.0-10.0)

7.0 (1.8-9.5)

-0.5 (-4.8-5.3)

0.152


17.0 (14.0-20.0)

17.0 (14.5-20.0)

15.5 (13.0-22.3)

17.0 (12.0-21.3)

0.855


11.0 (9.0-14.0)

11.0 (9.0-13.0)

9.0 (7.8-14.3)

17.0 (13.5-19.8)

0.037¹


3(2.0 – 5.0)

3(2.0 – 4.0)

4 (3.8 – 6.0)

4 (4.0 – 4.8)

0.050³

Lipid-lowering drugs N = 85 N = 67 N = 15 N = 3Differential median (IQR) 3.0

(0.0-7.0)4.0 (0.0-7.0)

2.0 (0.0-7.0)

-1.0† 0.459


15.0 (11.5-18.0)

15.0 (12.0-18.0)

14.0 (10.0-18.0)

11.0† 0.347


11.0(9.0-14.0)

11.0(9.0-14.0)

11.0 (8.0-14.0)

11.0† 0.779


2(2.0 – 2.0)

2(2.0 – 2.0)

2 (2.0 – 2.0)

2 (2.0 – 2.0)

0.884

IQR = Interquartile range; * = Kruskal-Wallis test; † = No IQR due to low numbers ¹ Significance due to difference between adherers and intentional non-adherers (P = 0.01) ² Significant difference between adherers and non-adherers (P = 0.04)³ Significant difference between adherers and non-adherers (P = 0.01)

R1R2R3R4R5R6R7R8R9


Chapter 6

150

DiscussionThis study shows that treatment complexity is related with non-adherence to glucose- and blood pressure-lowering drugs, which is similar for patients reporting mainly unintentional or partly intentional non-adherence. Patient beliefs have a more complex association with non-adherence. Concerns were higher in case of intentional non-adherence for blood pressure-lowering and glucose-lowering drugs but this finding was only significant for the blood pressure-lowering drugs. Beliefs about necessity did not show much difference among the adherence groups for glucose-lowering and blood pressure-lowering drugs. For lipid-lowering drugs, the difference in necessity beliefs among the adherence groups was larger but also non-significant. Concerns were more heavily weighed than necessity by intentional non-adherers to blood pressure- and lipid-lowering drugs, whereas necessity was more heavily weighed by intentional non-adherers to glucose-lowering drugs. Previous studies showed inconsistent associations between beliefs and (un)intentional non-adherence [4,13]. The current data suggests that this inconsistency may be due to differences across therapeutic groups.

Patients with type 2 diabetes differ in their concerns and beliefs about necessity across the three therapeutic groups. As has been observed before, necessity is believed to be higher for glucose-lowering drugs (median score 20) than for blood pressure-lowering drugs (median score 17) and lipid-lowering drugs (median score 15), whereas no differences are seen across the therapeutic groups for concerns, with median scores of 11 for all three groups [23]. These differences in necessity beliefs among the therapeutic groups may be explained using Leventhal’s self-regulatory model of illness cognitions [24]. Patients perceive more (personal) control, symptoms, and emotional distress in type 2 diabetes than either hypertension or hyperlipidemia [23].

Previously, it was found that for intentional non-adherers, concerns play a more important role than beliefs about the necessity of the drug [4]. This result was confirmed in the current study for the blood pressure-lowering drugs, where a negative necessity–concern differential corresponded with strong concern beliefs in the group of intentional non-adherers. For glucose-lowering drugs, our findings indicate a different mechanism with a positive necessity–concern differential and less strong concern beliefs in intentional non-adherers. This finding may be explained by a different type of intentional non-adherence for these drugs. We found that these patients often altered doses or took less than instructed, which could be driven by self-management. Self-measurement of glucose-levels was significantly related to intentional non-adherence to glucose-lowering drugs. One may doubt whether we should talk here of non-adherence, since the changes in drug use may be the result of the self-management. Consequently, intentional non-adherence to glucose-lowering drugs might be overestimated when self-management is not taken into account. For lipid-lowering drugs, a negative



151

6

necessity–concern differential was found for the intentional non-adherers. Concern beliefs were, however, similar in all adherence groups. Previously, it has been concluded that concerns about or experiences with adverse events are the most common reason for statin discontinuation [25]. Our findings suggest that also people who continue statin treatment may have similar concerns.

The current study confirms previous results that treatment complexity is related to non-adherence in general [10,17]. We did not observe any difference between patients who reported mainly unintentional non-adherence and those reporting partly intentional non-adherence. For lipid-lowering drugs, treatment complexity was not associated with non-adherence in general since complexity scores were low in all three groups.

Some limitations of this study need to be acknowledged. First of all, this is a cross-sectional study, which makes it impossible to draw causal inferences. Medication beliefs and treatment complexity can be expected to influence adherence but may also change in patients after they have become non-adherent. A self-report measure of adherence was used because self-report is the only method that can be used to distinguish between intentional and unintentional non-adherence [26,27]. Only the item on forgetting to take the drug is considered to reflect mainly unintentional non-adherence, whereas all other items are considered as reflecting largely intentional non-adherence [4,22]. The internal consistency of the intentional non-adherence items ranged from 0.6 to 0.7, indicating the need for better measures of intentional non-adherence. Currently, however, we lack better self-reported medication adherence measures [27]. The use of self-reported adherence measurement may lead to underestimations of non-adherence especially intentional non-adherence, because of socially desirable answering and recall bias when completing the questionnaire [26,28]. A study that compared self-reports of adherence with more objective instruments (e.g. pill counts, electronic monitors) found moderate to high agreement in adherence rates among the measures [29]. Although intentional non-adherence was assessed using four different types of non-adherence, only a small number of patients reported to be intentional non-adherent to the blood pressure- and lipid-lowering drugs. This low number limited the power to detect differences of less than 3 to 4 points on the belief scales for glucose- and blood pressure-lowering drugs and less than 5 points for lipid-lowering drugs as being significant. In addition, the results of intentional non-adherers should be interpreted with caution since patients who reported both unintentional and intentional non-adherence were classified as being intentional non-adherent. The low number of patients in this study made it impossible to classify this behaviour in a separate group. Possible selection bias of patients due to the moderate response rate could be another limitation of this study. Participating patients were, however, comparable in general patient characteristics to other patients with type 2 diabetes in Dutch primary care [30]. Patients with inadequate survey or

R1R2R3R4R5R6R7R8R9


Chapter 6

152

prescription data had to be excluded from the analyses. These patients did not differ from the other patients in age, gender, education level and body mass index, but for the glucose-lowering drugs included patients had longer diabetes duration. Furthermore, the use of other drugs then diabetes and cardiovascular risk management related drugs were not taken into account in the treatment complexity scores.

The strength of our study is that we evaluated the association between determinants and types of non-adherence for different therapeutic groups within the same population of patients. This approach prevents that differences in determinants are influenced by other differences in the patient population.

To conclude, addressing concerns about drugs appears to be more important than stressing the necessity of treatment in patients with diabetes. Concerns seem to be associated with intentional non-adherence to especially blood pressure-lowering drugs but not with unintentional non-adherence. Beliefs about necessity showed no clear association with either type of non-adherence. Treatment complexity was relevant for any non-adherence to glucose- and blood pressure-lowering drugs, and healthcare professionals should thus try to avoid complex regimens as much as possible. Finally, our study indicates that determinants do not only differ among types of non-adherence, but also differ across therapeutic groups. Fighting non-adherence asks for more than a one-size fits all approach.

AcknowledgementsThe study was funded by the Research Institute SHARE, University Medical Center Groningen, Groningen, The Netherlands and the Dutch Diabetes Foundation.



153

6

References[1] Vermeire E, Hearnshaw H, Van Royen P,

Denekens J. Patient adherence to treat-ment: three decades of research. A com-prehensive review. J Clin Pharm Ther 2001;26:331-42.

[2] Delamater AM. Improving patient adher-ence. Clin Diabetes 2006;24:71–7.

[3] Lehane E, McCarthy G. Intentional and unintentional medication non-adherence: a comprehensive framework for clinical research and practice? A discussion paper. Int J

Nurs Stud 2007;44:1468–77.[4] Clifford S, Barber N, Horne R. Understand-

ing different beliefs held by adherers, un-intentional nonadherers, and intentional nonadherers: application of the Necessi-ty–Concerns Framework. J Psychosom Res 2008;64:41–6.

[5] Schüz B, Marx C, Wurm S, Warner LM, Zie-gelmann JP, Schwarzer R, et al. Medication beliefs predict medication adherence in older adults with multiple illnesses. J Psy-chosom Res 2011;70:179–87.

[6] Wroe AL. Intentional and unintentional nonadherence: a study of decision making. J Behav Med 2002;25:355–72.

[7] Unni E, Farris KB. Determinants of different types of medication non-adherence in cho-lesterol lowering and asthma maintenance medications: a theoretical approach. Pati-ent Educ Couns 2011;83:382–90.

[8] Mann DM, Ponieman D, Leventhal H, Halm EA. Predictors of adherence to diabetes medications: the role of disease and med-ication beliefs. J Behav Med 2009;32:278–84.

[9] Rubin RR. Adherence to pharmacologic therapy in patients with type 2 diabetes mellitus. Am J Med 2005;118:27S–34S.

[10] Ingersoll KS, Cohen J. The impact of med-ication regimen factors on adherence to chronic treatment: a review of literature. J Behav Med 2008;31:213–24.

[11] Phatak HM, Thomas III J. Relationships between beliefs about medications and nonadherence to prescribed chronic medi-cations. Ann Pharmacother 2006;40:1737–42.

[12] Horne R, Weinman J, HankinsM. The Beliefs aboutMedicines Questionnaire: the devel-opment and evaluation of a new method for

assessing the cognitive representation of medication. Psychol Health 1999;14:1–24.

[13] Unni EJ, Farris KB. Unintentional non-ad-herence and belief in medicines in older adults. Patient Educ Couns 2011;83:265–8.

[14] George J, Phun YT, Bailey MJ, Kong DC, Stewart K. Development and validation of the Medication Regimen Complexity Index. Ann Pharmacother 2004;38:1369–76.

[15] Malmenas M, Bouchard JR, Langer J. Retro-spective real-world adherence in patients with type 2 diabetes initiating once-daily liraglutide 1.8 mg or twice-daily exenatide 10 mug. Clin Ther 2013;35:795–807.

[16] Dezii CM, Kawabata H, Tran M. Effects of once-daily and twice-daily dosing on ad-herence with prescribed glipizide oral therapy for type 2 diabetes. South Med J 2002;95:68–71.

[17] Pollack M, Chastek B, Williams SA, Moran J. Impact of treatment complexity on adher-ence and glycemic control: an analysis of oral antidiabetic agents. 2010;17:257–65.

[18] American Diabetes Association. Standards of medical care in diabetes—2013. Diabe-tes Care 2013;36:S11–66.

[19] Stack RJ, Bundy CE, Elliott RA, New JP, Gib-son M, Noyce PR. Intentional and uninten-tional non-adherence in community dwell-ing people with type 2 diabetes: the effect of varying numbers of medicines. Br J Dia-betes Vasc Dis 2010;10:148–52.

[20] Voorham J, Denig P. Computerized extrac-tion of information on the quality of diabe-tes care from free text in electronic patient records of general practitioners. J Am Med Inform Assoc 2007;14:349–54.

[21] Vink NM, Klungel OH, Stolk RP, Denig P. Comparison of various measures for as-sessing medication refill adherence using prescription data. Pharmacoepidemiol Drug Saf 2009;18:159–65.

[22] Horne R. The Medication Adherence Report Scale (MARS): a new measurement tool for eliciting patients’ reports of non-adher-ence. Unpublished working paper.

[23] Stack RJ, Bundy C, Elliott RA, New JP, Gibson JM, Noyce PR. Patient perceptions of treat-ment and illness when prescribed multiple medicines for co-morbid type diabetes. Diabetes Metab Syndr Obes 2011;4:127–35.

R1R2R3R4R5R6R7R8R9


Chapter 6

154

[24] Leventhal H, Diefenbach M, Leventhal EA. Illness cognition: using common sense to understand treatment adherence and af-fect cognition interactions. Cogn Ther Res 1992;16:143–63.

[25] Maningat P, Gordon BR, Breslow JL. How do we improve patient compliance and ad-herence to long-term statin therapy? Curr Atheroscler Rep 2013;15:291–8.

[26] Rees G, Leong O, Crowston JG, Lamoureux EL. Intentional and unintentional nonad-herence to ocular hypotensive treatment in patients with glaucoma. Ophthalmology 2010;117:903–8.

[27] Garfield S, Clifford S, Eliasson L, Barber N, Willson A. Suitability of measures of self-reported medication adherence for routine clinical use: a systematic review. BMC Med Res Methodol 2011;11:149–57.

[28] Holt EW, Muntner P, Joyce C, Morisky DE,Webber LS, Krousel-Wood M. Life events, coping, and antihypertensive med-ication adherence among older adults: the cohort study of medication adher-ence among older adults. Am J Epidemiol 2012;176:S64–71.

[29] Garber MC, Nau DP, Erickson SR, Aikens JE, Lawrence JB. The concordance of self-re-port with other measures of medication ad-herence: a summary of the literature. Med Care 2004;42:649–52.

[30] Sidorenkov G, Haaijer-Ruskamp FM, de Zeeuw D, Denig P. A longitudinal study Ex-amining adherence to guidelines in diabe-tes care according to different definitions of adequacy and timeliness. PLoS One 2011;6:e24278.

Summary and general discussion

R1R2R3R4R5R6R7R8R9



156

The focus of this thesis is on the patient perspective in the benefit-risk evaluation of drugs, particularly for patients with type 2 diabetes. Different aspects are highlighted, focusing on adverse drug event (ADE) assessment and on decisions to prescribe or take drug treatment. The first part of the thesis describes the development and validation of a patient-reported ADE questionnaire. The second part focuses on the role of patient characteristics and preferences on treatment decisions in clinical practice. Below, the main findings of the included studies are presented and implications of the findings for research and practice are discussed per part.

Part I. Development and validation of a patient-reported ADE questionnaire

In the past, the evaluation of the benefit-risk profile of a drug was mainly based on reports of healthcare professionals. Over time, direct patient-reporting of the benefits and risks of a drug has increased, since the added value of incorporating the patient perspective has generally been acknowledged [1-5]. In patient-reported outcome instruments, the patient is the direct source of information without an interpretation of their responses by a healthcare professional [6-8]. Many patient-reported outcome instruments have been developed to assess the effects of drug treatment [9], but a standard patient-reported outcome instrument to assess ADEs is lacking [3].

Therefore, the aims of the first part of the thesis were to:• develop a patient-reported ADE questionnaire;• assess the reliability and validity of this questionnaire.

Main findingsThe patient-reported ADE questionnaire is developed for research purposes and contains 1) questions about general patient characteristics, 2) drug use and diseases, 3) experienced ADEs using structured checklists based on body categories, and 4) additional questions about the nature and causality of the ADE (chapter 1).

The first step in the validation process was to assess the content validity of this new instrument (chapter 1). During the content validation, the understanding and interpretation of the items and answer options was assessed using cognitive debriefing interviews in 28 patients who used drugs for type 2 diabetes, asthma or chronic obstructive pulmonary disease (COPD). Items were adapted and answer options were added to improve the feasibility, understanding, and completeness of the questionnaire. After 14 revisions of the questionnaire, the content validity was confirmed. The final version contained 252 ADEs categorized in 16 body categories and 14 questions per



157

ADE regarding its nature and causality. All ADEs were linked to a Lowest Level Term of the Medical Dictionary for Regulatory Activities (MedDRA®).

For the content validation, a paper-based version of the questionnaire was used. Due to advantages of web-based questionnaires, such as direct storage of the data in a database and response checks [10-12], a web-based version of the questionnaire was made. In supplement I, a pilot study among 10 patients is presented showing sufficient user acceptance of this web-based version to allow its use in further studies assessing the reliability and validity of the instrument.

The web-based version was used to assess the test-retest reliability and feasibility of the questionnaire. In addition, the impact of the body category structure on the test-retest reliability and feasibility was assessed (chapter 1). A total of 135 patients receiving at least an oral glucose-lowering drug completed the questionnaire twice, with a one-week period in between. The test-retest reliability was sufficient for reporting any ADE, and similar ADEs at the primary System Organ Class level of the MedDRA®. However, the test-retest reliability of reporting the same ADE was insufficient. This finding implies that the questionnaire in its current form should not be used to quantify ADE rates at this specific level. The questionnaire was feasible for research purposes since around 75% of the patients found the questionnaire easy to use, and most of the patients who reported at least one ADE needed less than 60 minutes to complete the questionnaire. Finally, the use of the body category structure did not significantly influence the number of reported ADEs, the test-retest reliability, and the feasibility.

In addition, the construct and concurrent validity of the patient-reported ADE questionnaire were assessed (chapter 2). For this, the 135 patients completed two additional questionnaires. It was shown that patients who reported one or more ADEs in the patient-reported ADE questionnaire had a statistically significant and clinically relevant lower self-reported general quality of life and physical health than patients who did not report an ADE. This finding supports the construct validity of reporting any ADE versus no ADE in the questionnaire. The concurrent validity was demonstrated by the finding that the ADEs reported in the patient-reported ADE questionnaire as being associated with specific drugs were in 73% of the cases in agreement with the ADE information in the Summary of Product Characteristics of the drugs. This agreement increased to 76% when only cases with a patient-reported causality assessment score higher than or equal to the median were included. The agreement increased to 100% when only cases with a causality score higher than or equal to the third quartile were included. Additional concurrent validity assessment showed that ADEs which patients associated with metformin had sufficient positive predictive value (79%) when compared with metformin-related ADEs reported in an existing treatment satisfaction questionnaire. However, the sensitivity was insufficient (38%). This finding indicates that the patient-reported ADE questionnaire does not detect all experienced ADEs.

R1R2R3R4R5R6R7R8R9



158

During the cognitive debriefing interviews (chapter 1), it became clear that several patients found a recall period of 4 weeks relatively short for a patient-reported ADE questionnaire. Therefore, the concurrent validity was assessed in another study using a questionnaire with a recall period of 4 weeks and 3 months. Reported ADEs in the questionnaire were compared with ADEs reported in a 3-month daily diary (chapter 3). The study was completed by 78 patients on at least an oral glucose-lowering drug. The sensitivity (33% for both recall periods) and positive predictive value (10% and 51% for the 4-week and 3-month recall period respectively) in reporting ADEs at primary System Organ Class level of the MedDRA®, were low. In addition, the sensitivity was low when taking also secondary and tertiary System Organ Classes into account (33% for the 4-week recall period and 38% for the 3-month recall period), and when assessing the reports of specific ADEs (43% for the 4-week recall period and 41% for the 3-month period). Additional analyses showed that there may be differences in sensitivity among primary System Organ Classes since the sensitivity ranged from 50% for gastrointestinal disorders to 0% for metabolism and nutrition disorders, psychiatric disorders, and respiratory, thoracic and mediastinal disorders. Patients with more agreement in ADE reporting between the diary and the questionnaire seemed to be older and more often male.

Although the use of direct patient reporting is needed to increase the knowledge about the benefits and risks of a treatment [1,13], several types of biases may occur [14]. Supplement II presents some of the known biases of patient-reporting and less obvious problems that were encountered during the validation studies. It was shown that patients are not always consistent in their responses, may answer to questions which are not applicable, and may interpret items of previously validated questionnaires differently than intended. To reduce biases and problems, several solutions are proposed. Important is that patient-reported instruments are critically evaluated. This critical evaluation is an ongoing process that needs to be continued after the validity of the instrument has been demonstrated.

Implications for research and practiceImplications of these findings regarding questionnaire design and patients’ uncertainty of symptoms being an ADE will be subsequently discussed. Thereafter, the generaliza-bility of the findings will be discussed and future perspectives will be given.

Questionnaire design

Body category structureTo increase the feasibility of questionnaire completion, the 252 ADEs were divided in 16 body categories. Patients were asked to first check the relevant body categories and



159

then the ADE(s) within the selected body categories. A comparison of the questionnaire with the body category structure and without this structure showed that the number of reported ADEs and the test-retest reliability were similar for both versions (chapter 1). However, the feasibility was not improved by the structure, and almost half of the patients did not use this categorization or had difficulties in deciding in which body category their symptom was classified (chapter 1). On the other hand, some patients found the structure helpful and easy to use. Therefore, the body category structure was kept as supporting option in the paper-based version of the questionnaire. In the web-based version, however, the body category structure could not be used as supporting option, and patients were directly guided to the symptom checklists of selected body categories. For future use, it is proposed not to include such a structure in the web-based version of the questionnaire. Assessment of ADEs in checklistsIt was our aim to develop a checklist-based questionnaire since checklists are more sensitive in detecting potential ADEs compared to an open-ended question [15,16]. In the questionnaire, patients were additionally asked to describe the ADEs. Answers to this additional question revealed that patients often checked multiple symptoms related to one ADE. To improve the questionnaire, it is therefore advised to start with an open-ended question in which patients can cluster multiple, related symptoms and can indicate whether the symptoms are just symptoms or potential ADEs. Such a question should be accompanied by symptom lists to increase the sensitivity of detecting potential ADEs. A linkage of symptoms in a checklist to MedDRA®-terms is important to compare ADE data across studies [17]. The current linkage, where each ADE in the checklist was linked to a Lowest Level Term of the MedDRA®, was not optimal. Based on the ADE descriptions given by the patient, the ADE may be linked to different MedDRA® System Organ Classes (chapter 3). In an adapted version of the questionnaire, it would be advised to use advanced, interactive technology that can identify several overarching symptom descriptions based on the reported symptoms of the patient.

Our studies confirm the previously noted importance of involving patients in the construction of questionnaires [6,18,19]. The cognitive debriefing interviews with patients (chapter 1) revealed necessary insights in the lay-out and wording of the questionnaire. An important finding from these interviews is that ADE questionnaires should include the option for patients to distinguish ADEs from symptoms. When directly asking for ADEs, it turned out that patients also report symptoms which they do not see as real ADEs. Also adding the option ‘do not know’ is necessary since some patients are not sure about an association between the symptom and the drug.

