Unintended effects of anti-hyperglycaemic drugs studied in population-based cohorts Citation for published version (APA): Driessen, J. H. M. (2017). Unintended effects of anti-hyperglycaemic drugs studied in population-based cohorts. Gildeprint Drukkerijen. https://doi.org/10.26481/dis.20170421ad Document status and date: Published: 01/01/2017 DOI: 10.26481/dis.20170421ad Document Version: Publisher's PDF, also known as Version of record Please check the document version of this publication: • A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers. Link to publication General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal. If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement: www.umlib.nl/taverne-license Take down policy If you believe that this document breaches copyright please contact us at: [email protected] providing details and we will investigate your claim. Download date: 28 Sep. 2021 Unintended effects of anti‐hyperglycaemic drugs studied in population‐based cohorts Annemariek Driessen © Annemariek Driessen, Maastricht, 2017 Cover: Dorijn Otto Layout: Tiny Wouters Printing: Gildeprint B.V. ISBN: 978‐90‐9030183‐9 Printing of this thesis was kindly supported by: het Nederlands Bijwerkingen Fonds. Unintended effects of anti‐hyperglycaemic drugs studied in population‐based cohorts PROEFSCHRIFT ter verkrijging van de graad van doctor aan de Universiteit Maastricht, op gezag van de Rector Magnificus, Prof. dr. Rianne M. Letschert, volgens het besluit van het College van Decanen, in het openbaar te verdedigen op vrijdag 21 april 2017, om 12.00 uur door Johanna Hendrika Maria Driessen Promotor Prof. dr. J.P.W. van den Bergh Copromotoren Dr. F. de Vries Dr. R.M.A. Henry Dr. H.A.W. van Onzenoort (Radboudumc) Beoordelingscommissie Prof. dr. P.M.M. Geusens (voorzitter) Prof. dr. G.J.P. van Breukelen Prof. dr. G.J. Dinant Prof. dr. M. den Heijer (VU medisch centrum Amsterdam) Prof. dr. T.P. van Staa (University of Manchester, United Kingdom) Contents List of Abbreviations 8 Chapter 1 General Introduction 11 Chapter 2 Fracture risk and use of anti‐hyperglycaemic drugs 25 2.1 The epidemiology of fractures in Denmark in 2011 27 Osteoporos Int. 2016;27(6):2017‐25 2.2 Use of dipeptidyl peptidase 4 inhibitors for type 2 diabetes mellitus 43 and risk of fracture Bone. 2014;68:124‐30 2.3 Use of dipeptidyl peptidase 4 inhibitors and fracture risk compared 57 to use of other anti‐hyperglycaemic drugs Pharmacoepidemiol Drug Saf. 2015;24(10):1017‐25 2.4 Bone fracture risk is not associated with the use of glucagon‐like 73 peptide 1 receptor agonists: A population‐based cohort analysis Calcif Tissue Int. 2015;97(2):104‐12 2.5 Use of glucagon‐like peptide 1 receptor agonists and risk of 87 fracture as compared to use of other anti‐hyperglycaemic drugs Calcif Tissue Int. 2015;97(5):506‐15 2.6 The use of incretins and fractures — a meta‐analysis on 103 population‐based real life data Br J Clin Pharmacol. 2016; in press. 2.7 Long‐term use of dipeptidyl peptidase 4 inhibitors and risk of 113 fracture; a retrospective population‐based cohort study Diabetes Obes Metab. 2017;19(3):421‐428 Chapter 3 Anti‐hyperglycaemic drug use within the Maastricht Study 129 population and potential unintended effects 3.1 Drug utilization in the Maastricht Study: a comparison with 131 nationwide data 3.2 The association between insulin use and volumetric bone mineral 147 density, bone micro‐architecture and bone strength of the distal radius in patients with T2DM – the Maastricht Study 3.3 Metformin dosage and duration is not associated with aortic 161 stiffness – the Maastricht Study Chapter 4 Methodological Aspects 175 4.1 Evaluation of different missing data strategies in propensity 177 score analyses Chapter 5 General Discussion 189 Appendices Summary 211 Samenvatting 217 Valorisation Addendum 223 Dankwoord 227 Curriculum Vitae 229 List of Abbreviations AGE Advanced glycation end product ATC Anatomical Therapeutic Chemical BMD Bone mineral density BMI Body mass index CEL N(ε)‐(carboxyethyl)lysine cfPWV Carotid‐femoral pulse wave velocity CI Confidence interval CML N(ε)‐(carboxymethyl)lysine COPD Chronic obstructive pulmonary disease CPRD