Age-related physical health of older autistic adults in Sweden: a longitudinal, retrospective, population-based cohort study.

BACKGROUND

Research of health outcomes in older autistic adults (≥45 years) is concerningly scarce, and little is known about whether intellectual disability and sex affect the health outcomes of this population. The aim of this study was to investigate the association between autism and physical health conditions in older adults and to examine these associations by intellectual disability and sex.

METHODS

We conducted a longitudinal, retrospective, population-based cohort study of the Swedish population born between Jan 1, 1932, and Dec 31, 1967, using linked data from the nationwide Total Population Register and the National Patient Register. We excluded individuals who died or emigrated before the age of 45 years, or with any chromosomal abnormalities. Follow-up started at age 45 years for all individuals, and ended at emigration, death, or Dec 31, 2013 (the latest date of available follow-up), whichever was soonest. Diagnoses of autism, intellectual disability, 39 age-related physical conditions, and five types of injury (outcomes) were obtained from the National Patient Register. For each outcome, we calculated 25-year cumulative incidence and used Cox models to estimate hazard ratios (HRs). All analyses were repeated separately by intellectual disability and sex.

FINDINGS

Of 4 200 887 older adults (2 063 718 women [49·1%] and 2 137 169 men [50·9%]) in the study cohort, 5291 (0·1%) had a diagnosis of autism recorded in the National Patient Register. Older autistic adults (median follow-up 8·4 years [IQR 4·2-14·6]) had higher cumulative incidence and HRs of various physical conditions and injuries than their non-autistic counterparts (median follow-up 16·4 years [8·2-24·4]). In autistic individuals, the highest cumulative incidence was observed for bodily injuries (50·0% [95% CI 47·6-52·4]). Conditions that autistic adults were at higher risk of than were non-autistic adults included heart failure (HR 1·89 [95% CI 1·61-2·22]), cystitis (2·03 [1·66-2·49]), glucose dysregulation (2·96 [2·04-4·29]), iron deficiency anaemia (3·12 [2·65-3·68]), poisoning (4·63 [4·13-5·18]), and self-harm (7·08 [6·24-8·03]). These increased risks mainly persisted regardless of intellectual disability or sex.

INTERPRETATION

Our data indicate that older autistic adults are at substantially increased risk of age-related physical conditions and injuries compared with non-autistic adults. These findings highlight the need for collaborative efforts from researchers, health services, and policy makers to provide older autistic individuals with the necessary support to attain healthy longevity and a high quality of life.

FUNDING

Swedish Research Council, Servier Affaires Medicales.

TRANSLATION

For the Swedish translation of the abstract see Supplementary Materials section.