R1R2R3R4R5R6R7R8R9



160

Taken together, an advanced, interactive questionnaire may be the way forward. It is proposed that such a questionnaire includes 1) an open-ended question in which patients can describe the symptoms and can indicate whether the symptoms are symptoms or potential ADEs, 2) symptom lists to complement the open-ended question, and 3) an interactive system which presents some possible overarching terms for the symptom descriptions given by the patients. The patient should select the relevant term. These terms should be directly linked to a MedDRA® System Organ Class. Such adaptations are likely to increase the validity of the questionnaire.

Patient’s uncertainty of symptoms being an ADE Although it has been shown that patients are able to distinguish ADEs from symptoms [20,21], several findings from the studies in this thesis indicate that patients are not always certain about a symptom being an ADE: - in the patient-reported ADE questionnaire, patients sometimes checked a symptom

as a symptom at one point in time but as a possible ADE at another time (chapter 1);- in around half of the cases, patients did not mention a potential drug that they

believed was causing the ADE or they were not very sure about the causal relationship (chapter 1 – 3);

- reported symptoms in the diary (chapter 3) were sometimes indicated as an ADE and sometimes as ‘do not know’ whether the symptom was a symptom or an ADE.

This uncertainty has also been reported by others [22,23]. In the questionnaire, patients were asked to report the drug that they relate to the ADE. In chapter 2 it was shown that the agreement between such reported ADE-drug associations and the information in the Summary of Product Characteristics of the drug, increased when higher patient-reported causality scores were taken into account. This finding suggests that a patient-reported causality assessment score may be necessary in assessing patient-reported ADEs in observational studies. Opponents of direct patient-reporting may interpret the uncertainty of patients as indicating that such methods reveal inadequate information. However, ADE uncertainty has also been demonstrated in healthcare professional reports [24,25]. Therefore, the uncertainty can be understood as a problem subjected to ADE assessment and management in general (intermezzo).

GeneralizabilityThe generalizability of our findings may be limited in several ways. Firstly, the response rates in the studies was relatively low (around 20% in chapter 1 and chapter 2; and only 113 completed consent forms were returned after sending more than 1.150 letters in chapter 3). The responders were slightly younger than the non-responders when assessing the test-retest reliability and feasibility (chapter 1) and construct and



161

concurrent validity (chapter 2). This difference is likely to be due to the use of the web-based version of the questionnaire [26,27].

Furthermore, patients included in the studies had been prescribed at least an oral drug for type 2 diabetes (chapter 1 – 3) or asthma/COPD (chapter 1). This selection influences the type of ADEs that are reported and implies that the content validation cannot be guaranteed for all the ADEs in the questionnaire. However, patients with type 2 diabetes often use multiple drugs and have multiple diseases [28]. The questionnaire was not restricted to specific drugs or diseases. Polypharmacy and comorbidity are likely to increase the number and type of ADEs reported by patients but also the difficulty in relating them to a (specific) drug [22]. This difficulty may have negatively influenced the construct and concurrent validity of the questionnaire. The validity of an ADE questionnaire is expected to be better for serious ADEs and for ADEs that patients can distinguish more clearly from the symptoms related to the disease or occurring also as part of normal life or aging.

The validity of the questionnaire may further be influenced by the mode of questionnaire administration. Although the administration is not expected to affect the interpretation of the questions [29], other validity aspects (e.g. test-retest reliability) may differ between a web-based and a paper-based version of the questionnaire due to, for instance, differences in visual aspects and respondent required actions [11,30,31]. However, a meta-analysis showed a high agreement between both methods which indicates that they provide similar results for patient-reported outcomes [32].

It should be noted that the validation studies were conducted in observational settings. The questionnaire was developed to be used in observational studies and clinical trials, but the validity assessment in an observational setting cannot be generalized to the clinical trial setting. In clinical trials, patients are often new users and blinded to drug treatment which differs from observational studies in which patients know that they are using the drug and are familiar with it.

Future perspectivesBased on the findings of this part of the thesis, it is clear that the patient-reported ADE questionnaire needs further adaptations. Suggestions for adaptations have been presented. It should be noted that the validity and reliability of reporting ADEs in such an adapted version should be demonstrated in future studies.

In addition, it is advised to assess the instrument’s sensitivity to change. The ability of the instrument to detect within-individual change over time is an important validity aspect of an instrument [6,18,33]. This importance especially applies to a generic instrument such as the patient-reported ADE questionnaire since concerns have been documented that a generic instrument may negatively influence an instrument’s sensitivity to change [34,35].

R1R2R3R4R5R6R7R8R9



162

The use of a Dutch version of the questionnaire restricts the use of the questionnaire to Dutch-speaking countries. For a broader use of a patient-reported ADE questionnaire it is advised to translate the questionnaire in other languages. However, it should be noted that the use of the questionnaire in other languages requires pilot testing using cognitive debriefing methods [36].

The validity assessment of the questionnaire in a clinical trial setting is also advised. As discussed previously, the validity may differ between observational studies and clinical trials and is therefore needed to assess.

Finally, future studies are advised to assess whether there are differences in the validity among classes of ADEs and specific ADEs, whether or not patient characteristics influence the validity of the questionnaire, and whether or not the additional questions per ADE are valid. This information reveals whether or not all ADEs can be assessed by direct patient reporting and whether or not all patients can be involved in direct patient reporting.

In the end, a valid questionnaire which can be applied in observational studies and clinical trials is important to reveal information about ADEs experienced by patients. Such information can be used by regulatory authorities in the benefit-risk evaluation of drugs and by healthcare professionals and patients to make better informed decisions about preferred treatments [4,37].

Intermezzo

The intermezzo can be seen as a bridge between the two parts in this thesis. It describes a case-report on the assessment and management of ADEs in clinical practice from a patient’s perspective. ADE assessment can be complex and there are different ways to deal with the perspective of the patient. The assessment and management of ADEs in clinical practice may be improved by incorporating the perspective of both the patient and the healthcare professional.

Part II. The role of patient characteristics and preferences on treatment decisions in clinical practice

In clinical practice, the patient perspective is important to apply patient-centred care. In patient-centred care, treatment decisions and goals are individualised for which patient preferences and medication taking behaviour, and clinical aspects should be taken into account [38,39]. Patient-centred care has been recommended in guidelines of, for instance, the prevention and treatment of diabetes [40-42]. One related aspect is to take a patient’s age or life-expectancy into account when setting treatment goals [42,43]. Evidence of long-



163

term effects in aged patients is lacking [43-45] and patient preferences regarding drugs may differ among age groups [46,47]. Currently, little is known about the influence of age on actual prescribing behaviour. Furthermore, although non-adherence is common [48], there is uncertainty about the best approach to improve drug adherence in specific populations, such as patients with comorbidity [49]. More insight in the underlying processes of different types of non-adherence, especially in patients who need to take multiple drugs for different indications, may contribute to better tailored interventions for improving drug adherence [50].

The aims of the second part of this thesis were to provide insight in: • the decisions to start or intensify treatment with special attention for different

patient age groups;• the influence of age and medication beliefs on patients’ drug preferences;• the role of medication beliefs and treatment complexity on patients’ non-adherence

to drugs.

Main findingsIn chapter 4, potential under- and overtreatment over time of glucose- and blood pressure-lowering treatment was assessed for different patient age groups. In particular, the influence of the introduction of performance measures in 2008 was assessed in a dynamic cohort study using data from the Groningen Initiative to ANalyze Type 2 diabetes Treatment (GIANTT) database. Overtreatment at baseline was 7.4% for glucose-lowering treatment and 15.9% for blood pressure-lowering treatment. Undertreatment was more common with 49.2% for glucose-lowering treatment and 60.7% for blood pressure-lowering treatment. The introduction of performance measures reduced undertreatment for blood pressure-lowering treatment, which did not correspond with an increase in overtreatment. It seemed that the performance measures had little impact on improving glucose-regulating treatment. In addition, it was found that potential undertreatment –as defined by non-age specific recommendations– was more common in aged patients. This finding implies that healthcare professionals seem to be more restrictive in prescribing drugs in these patients. This restrictive prescribing behaviour possibly reflects prevailing concerns about the need for intensive drug treatment in aged patients as well as less willingness of aged patients to take additional drugs. In the study period, however, this was not yet incorporated in the guideline recommendations. Differences between non-aged and aged patients in their treatment preferences was assessed in chapter 5. More specifically, it was evaluated whether age affects 1) the patients’ willingness to add a blood pressure-lowering drug and 2) the importance they attach to specific treatment characteristics. In this study, 151 patients on at least

R1R2R3R4R5R6R7R8R9



164

an oral glucose- and blood pressure-lowering drug completed a survey with a discrete choice experiment in which they had to imagine that their blood pressure level was uncontrolled. The number of patients who were willing to add a blood pressure-lowering drug was significantly lower in aged patients (67%) than in non-aged patients (84%). In both aged and non-aged patients, the treatment’s effect on the risk of death within the next 5 years, on the blood pressure level, and on the risk of experiencing ADEs were important for choosing a drug. The effect on the risk of limitations due to a stroke was additionally important for non-aged patients. It turned out that the effect on the blood pressure level was less important for aged patients than for non-aged patients. An exploration of the role of medication beliefs on the association between age and patients’ willingness to add a blood pressure-lowering drug revealed that concern beliefs did not influence the association. Preferring additional drug treatment was, however, more common in non-aged patients with high necessity beliefs than non-aged patients with low necessity beliefs. A patient’s medication beliefs (e.g. beliefs regarding medication necessity and concern beliefs) are expected to be associated with non-adherence to a drug treatment. Non-adherence to a treatment can be intentional and unintentional [51-53], and may also be influenced by treatment complexity. In chapter 6, the association between medication beliefs and treatment complexity on intentional and unintentional non-adherence was assessed for glucose-, blood pressure-, and lipid-lowering drugs. These associations were studied within one group of patients to explore differences across therapeutic groups. In the study, 133 patients with type 2 diabetes completed a survey about adherence and beliefs towards these therapeutic groups. These data were combined with prescription data from the GIANTT-database to assess treatment complexity. Necessity beliefs did not significantly differ between adherers, unintentional non-adherers, and intentional non-adherers (differences smaller than 5 points on a scale from 5 to 25). Concern beliefs were higher in intentional non-adherers, but only significantly for the blood pressure-lowering drugs (8 point difference). Treatment complexity was related to both intentional and unintentional non-adherence for glucose- and blood pressure-lowering drugs. For lipid-lowering drugs the complexity was low in general. These findings imply that concern beliefs and treatment complexity are important for adherence. However, associations between concern beliefs and intentional non-adherence may differ among therapeutic groups.

Implications for research and practiceImplications of these findings regarding patient perspectives in treatment decisions and tailored interventions to improve treatment adherence will be subsequently discussed. Finally, the generalizability of the findings of the second part of this thesis will be discussed and future perspectives will be given.



165

Patient perspectives in treatment decisions Over time, several improvements have been observed in the care for patients with type 2 diabetes, which may have been stimulated by incorporating performance measures [54-56]. Previously, concerns have been raised that the introduction of performance measures can stimulate potential overtreatment, especially in patients who do not need aggressive treatment [57,58]. The study conducted in chapter 4 did not support these concerns since levels of potential overtreatment were relatively stable over time in all age groups.

Healthcare professionals seemed to be more restrictive in prescribing drugs in aged patients (chapter 4). A more restrictive prescribing in aged patients may be due to the lack of evidence of long-term effects in aged patients [43,44] and/or differences in patient preferences [46,47]. Chapter 5 showed indeed that preferences for drug treatment differ among age groups. The current perspective is that many aged or frail patients are likely to value quality of life over life extension [59-61]. This perspective is supported by the finding that aged patients are less willing to add a blood pressure-lowering drug than non-aged patients (chapter 5). However, the number of aged patients who were willing to add such a drug was still quite high (67%). Furthermore, the choices of both age groups for an additional drug were similarly influenced by quality of life aspects such as the risk of ADEs. The treatment’s effect on the risk of death within the next 5 years was also of similar importance between the age groups. Taken together, treatment decisions in clinical practice should focus on quality of life as well as life extension in both age groups, and the results underline the importance to tailor decisions to the individual patient’s preferences and willingness to take additional drugs.

The finding in chapter 5 that the effect on the blood pressure was less important for aged patients may indicate that aged patients prefer less strict targets than non-aged patients. This finding may have been driven by the advice in current guidelines and also the healthcare professionals to take a patient’s age or life-expectancy into account in setting blood pressure targets [42,62]. Tailored interventions to improve adherenceA drug should be taken as prescribed to have its intended effects, but non-adherence to drug treatment is a common problem in clinical practice [63-65]. Several factors have been associated with non-adherence [64,66] but no intervention can be expected to be effective across all patients, conditions and settings [49]. A meta-analysis showed that necessity and concern beliefs are associated with non-adherence but that these associations may differ among patients with different diseases [50]. In chapter 6 it was shown that even within one population, associations between medication beliefs and adherence may differ across therapeutic groups. This finding suggests that even for one patient several

R1R2R3R4R5R6R7R8R9



166

intervention approaches may be needed to improve adherence to all drugs needed for different indications. A review of interventions to improve adherence showed indeed that effective interventions in patients with long-term diseases were often complex [67]. GeneralizabilityThe studies in the second part of this thesis included only patients with type 2 diabetes. Type 2 diabetes is a chronic disease and although the incidence in younger people is increasing, it is typically a disease of older people [43]. It is expected that the findings differ for patients with other diseases since levels of overtreatment (chapter 4) and possible determinants of non-adherence (chapter 6) even differed among therapeutic groups within the study population. In chapter 5, it was shown that a large number of patients were willing to add a blood pressure-lowering drug and that the effect of the drug on life extension, experiencing ADEs and lowering the blood pressure is important for choosing a drug. Previous studies showed that patients with type 2 diabetes placed more importance on lowering glycohemoglobin (HbA1c) levels than on lowering the blood pressure [46,68]. Therefore, patients’ drug preferences are also expected to differ among therapeutic group. In chapter 4 and chapter 5, the results were presented according to different patient age groups. Aged patients were in both chapters defined as older than 75 years. The overall findings may be different when a higher age level would be used. The sensitivity analyses using different cut-off levels did not show a significant difference but this may be due to a lack of power (chapter 5). In both chapters, age was used as a proxy for patients with a limited life-expectancy. Age seemed to be a reasonable proxy (chapter 5), but future studies are needed to reveal better insight in other definitions of frail patients with a limited life-expectancy.

Future perspectivesNo support was found for concerns about an increase in potential overtreatment after the introduction of performance measures. Moreover, levels of potential undertreatment were much higher than overtreatment throughout the study period. High levels of undertreatment, especially in aged patients, have also been shown by others [69]. There may be good reasons to deviate from guideline recommendations for an individual patient, such as the preferences of a patient. However, percentages of undertreatment around 50-60% in patients with elevated HbA1c or systolic blood pressure levels suggest clinical inertia. This phenomenon has received much attention related to the treatment of diabetes and cardiovascular diseases, but it remains difficult to distinguish appropriate inaction from true clinical inertia [70]. Additional studies are advised to assess reasons for deviating from guideline recommendations at the individual patient level.



167

Looking at the role of the patient, one might speculate how medication beliefs affect the three components of medication adherence, that is, initiation, implementation and persistence [71]. Our studies showed that, for blood pressure-lowering drugs, concern beliefs were associated with intentional non-adherence and that necessity beliefs were to some extent associated with willingness to add such a drug. Future studies may focus on this difference in impact of medication beliefs on aspects of medication taking behaviour. Furthermore, which benefits and risks are important for each patient needs to be defined. In clinical practice, the benefits and risks of a drug and available treatment options are increasingly discussed using decision aids [72]. These decision aids use individualized benefit-risk information. Future studies are needed to compare the benefit-risk evaluation of patients with actual treatment decisions. In the end, it is all about the patient. The patient is the one experiencing ADEs, the one with treatment preferences, the one with specific beliefs, the one who is taking a drug as prescribed or not, and so one. Therefore, it is important to take the perspective of the patient into account in evaluating the benefits and risks of a drug and setting goals, making decisions, and evaluate decisions in clinical practice.

R1R2R3R4R5R6R7R8R9



168

References[1] Blenkinsopp A, Wilkie P, Wang M, Rout-

ledge PA. Patient reporting of suspected ad-verse drug reactions: a review of published literature and international experience. Br J Clin Pharmacol 2007;63(2):148-56.

[2] Black N, Jenkinson C. How can patients’ views of their care enhance quality im-provement? BMJ: British Medical Journal 2009;339(7714):202-5.

[3] Basch E, Reeve BB, Mitchell SA, Clauser SB, Minasian LM, Dueck AC, et al. Devel-opment of the National Cancer Institute’s patient-reported outcomes version of the common terminology criteria for adverse events (PRO-CTCAE). J Natl Cancer Inst 2014;106(9).

[4] Anker SD, Agewall S, Borggrefe M, Calvert M, Jaime Caro J, Cowie MR, et al. The im-portance of patient-reported outcomes: a call for their comprehensive integration in cardiovascular clinical trials. Eur Heart J 2014;35(30):2001-9.

[5] Weingart SN, Gandhi TK, Seger AC, Seger DL, Borus J, Burdick E, et al. Patient-re-ported medication symptoms in primary care. Arch Intern Med 2005;165(2):234-40.

[6] US Department of Health and Human Ser-vices, FDA, Center for Drug Evaluation and Research Center for Biologics Evaluation and Research Center for Devices and Ra-diological Health. Guidance for industry patient-reported outcome measures: use in medical product development to support labeling claims [online]. http://www.fda.gov/downloads/Drugs/GuidanceCompli-anceRegulatory Information/Guidances/UCM193282.pdf. Accessed Oct 28, 2014.

[7] Frost MH, Reeve BB, Liepa AM, Stauffer JW, Hays RD, Mayo/FDA Patient-Reported Out-comes Consensus Meeting Group. What is sufficient evidence for the reliability and validity of patient-reported outcome meas-ures? Value Health 2007;10(Suppl 2):S94-105.

[8] Patrick DL, Burke LB, Powers JH, Scott JA, Rock EP, Dawisha S, et al. Patient-report-ed outcomes to support medical prod-uct labeling claims: FDA perspective. Value Health 2007;10(Suppl 2):S125- 37.

[9] Patient-reported outcome and quality of life instrument database [online]. http://www.proqolid.org. Accessed Oct 28, 2014.

[10] Eysenbach G, Wyatt J. Using the Internet for surveys and health research. J Med Internet Res 2002;4(2):E13.

[11] Wyatt JC. When to use web-based surveys. J Am Med Inform Assoc 2000;7(4):426-9.

[12] van Gelder MMHJ, Bretveld RW, Roeleveld N. Web-based Questionnaires: The Future in Epidemiology? American Journal of Ep-idemiology 2010;172(11):1292-8.

[13] NHS. Guidance on the routine collection of Patient Reported Outcome Measures (PROMs) [online]. http://www.mstrust.org.uk/competencies/downloads/NHS-PROMS.pdf. Accessed Oct 28, 2014.

[14] Choi BC, Pak AW. A catalog of biases in questionnaires. Prev Chronic Dis 2005;2(1):A13.

[15] Bent S, Padula A, Avins AL. Brief commu-nication: Better ways to question patients about adverse medical events: a rand-omized, controlled trial. Ann Intern Med 2006;144(4):257-61.

[16] Sheftell FD, Feleppa M, Tepper SJ, Rapoport AM, Ciannella L, Bigal ME. Assessment of adverseevents associated with triptans--methods of assessment influence the re-sults. Headache 2004;44(10):978-82.

[17] US Department of Health and Human Ser-vices, FDA, Center for Drug Evaluation and Research Center for Biologics Evaluation and Research. Guidance for industry: Pre-marketing risk assessment [online]. http://www.fda.gov/downloads/RegulatoryInfor-mation/Guidances/ UCM126958.pdf. Ac-cessed Oct 28, 2014.

[18] Revicki DA, Gnanasakthy A, Weinfurt K. Documenting the rationale and psychomet-ric characteristics of patient reported out-comes for labeling and promotional claims: the PRO Evidence Dossier. Qual Life Res 2007;16(4):717-23.

[19] Terwee CB, Bot SD, de Boer MR, van der Windt DA, Knol DL, Dekker J, et al. Quality criteria were proposed for measurement properties of health status questionnaires. J Clin Epidemiol 2007;60(1):34-42.

[20] Krska J, Anderson C, Murphy E, Avery AJ. How Patient Reporters Identify Adverse Drug Reactions: A Qualitative Study of Re-



169

porting via the UK Yellow Card Scheme. Drug Saf 2011;34(5):429-36.

[21] De Smedt RH, Denig P, Haaijer-Ruskamp FM, Jaarsma T. Perceived medication ad-verse effects and coping strategies report-ed by chronic heart failure patients. Int J Clin Pract 2009;63(2):233-42.

[22] Chaipichit N, Krska J, Pratipanawatr T, Uchaipichat V, Jarernsiripornkul N. A qualitative study to explore how patients identify and assess symptoms as adverse drug reactions. Eur J Clin Pharmacol 2014;70(5):607-15.

[23] Britten N. Medication errors: the role of the patient. Br J Clin Pharmacol 2009;67(6):646-50.

[24] Forster AJ, O’Rourke K, Shojania KG, van Walraven C. Combining ratings from mul-tiple physician reviewers helped to over-come the uncertainty associated with ad-verse event classification. J Clin Epidemiol 2007;60(9):892-901.

[25] Forster AJ, Taljaard M, Bennett C, van Wal-raven C. Reliability of the peer-review pro-cess for adverse event rating. PLoS One 2012;7(7):e41239.

[26] Härmark L, Alberts S, van Puijenbroek E, Denig P, van Grootheest K. Representative-ness of diabetes patients participating in a web-based adverse drug reaction monitor-ing system. Pharmacoepidemiol Drug Saf 2013;22(3):250-5.

