Mortality in adults with learning disabilities with and without a health check: a cohort study

BackgroundAnnual health checks in primary care were introduced in Wales in 2006, for adults with learning disability on the social services register. The health check includes screening for conditions such as diabetes, general health measures, and medication review. We aimed to compare mortality rates in individuals who have a health check and those with no record of having a health check in their medical records.MethodsWe stratified general practice records of 24 474 people with a learning disability in Wales by health check status (read code for a health check vs no health check). The Secure Anonymised Data Linkage databank was used to link general practitioner (GP) data and Office for National Statistics mortality data between Jan 1, 2005, and Dec 31, 2017. We used survival analysis (Cox's regression) to calculate unadjusted and adjusted (for age, sex, and comorbidity) mortality hazard ratios (HRs).FindingsOf the 24 474 people with a learning disability in Wales, 7542 (30·8%) had a confirmed health check and 16 932 (69·2%) had no record of a health check. Mortality rates were higher for people with a learning disability who have never received a health check than those who had received a health check (3·55 deaths per 1000 individuals per year in the non-health check group vs 2·08 deaths per 1000 individuals in the health check group). Among individuals with autism, mortality was lower among individuals who had a health check than those who had no check (HR 0·61, 95% CI 0·39–0·96; adjusted HR 0·53, 0·33–0·83 [adjusted for age and sex]). Among individuals with epilepsy, mortality was also lower among individuals who had a health check than those who had no check (HR 0·72, 0·64–0.81; adjusted HR 0·65, 0·58–0·74 [adjusted for age and sex]) and the same trend was observed among individuals with diabetes (HR 0·77, 0·64–0·92; adjusted HR 0·71, 0·60–0·84). The largest differences in mortality between individuals with and without a health check were observed among individuals aged 18-50 years (adjusted HR 0·46, 0·39–0·50 [adjusted for comorbidity and sex]), with a smaller difference in mortality identified among individuals aged older than 50 years at their first health check (adjusted HR 0·81, 0·72–0·90 [adjusted for comorbidity and sex]).InterpretationMortality is lower among people with learning disabilities who have health checks than those who do not. Our findings suggest that health checks provide long-term benefit, especially for younger people.FundingWelsh Assembly Government.