Modelling the Effect of Multi-Indication Treatment in Patients With Multimorbidity Using Population-Scale Linked Electronic Health Records for Healthcare Policy and Decision-Making

OBJECTIVES: Healthcare decision-making has previously focussed on developing recommendations for single conditions. However, standardised care for each chronic condition in isolation can be inappropriate for individuals living with multiple long-term conditions known as multimorbidity, and may lead to unnecessary polypharmacy. This research aimed to develop a modelling framework to estimate the effect of a multi-indication treatment in multimorbid populations.

METHODS: Motivated by recent National Institute for Health and Care Excellence (NICE) recommendations on the use of a Sodium-Glucose Cotransporter-2 (SGLT2) inhibitor in three commonly co-existing long-term conditions: type 2 diabetes mellitus (T2DM), chronic kidney disease (CKD), and heart failure (HF), age-adjusted multistate models were used to estimate evolving temporal sequences of diseases and mortality. Baseline hazards were estimated using population-scale, individual-level, anonymised, linked electronic health record (EHR) data for 613,195 individuals aged 55 to 85 years in Wales, over a 20-year period (2000–2019). Hazard ratios for treatment effects were obtained from published randomised controlled trials and implemented in a patient-level simulation to estimate health-state transitions for different temporally ordered sequences of multimorbidity.

RESULTS: SGLT2 inhibitors increased the estimated mean life-expectancy from 0.02 (95%CI:-0.02,0.05) to 1.44 (95%CI:1.12,1.74) years in multimorbid T2DM, CKD and HF, with the largest gain in estimated mean life-expectancy for individuals with HF followed by T2DM. The estimated gain in life-expectancy was less than 0.5 years for individuals with T2DM, CKD and HF in different temporally ordered sequences. The estimated mean time to develop HF increased by 3.1 (95%CI:3.00,3.24) and 1.34 (95%CI:1.12,1.56) years in individuals with CKD, and CKD followed by T2DM, respectively.

CONCLUSIONS: There is an increasing need to appraise interventions in multiple long-term conditions to identify optimal treatment strategies and reduce polypharmacy in multimorbid populations. Multistate models applied to linked EHRs allowed for a more rigorous assessment of treatment effects in multimorbidity for healthcare decision-making.