[27] Schiotz M, Bogelund M, Willaing I. Chal-lenges using online surveys in a Danish population of people with type 2 diabetes. Chronic Illn 2012;8(1):56-63.

[28] Teljeur C, Smith SM, Paul G, Kelly A, O’Dowd T. Multimorbidity in a cohort of patients with type 2 diabetes. Eur J Gen Pract 2013;19(1):17-22.

[29] Leung DYP, Kember D. Comparability of data gathered from evaluation questionnaires on paper and through the internet. Res High Educ 2005;46(5):571-91.

[30] Manfreda KL, Batagelj Z, Vehovar V. Design of web survey questionnaires: Three basic experiments. J Comput Mediat Commun 2002;7(3).

[31] Couper MP. Review: Web Surveys: A Review of Issues and Approaches. The Public Opin-ion Quarterly 2000;64(4):464-94.

[32] Gwaltney CJ, Shields AL, Shiffman S. Equiv-alence of electronic and paper-and-pencil administration of patient-reported out-

come measures: a meta-analytic review. Value Health 2008;11(2):322-33.

[33] Revicki DA, Cella D, Hays RD, Sloan JA, Lenderking WR, Aaronson NK. Responsive-ness and minimal important differences for patient reported outcomes. Health Qual Life Outcomes 2006;4:70.

[34] McKenna SP. Measuring patient-report-ed outcomes: moving beyond misplaced common sense to hard science. BMC Med 2011;9:86.

[35] Fitzpatrick R, Davey C, Buxton MJ, Jones DR. Evaluating patient-based outcome meas-ures for use in clinical trials. Health Technol Assess 1998;2(14):i-iv,1-74.

[36] Coyne KS, Tubaro A, Brubaker L, Baven-dam T. Development and validation of patient-reported outcomes measures for overactive bladder: a review of concepts. Urology 2006;68(2 Suppl):9-16.

[37] Soreide K, Soreide AH. Using patient-re-ported outcome measures for improved decision-making in patients with gastroin-testinal cancer - the last clinical frontier in surgical oncology? Front Oncol 2013;3:157.

[38] Inzucchi SE, Bergenstal RM, Buse JB, Dia-mant M, Ferrannini E, Nauck M, et al. Man-agement of hyperglycemia in type 2 diabe-tes: a patient-centered approach: position statement of the American Diabetes Associ-ation (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care 2012;35(6):1364-79.

[39] Ismail-Beigi F, Moghissi E, Tiktin M, Hirsch IB, Inzucchi SE, Genuth S. Individualizing glycemic targets in type 2 diabetes melli-tus: implications of recent clinical trials. Ann Intern Med 2011;154(8):554-9.

[40] American Diabetes Association. Standards of medical care in diabetes--2014. Diabetes Care 2014;37(Suppl 1):S14-80.

[41] Rutten GEHM, De Grauw WJC, Nijpels G, Houweling ST, Van de Laar FA, Bilo HJ, Holleman F, Janssen PGH. The NHG guide-line Diabetes mellitus type 2. Huisarts Wet 2013;56(10):512-25.

[42] Verenso. Multidisciplinaire richtlijn diabe-tes. Verantwoorde diabeteszorg bij kwets-bare ouderen thuis en in verzorgings- of verpleeghuizen. Deel 1. [Multidisciplinary guideline diabetes. Responsible diabetes care in vulnerable elderly at home and in residential care or nursing homes. Part 1]. Utrecht, the Netherlands, Verenso, 2011.

R1R2R3R4R5R6R7R8R9



170

[43] American Diabetes Association. Standards of medical care for patients with diabetes mellitus. Diabetes Care 2003;26(Suppl 1):S33-50.

[44] van Hateren KJ, Landman GW, Kleefstra N, Houweling ST, van der Meer K, Bilo HJ. Time for considering other blood pressure target values in elderly patients with type 2 diabetes? Int J Clin Pract 2012;66(2):125-7.

[45] Aagaard L, Hansen EH. Consumers’ reports of suspected adverse drug reactions volun-teered to a consumer magazine. Br J Clin Pharmacol 2010;69(3):317-8.

[46] Dibonaventura MD, Wagner JS, Girman CJ, Brodovicz K, Zhang Q, Qiu Y, et al. Multi-national Internet-based survey of patient preference for newer oral or injectable Type 2 diabetes medication. Patient Prefer Adherence 2010;4:397-406.


[48] Cramer JA. A systematic review of adher-ence with medications for diabetes. Diabe-tes Care 2004;27(5):1218-24.

[49] Ryan R, Santesso N, Lowe D, Hill S, Grim-shaw J, Prictor M, et al. Interventions to improve safe and effective medicines use by consumers: an overview of systemat-ic reviews. Cochrane Database Syst Rev 2014;4:CD007768.

[50] Horne R, Chapman SC, Parham R, Freeman-tle N, Forbes A, Cooper V. Understanding patients’ adherence-related beliefs about medicines prescribed for long-term con-ditions: a meta-analytic review of the Ne-cessity-Concerns Framework. PLoS One 2013;8(12):e80633.

[51] Clifford S, Barber N, Horne R. Understand-ing different beliefs held by adherers, un-intentional nonadherers, and intentional nonadherers: application of the Necessi-ty-Concerns Framework. J Psychosom Res 2008;64(1):41-6.

[52] Schüz B, Marx C, Wurm S, Warner LM, Zie-gelmann JP, Schwarzer R, et al. Medication beliefs predict medication adherence in older adults with multiple illnesses. J Psy-chosom Res 2011;70(2):179-87.

[53] Wroe AL. Intentional and unintentional nonadherence: a study of decision making. J Behav Med 2002;25(4):355-72.

[54] British Medical Association (BMA) and NHS Employers. Quality and Outcomes Framework Guidance for GMS Contract 2013/14. London, UK: British Medical Association, National Health Service Con-federation, 2013 [online]. http://bma.org.uk/practical-support-at-work/contracts/ independent-contractors/qof-guidance. Accessed Oct 28, 2014.

[55] National Quality Forum (NQF). National Voluntary Consensus Standards for Ambu-latory Care Using Clinically Enriched Data: A Consensus Report, Washington, DC: NQF; 2010 [online]. http://www.qualityforum.org/Publications/2010/08/National_Vol-untary_Consensus_Standards_ for_Ambu-latory_Care_Using_Clinically_Enriched_Ad-ministrative_Data.aspx. Accessed Oct 28, 2014.

[56] O’Connor PJ, Bodkin NL, Fradkin J, Glas-gow RE, Greenfield S, Gregg E, et al. Di-abetes performance measures: current status and future directions. Diabetes Care 2011;34(7):1651-9.

[57] Lehman R, Krumholz HM. Tight control of blood glucose in long standing type 2 dia-betes. BMJ 2009;338:b800.

[58] Choe HM, Bernstein SJ, Standiford CJ, Hayward RA. New diabetes HEDIS blood pressure quality measure: potential for overtreatment. Am J Manag Care 2010;16(1):19-24.


[60] Tsevat J, Dawson NV, Wu AW, Lynn J, Soukup JR, Cook EF, et al. Health values of hospital-ized patients 80 years or older. HELP Inves-tigators. Hospitalized Elderly Longitudinal Project. JAMA 1998;279(5):371-5.

[61] Sherbourne CD, Keeler E, Unutzer J, Lenert L, Wells KB. Relationship between age and patients’ current health state preferences. Gerontologist 1999;39(3):271-8.

[62] NICE National Institute for Health and Care Excellence. Hypertension. Clinical manage-ment of primary hypertension in adults [online]. http://www.nice.org.uk/guidance /cg127/resources/ guidance-hypertension -pdf. Accessed Nov 7, 2014.

[63] Vermeire E, Hearnshaw H, Van Royen P, Denekens J. Patient adherence to treat-ment: three decades of research. A com-



171

prehensive review. J Clin Pharm Ther 2001;26(5):331-42.

[64] Lehane E, McCarthy G. Intentional and un-intentional medication non-adherence: a comprehensive framework for clinical re-search and practice? A discussion paper. Int J Nurs Stud 2007;44(8):1468-77.

[65] Delamater AM. Improving Patient Adher-ence. Clinical Diabetes 2006;24(2):71-7.

[66] Casula M, Tragni E, Catapano AL. Adher-ence to lipid-lowering treatment: the pa-tient perspective. Patient Prefer Adherence 2012;6:805-14.

[67] Haynes RB, Ackloo E, Sahota N, McDon-ald HP, Yao X. Interventions for enhancing medication adherence. Cochrane Database Syst Rev 2008;(2):CD000011.

[68] Gelhorn HL, Stringer SM, Brooks A, Thomp-son C, Monz BU, Boye KS, et al. Preferences for medication attributes among patients with type 2 diabetes mellitus in the UK. Di-abetes Obes Metab 2013;15(9):802-9.

[69] Kasteleyn MJ, Wezendonk A, Vos RC, Nu-mans ME, Jansen H, Rutten GE. Repeat pre-scriptions of guideline-based secondary prevention medication in patients with type 2 diabetes and previous myocardial infarction in Dutch primary care. Fam Pract 2014 [Epub ahead of print].

[70] Aujoulat I, Jacquemin P, Rietzschel E, Scheen A, Trefois P, Wens J, et al. Factors as-sociated with clinical inertia: an integrative review. Adv Med Educ Pract 2014;5:141-7.

[71] Vrijens B, De Geest S, Hughes DA, Przemy-slaw K, Demonceau J, Ruppar T, et al. A new taxonomy for describing and defining ad-herence to medications. Br J Clin Pharma-col 2012;73(5):691-705.

[72] Charles C, Gafni A, Whelan T, O’Brien MA. Treatment decision aids: conceptual is-sues and future directions. Health Expect 2005;8(2):114-25.

Nederlandse samenvatting

R1R2R3R4R5R6R7R8R9



174

Dit proefschrift richt zich op het patiëntenperspectief bij de evaluatie van medicijnen. De focus ligt met name op patiënten met type 2 diabetes. Verschillende aspecten worden belicht die gerelateerd zijn aan 1) het in kaart brengen van bijwerkingen en 2) het nemen van beslissingen om medicijnen voor te schrijven en te gebruiken. In het eerste deel van het proefschrift wordt de ontwikkeling en validatie van een patiëntgerapporteerde vragenlijst over bijwerkingen weergegeven. Het tweede deel richt zich op de rol van patiëntkenmerken en -voorkeuren in behandelbeslissingen die in de klinische praktijk genomen worden. Hieronder wordt per deel een korte introductie gegeven met daaropvolgend de belangrijkste bevindingen van de uitgevoerde onderzoeken.

Deel I. Ontwikkeling en validatie van een patiëntgerapporteerde vragenlijst over bijwerkingen

De evaluatie van de positieve effecten en de bijwerkingen van een medicijn (de benefit-risk evaluatie) werd voorheen met name gebaseerd op de beoordeling van professionals. In de afgelopen decennia is de aandacht voor de evaluatie door de patiënt toegenomen. Uit eerdere onderzoeken is gebleken dat rapportage over de positieve effecten en de bijwerkingen van een medicijn door de patiënt van toegevoegde waarde kan zijn. In patiëntgerapporteerde instrumenten is de patiënt de directe bron van informatie en zijn de gegeven antwoorden niet onderhevig aan de interpretatie van een professional. Inmiddels zijn vele patiëntgerapporteerde instrumenten ontwikkeld om de positieve effecten van een medicijn te meten maar een standaard patiëntgerapporteerd instrument om de bijwerkingen in kaart te brengen is niet voorhanden.

De doelen van het eerste deel van dit proefschrift zijn om:• een patiëntgerapporteerde vragenlijst over bijwerkingen te ontwikkelen;• de betrouwbaarheid en validiteit van deze vragenlijst te evalueren.

De patiëntgerapporteerde vragenlijst over bijwerkingen is ontwikkeld voor onderzoeksdoeleinden en bevat 1) vragen over algemene patiëntkenmerken, 2) medicijngebruik en ziekten, 3) ervaren bijwerkingen die getoond worden in een checklijst en ingedeeld zijn in lichaamscategorieën, en 4) aanvullende vragen over de aard en causaliteit van de bijwerking (hoofdstuk 1).

Als eerste stap in het validatieproces is de content validiteit van deze nieuwe vragenlijst beoordeeld (hoofdstuk 1). Voor de content validiteit zijn het begrip en de interpretatie van de vragen en antwoordopties geëvalueerd met behulp van cognitieve debriefing interviews. Deze interviews zijn gehouden met 28 patiënten die medicijnen gebruiken voor de behandeling van type 2 diabetes, astma of Chronische Obstructieve



175

Long Ziekte (COPD). Tijdens dit proces zijn de vragen aangepast en antwoordopties toegevoegd om de bruikbaarheid, het begrip en de volledigheid van de vragenlijst te verbeteren. Na veertien herzieningen van de vragenlijst is de content validiteit voldoende bevonden. De uiteindelijke versie bevat 252 bijwerkingen verdeeld over zestien lichaamscategorieën. Daarnaast bevat de vragenlijst veertien vragen per bijwerking over de aard en causaliteit. Alle bijwerkingen zijn gekoppeld aan een ‘Lowest Level Term’ in de Medical Dictionary for Regulatory Activities (MedDRA®). Tijdens de beoordeling van de content validiteit is gebruik gemaakt van een papieren versie van de vragenlijst. Vanwege de voordelen van een digitale vragenlijst, zoals directe opslag van de gegevens in een databestand en controles op gegeven antwoorden, is een digitale versie gemaakt. In supplement I is een pilotstudie naar de gebruikersacceptatie onder tien patiënten weergegeven. De gebruikersacceptatie bleek voldoende hoog te zijn om de digitale versie te gebruiken in vervolgonderzoeken naar de betrouwbaarheid en validiteit van het instrument. De digitale versie is vervolgens gebruikt om de test-hertest betrouwbaarheid en de bruikbaarheid van de vragenlijst te evalueren. Daarnaast is de invloed van de lichaamscategorieënstructuur op de test-hertest betrouwbaarheid en bruikbaarheid geëvalueerd (hoofdstuk 1). Voor deze evaluaties hebben 135 patiënten, die in ieder geval een oraal glucoseverlagend medicijn gebruiken, de vragenlijst tweemaal ingevuld. Tussen het invullen van beide vragenlijsten zat een periode van een week. De test-hertest betrouwbaarheid bleek voldoende te zijn voor het rapporteren van wel of niet een bijwerking. Dit betekent dat patiënten voldoende consistent waren in het rapporteren van bijwerkingen op dit niveau. De consistentie in het rapporteren van soortgelijke bijwerkingen was eveneens voldoende. Patiënten waren voldoende consistent in het rapporteren van bijwerkingen, wanneer dit beoordeeld wordt op het primaire MedDRA® System Organ Class niveau. De test-hertest betrouwbaarheid van het rapporteren van dezelfde specifieke bijwerkingen was echter onvoldoende. Dit betekent dat de vragenlijst in zijn huidige vorm niet gebruikt moet worden om bijwerkingen op dit niveau te kwantificeren. De vragenlijst bleek bruikbaar te zijn voor onderzoeksdoeleinden omdat ongeveer 75% van de patiënten rapporteerde dat de vragenlijst eenvoudig in te vullen was. Bovendien had het merendeel van de patiënten, die minstens één bijwerking rapporteerde, minder dan zestig minuten nodig om de vragenlijst te voltooien. Het gebruik van de lichaamscategorieënstructuur bleek het aantal gerapporteerde bijwerkingen, de test-hertest betrouwbaarheid en de bruikbaarheid niet significant te beïnvloeden. In hoofdstuk 2 is een onderzoek naar de construct validiteit en concurrent validiteit weergegeven. Voor dit onderzoek hebben de 135 patiënten twee aanvullende vragenlijsten ingevuld. Hieruit bleek dat patiënten die één of meer bijwerkingen rapporteerden in de patiëntgerapporteerde vragenlijst over bijwerkingen, een significant en klinisch relevant

R1R2R3R4R5R6R7R8R9



176

lagere algehele kwaliteit van leven en fysieke gezondheid hadden dan patiënten die geen bijwerking rapporteerden. Deze bevinding bevestigt de construct validiteit van de vragenlijst voor het rapporteren van wel of niet een bijwerking. De concurrent validiteit is aangetoond door de bevinding dat gerapporteerde bijwerkingen in de vragenlijst in 73% van de gevallen overeenkwamen met de informatie in de samenvatting van de productkenmerken van het gerapporteerde medicijn (de Summary of Product Characteristics). De overeenstemming was 76% wanneer alleen de bijwerkingen met een patiëntgerapporteerde causaliteits-score hoger dan of gelijk aan de mediaan werden geïncludeerd. De overeenstemming was 100% wanneer alleen de bijwerkingen met een causaliteits-score hoger dan of gelijk aan het derde kwartiel werden geïncludeerd. Aanvullende beoordeling van de concurrent validiteit toonde aan dat bijwerkingen die patiënten toewezen aan metformine, voldoende positieve voorspellende waarde hadden (79%) wanneer zij vergeleken werden met bijwerkingen gemeld bij metformine in een bestaande vragenlijst over tevredenheid met de behandeling. De sensitiviteit bleek echter onvoldoende te zijn (38%). Deze bevinding impliceert dat de patiëntgerapporteerde vragenlijst over bijwerkingen niet alle bijwerkingen opspoort. Tijdens de cognitieve debriefing interviews (hoofdstuk 1) werd duidelijk dat meerdere patiënten een recall periode (de periode waarover de informatie in de vragenlijst uitgevraagd wordt) van vier weken relatief kort vinden voor een patiëntgerapporteerde vragenlijst over bijwerkingen. In een aanvullend onderzoek is daarom de concurrent validiteit bepaald voor de vragenlijst met een recall periode van vier weken en drie maanden. Gerapporteerde bijwerkingen in de vragenlijst zijn vergeleken met gerapporteerde bijwerkingen in een dagboek dat gedurende een periode van drie maanden dagelijks bijgehouden diende te worden (hoofdstuk 3). Het onderzoek is voltooid door 78 patiënten die in ieder geval een oraal glucoseverlagend medicijn gebruiken. De sensitiviteit (33% voor beide recall periodes) en positieve voorspellende waarde (10% voor de 4 weken en 51% voor de 3 maanden recall periode) waren laag voor het rapporteren van bijwerkingen op het primaire System Organ Class niveau van de MedDRA®. De sensitiviteit was eveneens laag wanneer ook secundaire en tertiaire System Organ Classes in ogenschouw werden genomen (33% voor de 4 weken en 38% voor de 3 maanden recall periode) en bij het beoordelen van de rapportage van specifieke bijwerkingen (43% voor de 4 weken en 41% voor de 3 maanden recall periode). Uit aanvullende analyses is gebleken dat de sensitiviteit kan verschillen tussen primaire System Organ Classes omdat de sensitiviteit varieerde van 50% voor gastro-intestinale klachten tot 0% voor klachten gerelateerd aan metabolisme en voeding, psychiatrische klachten en klachten aan het ademhalingsstelsel, de borstkas en het mediastinum. Patiënten met meer overeenstemming tussen het rapporteren van bijwerkingen in de vragenlijst en het dagboek leken ouder en vaker man te zijn.



177

Ondanks dat het gebruik van rapportage door patiënten noodzakelijk is om de kennis over de benefit-risk evaluatie van een behandeling te vergroten kunnen een aantal problemen ontstaan. In supplement II zijn een aantal bekende maar ook minder bekende problemen van rapportage door patiënten gepresenteerd die naar voren zijn gekomen tijdens de onderzoeken naar de validiteit van de vragenlijst. Het bleek dat patiënten niet altijd consistent zijn in hun antwoorden, dat ze soms antwoord geven op vragen die niet op hun van toepassing zijn en dat ze vragen van eerder gevalideerde vragenlijsten anders kunnen interpreteren dan bedoeld. Een aantal mogelijke oplossingen zijn voorgesteld om problemen te reduceren. Belangrijk is dat patiëntgerapporteerde instrumenten kritisch geëvalueerd worden. Deze kritische evaluatie is een continu proces dat voortgezet moet worden nadat de validiteit van een instrument is aangetoond. Op basis van de bevindingen in het eerste deel van dit proefschrift kan worden geconcludeerd dat de patiëntgerapporteerde vragenlijst over bijwerkingen verdere aanpassingen behoeft. Suggesties voor aanpassingen zijn genoemd aan het eind van het proefschrift (summary and general discussion). Het uiteindelijke doel is te komen tot een valide patiëntgerapporteerde vragenlijst over bijwerkingen, die gebruikt kan worden in observationele en klinische studies om informatie te verkrijgen over bijwerkingen vanuit het patiëntenperspectief. Deze informatie is relevant voor de registratieautoriteiten in de benefit-risk evaluatie van medicijnen en voor professionals en patiënten om beter geïnformeerde beslissingen te kunnen nemen over de behandeling van voorkeur.

Intermezzo

Het intermezzo kan gezien worden als een brug tussen de twee delen in dit proefschrift. Aan de hand van een case-report wordt het toewijzen van klachten aan een medicijn en de ondernomen acties ten aanzien van bijwerkingen in de klinische praktijk beschreven vanuit het patiëntenperspectief. De beoordeling van bijwerkingen kan complex zijn en professionals kunnen op verschillende manieren met het perspectief van de patiënt omgaan. Door het perspectief van zowel de patiënt als de professional in ogenschouw te nemen, kan het toewijzen van klachten aan een medicijn en het ondernemen van acties ten aanzien van bijwerkingen verbeterd worden.

