- Get Involved
- COVID-19 Resources
Polypharmacy (the concurrent use of multiple medications) can be problematic, and particularly affects those with multiple comorbidities. Polypharmacy increases the risk of adverse drug interactions, and patients may find it difficult to adhere to complex treatment regimens. Furthermore, there is evidence that many such patients will not take all of their medications as prescribed, and will attempt to ‘self-optimise’ their medication by trial-and-error, in order to minimise the number of medications they take whilst maximising their perceived clinical benefit. This can result in significant wastage of prescribed medication, loss of therapeutic benefit, or unsafe use of medicines.
In 2009, the prevalence of diabetes in England was 5.1%. Of those, 90% have type 2 diabetes. Coronary heart disease (CHD) is a major cause of morbidity and mortality in patients with type 2 diabetes. The sub-population of those with both type 2 diabetes and coronary heart disease is significant. It would be hugely beneficial to understand how prescriptions for this sub-population might be rationalised to minimise patient self-optimisation (and therefore potential wastage) whilst maximising real clinical benefit.
The aim of this project was exploratory, in order to build a simple but effective proof-of-concept model that attempts to find the optimal combination of medicines for patients with both type 2 diabetes and CHD. Agent-Based Simulation is a way of modelling the behaviours and interactions of individuals within a system, and so could be theoretically applied to capture the self-optimising behaviours of people prescribed with multiple medications. This was a new application of Agent-Based Simulation, and therefore the main anticipated outcome of the project was a paper that assessed the potential of Agent-Based Simulation to tackle this problem. However, this project was also suggested to act as a proof-of-concept for potential application to other sub-populations, and those with more than two comorbidities.
What has this project achieved?
Currently we are exploring the option of a student taking over the project to see it through to completion (including validation against real-world data sets). This may be a BSc Medical Sciences Undergraduate in their professional training year, or a longer term PhD project (perhaps in collaboration with a pharmaceutical company) looking to extend and apply the model more widely, or both.
There is potential for the model to be made more generically applicable to other conditions to be used as a tool to better estimate adherence rates in cost-effectiveness analyses. This could be explored by a student (as above), or by a cost-effectiveness researcher with programming experience elsewhere in the CLAHRC.
Prof Nicky Britten, Dr Janet Heaton, Dr Bettina Kluettgens - AHSN, Rati Magura - Royal Devon and Exeter NHS Foundation Trust