Geographic variation in long-term health outcomes following myocardial infarction in England
C Sturley, T del Toro, J A Batty, E Booth, M HallAbstract
Background
Survival after myocardial infarction (MI) has improved substantially, resulting in a growing population at risk of subsequent adverse health outcomes. However, the extent to which the incidence of these outcomes varies geographically within the English National Health Service is not well understood.
Purpose
Hospital Episode Statistics data was used to determine the incidence of nine major health outcomes and all-cause mortality following MI at a sub-regional level in England.
Methods
All adults ≥18 years of age with index MI admitted to hospital within all 106 integrated care board (ICB) subpopulations were included (2008-2017). The cumulative incidence of all-cause mortality and non-fatal health outcomes (including subsequent MI, first hospitalisation for heart failure, atrial fibrillation, cerebrovascular disease, peripheral arterial disease, severe bleeding, renal failure, diabetes mellitus and depression) over 9 years of follow-up were estimated by ICB subpopulation from flexible parametric survival models adjusted for age, sex, year of MI and socioeconomic deprivation. Choropleth maps and funnel plots were used to visualise the degree of variation in incidence of each outcome. The intraclass correlation co-efficient (ICC) was used to determine the degree of variation explained at the area versus patient level.
Results
A total of 381,518 individuals with MI were included in the study. The crude incidence of post-MI health outcomes varied considerably between geographic areas, including heart failure (range 20.2-47.9%), severe bleeding (14.3-34.6%), renal failure (23.7-38.0%) and all-cause mortality (34.3-47.7%). The incidence of each outcome fell within expected range (between 95% control limits) after adjusting for patient demographics, with the exception of two ICB subpopulations in northern England, where rates were consistently higher than expected. The variation observed was predominantly explained by area-specific patient demographic profiles rather than at the ICB subpopulation level (ICC<5% for all outcomes).
Conclusions
There was substantial geographic variation in the incidence of post-MI health outcomes across England, despite free and equitable access to healthcare. Understanding variation in post-MI health outcomes by health commissioning area is important to enable targeted resource allocation to areas of greatest need, in order to reduce the significant burden of disease for the growing population of survivors of MI.Cumulative incidence of heart failure