R1R2R3R4R5R6R7R8R9



178

Deel II. De rol van patiëntkenmerken en –voorkeuren in behandelbeslissingen in de klinische praktijk

In de klinische praktijk is het patiëntenperspectief belangrijk om patiëntgerichte zorg te leveren. In patiëntgerichte zorg worden behandelbeslissingen en –doelen geïndividualiseerd waarbij de voorkeuren van de patiënt en zijn/haar innamegedrag ten aanzien van medicijnen samen met klinische aspecten in ogenschouw worden genomen. Patiëntgerichte zorg wordt geadviseerd in richtlijnen voor bijvoorbeeld de preventie en behandeling van diabetes. Daarbij hoort ook het rekening houden met de leeftijd of levensverwachting van een patiënt bij het stellen van behandeldoelen. Bewijs van langetermijneffecten van geneesmiddelen bij oudere patiënten ontbreekt vaak en patiëntenvoorkeuren ten aanzien van medicijnen kunnen verschillen tussen leeftijdsgroepen. Op dit moment is weinig bekend over de invloed van leeftijd op het voorschrijfgedrag van artsen. Therapie-ontrouw (non-adherence) is een veelvoorkomend probleem maar het is onduidelijk wat de beste manier is om therapietrouw (adherence) te verbeteren in een populatie van patiënten met comorbiditeit. Meer inzicht in de onderliggende processen van verschillende vormen van therapie-ontrouw voor patiënten die meerdere medicijnen voor meerdere indicaties in dienen te nemen, kan bijdragen aan betere, op maat gemaakte interventies om therapietrouw te verbeteren.

De doelen van het tweede deel van dit proefschrift zijn om inzicht te verschaffen in:• de beslissingen om een behandeling te starten of te intensiveren met speciale

aandacht voor verschillende leeftijdsgroepen van patiënten;• de invloed van leeftijd en medicatiepercepties (overtuigingen) op patiënten-

voorkeuren voor medicijnen;• de rol van medicatiepercepties en behandelcomplexiteit op de therapietrouw van

patiënten.

In hoofdstuk 4 zijn mogelijke onder- en overbehandeling van glucose- en bloeddrukverlagende behandeling over de tijd geëvalueerd voor verschillende leeftijdsgroepen van patiënten met diabetes. Daarbij is ook de invloed van de introductie van kwaliteitsindicatoren in 2008 geëvalueerd. Voor deze evaluatie is gebruik gemaakt van een dynamisch cohort onderzoek met data van de Groningen Initiative to ANalyze Type 2 diabetes Treatment (GIANTT) database. Mogelijke overbehandeling op baseline was 7,4% voor glucoseverlagende behandeling en 15,9% voor bloeddrukverlagende behandeling. Deze percentages bleven relatief stabiel na de introductie van kwaliteitsindicatoren. Onderbehandeling, dat wil zeggen het niet starten of intensiveren van medicatie wanneer dat wel geïndiceerd lijkt te zijn, kwam vaker voor met 49,2% voor glucoseverlagende



179

en 60,7% voor bloeddrukverlagende behandeling op baseline. Na de introductie van kwaliteitsindicatoren steeg dit percentage voor de glucoseverlagende behandeling tot maximaal 56,4%. De invoering van kwaliteitsindicatoren leek dus niet te leiden tot verbetering van de glucoseverlagende behandeling. Voor de bloeddrukverlagende behandeling was een daling in mogelijke onderbehandeling waargenomen tot maximaal 50,7%. Uit deze cijfers komt naar voren dat de daling in onderbehandeling niet gepaard bleek te gaan met een toename in overbehandeling. Verder bleek dat mogelijke onderbehandeling –gedefinieerd op basis van niet-leeftijdsspecifieke aanbevelingen– in het algemeen vaker voorkwam bij oudere patiënten. Deze bevinding impliceert dat artsen in mindere mate (extra) medicijnen voorschrijven in de oudere populatie. Dit behoudende voorschrijfgedrag weerspiegelt mogelijk de bezorgdheid over de noodzaak van intensieve medicatiebehandeling in de oudere populatie evenals een geringere bereidheid van oudere patiënten om extra medicijnen te gebruiken. Verschillen tussen jongere en oudere patiënten in hun behandelvoorkeuren is geëvalueerd in hoofdstuk 5. In deze evaluatie is gekeken naar 1) of leeftijd de bereidheid van patiënten om een extra bloeddrukverlagend medicijn te gebruiken beïnvloedt en 2) of leeftijd het belang van specifieke medicijnkenmerken beïnvloedt. In dit onderzoek hebben 151 patiënten die in ieder geval een oraal glucoseverlagend en een bloeddrukverlagend medicijn gebruiken een vragenlijst ingevuld. Deze vragenlijst bevatte een discrete choice experiment waarin patiënten zich moesten voorstellen dat hun bloeddruk onvoldoende gereguleerd was. Het aantal patiënten dat bereid was een bloeddrukverlagend medicijn toe te voegen was significant lager onder oudere patiënten (67%) dan onder jongere patiënten (84%). Voor zowel de jongere als de oudere patiënten waren de effecten van het medicijn op a) het risico om te overlijden in de komende 5 jaar, b) de bloeddruk en c) het risico van bijwerkingen belangrijk voor het kiezen van een medicijn. Voor de jongere patiënten was het effect van het medicijn op het risico van beperkingen in het dagelijks leven door een beroerte daarnaast ook belangrijk. Het effect van het medicijn op de bloeddruk bleek minder belangrijk te zijn voor oudere patiënten dan voor jongeren patiënten. Een verkenning van de rol van medicatiepercepties liet zien dat de zorgen die mensen hebben over hun bloeddrukverlagende medicijnen (bezorgdheidpercepties) niet van invloed zijn op hun bereidheid een bloeddrukverlagend medicijn aan de behandeling toe te voegen. Jongere patiënten die meer het nut zagen van hun bloeddrukverlagende medicijnen (noodzaakpercepties) bleken echter vaker een voorkeur voor een extra medicijn te hebben dan jongere patiënten die minder het nut zagen van hun bloeddrukverlagende medicijnen. De medicatiepercepties van een patiënt zijn mogelijk van invloed op de therapietrouw van medicatiebehandeling. Daarnaast speelt ook de complexiteit van de behandeling een rol. Bekend is dat therapie-ontrouw intentioneel en niet-intentioneel kan zijn.

R1R2R3R4R5R6R7R8R9



180

In hoofdstuk 6 is de associatie tussen medicatiepercepties en zowel intentionele als niet-intentionele therapie-ontrouw geëvalueerd voor glucose-, bloeddruk- en lipidenverlagende medicijnen. Daarnaast is de associatie tussen behandelcomplexiteit en beide vormen van therapie-ontrouw voor de drie therapeutische groepen geëvalueerd. Deze associaties werden bestudeerd binnen dezelfde groep van patiënten om een indicatie te krijgen van verschillen tussen therapeutische groepen. In het onderzoek hebben 133 patiënten met type 2 diabetes een vragenlijst over therapietrouw en medicatiepercepties ten aanzien van de drie therapeutische groepen ingevuld. Deze gegevens zijn gecombineerd met voorschrijfgegevens uit de GIANTT-database om de behandelcomplexiteit te bepalen. Noodzaakpercepties ten aanzien van de medicijnen waren niet significant verschillend tussen de therapietrouwe, niet-intentionele therapie-ontrouwe en intentionele therapie-ontrouwe patiënten (verschillen kleiner dan 5 punten op een schaal van 5 tot 25). Bezorgdheidpercepties ten aanzien van de medicijnen waren hoger voor de intentionele therapie-ontrouwe patiënten maar alleen significant voor de bloeddrukverlagende medicijnen (8 punten verschil). Behandelcomplexiteit was gerelateerd aan zowel intentionele als niet-intentionele therapie-ontrouw voor glucose- en bloeddrukverlagende medicijnen. Voor lipidenverlagende medicijnen was de behandelcomplexiteit in het algemeen laag. De bevindingen bevestigen een deel van de verwachtingen, maar laten zien dat associaties tussen percepties en intentionele therapie-ontrouw kunnen verschillen tussen therapeutische groepen. In het tweede deel van dit proefschrift is geen bevestiging gevonden voor de vrees dat de introductie van kwaliteitsindicatoren in de klinische praktijk overbehandeling stimuleert. Verder is gebleken dat voor zowel jongere als oudere patiënten het risico op overlijden en bijwerkingen belangrijke aspecten zijn in het kiezen van een medicijn. Behandelbeslissingen moeten worden afgestemd op de voorkeuren van de individuele patiënt en hun bereidheid extra medicijnen te nemen. Tot slot is gebleken dat binnen dezelfde populatie de relatie van medicatiepercepties met therapietrouw mogelijk verschilt tussen therapeutische groepen. Deze bevinding suggereert dat zelfs voor een individuele patiënt verschillende soorten interventies nodig kunnen zijn om therapietrouw voor verschillende medicijnen te verbeteren. Uiteindelijk gaat het allemaal om de patiënt. De patiënt is onder andere degene die mogelijke bijwerkingen ervaart, eigen behandelvoorkeuren heeft, eigen medicatiepercepties heeft en het medicijn inneemt zoals voorgeschreven of niet. Daarom is het belangrijk het patiëntenperspectief in ogenschouw te nemen in de benefit-risk evaluatie van een medicijn en bij het bepalen van behandeldoelen, het nemen van behandelbeslissingen en de evaluatie van dergelijke beslissingen in de klinische praktijk.

Fryske gearfetting

R1R2R3R4R5R6R7R8R9


Fryske gearfetting

182

Dit proefskrift rjochtet him op it pasjinteperspektyf yn de benefit-risk evaluaasje fan medisinen. De fokus leit benammen op pasjinten mei type 2 diabetes. Ferskillende aspekten wurde beljochte dy’t relatearre binne oan 1) it yn kaart bringen fan bywurkingen en 2) it nimmen fan beslissingen om medisinen foar te skriuwen en te brûken. Yn it earste diel fan it proefskrift wurde de ûntwikkeling en falidaasje fan in pasjintrapportearre fragelist oer bywurkingen werjûn. It twadde diel rjochtet him op de rol fan pasjintskaaimerken en -foarkarren yn behannelbesluten dy’t yn de klinyske praktyk nommen wurde. Hjirûnder wurdt per diel in koarte yntroduksje jûn mei dêrnei de wichtichste befiningen fan de útfierde ûndersiken.

Diel I. Ûntwikkeling en falidaasje fan in pasjintrapportearre fragelist oer bywurkinge

De benefit-risk evaluaasje fan in medisyn waard foarhinne benammen basearre op de beoardieling fan professionals. Nei ferrin fan tiid is de oandacht foar rapportaazje troch de pasjint oangeande de benefit-risk evaluaasje tanommen. Út eardere ûndersiken hat nammentlik bliken dien, dat rapportaazje troch de pasjint fan tafoege wearde wêze kin. Yn pasjintrapportearre ynstruminten is de pasjint de direkte boarne fan ynformaasje en binne de antwurden dy’t jûn binne net bleatsteld oan de ynterpretaasje fan in professional. Ûnderwilens binne in protte pasjintrapportearre ynstruminten ûntwikkele om de positive effekten (de benefits) fan in medisyn te mjitten, mar in standert pasjintrapportearre ynstrumint om de bywurkingen (de risks) yn kaart te bringen is net foarhannen. De doelen fan it earste diel fan dit proefskrift wiene dêrom om: • in pasjintrapportearre fragelist oer bywurkingen te ûntwikkeljen;• de betrouberens en faliditeit fan dizze fragelist te evaluearjen. De pasjintrapportearre fragelist oer bywurkingen is ûntwikkele foar ûndersyksdoelen en befettet 1) fragen oer algemiene pasjintskaaimerken, 2) fragen oer medisyngebrûk en syktes, 3) ûnderfûne bywurkingen dy’t toand wurde yn in tsjeklist en yndield binne yn lichemskategoryen, en 4) oanfoljende fragen oer de aard en kausaliteit fan de bywurking (haadstik 1).

As earste stap yn it falidaasjeproses is de content-faliditeit fan dizze nije fragelist beoordiele (haadstik 1). Foar de content-faliditeit binne it begryp en de ynterpretaasje fan de fragen en antwurdopsjes evaluearre mei help fan kognitive debriefing ynterviews. Dizze ynterviews binne hâlden mei 28 pasjinten dy’t medisinen brûke foar de behanneling fan type 2 diabetes, astma of Groanyske Obstruktive Long Sykte (COPD). Fragen binne oanpast en antwurdopsjes binne tafoege om de brûkberens, it begryp en


Fryske gearfetting

183

de folsleinens fan de fragelist te ferbetterjen. Nei fjirtjin oanpassingen fan de fragelist is de content-faliditeit foldwaande befûn. De úteinlike ferzje befettet 252 bywurkingen, ferdield oer sechtjin lichemskategoryen. Dêrnjonken befettet de fragelist fjirtjin fragen per bywurking oer de aard en kausaliteit. Alle bywurkingen binne keppele oan in ‘Lowest Level Term’ yn de Medical Dictionary for Regulatory Activities (MedDRA®).

Tidens de beoardieling fan de content-faliditeit is gebrûk makke fan in papieren ferzje fan de fragelist. Fanwege de foardielen fan in digitale fragelist, lykas direkte opslach fan de gegevens yn in databestân en kontrôles op de antwurden dy’t jûn binne, is in digitale ferzje makke. Yn supplement I is in pilotstúdzje nei de brûkersakseptaasje ûnder tsien pasjinten werjûn. Út dizze stúdzje die bliken dat de brûkersakseptaasje foldwaande heech wie om de digitale ferzje te brûken yn ferfolchûndersiken nei de betrouberens en faliditeit fan it ynstrumint.

De digitale ferzje is dêrnei brûkt om de test-hertest betrouberens en de brûkberens fan de fragelist te evaluearjen. Dêrnjonken is de ynfloed fan de lichemskategoryenstruktuer op de test-hertest betrouberens en brûkberens evaluarre (haadstik 1). Foar dizze evaluaasjes hawwe 135 pasjinten, dy’t yn alle gefallen in oraal glukoazeferleegjend medisyn brûkten, de fragelist twaris ynfold. Tusken it ynfoljen fan beide fragelisten siet in perioade fan in wike. It die bliken dat de test-hertest betrouberens foldwaande wie foar it rapportearjen fan wol of net in bywurking. Dit betsjut dat pasjinten foldwaande konsistint wiene yn it rapportearjen fan bywurkingen op dit nivo. De konsistinsje yn it rapportearjen fan soartgelikense bywurkingen wie ek foldwaande. Pasjinten wiene foldwaande konsistint yn it rapportearjen fan bywurkingen op it primêre MedDRA® System Organ Class nivo. De test-hertest betrouberens fan it rapportearjen fan deselde spesifike bywurkingen wie lykwols ûnfoldwaande. Dit betsjut dat de fragelist yn syn hjoeddeistige foarm net brûkt wurde moat om bywurkingen op dit nivo te kwantifisearjen. De fragelist die bliken brûkber te wêzen foar ûndersyksdoelen, omdat likernôch 75% fan de pasjinten rapportearre dat de fragelist ienfâldich yn te foljen wie. Boppedat hie it grutste diel fan de pasjinten, dy’t yn alle gefallen in bywurking rapportearren, minder dan sechstich minuten nedich om de fragelist dien te meitsjen. It die bliken dat it gebrûk fan de lichemskategoryenstruktuer oantal rapportearre bywurkingen, de test-hertest betrouberens en de brûkberens net signifikant beynfloeden.

Yn haadstik 2 is in ûndersyk nei de construct faliditeit en concurrent faliditeit werjûn. Foar dit ûndersyk hawwe de 135 pasjinten twa oanfoljende fragelisten ynfold. Hjirút die bliken dat pasjinten dy’t ien of mear bywurkingen rapportearren yn de pasjintrapportearre fragelist oer bywurkingen, in signifikant en klinysk relevant legere algehiele kwaliteit fan libjen en fisike sûnens hiene dan pasjinten dy’t gjin bywurking rapportearren. Dizze befining befêstige de construct faliditeit fan de fragelist foar it rapportearjen fan wol of net in bywurking. De concurrent faliditeit is oantoand troch

R1R2R3R4R5R6R7R8R9


Fryske gearfetting

184

de befining dat rapportearre bywurkingen yn de fragelist dy’t pasjinten assosjearren mei spesifike medisinen, yn 73% fan de gefallen oerien kamen mei de ynformaasje yn de gearfetting fan de produktskaaimerken fan it medisyn (de Summary of Product Characteristics). Dizze oerienstimming wie 76% wannear’t allinne de bywurkingen mei in pasjintrapportearre kausaliteits-skoare heger as of gelyk oan de mediaan besjoen waarden. De oerienstimming wie 100% wannear’t allinne de bywurkingen mei in kausaliteits-skoare heger as of gelyk oan it tredde kwartyl besjoen waarden. Oanfoljende beoardieling fan de concurrent faliditeit toant oan dat bywurkingen dy’t pasjinten assosjearren mei metformine, foldwaande positive foarsizzende wearde hiene (79%) wannear’t sy fergelike waarden mei bywurkingen relatearre oan metformine yn in besteande fragelist oer tefredenens mei de behanneling. De sensitiviteit die lykwols bliken ûnfoldwaande te wêzen (38%). Dizze befining hâldt yn dat de pasjintrapportearre fragelist oer bywurkingen net alle bywurkingen opspoart. Tidens de kognitive debriefing ynterviews (haadstik 1) waard dúdlik dat meardere pasjinten in recall-perioade (de perioade dêr’t de ynformaasje yn de fragelist oer útfrege wurdt) fan fjouwer wiken relatyf koart fine foar in pasjintrapportearre fragelist oer bywurkingen. Yn in oanfoljend ûndersyk is dêrom de concurrent faliditeit bepaald foar de fragelist mei in recall-perioade fan fjouwer wiken en trije moannen. Rapportearre bywurkingen yn de fragelist binne fergelike mei rapportearre bywurkingen yn in deiboek dat oer in perioade fan trije moannen alle dagen byhâlden wurde moast (haadstik 3). It ûndersyk is ôfsletten troch 78 pasjinten dy’t yn alle gefallen in oraal glukoazeferleegjend medisyn brûke. De sensitiviteit (33% foar beide recall-perioades) en positive foarsizzende wearde (10% foar de 4 wiken en 51% foar de 3 moannen recall-perioade) wiene leech foar it rapportearjen fan bywurkingen op it primêre System Organ Class nivo fan de MedDRA®. De sensitiviteit wie ek leech wannear’t ek sekundêre en tertiêre System Organ Classes beskôge waarden (33% foar de 4 wiken en 38% foar de 3 moannen recall-perioade) en by it beoardieljen fan de rapportaazje fan spesifike bywurkingen (43% foar de 4 wiken en 41% foar de 3 moannen recall-perioade). Út oanfoljende analyses die bliken dat de sensitiviteit ferskille kin tusken primêre System Organ Classes, omdat de sensitiviteit fariearre fan 50% foar gastro-intestinale oandwaningen oant 0% foar oandwaningen relatearre oan metabolisme en fieding, psychiatryske oandwaningen en oandwaningen oan it sykheljensstelsel, de boarstkas en it mediastinum. Pasjinten mei mear oerienstimming tusken it rapportearjen fan bywurkingen yn de fragelist en it deiboek lykje âlder en faker man te wêzen. Nettsjinsteande dat it gebrûk fan rapportaazje troch pasjinten needsaaklik is om de kennis oer de benefit-risk evaluaasje fan in behanneling te fergrutsjen, kinne der in oantal problemen ûntstean. Yn supplement I binne in tal bekende problemen fan rapportaazje troch pasjinten en minder foar de hân lizzende problemen presintearre


Fryske gearfetting

185

dy’t nei foaren kommen binne yn de ûndersiken nei de faliditeit fan de fragelist. It die bliken dat pasjinten net altyd konsistint binne yn harren antwurden, dat se soms antwurd jouwe op fragen dy’t net op har fan tapassing binne en dat se fragen fan earder falidearre fragelisten oars ynterpretearje kinne dan bedoeld. Der is in tal mooglike oplossingen foarsteld om problemen te redusearjen. Wichtich is dat pasjintrapportearre ynstruminten kritysk evaluearre wurde. Dizze krityske evaluaasje is in kontinu proses dat kontinuearre bliuwe moat neidat de faliditeit fan ien ynstrumint oantoand is. Op basis fan de ûnderfiningen yn it earste diel fan dit proefskrift kin konkludearre wurde dat de pasjintrapportearre fragelist oer bywurkingen fierdere oanpassingen nedich hat. Suggestjes foar oanpassingen binne neamd yn it proefskrift (summary and general discussion). Ta beslút is it brûken fan in falidearre pasjintrapportearre fragelist oer bywurkingen yn observasjonele ûndersiken en klinyske studzjes wichtich om ynformaasje te krijen oer bywurkingen út it pasjinteperspektyf wei. Dizze ynformaasje kin brûkt wurde troch de registraasjeautoriteiten yn de benefit-risk evaluaasje fan medisinen en troch professionals en pasjinten om better ynformearre beslissingen te nimmen oer de behanneling fan foarkar.

Yntermezzo

It yntermezzo kin sjoen wurde as in brêge tusken de twa dielen yn dit proefskrift. Oan de hân fan in case-report wurde de beoardieling fan en de ûndernommen aksjes oangeande bywurkingen yn de klinyske praktyk beskreaun út it pasjinteperspektyf wei. De beoardieling fan bywurkingen kin kompleks wêze en op ferskillende manieren kin mei it perspektyf fan de pasjint omgien wurde. De beoardieling fan en de ûndernommen aksjes oangeande bywurkingen yn de klinyske praktyk kinne ferbettere wurde troch it perspektyf fan sawol de pasjint as de professional te beskôgjen.

