Adherence to medication among type 2-diabetes patients

Multi-state model to map patterns of adherence to medication

The challenge of non-adherence
Adherence to medication – or the extent to which a patient takes medication corresponding to the recommendations of his or her health care provider – has been explored extensively for more than four decades, yet it remains a major challenge.

The World Health Organisation estimates adherence to long-term therapy for chronic illnesses in developed countries averages 50%. Studies have shown that poor adherence is associated with worsening of the patients’ clinical status/health, higher risk of hospitalisation, risk of preventable medication-related hospital admissions and higher mortality risk. Patient’s loss of adherence has also been shown to lead to higher outcome of hospitalizations and emergency department visits.

Understanding the broken chain
It can be argued that poor adherence among some patients is probably a relapsing condition over time, similar to relapses in attempts to stop smoking or lose weight. Whether intentionally or unintentionally on the part of the patient, these lapses represent a mismatch between the healthcare professionals’ assumptions about the medication intake based on what has been prescribed and what in reality has been collected and ingested by the patient. This signifies a problem in the pharmacotherapeutic chain.

Healthcare professionals have called for extended pharmacoepidemiological research in methods identifying therapeutic lapses and estimating and predicting the extent and magnitude of discrepancies between written prescriptions issued by the physician and prescriptions filled by the patient.

Multi-state model
Our research project aims to develop a multi-state model to calculate and measure adherence to medication in terms of acceptance, sufficient supply and persistence among a sample of patients with a history of type 2-diabetes. This information will be used to predict poor adherence among patients with type 2 diabetes using socio-economic measures and health status. This can provide healthcare providers with information on which type of patients to target with advice and counselling in order for these patients to obtain better medication adherence and hence better self-management and control of the disease.

Design and method 
This study is performed in Denmark. The Danish health care system is substantially different from those in the USA, The Netherlands and The United Kingdom, where many previous studies have been done on adherence to medication.

Denmark’s national health care system allows for extended register research opportunities including access to the Register of Medicinal Product Statistics containing information about consumption of medicinal products, which has covered the entire Danish population since 1995.

The analyses in this project will be based on these Danish national health records and information on socio-economic status in combination with the electronic patient register at Steno Diabetes Center A/S.

Target group
The research project targets patients with type 2-diabetes. The methodology and analyses can be applied to other chronic diseases with a life-long need for medication.

Mapping social patterns and health outcomes
The multi-state model is to be used for mapping the variation of adherence to medication among a sample of patients with type 2-diabetes. Furthermore, the model will take variation over time and duration of diabetes into account.

We will characterise the differences between Danish patients with type-2 diabetes with different patterns of adherence by socio-economic measures and health status.

We will estimate the associations between adherence to medication and health outcomes (nephropathy and death).

A predictive tool
The study will thus provide detailed information on medication adherence among patients with diabetes and on associations between medication adherence and intermediate clinical/health outcomes. This information can help clinicians predict which patients to target with advice and counselling for better medication adherence and hence better self-management and control of the disease.

Information on how to assess adherence to medication among diabetes patients can be translated to other areas where long-term medication is needed: asthma, therapy with immunosuppressants in relation to organ transplants, antiretroviral therapy in relation to human immunodeficiency virus, therapy for cancer patients and others.

Funding and collaborators 

The PhD project is funded in part by The Danish Agency for Science, Technology and Innovation at The Ministry of Science, and “danmark” sygeforsikring, a mutual health insurance company.

Papers published related to this project
ML Jensen, ME Jørgensen, EH Hansen, L Aagaard, B Carstensen: “A Multi-State Model and Algorithm for Measuring Long-Term Adherence to Medication: A Case of Diabetes Mellitus”, (2014), Value in Health, (DOI) 10.1016/j.jval.2013.11.014.

Key collaborators

Professor Ebba Home Hansen, M.Sc. (Pharm.), University of Copenhagen, Faculty of Health and Medical Sciences, Copenhagen, Denmark.

Professor Lise Aagaard, M.Sc. (Pharm.), PhD, University of Southern Denmark, Faculty of Health, Odense, Denmark.


Sidst opdateret 14-02-2017


Majken Linnemann Jensen 
Post.doc., PhD, M.Sc.(Econ.), M. Applied Statistics
Clinical Epidemiology Research


Inspirational articles

Vrijens B, De Geest S, Hughes DA, Przemyslaw K, Demonceau J, Ruppar T et al. A new taxonomy for describing and defining adherence to medications. Br J Clin Pharmacol 2012;73(5):691-705.

Blaschke T, Osterberg L, Vrijens B, Urquhart J. Adherence to Medications: Insights Arising from Studies on the Unreliable Link Between Prescribed and Actual Drug Dosing Histories. Annual Review of Pharmacology and Toxicology 2012;52:275-301.