Simulation modelling of pathways of management for polymyalgia rheumatica

Polymyalgia rheumatica (PMR) is the commonest inflammatory musculoskeletal condition affecting 1.7% of those aged older than 55yrs and is treated with glucocorticoids. National guidelines endorsed by NICE recommend that patients take glucocorticoids for 1 to 2 years and that those who require longer treatment be referred to rheumatologists for specialist input. Meta-analyses from observational studies and routinely collected primary care data show that around half of patients take glucocorticoids longer than this.  

However, the reasons why and the patterns of prescribing for patients are not clear. In addition, it is hypothesed that there are sub-groups of patients with PMR whose glucocorticoid requirements differ. Analyses of patients recruited through secondary care have failed to demonstrate consistently which factors make for a more favourable disease course. Secondly, since most patients with PMR are managed in the community by primary care physicians, with only around 20 to 25% being referred to rheumatologists. Modelling different treatment options, pathways and strategies based on real-world data allows for testing of “what-if” scenarios. 

This PhD project will make use of 23 million individual primary care records in which 150,000 have been treated for PMR to carry out real world analysis of glucocorticoid exposure, patterns, and pathways. Secondly, we will refine these data by supplementing results from patients recruited to the recently funded STERLING trial (£1.8m NIHR HTA) testing the use of methotrexate on glucocorticoid sparing. Using these data we plan to simulate the health care delivery for patients with PMR and assess scenarios of how it might be better managed. Following this we plan a health economic decision analysis to evaluate the costs associated with the various pathways. Together these models will provide important information for policy decision makers about PMR management.