Diel II. De rol fan pasjintskaaimerken en -foarkarren yn behannelbesluten yn de klinyske praktyk

Yn de klinyske praktyk is it pasjinteperspektyf wichtich om op de pasjint rjochte soarch ta te passen. Yn op de pasjint rjochte soarch wurde behannelbesluten en -doelen yndividualisearre, wêrby’t de foarkar fan de pasjint en syn/har fertoande gedrach oangeande medisinen tegearre mei klinyske aspekten beskôge wurde moatte. Op de pasjint rjochte soarch wurdt advisearre yn rjochtlinen fan bygelyks de previnsje en behanneling fan diabetes. In relatearre aspekt is it rekkening hâlden mei de leeftyd of libbensferwachting fan in pasjint yn it stellen fan behanneldoelen. Bewiis fan langetermyneffekten by âldere pasjinten ûntbrekt en pasjintfoarkarren oangeande

R1R2R3R4R5R6R7R8R9


Fryske gearfetting

186

medisinen kinne ferskillen tusken leeftydsgroepen sjen litte. Op dit momint is net folle bekend oer de ynfloed fan leeftyd op it foarskriuwgedrach fan professionals. Hoewol terapy-ûntrou (non-adherence) in faak foarkommend probleem is yn de klinyske praktyk, is it ûndúdlik wat de bêste manier is om terapytrou (adherence) te ferbetterjen yn spesifike populaasjes, wêrûnder pasjinten mei komorbiditeit. Mear ynsjoch yn de ûnderlizzende prosessen fan ferskillende foarmen fan terapy-ûntrou by benammen pasjinten dy’t meardere medisinen foar meardere yndikaasjes ynnimme moatte, kin bydrage oan bettere, op maat makke yntervinsjes om terapytrou te ferbetterjen. De doelen fan it twadde diel fan dit proefskrift wiene om ynsjoch te ferskaffen yn: • de besluten om in behanneling te begjinnen of te intensivearjen mei spesjale

oandacht foar ferskillende leeftydsgroepen fan pasjinten;• de ynfloed fan leeftyd en medikaasjepersepsjes (beliefs) op pasjintfoarkarren

oangeande medisinen;• de rol fan medikaasjepersepsjes en behannelkompleksiteit op de terapy-ûntrou fan

pasjinten oangeande medisinen. Yn haadstik 4 binne mooglike ûnder- en oerbehanneling fan glukoaze- en bloeddrukferleegjende behanneling oer de tiid evaluearre foar ferskillende leeftydsgroepen fan pasjinten. Benammen de ynfloed fan de yntroduksje fan kwaliteitsindikatoaren yn 2008 is evaluearre. Foar dizze evalueasjes is gebrûk makke fan in dynamysk kohortûndersyk mei data fan de Groningen Initiative to ANalyze Type 2 diabetes Treatment (GIANTT) database. Oerbehanneling op baseline wie 7,4% foar glukoazeferleegjende behanneling en 15,9% foar bloeddrukferleegjende behanneling. Dizze persintaazjes bleaune relatyf stabyl nei de yntroduksje fan kwaliteitsindikatoaren. Ûnderbehanneling kaam faker foar mei 49,2% foar glukoazeferleegjende en 60,7% foar bloeddrukferleegjende behanneling op baseline. Nei de yntroduksje fan kwaliteitsindikatoaren wie dit persintaazje ferhege foar de glukoazeferleegjende behanneling (op syn heechst 56,4%). Foar de bloeddrukferleegjende behanneling is in ferleging te sjen (op syn leechst 50,7%). Dizze ferleging fan ûnderbehanneling die bliken net gear te hingjen mei in taname yn oerbehanneling. De kwaliteitsindikatoaren diene bliken in bytsje effekt te hawwen op it ferbetterjen fan de glukoazeferleegjende behanneling. Fierder die bliken dat mooglike ûnderbehanneling – definiearre op basis fan net-leeftydsspesifike oanbefellingen – yn it algemien faker foarkaam yn âldere pasjinten. Dizze befining hâldt yn dat professionals minder medisinen foarskriuwe yn de âldere populaasje. Dit legere foarskriuwgedrag wjerspegelet mooglik de soargen oer de needsaak fan yntensive medikaasjebehanneling yn de âldere populaasje, lykas de mindere reewilligens fan âldere pasjinten om ekstra medisinen te brûken. Tidens de ûndersyksperioade wie dit lykwols noch net opnommen yn de rjochtlinen.


Fryske gearfetting

187

Ferskillen tusken jongere en âldere pasjinten yn har behannelfoarkarren binne evaluearre yn haadstik 5. Yn dizze evaluaasje is sjoen nei 1) of leeftyd de reewilligens fan pasjinten om in ekstra bloeddrukferleegjend medisyn te brûken beynfloedet en 2) of leeftyd it belang fan spesifike medisynskaaimerken beynfloedet. Yn dit ûndersyk hawwe 151 pasjinten dy’t yn alle gefallen in oraal glukoaze- en bloeddrukferleegjend medisyn brûke in fragelist ynfold. Dizze fragelist befettet in discrete choice experiment dêr’t pasjinten harren yn yntinke moasten dat de bloeddruk ûnfoldwaande regulearre wie. It oantal pasjinten dy’t ree wiene in bloeddrukferleegjend medisyn ta te foegjen wie signifikant leger yn âldere pasjinten (67%) as yn jongere pasjinten (84%). Foar sawol de jongere as de âldere pasjinten wiene it effekt fan it medisyn op it risiko om earder te ferstjerren, it effekt op de bloeddruk en it effekt op it risiko fan bywurkingen wichtich foar it kiezen fan in medisyn. Foar de jongere pasjinten wie it effekt fan it medisyn op it risiko fan beheiningen yn it deistisch libben troch in oerhaal dêrnjonken ek wichtich. It effekt fan it medisyn op de bloeddruk die bliken minder wichtich te wêzen foar âldere pasjinten as foar jongere pasjinten. In eksploraasje fan de rol fan medikaasjepersepsjes yn de assosjaasje tusken leeftyd en reewilligens om in bloeddrukferleegjend medisyn ta te foegjen liet sjen dat soarchpersepsjes oangeande bloeddrukferleegjende medisinen (concern beliefs) de assosjaasje net beynfloede. Jongere pasjinten mei hegere needsaakpersepsjes foar bloeddrukferleegjende medisinen (necessity beliefs) diene lykwols bliken faker in foarkar te hawwen foar in ekstra medisyn as jongere pasjinten mei legere needsaakpersepsjes foar bloeddrukferleegjende medisinen.

De medikaasjepersepsjes fan in pasjint (sa as soargen en needsaak oangeande medisinen) wurde ferwachte relatearre te wêzen oan terapy-ûntrou fan medikaasjebehanneling. Terapy-ûntrou kin yntensjoneel en net-yntensjoneel wêze en kin ek beynfloede wurde troch de kompleksiteit fan in behanneling. Yn haadstik 6 is de assosjaasje tusken sawol medikaasjepersepsjes as behannelkompleksiteit en sawol yntensjonele as net-yntensjonele terapy-ûntrou evaluearre foar glukoaze-, bloeddruk- en lipideferleegjende medisinen. Dizze assosjaasjes waarden ûndersocht binnen deselde groep fan pasjinten om in yndikaasje te krijen fan ferskillen tusken terapeutyske groepen. Yn it ûndersyk hawwe 133 pasjinten mei type 2 diabetes in fragelist oer terapytrou en medikaasjepersepsjes oangeande de trije terapeutyske groepen ynfold. Dizze gegevens binne kombinearre mei foarskriuwgegevens út de GIANTT-database om de behannelkompleksiteit te bepalen. Needsaakpersepsjes oangeande de medisinen wiene net signifikant ferskillend tusken de terapytrouwe, net-yntensjonele terapy-ûntrouwe en yntensjonele terapy-ûntrouwe pasjinten (ferskillen lytser as 5 punten op in skaal fan 5 oant 25). Soarchpersepsjes oangeande de medisinen wiene heger foar de yntensjonele terapy-ûntrouwe pasjinten, mar allinne signifikant foar de bloeddrukferleegjende medisinen (8 punten ferskil). Behannelkompleksiteit

R1R2R3R4R5R6R7R8R9


Fryske gearfetting

188

wie relatearre oan sawol yntensjonele as net-yntensjonele terapy-ûntrou foar glukoaze- en bloeddrukferleegjende medisinen. Foar lipideferleegjende medisinen wie de behannelkompleksiteit yn it algemien leech. Dizze ûnderfiningen hâlde yn dat soarchpersepsjes en behannelkompleksiteit wichtich binne foar terapytrou. Assosjaasjes tusken soarchpersepsjes en yntinsjonele terapy-ûntrou kinne lykwols ferskillen tusken terapeutyske groepen.

Yn it twadde diel fan dit proefskrift is gjin befêstiging fûn foar de soargen dat de yntroduksje fan kwaliteitsindikatoaren yn de klinyske praktyk mooglike oerbehanneling stimulearje kin. Fierder die bliken dat behannelbesluten harren rjochtsje moatte op sawol kwaliteit fan libjen as libbensferlinging en dat dit net ferskilt tusken jongere en âldere pasjinten. Behannelbesluten moatte ôfstimd wurde op de foarkarren fan de yndividuele pasjint en syn/har reewilligens om ekstra medisinen te nimmen. Fierder die bliken dat binnen deselde populaasje it ferbân tusken medikaasjepersepsjes en terapytrou mooglik ferskilt tusken terapeutyske groepen. Dizze befining suggerearret dat sels foar ien yndividuele pasjint meardere yntervinsjes nedich binne om terapytrou oangeande alle medisinen foar ferskillende indikaasjes te ferbetterjen.

Úteinlik giet it allegearre om de pasjint. De pasjint is dejinge dy’t bywurkingen ûnderfynt, dejinge mei behannelfoarkarren, dejinge mei spesifike medikaasjepersepsjes, dejinge dy’t it medisyn ynnimt lykas foarskreaun of net, ensafuorthinne. Dêrom is it wichtich it pasjinteperspektyf te beskôgjen yn de benefit-risk evaluaasje fan in medisyn en it bepalen fan behanneldoelen, it nimmen fan behannelbesluten en de evaluaasje fan soksoarte besluten yn de klinyske praktyk.

Appendices

R1R2R3R4R5R6R7R8R9



Appendix 1

191

Appendix 1. Patient-reported adverse drug event questionnaire

Questionnaire

Drug use and side effects experienced by patients

This questionnaire contains questions about the drugs that you take and any side effects (adverse effects) that you experience from these drugs. The questionnaire is made up of two parts:

Part A: general information, your drug use and the symptoms you experiencePart B: side effects that you have experienced during the past four weeks

Your details will remain confidential at all times.

InstructionsMost of the questions can be answered by checking the box next to the most applicable answer. There are no right or wrong answers and there will generally only be one possible answer, unless stated otherwise. There are also a number of questions that ask you to provide additional information on the dotted lines.

If you check the box next to the wrong answer, you can color that box black and then check the box next to the right answer. For example:Are you married? ¢ No T Yes The intended answer in this example was ‘Yes’. Most respondents take between 20 and 40 minutes to complete the questionnaire. You may need more time or less time to complete it yourself.

Please feel free to take a break during the questionnaire, but we do ask that you complete it at a later time, as incomplete questionnaires cannot be used for this research.

Thank you very much for your cooperation

R1R2R3R4R5R6R7R8R9


Appendices

192

Part AGeneral Information

1. What is your gender? MaleFemale

2. How old are you? years

3. What town/city do you live in? ..........................................

4. What is your highest level of completed education?No education completedElementary school, special educationJunior secondary vocational education, pre-vocational education (for example VMBO, LTS, LEAO)Junior general secondary education (for example MAVO, MULO, ULO, VMBO-t)Senior secondary vocational education, other vocational education (for example MBO, MEAO, MTS, BBL)Senior general secondary education (for example HAVO, VWO, Athenaeum, HBS)Higher professional education (for example HBO, HTS, HEAO)University education (research university)Other (please specify) ...............................................................................................................

5. What is your country of birth?

The NetherlandsOther (please specify) ...............................................................................................................

6. What is your father’s country of birth?The NetherlandsOther (please specify) ...............................................................................................................Don’t know

7. What is your mother’s country of birth?The NetherlandsOther (please specify) ...............................................................................................................Don’t know

8. How would you describe your general health?ExcellentVery goodGoodFairPoor Page 2.


Appendix 1

193

Part ADrug use

9. Which prescription drugs did you take during the past 4 weeks?

Example: I took metformin 500mg for diabetes. Example name + strength

metformin 500mg

Example disease/disorder

diabetes

Please enter all prescription drugs that you took during the past 4 weeks below:

Name + strength of the drug For which disorder/disease /ailment did or do you take this drug?

Page 3.

R1R2R3R4R5R6R7R8R9


Appendices

194

Question 9. Drug use, continued

Name + strength of the drug For which disorder/disease /ailment did or do you take this drug?

Page 4.


Appendix 1

195

10. Do you suffer from other disorders or diseases besides those mentioned above?NoYes (please specify) ....................................................................................................................

11. Did you use drugs during the past 4 weeks for which you did not require a prescription (for example self-help drugs, incidental drugs or alternative, homeopathic or natural drugs)?

NoYes (please specify) ....................................................................................................................

Page 5.

R1R2R3R4R5R6R7R8R9


Appendices

196

Part ASymptoms

12. Did you experience symptoms during the past 4 weeks? • If yes, you can find lists of symptoms per body part on the following pages. You should first consider in which part of your body you experienced the symptoms. You can then go to the page mentioned for each of the body parts. There you can check the box next to the applicable symptom.

Yes, I experienced symptoms in the following parts of my body: Eyes and/or eyelids ................................................................................................ go to page 7Throat, nose, ears (hearing) and/or swallowing ....................................... go to page 8Sweating, blushing, temperature increase or decrease, colds and/or flu .................................................................................................................... go to page 9Mouth, lips, speech and/or voice ...................................................................... go to page 10Tongue, teeth, gums and/or taste .................................................................... go to page 11Lungs, heart, chest, breathing (including sleep apnea) and/or blood (blood pressure, bleeding) .................................................................................. go to page 12Bladder and/or urination .................................................................................... go to page 13Intestines, stomach, vomiting (including vomiting blood), stool and/or bowel movements ................................................................................... go to page 14Skin (including wounds, bruises, rashes), hair and/or nails ................. go to page 15Genitals, sexuality, menopause, breasts and/or menstruation ............ go to page 16Muscles, bones, joints and/or bodily complaints ....................................... go to page 17Dizziness, falling and/or balance ...................................................................... go to page 18Head, brain, moods and/or emotions ............................................................. go to page 19Sleep, dreams, fatigue, yawning and/or more energy or less energy go to page 20Eating, drinking, weight and/or blood sugar ............................................... go to page 20Inflammation, edema, fungal infection and/or All other complaints go to page 21

No, I did not experience any symptoms ................................... go to page 26 (in Part B)

Page 6.


Appendix 1

197

Part AEyes and/or eyelids

13. Which symptoms involving your ‘eyes and/or eyelids’ did you experience during the past 4 weeks (you may give more than one answer)?

Yes, I experienced this symptom and ...I don’t think the drug caused it or I’m not sure

I do think that it is, or could be, a side effect of my drug

Blurred vision ----------------------------------- ---------------- -----------------------

Double vision ------------------------------------ ---------------- -----------------------

Seeing less or poorer vision ------------------ ---------------- -----------------------

Seeing (black) spots ---------------------------- ---------------- -----------------------

Night blindness --------------------------------- ---------------- -----------------------

Flashes of light ---------------------------------- ---------------- -----------------------

Painful eyes -------------------------------------- ---------------- -----------------------

Teary, watery eyes ------------------------------ ---------------- -----------------------

Dry eyes ------------------------------------------ ---------------- -----------------------

Burning, itchy or irritated eyes -------------- ---------------- -----------------------

Inflamed eyes ------------------------------------ ---------------- -----------------------

Itchy or irritated eyelids----------------------- ---------------- -----------------------

Inflamed eyelids -------------------------------- ---------------- -----------------------

Puffy or swollen eyes or eyelids ------------- ---------------- -----------------------

Enlarged pupils --------------------------------- ---------------- -----------------------

Pressure on the eyes --------------------------- ---------------- -----------------------

Burst eye vessels -------------------------------- ---------------- -----------------------

Inability to move eyes ------------------------- ---------------- -----------------------

Unusual eye movements ---------------------- ---------------- -----------------------

Other (please specify) . . . . . . . . . . . . . . . . . . . . ---------------- -----------------------

Page 7.

R1R2R3R4R5R6R7R8R9


Appendices

198

Part AThroat, nose, ears (hearing) and/or swallowing

14. Which symptoms involving your ‘throat, nose, ears (hearing) and/or swallowing’ did you experience during the past 4 weeks (you may give more than one answer)?



Painful throat, throat-ache -------------------- ---------------- ------------------------

Inflamed throat --------------------------------- ---------------- ------------------------

Dry throat ---------------------------------------- ---------------- ------------------------Difficulty swallowing, food sticks in the throat --------------------------------------------- ---------------- ------------------------Choking ------------------------------------------- ---------------- ------------------------Changed sense of smell (for example sensitivity to odors) ---------------------------- ---------------- ------------------------Bloody nose, nosebleed ----------------------- ---------------- ------------------------

Dry nostrils -------------------------------------- ---------------- ------------------------

Blocked nose ------------------------------------- ---------------- ------------------------

Runny nose--------------------------------------- ---------------- ------------------------

Ear infection ------------------------------------- ---------------- ------------------------

Earache ------------------------------------------- ---------------- ------------------------

Buzzing or ringing in the ear or ears ------- ---------------- ------------------------Impaired hearing, difficulty hearing or deafness ------------------------------------------ ---------------- ------------------------

Other (please specify) . . . . . . . . . . . . . . . . . . . . ---------------- ------------------------

Page 8.


Appendix 1

199

Part ASweating, blushing, temperature increase or decrease, colds and/or flu

15. Which symptoms involving ‘sweating, blushing, temperature, colds and/or the flu’ did you experience during the past 4 weeks (you may give more than one answer)?



Shivering, shivery ------------------------------ ---------------- ------------------------

Goose bumps ----------------------------------- ---------------- ------------------------Cold limbs (for example cold feet and/or hands) -------------------------------------------- ---------------- ------------------------

Often cold --------------------------------------- ---------------- ------------------------

Lower body temperature -------------------- ---------------- ------------------------

Higher body temperature (not fever) ----- ---------------- ------------------------Fever (temperature above 38 degrees Celsius) ------------------------------------------ ---------------- ------------------------

Insufficient sweating/transpiration ------- ---------------- ------------------------

Excessive sweating/transpiration ---------- ---------------- ------------------------

Blushing ----------------------------------------- ---------------- ------------------------

Cold ----------------------------------------------- ---------------- ------------------------

Flu-like symptoms ----------------------------- ---------------- ------------------------

Coughing, barking, hawking ----------------- ---------------- ------------------------

Sneezing ----------------------------------------- ---------------- ------------------------

Swollen glands --------------------------------- ---------------- ------------------------

Other (please specify) . . . . . . . . . . . . . . . . . . . . ---------------- ------------------------Page 9.

R1R2R3R4R5R6R7R8R9


Appendices

200

Part AMouth, lips, speech and/or voice

16. Which symptoms involving your ‘mouth, lips, speech and/or voice’ did you experience during the past 4 weeks (you may give more than one answer)? Yes, I experienced this symptom and ...

I don’t think the drug caused it or I’m not sure


Ulcers or bumps in the mouth and/or on the roof of the mouth --------------------------------- ---------------- ----------------------

Increased saliva in the mouth ----------------- ---------------- ----------------------

Dry mouth, less saliva in the mouth ---------- ---------------- ----------------------

Painful or sensitive mouth ---------------------- ---------------- ----------------------

Lockjaw --------------------------------------------- ---------------- ----------------------

Bad breath ----------------------------------------- ---------------- ----------------------

Inflamed lips --------------------------------------- ---------------- ----------------------

Painful or sensitive lips -------------------------- ---------------- ----------------------

Dry lips --------------------------------------------- ---------------- ----------------------

Swollen lips ---------------------------------------- ---------------- ----------------------Voice change (for example hoarseness, huskiness) ----------------------------------------- ---------------- ----------------------Unclear speech, mumbling, speech difficulties ------------------------------------------ ---------------- ----------------------

Word-finding problems, stumbling speech -- ---------------- ----------------------

Other (please specify) . . . . . . . . . . . . . . . . . . . . . . ---------------- ----------------------

Page 10.


Appendix 1

201

Part ATongue, teeth, gums and/or taste

17. Which symptoms involving your ‘tongue, teeth, gums and/or taste did you experience during the past 4 weeks (you may give more than one answer)?

Yes, I experienced this symptom and ...



Tooth discoloration------------------------------- ---------------- -----------------------

Plaque ----------------------------------------------- ---------------- -----------------------

Caries, tooth decay ------------------------------- ---------------- -----------------------

Toothache ------------------------------------------ ---------------- -----------------------

Teeth grinding ------------------------------------- ---------------- -----------------------

Inflamed or irritated gums --------------------- ---------------- -----------------------

Bleeding gums ------------------------------------- ---------------- -----------------------

Sensitive gums ------------------------------------ ---------------- -----------------------

Painful or sensitive tongue --------------------- ---------------- -----------------------

Swollen tongue ------------------------------------ ---------------- -----------------------

Tingling tongue ----------------------------------- ---------------- -----------------------

Changed sense of taste -------------------------- ---------------- -----------------------

Tongue blistering --------------------------------- ---------------- -----------------------

Dry tongue ----------------------------------------- ---------------- -----------------------

Tongue discoloration ---------------------------- ---------------- -----------------------

Other (please specify) . . . . . . . . . . . . . . . . . . . . . . ---------------- -----------------------

Page 11.

R1R2R3R4R5R6R7R8R9


Appendices

202

Part ALungs, heart, chest, breathing (including sleep apnea) and/or blood (blood pressure, bleeding)

18. Which symptoms involving your ‘lungs, heart, chest, breathing and/or blood’ did you experience during the past 4 weeks (you may give more than one answer)?



