Title of project

Medication use patterns and individual-level effects in type 2 diabetes

Abstract

Type 2 diabetes mellitus (T2D) is estimated to afflict 590 million people worldwide, nearly half of whom are undiagnosed. With this project, we aim to make important strides toward better precision medicine in T2D. The project will consist of three work packages and will be based on Danish national health registries.
Firstly, we will develop a comprehensive tool for characterising complex medication use patterns, including relations between multiple medication types, levels of medication burden, and trajectories over time; i.e., this goes beyond characterising use of each medication individually, but rather provides in-depth characterisation of sequences of different medications and allows evaluation of common discontinuation patterns, treatment stability, polypharmacy and more. This tool will be published as an open-source R package. Secondly, we will apply this tool to the Danish population and compare individuals with T2D, individuals with prediabetes, and individuals without diabetes. We will identify common medication use patterns, including sequences in which medications are initiated or discontinued. Particular attention will be given to identifying areas in which individuals with T2D or prediabetes are more likely to receive insufficient treatments, and to identify predictors of treatment insufficiency. The results from characterising complex medication use patterns in the full Danish population will be published in an open-access website which will be continually updated as new data become available. Thirdly, we will validate in the Danish population a UK-developed model for selection of optimal second-line glucose-lowering medication—and identify predictors of treatment discontinuation. Finally, we will examine how such a treatment selection model could be implemented as a decision support tool in an electronic patient overview and investigate how primary care physicians view and interact with such a tool, in order to understand the utility of such tools and how they could be implemented to best meet clinicians’ needs.
These studies will advance the state of precision medicine in T2D by identifying under-treated individuals and groups and improving the choice of glucose-lowering treatment on an individual level. In addition, they will provide researchers and clinicians with an interactive tool to view upto- date summaries of simple and medication use patterns in the Danish population along with a tool to perform new analyses on prescription data in a streamlined and reproducible manner.