Clinical Practice Research Datalink CVD Cardiovascular disease DDD Defined daily dose DPP4‐I Dipeptidyl peptidase 4 inhibitor DXA Dual‐energy X‐ray absorptiometry EGFR Estimated glomerular filtration rate EHD Electronic healthcare database FRAX Fracture risk assessment tool GI Gastrointestinal GIP Gastric inhibitory polypeptide GIP Glucose dependent insulinotropic polypeptide GLP‐1 Glucagon‐like peptide 1 GLP1‐RA Glucagon‐like peptide 1 receptor agonist GP General practitioner GPRD General practice research database HbA1c Glycosylated hemoglobin type A1c HR Hazard ratio HR‐pQCT High‐resolution peripheral quantitative computed tomography ICD International classification of diseases and related health problems IOF International osteoporosis foundation IQR Interquartile range IR Incidence rate IRR Incidence rate ratio MAP Mean arterial pressure MAR Missing at random MCAR Missing completely at random MI Multiple imputation MPR Medication possession rate NHDR National Hospital Discharge Register NHSR National Health Services Register 8 NIAD Non‐insulin anti‐diabetic drug NSAID Non‐steroidal anti‐inflammatory drugs OGTT Oral glucose tolerance test OR Odds ratio PS Propensity score PPAR‐γ Peroxisome proliferator‐activated receptor‐γ RAAS Renin angiotensin aldosterone system RCT Randomised controlled trial RR Relative risk SAF Skin autofluorescence SD Standard deviation SU Sulfonylurea derivative T1DM Type 1 diabetes mellitus T2DM Type 2 diabetes mellitus TBS Trabecular bone score TZD Thiazolidinedione UK United Kingdom vBMD Volumetric bone mineral density 9 10 Chapter 1 General Introduction Chapter 1 1 12 General Introduction Pharmacoepidemiology 1 Pharmacoepidemiology studies the use and effects of medications in large numbers of people. It frequently utilizes large “real‐world” electronic healthcare databases (EHDs) that often include millions of patients and multiple years of observations. These EHDs contain longitudinal data on patient characteristics as well as medical diagnoses and sometimes also life‐style variables, like body mass index (BMI), smoking status and alcohol use. More recently, longitudinal drug prescription history has also been added to these databases. With these databases population‐ based cohorts can be created to study exposure ‐ outcome associations. Examples of EHDs include the Clinical Practice Research Datalink (CPRD) (www.cprd.com), which is representative for the United Kingdom (UK) population, and the Danish National Health Services Register (NHSR)1. The NHSR contains data on 98% of all Danish residents and can be linked to the national hospital register2 as well as to the national prescription registry3. The PHARMO Record Linkage System database (www.pharmo.nl) is a Dutch EHD, which holds drug dispensing data of more than four million Dutch residents, representing 25% of the Dutch population. This database is linked to different registers, including the national hospital discharge register. EHDs are particularly useful when studying rare adverse events which cannot be detected in randomised controlled trials (RCTs), due to the restricted number of included patients. Consequently, with use of EHDs adverse effects with a low incidence rate can be detected. Moreover, the longer follow‐up time in EHDs makes it possible to study long‐term adverse effects. Another advantage of EHDs over RCTs is that when studying real‐life data, adherence to treatment rates may better represent actual medication intake, while RCTs tend to overestimate adherence4. Additionally, RCTs often apply strong in‐ and exclusion criteria which consequently result in a selected population. This therefore not always reflects the real‐life situation where patients have comorbidities and use other drugs concomitantly. With the use of EHDs this can be taken into account. One of the drawbacks of using observational data is the possibility for bias and confounding, which might seriously distort the effect estimates. RCTs suffer less from this problem as patients are randomised to one of the investigated treatment arms. When interpreting the results from a pharmacoepidemiological study, one has to take the potential sources of bias and