Hiccups ------------------------------------------- ----------------- ----------------------Lung disorder ----------------------------------- ----------------- ----------------------Pneumonia --------------------------------------- ----------------- ----------------------Pneumothorax----------------------------------- ----------------- ----------------------Respiratory infection -------------------------- ----------------- ----------------------Hyperventilation -------------------------------- ----------------- ----------------------Apnea (gap between breaths longer than 10 seconds) ----------------------------------------- ----------------- ----------------------Sleep apnea (gaps or pauses between breaths while sleeping) ----------------------- ----------------- ----------------------Slow breathing ---------------------------------- ----------------- ----------------------Rapid breathing --------------------------------- ----------------- ----------------------Shortness of breath, wheeziness, difficulty breathing, quickly out of breath ------------- ----------------- ----------------------Panting, puffing, wheezing, whistling (heavy breath)----------------------------------- ----------------- ----------------------Palpitations -------------------------------------- ----------------- ----------------------Rapid heartbeat --------------------------------- ----------------- ----------------------Slow heartbeat ---------------------------------- ----------------- ----------------------Irregular heartbeat, arrhythmia ------------- ----------------- ----------------------Chest pain or pressure ------------------------- ----------------- ----------------------Blood poisoning --------------------------------- ----------------- ----------------------Hemorrhage ------------------------------------- ----------------- ----------------------High blood pressure ---------------------------- ----------------- ----------------------Low blood pressure ---------------------------- ----------------- ----------------------Anemia -------------------------------------------- ----------------- ----------------------Thrombosis -------------------------------------- ----------------- ----------------------Other (please specify) . . . . . . . . . . . . . . . . . . . . ----------------- ----------------------

Page 12.


Appendix 1

203

Part ABladder and/or urination

19. Which symptoms involving your ‘bladder and/or urination’ did you experience during the past 4 weeks (you may give more than one answer)?



Blood in urine --------------------------------------- ---------------- ------------------------

Urine discoloration --------------------------------- ---------------- ------------------------

Pain when urinating-------------------------------- ---------------- ------------------------

Burning sensation when urinating -------------- ---------------- ------------------------

Less frequent and/or difficulty urinating ------ ---------------- ------------------------

More frequent need to urinate ------------------- ---------------- ------------------------

Less urine per toilet visit ------------------------- ---------------- ------------------------

Urine incontinence (involuntary urine loss) -- ---------------- ------------------------

Pressure on the bladder --------------------------- ---------------- ------------------------

Bladder infection ----------------------------------- ---------------- ------------------------

Other (please specify) . . . . . . . . . . . . . . . . . . . . . . . ---------------- ------------------------

Page 13.

R1R2R3R4R5R6R7R8R9


Appendices

204

Part AIntestines, stomach, vomiting (including vomiting blood), stool and/or bowel movements

20. Which symptoms involving your ‘intestines, stomach, vomiting, feces and/or bowel movements’ did you experience during the past 4 weeks (you may give more than one answer)?



(Excessive) burping, belching ------------------ ---------------- ----------------------Nauseous, sick ------------------------------------- ---------------- ----------------------Acid indigestion, stomach acid, heartburn -- ---------------- ----------------------Vomiting reflex ------------------------------------ ---------------- ----------------------Vomiting -------------------------------------------- ---------------- ----------------------Vomiting blood ------------------------------------ ---------------- ----------------------Bloated feeling ------------------------------------ ---------------- ----------------------Bloated stomach ---------------------------------- ---------------- ----------------------Intestinal, stomach, abdominal cramps and/or pain ---------------------------------------------- ---------------- ----------------------Gurgling or rumbling in the intestines and/or stomach --------------------------------------------- ---------------- ----------------------Flatulence (gas) ----------------------------------- ---------------- ----------------------Hemorrhoids -------------------------------------- ---------------- ----------------------Fecal incontinence (involuntary loss of feces) ------------------------------------------------ ---------------- ----------------------Diarrhea -------------------------------------------- ---------------- ----------------------Runnier, softer feces (not diarrhea) ----------- ---------------- ----------------------Mucus in feces ------------------------------------- ---------------- ----------------------Blood with feces ---------------------------------- ---------------- ----------------------Blood in feces -------------------------------------- ---------------- ----------------------Blockage, constipation, hard feces ------------ ---------------- ----------------------Black feces ----------------------------------------- ---------------- ----------------------More frequent bowel movements ------------- ---------------- ----------------------Other (please specify) . . . . . . . . . . . . . . . . . . . . . . ---------------- ----------------------

Page 14.


Appendix 1

205

Part ASkin (including wounds, bruises, rashes), hair and/or nails

21. Which symptoms involving your ‘skin, hair and/or nails’ did you experience during the past 4 weeks (you may give more than one answer)?



Greasy skin ----------------------------------------- ---------------- ----------------------Warm/burning skin ------------------------------ ---------------- ----------------------Dry, rough skin ------------------------------------ ---------------- ----------------------Painful skin ---------------------------------------- ---------------- ----------------------Itchiness -------------------------------------------- ---------------- ----------------------Flaking ---------------------------------------------- ---------------- ----------------------Acne ------------------------------------------------- ---------------- ----------------------Blisters ---------------------------------------------- ---------------- ----------------------Rashes (for example red patches, pimples) - ---------------- ----------------------Spot (painful), ulcer, wound -------------------- ---------------- ----------------------Skin discoloration (for example yellow or pale skin) ------------------------------------------ ---------------- ----------------------Pigment stains ------------------------------------- ---------------- ----------------------Patches of little or no skin pigment (pale patches of skin) ----------------------------------- ---------------- ----------------------Bruises, contusions ------------------------------- ---------------- ----------------------Increased sensitivity of the skin to light ----- ---------------- ----------------------Weak hair ------------------------------------------- ---------------- ----------------------Loss of hair ----------------------------------------- ---------------- ----------------------Increased hair growth --------------------------- ---------------- ----------------------Nail discoloration --------------------------------- ---------------- ----------------------Wrinkled nails ------------------------------------- ---------------- ----------------------Brittle, fragile nails ------------------------------- ---------------- ----------------------Other (please specify) . . . . . . . . . . . . . . . . . . . . . . ---------------- ----------------------

Page 15.

R1R2R3R4R5R6R7R8R9


Appendices

206

Part AGenitals, sexuality, menopause, breasts and/or menstruation

22. Which symptoms involving your ‘genitals, sexuality, menopause, breasts and/or menstruation’ did you experience during the past 4 weeks (you may give more than one answer)?




Vaginal discharge -------------------------------- ---------------- ----------------------Vaginal bleeding --------------------------------- ---------------- ----------------------Vaginal dryness (insufficient moisture production in the vagina) ---------------------- ---------------- ----------------------Burning sensation in the vagina -------------- ---------------- ----------------------Irritated vagina ----------------------------------- ---------------- ----------------------Menstruation pain ------------------------------- ---------------- ----------------------Irregular menstruation ------------------------- ---------------- ----------------------Heavy menstruation (excessive loss of blood or excessively long menstruation) --- ---------------- ----------------------Absence of menstruation ---------------------- ---------------- ----------------------Painful breasts or breast ----------------------- ---------------- ----------------------Prostate complaints ----------------------------- ---------------- ----------------------Sperm discoloration ----------------------------- ---------------- ----------------------Erection problems, impotence ---------------- ---------------- ----------------------Painful erection ---------------------------------- ---------------- ----------------------Painful penis -------------------------------------- ---------------- ----------------------Irritated penis ------------------------------------ ---------------- ----------------------Testicle pain -------------------------------------- ---------------- ----------------------Breast growth (in men) ------------------------ ---------------- ----------------------Pain during and/or after intercourse -------- ---------------- ----------------------Reduced sexual desire/interest in sex ------- ---------------- ----------------------Increased sexual desire/interest in sex ----- ---------------- ----------------------Difficulty having an orgasm ------------------- ---------------- ----------------------No orgasm ----------------------------------------- ---------------- ----------------------Painful orgasm ----------------------------------- ---------------- ----------------------Hot flushes ---------------------------------------- ---------------- ----------------------Premature menopause ------------------------- ---------------- ----------------------Other (please specify) . . . . . . . . . . . . . . . . . . . . . ---------------- ----------------------

Page 16.


Appendix 1

207

Part AMuscles, bones, joints and/or bodily complaints

23. Which symptoms involving your ‘muscles, bones, joints and/or bodily complaints’ did you experience during the past 4 weeks (you may give more than one answer)?



Muscle cramps (for example leg cramp) ---- ---------------- ----------------------Muscle pain, susceptible to muscle ache ---- ---------------- ----------------------Muscle contractions ----------------------------- ---------------- ----------------------Weak muscles, decreased muscular strength -------------------------------------------- ---------------- ----------------------Tired, heavy muscles ---------------------------- ---------------- ----------------------Stiff muscles, stiffness (for example stiff neck) ----------------------------------------------- ---------------- ----------------------Muscle tension ----------------------------------- ---------------- ----------------------Quivering, trembling, shaking muscles ------ ---------------- ----------------------Restless legs -------------------------------------- ---------------- ----------------------Bone fracture or fractures --------------------- ---------------- ----------------------Bone pain ------------------------------------------ ---------------- ----------------------Painful joints -------------------------------------- ---------------- ----------------------Inflammatory arthritis (gout) ---------------- ---------------- ----------------------Stiff joints ----------------------------------------- ---------------- ----------------------Unusual and/or involuntary movements or twitches -------------------------------------------- ---------------- ----------------------Difficulty walking -------------------------------- ---------------- ----------------------No or numb sensation in . . . . . . . . . . . . . . . . . . ---------------- ----------------------Tingling or prickling sensation in . . . . . . . . . . ---------------- ----------------------Pain in . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ---------------- ----------------------Bruised or damaged muscle, bone, body part . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ---------------- ----------------------Other (please specify) . . . . . . . . . . . . . . . . . . . . . ---------------- ----------------------

Page 17.

R1R2R3R4R5R6R7R8R9


Appendices

208

Part ADizziness, falling and/or balance

24. Which symptoms involving ‘dizziness, falling and/or balance’ did you experience during the past 4 weeks (you may give more than one answer)?



Unsteadiness, insecure, unsteady feeling ---- ---------------- ------------------------Balance problems --------------------------------- ---------------- ------------------------Falling ----------------------------------------------- ---------------- ------------------------Fainting --------------------------------------------- ---------------- ------------------------Light-headedness --------------------------------- ---------------- ------------------------Dizziness ------------------------------------------- ---------------- ------------------------Poorer coordination ----------------------------- ---------------- ------------------------Other (please specify) . . . . . . . . . . . . . . . . . . . . . . ---------------- ------------------------

Page 18.


Appendix 1

209

Part AHead, brain, moods and/or emotions

25. Which symptoms involving your ‘head, brain, moods and/or emotions’ did you experience during the past 4 weeks (you may give more than one answer)?




Stroke -------------------------------------------- ---------------- ----------------------Headache ---------------------------------------- ---------------- ----------------------Migraine ----------------------------------------- ---------------- ----------------------High, drunken sensation --------------------- ---------------- ----------------------Impaired consciousness ---------------------- ---------------- ----------------------Lack of concentration ------------------------- ---------------- ----------------------Lower reaction time --------------------------- ---------------- ----------------------Memory loss ------------------------------------ ---------------- ----------------------Forgetfulness ----------------------------------- ---------------- ----------------------Black-out ---------------------------------------- ---------------- ----------------------Confusion ---------------------------------------- ---------------- ----------------------Over-sensitive, irritable ---------------------- ---------------- ----------------------Restless ------------------------------------------ ---------------- ----------------------Aggressive --------------------------------------- ---------------- ----------------------Nervous, tense ---------------------------------- ---------------- ----------------------Anxious, fretful, worried --------------------- ---------------- ----------------------Absent-minded --------------------------------- ---------------- ----------------------Disorientated ----------------------------------- ---------------- ----------------------Lack of emotions ------------------------------- ---------------- ----------------------Over-emotional -------------------------------- ---------------- ----------------------Depressed, somber ---------------------------- ---------------- ----------------------Crying fits --------------------------------------- ---------------- ----------------------Changed mood --------------------------------- ---------------- ----------------------Mood swings ------------------------------------ ---------------- ----------------------Changed personality -------------------------- ---------------- ----------------------Voices in the head ----------------------------- ---------------- ----------------------Hallucinations ---------------------------------- ---------------- ----------------------Psychosis --------------------------------------- ---------------- ----------------------Other (please specify) . . . . . . . . . . . . . . . . . . . . ---------------- ----------------------

Page 19.

R1R2R3R4R5R6R7R8R9


Appendices

210

Part ASleep, dreams, fatigue, yawning and/or more or less energy

26. Which symptoms involving ‘sleep, dreams, tiredness, yawning and/or energy’ did you experience during the past 4 weeks (you may give more than one answer)?



Sleep attacks -------------------------------------- ----------------- ----------------------Sleep problems, sleeplessness ---------------- ----------------- ----------------------Sleepiness, dullness, heavy eyelids ----------- ----------------- ----------------------Dreams, nightmares ----------------------------- ----------------- ----------------------Fatigue --------------------------------------------- ----------------- ----------------------Yawning -------------------------------------------- ----------------- ----------------------More energetic ----------------------------------- ----------------- ----------------------Listlessness, dullness, lethargy, lack of energy ---------------------------------------------- ----------------- ----------------------Other (please specify) . . . . . . . . . . . . . . . . . . . . . ----------------- ----------------------

Part AEating, drinking, weight and/or blood sugar

27. Which symptoms involving ‘eating, drinking, weight and/or blood sugar did you experience during the past 4 weeks (you may give more than one answer)?



Decreased appetite ----------------------------- ----------------- ----------------------Increased appetite ------------------------------ ----------------- ----------------------Excessive thirst ---------------------------------- ----------------- ----------------------Sensitive to alcohol (less able or unable to handle alcohol) ---------------------------------- ----------------- ----------------------Increased weight -------------------------------- ----------------- ----------------------Loss of weight ----------------------------------- ----------------- ----------------------Low blood sugar level (hypoglycemia) ----- ----------------- ----------------------High blood sugar level (hyperglycemia) --- ----------------- ----------------------Unstable blood sugar level -------------------- ----------------- ----------------------Other (please specify) . . . . . . . . . . . . . . . . . . . . ----------------- ----------------------

Page 20.


Appendix 1

211

Part AInflammation, edema, fungal infection and/or all other complaints

28. Which symptoms involving ‘infection, fungus infection edema and/or all other complaints’ did you experience during the past 4 weeks (you may give more than one answer)?



Fungal infection -------------------------------- ----------------- ----------------------Edema, bloating -------------------------------- ----------------- ----------------------Inflammation of . . . . . . . . . . . . . . . . . . . . . . . . ----------------- ----------------------Other symptoms:. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ----------------- ----------------------. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ----------------- ----------------------. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ----------------- ----------------------. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ----------------- ----------------------. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ----------------- ----------------------

Page 21.

R1R2R3R4R5R6R7R8R9


Appendices

212

Part BPart B. Questions per side effect

In Part A of the questionnaire you were asked which symptoms you experienced in the past 4 weeks. In each case you indicated whether you thought it could be a side effect of your drugs.

In Part B you are asked to provide more information about these possible side effects. Please answer the questions 29 to 43 for each side effect.

In other words, you answer the questions on the first side effect on pages 23 to 25, the questions on the second side effect on the next three pages, etc.

Feel free to refer to Part A to see which side effects you checked.

Did you not experience any side effects of your drugs during the past 4 weeks? Ú Please go to page 26

Page 22.


Appendix 1

213

Part BSide effect 1.

29. Can you describe the side effect in your own words? ............................................................................................................................................................. ............................................................................................................................................................. ............................................................................................................................................................. 30. When did you first experience this side effect of your drugs?

Today Between 1 and 6 months agoYesterday Between 6 and 12 months ago2-7 days ago More than 12 months agoBetween 1 week and 1 month ago

31. Has this side effect gone away by now or improved?

No, the side effect has not gone away yetNo, but the side effect has clearly improvedNo, but the side effect was treated and has now improvedYes, the side effect:went away by itselfwent away after I stopped taking the drugwent away after treatment other (please specify) ................................................................................................

32. How often did you experience this side effect during the past 4 weeks (on how many or which days)?

..............................................................................................................................................................

33. On the days that you experienced this side effect, how much did it bother you (how bad or intense was it)?

Not at allOnly a bitSomewhatQuite a lotVery much

34. On the days that you experienced this side effect, how much influence did it have on your daily functioning?

NoneOnly a bitSomewhatQuite a lotVery much

35. Did this side effect result in serious medical situations for yourself during the past 4 weeks?

NoYes, please specify (you may select more than one answer):Admitted to hospital Permanent incapacity to workLife-threatening situationOther (please specify) ....

Page 23.

R1R2R3R4R5R6R7R8R9


Appendices

214

Part B

36. What action did you take in relation to this side effect during the past 4 weeks?NothingIn consultation with a healthcare professional, the drug dosage was reduced I reduced the dosage of the drug by myself In consultation with a healthcare professional, I stopped taking the drug temporarilyI stopped taking the drug temporarily by myselfIn consultation with a healthcare professional, I stopped taking the drug permanentlyI stopped taking the drug by myselfA drug and/or remedy has been prescribed to reduce/relieve the side

effect, please specify ........................................................................................................... I started using other drugs and/or remedy by myself to reduce/relieve

the side effect, please specify ................................................................................................Other, please specify ..................................................................................................................

37. Why do you think this symptom was caused by your drug (you may give more than one answer)?

I did not experience this symptom before I started taking the drugThe symptom started soon after I started taking the drug I experienced this symptom less often before I started taking the drugThe symptom was less serious before I started taking the drugThe symptom went away when I stopped taking the drug and came back

when I started taking it againThe symptom went away when I stopped taking the drugThe symptom started or grew worse when the drug dosage was increasedThe symptom decreased or went away when the drug dosage was

decreasedA healthcare professional (for example a doctor or pharmacist) confirmed

thisThe symptom is described in the patient leafletOther (please specify) ........................................................................................................

38. Which drug or drugs do you think caused this side effect?One drug that I use (please specify): ...........................................................................More than one drug that I use (please specify): ......................................................................................................................................................I don’t know Ú please go to question 42

39. How sure are you that this side effect is caused by this drug or these drugs?Very sureQuite sureNot very sureVery unsure

40. How long had you been using this drug or these drugs before this side effect started occurring?

......................................................................................................................................................

Page 24.


Appendix 1

215

Part B

41. How satisfied are you with the drug (or drugs) described in question 38 when you consider both this particular side effect and the effect of the drug or drugs?

Very satisfiedSatisfiedNeither satisfied or dissatisfiedDissatisfiedVery dissatisfied

42. Do you think there are other reasons for your experiencing this side effect (other than your drugs)?

NoYes (please specify):

..................................................................................................................................................... 43. Have you experienced this side effect in the past in combination with other drugs?

NoYes (please specify which drug):

Page 25.

R1R2R3R4R5R6R7R8R9


Appendices

216

Part B

Please note!

Have you completed Part B for all the side effects you experienced in the last 4 weeks?

Are there not enough forms to complete the questions? Please ask the researcher for more copies.

This is the end of the questionnaire. Please check whether you have answered all the questions.

You may make any further remarks below: ......................................................................................................................................................................

......................................................................................................................................................................

......................................................................................................................................................................

......................................................................................................................................................................

......................................................................................................................................................................

Once again, thank you very much for your cooperation

Page 26.


Appendix 2

217

Appendix 2. Supplemental tables chapter 1

Supplemental table 1. 2x2 tables for test-retest reliability at Patient level, MedDRA® level and ADE specific level

T2 TotalPatient level No ADE ≥1 ADE

T1 No ADE 28 5 33≥1 ADE 4 8 12

Total 32 13 45 MedDRA level No ADE in MedDRA ≥1 ADE in MedDRA

T1 No ADE in MedDRA 764 13 777≥1 ADE in MedDRA 16 17 33

Total 780 30 810ADE specific level ADE not reported ADE reported

T1 ADE not reported 11247 29 11276ADE reported 42 22 64

Total 11289 51 11340T1 = First measurement; T2 = Second measurement after one week period

Supplemental table 2. 2x2 tables for reliability of body categories at patient level, MedDRA® level and ADE specific level

T2 TotalPatient level No ADE ≥1 ADE


T1 No ADE 31 3 34≥1 ADE 3 8 11

Total 34 11 45Group with body categories at T2

T1 No ADE 28 6 34≥1 ADE 4 7 11

Total 32 13 45MedDRA level No ADE in MedDRA ≥1 ADE in MedDRA





Total 776 34 810ADE specific level ADE not reported ADE reported





Total 11279 61 11340T1 = First measurement; T2 = Second measurement after one week period

R1R2R3R4R5R6R7R8R9


Appendices

218

Supplemental table 3. 2x2 tables and kappa values for test-retest reliability at patient level, MedDRA® level and ADE specific level (only patients who completed second questionnaire within 10 days)

T2 Total Kappa value (95% CI)

Patient level No ADE ≥1 ADE 0.510 (0.16-0.86)

T1 No ADE 21 5 26≥1 ADE 1 5 6

Total 22 10 32MedDRA level No ADE in

MedDRA≥1 ADE in MedDRA

0.453(0.19-0.72)


Total 559 17 576ADE specific level ADE not

reportedADE reported

0.300(0.07-0.52)


Total 8034 30 8064T1 = First measurement; T2 = Second measurement after one week period. CI = Confidence interval.

Supplemental table 4. 2x2 tables and kappa values for reliability of body categories at patient level, MedDRA® level and ADE specific level (only patients who completed second questionnaire within 10 days)

T2 Total Kappa value (95% CI)

Patient level No ADE ≥1 ADEGroup with body categories at T1

T1 No ADE 22 3 25 0.633 ≥1 ADE 1 5 6 (0.30-0.97)


T1 No ADE 19 4 23 0.439 ≥1 ADE 2 4 6 (0.04-0.84)

Total 21 8 29MedDRA level No ADE in

MedDRA≥1 ADE in MedDRA


T1 No ADE in MedDRA 532 10 542 0.541 ≥1 ADE in MedDRA 6 10 16 (0.32-0.76)


T1 No ADE in MedDRA 490 16 506 0.335 ≥1 ADE in MedDRA 9 7 16 (0.08-0.59)

Total 499 23 522ADE specific level ADE not

reportedADE reported


T1 ADE not reported 7773 15 7788 0.373 ADE reported 15 9 24 (0.15-0.60)


T1 ADE not reported 7247 35 7282 0.176 ADE reported 20 6 26 (0.00-0.39)

Total 7267 41 7308T1 = First measurement; T2 = Second measurement after one week period. CI = Confidence interval.


Appendix 3

219


Supplemental table 1. Algorithm of the patient-reported causality assessmentItem Description Answer Scores

Causality 1 I did not experience this symptom before I started taking the medication

Yes 1

Causality 2 The symptom started soon after I started taking the medication

Yes 1

Causality 3 I experienced this symptom less often before I started taking the medication

Yes 1

Causality 4 The symptom was less serious before I started taking the medication

Yes 1

Causality 5 The symptom went away when I stopped taking the medication and came back when I started taking it again

Yes 2

Causality 6 The symptom went away when I stopped taking the medication

Yes 1

Causality 7 The symptom started or grew worse when the medication dosage was increased

Yes 1

Causality 8 The symptom decreased or went away when the medication dosage was decreased

Yes 1

Other reasons Do you think there are other reasons for your experiencing this side effect (other than your medication)?

Yes -1

Which drug + certainty

No drug reported I don’t know -1

One or more drugs reported

+Patient’s certainty

One drug More than one

Very sureQuite sure

1

One or more drugs reported +Patient’s uncertainty

One drug More than one

Not very sureVery unsure

0

R1R2R3R4R5R6R7R8R9


Appendices

220

Supplemental table 2. Characteristics of patients who mentioned ≥1 particular drugs and who mentioned no particular drug for their ADE

Patients who mentioned ≥1 particular drugs to ≥1 ADEs (N=25)

Patients who did not mention a particular drug to any of their ADEs (N=12)

P-value

Mean age in years (SD) 61 (10) 67 (10) 0.118*

Education (%) 1.000†

Lower education 7 (28) 4 (33) Middle education 8 (32) 4 (33) Higher education 9 (36) 4 (33) Other 1 (4) 0 (0)Female (%) 8 (32) 3 (25) 0.663‡

ADE = Adverse drug event; SD = Standard deviation* T-test; † Fisher-Freeman-Halton test; ‡ χ²-test


Appendix 3

221

Supplemental table 3. Characteristics of ADEs for which ≥1 particular drugs and no particular drugs were mentioned

ADEs for which ≥1 particular drug(s) were mentioned(N=78)

ADEs for which no particular drugs were mentioned(N=68)

P-value

First time experiencing the ADE (%) 0.021†

Today 2 (3) 4 (6) Yesterday 0 (0) 1 (2) 2-7 days ago 5 (6) 0 (0) Between 1 week and 1 month ago 12 (15) 2 (3) Between 1 and 6 months ago 8 (10) 6 (9) Between 6 and 12 months ago 10 (13) 11 (16) More than 12 months ago 41 (53) 44 (65)ADE gone away or improved (%) 0.041†

Not yet 57 (73) 63 (93) Clearly improved 11 (14) 3 (4) ADE was treated and has improved 3 (4) 1 (2) ADE went away by itself 0 (0) 0 (0) ADE went away after quitting medication 1 (1) 0 (0) ADE went away after treatment 1 (1) 0 (0) Other 5 (6) 1 (2)How much bothersome (%) 0.239†

Not at all 5 (6) 6 (9) Only a bit 9 (12) 13 (19) Somewhat 37 (47) 34 (50) Quite a lot 18 (23) 13 (19) Very much 9 (12) 2 (3)Influence daily functioning (%) 0.090†

None 29 (37) 18 (27) Only a bit 7 (9) 17 (25) Somewhat 31 (40) 23 (34) Quite a lot 10 (13) 8 (12) Very much 1 (1) 1 (2)Causality assessment (%) I did not experience this symptom before I started taking the medication

47 (60) 40 (59) 0.860‡

The symptom started soon after I started taking the medication

26 (33) 7 (10) 0.001‡

I experienced this symptom less often before I started taking the medication

4 (5) 13 (19) 0.010†

The symptom was less serious before I started taking the medication

3 (4) 5 (7) 0.473†

The symptom went away when I stopped taking the medication and came back when I started taking it again

2 (3) 0 (0) 0.499†

The symptom went away when I stopped taking the medication

0 (0) 1 (2) 0.466†

The symptom started or grew worse when the medication dosage was increased

9 (12) 0 (0) 0.004†

The symptom decreased or went away when the medication dosage was decreased

1 (1) 0 (0) 1.000†

A healthcare professional confirmed this 18 (23) 4 (6) 0.005†

The symptom is described in the patient leaflet 12 (15) 10 (15) 0.909‡

ADE = Adverse drug event; † Fisher-Freeman-Halton test; ‡ χ²-test

R1R2R3R4R5R6R7R8R9


Appendices

222

Supplemental table 4. Patient-reported ADE–drug association and information in Summary of Product Characteristics (SPC)Medical Dictionary for Regulatory Activities (MedDRA®) terminology of the reported ADE

Name of particular drugs mentioned for the ADE (RVGe/EU numberf; date of the SPC)

ADE in SPC

Clustered ADEsªDry throat + Dry mouth + Tongue dry + Thirst

Amitriptyline (RVG 52947-52948; date 09-2010) Yes

Blurred vision + Vision decreased

Rosuvastatin (RVG 26872; date 09-07-2012) No

Bloated feeling + Abdominal distension

Macrogol, combination (RVG 100287; date 16-05-2012) Yes

Dry mouth + Fall + Libido decreased

- Formoterol/budesonide (RVG 25887; date 01-11-2011)- Simvastatin (RVG 25536-25539, 33211; date 18-04-2012) - Metformin (RVG 25368, 21698; date 08-2011)- Acenocoumarol (RVG 04464; date 01-2012)

1x Yes based on description3x No

Hypoglycaemia + Blood glucose Fluctuation

Quetiapine (RVG 20826-20828, 25602-25603, 25128; date 22-02-2012) No

Dizziness + Balance Difficulty

Irbesartan (EU/1/97/046/001-003, 010, 013; date 22-04-2009) Yes

Dysphasia + Dysarthria + Sore Mouth

- Quetiapine (RVG 20826-20828, 25602-25603, 25128; date 22-02-2012)- Valproic acid (RVG 06175-06176, 07419, 08659-08661; date 16-04-2012)

1x Yes1x No

Gastrointestinal pain + Flatulence

Metformin (RVG 25368, 21698; date 08-2011) Yes

Diarrhoea + Increased stool frequency

Metformin (RVG 25368, 21698; date 08-2011) Yes

Myalgia + Muscle stiffness

Atorvastatin (RVG 21081-21083, 27148; date 14-06-2012) Yes

Vomiting reflex + Gastrointestinal pain

- Metformin (RVG 25368, 21698; date 08-2011) - Tolbutamide (RVG 15523-15524; date 30-10-2012)

2x Yes

Diarrhoea + Faecal incontinence


2x Yes

Taste alteration + Decreased appetite


1x Yes1x No

Mycosis + Other classified as itching

Metformin (RVG 25368, 21698; date 08-2011) No

Listlessness + Insomnia + Fatigue

- Atorvastatin (RVG 21081-21083, 27148; date 14-06-2012)- Rosuvastatin (RVG 26872; date 09-07-2012)

2x Yes

Balance difficulty + Balance disorder

- Quetiapine (RVG 20826-20828, 25602-25603, 25128; date 22-02-2012) - Valproic acid (RVG 06175-06176, 07419, 08659-08661; date 16-04-2012)

2x Yes


Appendix 3

223

Single ADEsHypohidrosisb,c Metformin (RVG 25368, 21698; date 08-2011) NoHeartburnb,c,d Metformin (RVG 25368, 21698; date 08-2011) YesHearing impairedb,c,d

Hydroquinine )RVG 03166; date 09-03-2009) Yes

Stools looseb,c,d Metformin (RVG 25368, 21698; date 08-2011) YesIncreased appetiteb,c,d

Pramipexole (RVG 101918-101920; date 05-10-2010) Yes

Flatulenceb,c,d Metformin (RVG 25368, 21698; date 08-2011) YesBlack stoolsb,c,d Ferrous fumarate (RVG 50165; date 10-09-2012) YesFlu like symptomsb,c

Rosuvastatin (RVG 26872; date 09-07-2012) Yes based on description

Other classified as numbnessb,c,d

Perindopril (RVG 33327-33328; date 12-2010) Yes

Urine discolorationb,c

Vildagliptin (EU/1/07/414/001-010, 018; date 17-09-2009) Yes based on description

Sore throatb Lisinopril (RVG 28424-6; date 11-2012) Yes based on description

Diarrhoeab,c Macrogol, combination (RVG 100287; date 16-05-2012) YesTongue dryb,c Metformin (RVG 25368, 21698; date 08-2011) NoAbdominal discomfortb,c


Hyperhidrosis - Metformin (RVG 25368, 21698; date 08-2011) - Candesartan (RVG 21704-21706, 30755; date 31-12-2011)

2x No

Tingling tongueb,c Candesartan (RVG 21704-21706, 30755; date 31-12-2011) NoIncreased stool frequencyb,c


Diarrhoeab Liraglutide (EU/1/09/529/001-005; date 08-07-2009) YesAbdominal discomfortb

Liraglutide (EU/1/09/529/001-005; date 08-07-2009) Yes

Weight decreasedb Liraglutide (EU/1/09/529/001-005; date 08-07-2009) YesHyperglycaemiab Liraglutide (EU/1/09/529/001-005; date 08-07-2009) NoFlatulenceb Liraglutide (EU/1/09/529/001-005; date 08-07-2009) YesWeight increasedb,c Insulin aspart (EU/1/00/142/004, 005; date 11-05-2012) Yes

Other classified as concentration impairment

- Tramadol (RVG 21626; date 28-09-2012) - Amitriptyline (RVG 52947-52948; date 09-2010) - Naproxen (11195-6; 08-2012)

3x Yes

Hair loss - Tramadol (RVG 21626; date 28-09-2012) - Amitriptyline (RVG 52947-52948; date 09-2010) - Naproxen (11195-6; 08-2012)

1x No2x Yes

Weight increased - Glimepiride (RVG 31961-31965; date 21-10-2010) - Simvastatin (RVG 25536-25539, 33211; date 18-04-2012)

2x No

Increased appetiteb,c


Haemorrhageb,c Acetylsalicyclic acid (RVG 16466; date 10-11-2008)http://www.whocc.no/atc_ddd_index/?code=N02BA01&showdescription=yes

Yes

R1R2R3R4R5R6R7R8R9


Appendices

224

Nauseab,c Methotrexate (RVG 28636-28638; date 03-2012) YesSleep apnea - Irbesartan (EU/1/97/046/001-003, 010, 013; date 22-04-2009)

- Hydrochlorothiazide (RVG 09640-09641; date 20-03-2012) - Sotalol (RVG 16723-16724; date 24-12-2008)

1x No2x Yes

Weight increasedb Quetiapine (RVG 20826-20828, 25602-25603, 25128; date 22-02-2012) YesConstipationb,c Paracetamol/codeine (RVG 12030; date 31-07-2012) YesHypertoniab Metformin (RVG 25368, 21698; date 08-2011) NoAbdominal discomfortb,c


Depressed mood - Metformin (RVG 25368, 21698; date 08-2011) - Tolbutamide (RVG 15523-15524; date 30-10-2012)

1x No1x Yes

Irritability - Metformin (RVG 25368, 21698; date 08-2011) - Tolbutamide (RVG 15523-15524; date 30-10-2012)

1x No1x Yes

Hyperglycaemia - Metformin (RVG 25368, 21698; date 08-2011) - Tolbutamide (RVG 15523-15524; date 30-10-2012)

2x No

Increased stool frequency


2x Yes

Tooth discolouration

Metformin (RVG 25368, 21698; date 08-2011) or perindopril (RVG 33327-33328; date 12-2010)

No

Vaginal irritation Glibenclamide (RVG 56114-56115; date 10-2010) or metformin (RVG 25368, 21698; date 08-2011)

No

ADE = Adverse drug eventa Clustered ADEs are multiple related ADEs reported by a patient that were clustered into one overall ADE by the researchers b Single ADE-single drug = included in patient-reported causality assessmentc Single ADE-single drug associations with a causality score higher than or equal to the median d Single ADE-single drug associations with a causality score higher than or equal to the third quartilee RVG number: Drug registration has been conducted in the Netherlands f EU number: Drug registration is for the whole European Union


Appendix 4

225


Supplemental table 1. Sensitivity analyses of the validity of reporting adverse drug events at MedDRA® primary class level in the questionnaire with a recall period of 4 weeks and 3 months

TP FP TN FN Se (95% CI) PPV (95% CI)Excluding delayed diary completersa

4-week recall; last 4 weeks of diary (N=30*18)

2 16 518 4 33% (4-78) 11% (1-35)

3-month recall; full 3-month diary (N=31*18)

16 16 491 35 32% (19-46) 50% (32-68)

Excluding delayed questionnaire completersb

4-week recall; last 4 weeks of diary (N=37*18)

2 19 641 4 33% (4-78) 10% (1-30)


17 16 548 31 35% (22-51) 52% (34-69)

Excluding both the delayed diary and questionnaire completers4-week recall; last 4 weeks of diary (N=29*18)

2 16 500 4 33% (4-78) 11% (1-35)


15 15 409 29 34% (20-50) 50% (31-69)

a Patients who returned the diary after >14 days. The diary was returned within 2 to 121 days (median: 8, interquartile range: 5-11) from the last date reported in the 3-month diary, which was not significantly different between the two recall groups (P=0.09).b Patients who completed the questionnaire after >14 days. The questionnaire was completed within 0 to 42 days (median: 1, interquartile range: 0-4) after sending, which was not significantly different between the two recall groups (P=0.65). TP = True positive; FP = False positive; TN = True negative; FN = False negative; Se = Sensitivity; PPV = Positive Predictive Value; CI = Confidence interval.

R1R2R3R4R5R6R7R8R9


Appendices

226


Supplemental table 1. Comparison of possible adverse drug events (ADEs) related to blood pressure-lowering or glucose-lowering treatment between patients with or without potential overtreatment Blood pressure lowering treatment – possible ADEs1

Patients not overtreated Patients with overtreatmentY_m1, N (%) 413 (3.7) 15 (3.9)Y, N (%) 518 (4.1) 19 (4.6)Y_p1, N (%) 663 (4.7) 29 (6.1)Y_p2, N (%) 540 (4.8) 37 (8.4)***

Y_p3, N (%) 304 (6.9) 19 (11.7)*

Glucose lowering treatment – possible ADEs2

Patients not overtreated Patients with overtreatmentY_m1, N (%) 363 (3.1) 17 (5.7)**

Y, N (%) 521 (3.9) 15 (4.8)Y_p1, N (%) 607 (4.1) 16 (4.7)Y_p2, N (%) 670 (5.8) 23 (9.6)*

Y_p3, N (%) 432 (9.8) 11 (10.7)1 ADEs included for blood pressure-lowering treatment were hypotension, dizziness, headache as well as unspecified coded ADEs (ICPC code A85 or WCIA code 1830) assigned to blood pressure-lowering drugs in the electronic medical record during the corresponding year.2 ADEs included for glucose-lowering treatment were hypoglycaemia as well as unspecified coded ADEs (ICPC code A85 or WCIA code 1830) assigned to glucose-lowering drugs in the electronic medical record during the corresponding year.* P <0.05; ** P<0.01; *** P=0.001.


Appendix 5

227

Sup

ple

men

tal t

able

2. S

ensi

tivity

ana

lyse

s usi

ng m

ore

rela

xed

defin

ition

s of o

vert

reat

men

t.Ba

selin

e En

try

DM-

prog

ram

Follo

w-u

p ye

ar 1

Follo

w-u

p ye

ar 2

Follo

w-u

p ye

ar 3

Blo

od p

ress

ure-

low

erin

g tr

eatm

ent

With

SBP

mea

sure

men

t (%

of a

ll pa

tient

s)11

,522

(84.

4)12

,929

(86.

9)14

,580

(89.

8)11

,706

(91.

2)4,

575

(94.

1) N

SBP

<12

0 m

mH

g 88

893

497

787

234

3 N

SBP

<12

0 m

mH

g w

ith p

oten

tial

o

vert

reat

men

t (%

of e

ligib

le p

atie

nts)

115

(13.

0)12

6 (1

3.5)

141

(14.

4)12

7 (1

4.6)

54 (1

5.7)

N cl

asse

s >=

3†10

0 (1

1.3)

113

(12.

1)12

4 (1

2.7)

116

(13.

3)46

(13.

4)

N

inte

nsifi

ed†

21 (2

.4)

24 (2

.6)

25 (2

.6)

16 (1

.8)

13 (3

.8)

Gluc

ose-

low

erin

g tr

eatm

ent

With

HbA

1c m

easu

rem

ent (

% o

f all

patie

nts)

12,1

21 (8

8.8)

13,5

49 (9

1.1)

15,0

02 (9

2.4)

11,8

86 (9

2.6)

4,50

3 (9

2.6)

N H

bA1c

<6%

(42

mm

ol/m

ol)

1,30

01,

208

1,26

798

434

9 N

HbA

1c <

6% (4

2 m

mol

/mol

) with

p

oten

tial o

vert

reat

men

t (%

of e

ligib

le

pat

ient

s)

77 (5

.9)

71 (5

.9)

72 (5

.7)

52 (5

.3)

32 (9

.2)

N cl

asse

s>=3

†11

(0.8

)5

(0.4

)5

(0.4

)2

(0.2

)1

(0.3

)

N

insu

lin u

se†

47 (3

.6)

39 (3

.2)

45 (3

.6)

38 (3

.9)

22 (6

.3)

N in

tens

ified

†20

(1.5

)28

(2.3

)24

(1.9

)12

(1.2

)11

(3.2

)Ba

selin

e =

Year

bef

ore

entr

y to

the

dise

ase

man

agem

ent p

rogr

am; E

ntry

DM

-pro

gram

= E

ntry

to d

isea

se m

anag

emen

t pro

gram

; Fol

low

-up

year

1 =

1 y

ear

afte

r ent

ry; F

ollo

w-u

p ye

ar 2

= 2

yea

rs a

fter e

ntry

; Fol

low

-up

year

3 =

3 y

ears

afte

r ent

ry; S

BP =

Sys

tolic

blo

od p

ress

ure;

HbA

1c =

gly

cohe

mog

lobi

n.

† Per

cent

ages

do

not s

um to

the

perc

enta

ges o

f pat

ient

s with

pot

entia

l ove

rtre

atm

ent b

ecau

se p

atie

nts c

an b

e in

clud

ed in

mul

tiple

cate

gori

es o

f ov

ertr

eatm

ent.

R1R2R3R4R5R6R7R8R9


Appendices

228

Supplemental table 3. General practices with a ≥5% increase, ≥5% decrease or stable percentage of under- and overtreated patients to systolic blood pressure (SBP) or glycohemoglobin (HbA1c) at one year after entry to the disease management program compared to baseline

SBP overtreatment (% within SBP undertreatment)1

Decrease Stable Increase Row total (% of total) P-value*SBP undertreatment 0.02 Decrease 26 (45.6) 11 (19.3) 20 (35.1) 57 (43.5) Stable 15 (36.6) 20 (48.8) 6 (14.6) 41 (31.3) Increase 13 (39.4) 8 (24.2) 12 (36.4) 33 (25.2) Row total (% of total) 54 (41.2) 39 (29.8) 38 (29.0)

HbA1c overtreatment (% within HbA1c undertreatment)Decrease Stable Increase Row total (% of total) P-value*

HbA1c undertreatment 0.13 Decrease 3 (9.1) 17 (51.5) 13 (39.4) 33 (24.8) Stable 10 (27.0) 19 (51.4) 8 (21.6) 37 (27.8) Increase 14 (22.2) 37 (58.7) 12 (19.0) 63 (47.4) Row total (% of total) 27 (20.3) 73 (54.9) 33 (24.8)* χ²-test; 1 Numbers do not sum to 133 GPs because two GPs were excluded due to no denominator (no patients with an SBP <130mmHg).


Appendix 6

229


Supplemental table 1. Association between self-reported life-expectancy and age†

Self-reported life-expectancy*‡

≤2 years >2 and ≤5 years >5 and ≤10 years >10 yearsAge <65 years 0 (0%) 0 (0%) 2 (4%) 45 (96%)

≥65 and <75 years 0 (0%) 1 (2%) 12 (24%) 36 (73%)≥75 and <85 years 2 (8%) 12 (48%) 6 (24%) 5 (20%)≥85 years 0 (0%) 3 (75%) 0 (0%) 1 (25%)

* N = 126 since 26 patients did not report their life-expectancy‡ The age of how old patients expect they will become ranged from 70 to 100 for both males and females. Most patients reported to become 80 years (30% of the males, 37% of the females)† Fisher freeman-halton test revealed a P-value of <0.001

Supplemental table 2. Preferences of patients aged <75 years and ≥75 years including patients who failed the dominant choice setConstant and attributes <75 yearsa ≥75 yearsb



P-value

Constant (additional drug) -0.79 (-1.31 – -0.26)

0.003 -1.19(-1.98 – -0.39)

0.003

Blood pressure -0.09(-0.10 – -0.07)

0.000 -0.05(-0.07 – -0.03)

0.000

Death within the next 5 years -20.67(-28.18 – -13.16)

0.000 -21.58(-33.22 – -9.94)

0.000

Limitations heart attack -6.24(-21.15 – 8.66)

0.412 -8.55(-31.67 – 14.57)

0.469

Limitations stroke -25.55(-40.55 – -10.56)

0.001 -10.12(-33.26 – 13.03)

0.392

Adverse drug events -15.22 (-18.36 – -12.07)

0.000 -9.52(-14.22 – -4.82)

0.000

Additional tablet in the evening 0.13(-0.07 – 0.33)

0.193 0.04(-0.26 – 0.35)

0.778

Combination tablet 0.13(-0.08 – 0.33)

0.224 0.16(-0.14 – 0.46)

0.302

a Number of observations 3,330 (111 patients * 10 choice sets * 3 alternatives per choice set)b Number of observations 1,500 (50 patients * 10 choice sets * 3 alternatives per choice set)

R1R2R3R4R5R6R7R8R9


Appendices

230

Supplemental table 3. Preferences of patients aged <65 years and ≥80 years Constant and attributes <65 yearsa ≥80 yearsb



P-value

Constant (additional drug) -0.09(-0.83 – 0.65)

0.813 -0.24(-1.82 – 1.34)

0.765

Blood pressure -0.07(-0.09 – -0.04)

0.000 -0.06(-0.11 – -0.01)

0.017

Death within the next 5 years

-22.41(-33.25 – -11.58)

0.000 -15.81(-40.24 – 8.62)

0.205

Limitations heart attack -10.76 (-32.07 – 10.55)

0.323 29.90(-18.91 – 78.72)

0.230

Limitations stroke -27.92(-49.29 – -6.54)

0.010 -20.15(-69.44 – 29.15)

0.423

Adverse drug events -18.42(-22.99 – -13.86)

0.000 -21.60(-32.02 – -11.19)

0.000

Additional tablet in the evening 0.09(-0.19 – 0.38)

0.532 -0.40(-1.05 – 0.24)

0.219

Combination tablet 0.07(-0.22 – 0.36)

0.645 0.07(-0.54 – 0.68)

0.828

a Number of observations 1,560 (52 patients * 10 choice sets * 3 alternatives per choice set)b Number of observations 450 (15 patients * 10 choice sets * 3 alternatives per choice set)


Appendix 7

231


Supplemental table 1. Differences between patients included and excluded from analysesIncluded Excluded P-value

Glucose-lowering drugsFemales (%) 38 (44.7) 28 (58.3) 0.131*

Mean age in years (SD) 65.8 (9.5) 67.1 (9.7) 0.471†

Education (%) 0.261*

Low education 41 (48.8) 29 (60.4) Middle education 23 (27.4) 6 (12.5) High education 15 (17.9) 10 (20.8) Other 5 (6.0) 3 (6.3)Mean BMI (SD) 29.0 (4.0) 29.9 (4.9) 0.252†

Median diabetes duration (IQR) 7 (5.0 – 11.0) 5 (3.0 – 9.0) 0.039‡

Blood pressure-lowering drugsFemales (%) 34 (50.7) 32 (48.5) 0.794*




Median diabetes duration (IQR) 7 (5.0 – 11.0) 6 (3.0 – 10.0) 0.206‡

Lipid-lowering drugsFemales (%) 39 (45.9) 27 (56.3) 0.251*




Median diabetes duration (IQR) 7 (3.0 – 10.0) 6.5 (4.0 – 10.8) 0.609‡

BMI = Body mass index; SD = Standard deviation; IQR = Interquartile range * Pearson χ²-test; † T-test; ‡ Mann-Whitney U test

R1R2R3R4R5R6R7R8R9


Appendices

232

Supplemental table 2. Patient characteristics of the adherers, unintentional non-adherers and intentional non-adherers to glucose-, blood pressure-, and lipid-lowering drugs

Adherers Unintentional non-adherers

Intentional non-adherers

P- value

Glucose-lowering drugsN 53 22 10Females (%) 28 (52.8) 8 (36.4) 2 (20.0) 0.105* Mean age in years (SD) 67.2 (10.0) 64.1 (7.4) 62.3 (10.0) 0.209†


Low education 31 (59.6) 7 (31.8) 3 (30.0) Middle education 11 (21.2) 9 (40.9) 3 (30.0) High education 8 (15.4) 4 (18.2) 3 (30.0) Other 2 (3.8) 2 (9.1) 1 (10.0)Mean BMI (SD) 29.1 (4.2) 27.7 (3.2) 31.6 (3.2) 0.033† ¹Median diabetes duration (IQR) 7 (3.0-10.0) 8 (6.0-12.5) 7 (5.5-14.5) 0.196‡

Measurement outside GPs office 0.012*² Yes (%) 8 (16.0) 9 (40.9) 5 (55.6)Blood pressure-lowering drugsN 53 10 4Females (%) 26 (49.1) 5 (50.0) 3 (75.0) 0.605*

Mean age in years (SD) 66.1 (0.9) 67.9 (10.1) 58.3 (16.5) 0.259†


Low education 29 (55.8) 2 (20.0) 1 (25.0) Middle education 12 (23.1) 4 (40.0) 1 (25.0) High education 9 (17.3) 2 (20.0) 2 (50.0) Other 2 (3.8) 2 (20.0) 0 (0.0)Mean BMI (SD) 30.1 (0.5) 27.6 (2.7) 28.9 (6.3) 0.290†

Median diabetes duration (IQR) 7 (5.0-9.0) 12 (6.8-14.5) 4.5 (2.5-6.5) 0.030‡³Measurement outside GPs office 0.078*

Yes (%) 13 (24.5) 2 (20.0) 3 (75.0)Lipid-lowering drugsN 67 15 3Females (%) 31 (46.3) 7 (46.7) 1 (33.3) 0.906*

Mean age in years (SD) 66.4 (9.1) 62.5 (9.1) 66.7 (25.8) 0.394†


Low education 38 (57.6) 3 (20.0) 1 (33.3) Middle education 16 (24.2) 5 (33.3) 1 (33.3) High education 7 (10.6) 6 (40.0) 1 (33.3) Other 5 (7.6) 1 (6.7) 0 (0.0)Mean BMI (SD) 29.5 (4.5) 28.3 (3.3) 27.9 (1.7) 0.515†

Median diabetes duration (IQR) 6 (3.0-9.0) 8 (5.8-12.3) 3a 0.090‡

Measurement outside GPs office 0.741*

Yes (%) 6 (9.0) 2 (13.3) 0 (0.0)GPs = General practitioners; BMI = Body mass index; SD = Standard deviation; IQR = Interquartile range; * Pearson χ²-test; † One-way analysis of variance; ‡ Kruskal-Wallis test a No IQR due to low numbers ¹ Significance due to difference between unintentional and intentional non-adherers (P = 0.003) ² Significance due to difference between adherers and intentional non-adherers (P = 0.008) ³ Post-hoc analysis did not reveal any significant differences

Curriculum vitae

R1R2R3R4R5R6R7R8R9



Curriculum vitae

235

Sieta de Vries was born on July 27th, 1986 in Drachten, the Netherlands. After finishing secondary school in 2003, she started her study, Sport & Health, at the Hanze University in Groningen. In 2007, she graduated from this educational program and continued with a pre-master program Psychology at the University of Groningen. She combined this educational program with a research project at the Hanze University Sports Science Research Group, where she evaluated the Groninger Sport Model. In 2009, she started the master program Social Psychology at the University of Groningen, and obtained her master’s degree in 2010. At the department of Health Psychology of the University Medical Center Groningen (UMCG), Sieta wrote her master thesis “The reliability and validity of an implicit coping- and wellbeing-test”. From December 2010 to December 2014, she worked at the department of Clinical Pharmacy and Pharmacology of the UMCG on her PhD-project on patient perspectives in the benefit-risk evaluation of drugs. In the period 2011 – 2013, she was a member of the PhD council of the graduate school SHARE of the University of Groningen. In December 2014, she started working as a post-doc researcher on the Innovative Medicines Initiative (IMI) Web-RADR project and the Strengthening Collaboration for Operating Pharmacovigilance in Europe (SCOPE) project at the department of Clinical Pharmacy and Pharmacology of the UMCG.

Dankwoord

R1R2R3R4R5R6R7R8R9



Dankwoord

239

Ik herinner het me nog goed: de eerste promotie die ik ooit bijgewoond heb. Samen met oud-studiegenootje Eveline Hamstra keek ik met grote overweldiging naar de verdediging van onze oud-leraar dr. J. de Jong. Wat een happening! Toen dr. J. de Jong zei: “Misschien jij ooit ook?” wist ik zeker dat dát nooit zou gaan gebeuren. Tsja… het kan soms gek lopen. Johan, ontzettend bedankt voor je begeleiding tijdens mijn allereerste onderzoek. Door jou is mijn interesse in het doen van onderzoek gewekt.

Mijn interesse in onderzoek is verder gegroeid door het master-onderzoek dat ik op de afdeling Gezondheidswetenschappen van het UMCG uit mocht voeren. Beste Adelita (Prof. dr. A.V. Ranchor), Maya (dr. M.J. Schroevers) en Robbert (Prof. dr. R. Sanderman), bedankt voor de begeleiding, het vertrouwen en het in contact brengen met Prof. dr. P. Denig en Prof. dr. F.M. Haaijer-Ruskamp wat geleid heeft tot mijn promotietraject.

Voor de totstandkoming van dit proefschrift ben ik vele mensen mijn dank verschuldigd. Een aantal mensen wil ik via deze weg graag in het bijzonder bedanken.

Prof. dr. P. Denig: Beste Petra, de meeste dank ben ik aan jou verschuldigd. Als eerste promotor en dagelijks begeleider moet je het behoorlijk druk met me hebben gehad. Ik waardeer je behulpzaamheid, passie voor de wetenschap en je altijd kritische blik die mij geholpen hebben mij verder als onderzoeker te ontwikkelen. Bedankt voor de prettige samenwerking die zich hopelijk nog lang mag voortzetten.

Prof. dr. F.M. Haaijer-Ruskamp: Beste Floor, jij doet me vaak denken aan m’n oma; geen blad voor de mond nemen en zeggen waar het op staat. Bedankt voor het vertrouwde gevoel dat dit gaf.

Prof. dr. D. de Zeeuw: Beste Dick, met je medische kennis kwam je vaak met een heel andere kijk op het geheel, dank hiervoor. Bedankt ook voor je bezorgdheid over de effecten van deze andere kijk op eventuele slaapproblemen. Gelukkig lijkt dit effect er niet te zijn.

Alle co-auteurs wil ik bedanken voor hun bijdrage aan de studies. Een speciaal dankwoord voor dr. P.G.M. Mol, dr. J. Voorham en dr. T. Dekker: Beste Peter, bedankt voor je regulatoire kijk op het geheel en de kans die je me geeft om te werken aan het IMI Web-RADR project en het SCOPE project.Beste Jaco, bedankt voor je statistische input en het me leren werken met Stata. Beste Thijs, bedankt voor al je hulp met het discrete choice experiment.

R1R2R3R4R5R6R7R8R9


Appendices

240

De beoordelingscommissie, bestaande uit Prof. dr. G. Nijpels, Prof. dr. E.P. van Puijenbroek en Prof. dr. B.H. Stricker, dank ik voor hun bereidheid het proefschrift te lezen en te beoordelen.

Dit proefschrift maakt deel uit van de SHARE graduate school van de Rijksuniversiteit Groningen en het Escher project van Top Institute Pharma. Betrokkenen van de SHARE graduate school, waaronder Truus van Ittersum en Renate Kroese, en van het Escher project, in het bijzonder Prof. dr. H.G.M. Leufkens en dr. P. Stolk, wil ik bedanken voor het mogelijk maken van de studies in dit proefschrift.De aanwezigen tijdens de Monday Morning Meetings, dRUGs/M2O meetings, het gangoverleg en de Escher-bijeenkomsten wil ik bedanken voor de interessante discussies.

Mijn directe collega’s: Arna, Yan, Paul, Mathijs, Sigrid, Grigory, Yunyu, Michelle, Tobias, Derbew, Sergei, Ellen, Mishgina, Bauk(j)e, Kirsten, Tomoko, Sophie, Ruben, Rudolf, Skander, Sara, Jarno, Maartje, Frank, Margje, Taco, Patrick, Judith, Wessel, Janny, Manfred, Ronald, Ardy, Alexandra, Marja, Hiddo en alle andere medewerkers en (PhD)studenten van de afdeling Klinische Farmacie en Farmacologie wil ik bedanken voor onder andere: - de geboden adviserende en/of praktische hulp; - the help with English texts; - de ‘daily doses of rock’; - het zingen van diverse liedjes; - de hulp bij etiketjes plakken en vragenlijsten versturen; - de consequente consistentie; - het bouwen van een fort met verhuisdozen; - saving my life several times; - het gipsen van een buik; - de noodzakelijke quarantaine situatie; - het Friese tintje in Groningen; - het constante herinneren aan de deadline van het promotie-traject; - en voor alle andere geboden hulp en onvergetelijke momenten. Daarnaast wil ik alle professionals die betrokken zijn geweest bij het benaderen van patiënten heel hartelijk danken voor hun hulp. Een speciaal dankwoord uiteraard naar alle patiënten die deelgenomen hebben.

Dianna, bedankt voor de prettige samenwerking tijdens de discrete choice studie en dat je mijn paranimf wilt zijn.


Dankwoord

241

Lieve Marrit, Freya, Hanne en Marlijn, onze vriendschap is er al vanaf de basisschool/middelbare school en betekent veel voor me. Ondanks dat we elkaar tegenwoordig iets minder vaak zien dan vroeger is het altijd goed. Bedankt dat jullie mijn (in)officiële paranimfen willen zijn.

Lieve Gerard, Gretha, Nola, Henk, Anneke, Edwin, Judith, Stijn, Marcel, Marijke, alle andere familieleden en (voetbal)vrienden: bedankt voor de andere belangrijke dingen in het leven. Een speciaal dankwoord aan Alberta en Beppie: bedankt voor jullie tekstuele feedback.

Leave pake Tjibbe, beppe Sytske, pake Geert en beppe Gryt, spitigernôch kinne jimme dit momint net meimeitse mar ik bin der wis fan dat jimme it prachtich fûn hiene.Leave heit en mem, ik bin jimme hiel tankber dat jimme altyd foar my klear stean en it altyd mooglik hawwe makke om the studearjen. Tige, tige tank.

Tot slot mijn grote steun en toeverlaat: Lieve Jeroen, je bent er altijd voor me en haalt het beste in me naar boven. Jouw relativeringsvermogen is bovendien een noodzakelijke vereiste in mijn leven. Ik hoop dat we samen een gelukkige toekomst tegemoet mogen gaan.

Groningen, December 2014 Sieta de Vries

SHARE publications

R1R2R3R4R5R6R7R8R9



SHARE publications

245

Research Institute SHARE

This thesis is published within the Research Institute SHARE (Science in Healthy Ageing and healthcaRE) of the University Medical Center Groningen / University of Groningen. Further information regarding the institute and its research can be obtained from our internetsite: www.share.umcg.nl.

More recent theses can be found in the list below.((co-) supervisors are between brackets)

2015

Febrianna SASkin problems related to Indonesian leather & shoe production and the use of footwear in Indonesia(prof PJ Coenraads, prof H Soebono, dr MLA Schuttelaar)

2014

Schneeberger CAsymptomatic bacteruiria and urinary tract infections in women: focus on diabetes mellitus and pregnancy(prof RP Stolk, prof JJHM Erwich, dr SE Geerlings)

Skorvanek, MFatigue, apathy and quality of life in patients with Parkinson’s disease(prof JW Groothoff, prof Z Gdovinova, dr JP van Dijk, dr J Rosenberger)

Kolvek GEtiology and prognosis of chronic kidney disease in children: Roma ethnicity and other risk factors(prof SAReiijneveld, prof L Podracka, dr JP van Dijk, dr J Rosenberger)

Mikula PHealth related quality of life in people with multiple sclerosis; the role of coping, social participation and self-esteem(prof JW Groothoff, prof Z Gdovinova, dr JP van Dijk, dr I Nagyova)

R1R2R3R4R5R6R7R8R9


Appendices

246

Amalia RImproving a school-based dental programme through a sociodental risk group approach(prof RMH Schaub, prof JW Groothoff, prof N Widyanti)

Christoffers WAHand eczema; interventions and contact allergies(prof PJ Coenraads, dr MLA Schuttelaar)

Troquete NACSTART-ing risk assessment and shared care planning in out-patient forensic psychiatry; results from a cluster randomized controlled trial(prof D Wiersma, prof RA Schoevers, dr RHS van den Brink)

Golea EFunctioning of young individuals with upper limb reduction deficiencies(prof CK van der Sluis, dr RM Bongers, dr HA Reinders-Messelink)

Nguyen HTMedication safety in Vietnamese hospitals; a focus on medication errors and safety culture(prof K Taxis, prof FM Haaijer-Ruskamp, prof JRBJ Brouwers, dr TD Nguyen)

Lehmann VSinglehood and partnerships in healthy people and childhood cancer survivors; a focus on satisfaction(prof M hagedoorn, prof R Sanderman, dr MA Tuinman)

Jaarsma EASports participation and physical disabilities: taking the hurdle?!(prof JHB Geertzen, prof PU Dijkstra, dr R Dekker)

Ockenburg SL vanPsycholopgical states and physical fates; studying the role of psychosocial stress in the etiology of cardiovascular disease: a nomothetic versus an idiographic approach(prof JGM Rosmalen, prof P de Jonge, prof ROB Gans)


SHARE publications

247

Beijers CGHMUnhealthy behaviors during pregnancy; who continues to smoke and consume alcohol, and is treatment of anxiety and depressive symptoms effective?(prof J Ormel, prof CLH Bockting, dr H Burger)

Kerdijk WStrategic choices in curriculum design to facilitate knowledge and competency development(prof J Cohen-Schotanus, prof JW Snoek, dr R Tio)

Spaans FHemopexin activity and extracellular ATP in the pathogenesis of preeclampsia(prof H van Goor, dr MM Faas, dr WW Bakker)

Brinksma ANutritional status in children with cancer(prof PF Roodbol, prof R Sanderman, prof ESJM de Bont, dr WJE Tissing)

Prihodova LPsychological and medical determinants of long-term patient outcomes; a specific focus on patients after kidney transplantation and with haemophilia(prof JW Groothoff, dr JP van Dijk, dr I Rajnicova-Nagyova, dr J Rosenberger)

Snippe EUnderstanding change in psychological treatments for depressive symptoms; the individual matters(prof R Sanderman, prof PMG Emmelkamp, dr MJ Schroevers, dr J Fleer)

Groen B Complications in diabetic pregnancy; role of immunology and Advanced Glycation End products(prof TP Links, prof PP van den Berg, dr MM Faas)

Visser LEarly detection and prevention of adolescent alcohol use; parenting and psychosocial factors(prof SA Reijneveld, dr AF de Winter)

R1R2R3R4R5R6R7R8R9


Appendices

248

Tovote KAAcceptance or challenge? Psychological treatments for depressive symptoms in patients with diabetes(prof R Sanderman, prof PMG Emmelkamp, prof TP Links)

Trippolini MEvaluation of functioning in workers with whiplash-associated disorders and back pain(prof MF Reneman, prof PU Dijkstra, prof JHB Geertzen)

Eriks-Hoogland IEShoulder impairment in persons with a spinal cord injury & associations with activities and participation(prof LHV van der Woude, porf G Stucki, prof MWM Post, dr S de Groot)

Suwantika AAEconomic evaluations of non-traditional vaccinations in middle-income countries: Indonesia as a reference case(prof MJ Postma, dr K Lestari)

Behanova MArea- and individual-level socioeconomic differences in health and health-risk behaviours; a comparison of Slovak and Dutch cities(prof SA Reijneveld, dr JP van Dijk, dr I Rajnicova-Nagyova, dr Z Katreniakova)

Dekker HTeaching and learning professionalism in medical education(prof J Cohen-Schotanus, prof T van der Molen, prof JW Snoek)

Dontje MLDaily physical activity in patients with a chronic disease(prof CP van der Schans, prof RP Stolk)

Gefenaite GNewly introduced vaccines; effectiveness and determinants of acceptance(prof E Hak, prof RP Stolk)


SHARE publications

249

Dagan MThe role of spousal supportive behaviors in couples’ adaptation to colorectal cancer(prof M Hagedoorn, prof R Sanderman)

Monteiro SPDriving-impairing medicines and traffic safety; patients’perspectives(prof JJ de Gier, dr L van Dijk)

Bredeweg SRunning related injuries(prof JHB Geertzen, dr J Zwerver)

Mahmood SISelection of medical students and their specialty choices(prof JCC Borleffs, dr RA Tio)

For more 2014 and earlier theses visit our website.


Uitnodiging

voor het bijwonen van de openbare verdediging van

mijn proefschrift


evaluation of drugs

op woensdag 22 april 2015om 16.15 uur in de aula

van het academiegebouw vande Rijksuniversiteit Groningen,

Broerstraat 5 te Groningen.

Aansluitend op de promotieis er een receptie in het

Academiegebouw.

Sieta de VriesHunze 12,

9204 BP Drachten06-14227380

[email protected]

Paranimfen

Marrit Groen ([email protected])

Freya Hornyák ([email protected])

Dianna de Vries ([email protected])

Sieta T. d

e Vries


evaluation of drugs

Sieta T. de Vries


The patient perspective in the process of drug evaluation and drug use is high on the agenda, which is demonstrated by an increased use of patient-reported outcome instruments to evaluate drugs and a shift towards patient-centred care in clinical practice. This thesis contains studies focusing on 1) the development and validation of a patient-reported outcome instrument to assess adverse drug events (ADEs), and 2) the role of patient characteristics and preferences on treatment decisions in clinical practice. The first part presents the development of a generic questionnaire to assess ADEs from the patient perspective. Although this questionnaire showed sufficient content and concurrent validity to detect ADEs at a general level, it was not sensitive enough to detect all ADEs perceived by patients. Suggestions are provided to improve the questionnaire for future use. In the second part, insight in decisions to start or intensify treatment with special attention for different patient age groups is provided. It was found that age influenced prescribing behaviour as well as the patient’s willingness to add a drug. For all patients, preventing death and ADEs were important considerations when choosing an additional drug. The influence of beliefs about benefits and risks on patients’ drug adherence, however, differed among types of drugs. These findings can be used to improve the assessment of ADEs from the patient perspective, to incorporate the patient perspective in treatment decisions and to develop better tailored interventions for improving drug adherence.

Proefschrift de Vries

Documents

Transcript of Proefschrift de Vries