Association between co-occurring psychiatric conditions and premature mortality in autistic people: population-based cohort study
Mark J. Taylor, Miriam I. Martini, Ralf Kuja-Halkola, Sarah Cassidy, Angelica Ronald, Magnus Tideman, Isabell Brikell, Zheng Chang, Brian M. D’Onofrio, Paul Lichtenstein, Agnieszka Butwicka, Tatja Hirvikoski, Henrik LarssonBackground
Autistic people are at an elevated risk of premature mortality; however, few studies have identified factors that explain this association. Co-occurring psychiatric conditions are a strong candidate, given that autistic people are at a high risk of such conditions and these conditions are broadly associated with premature mortality.
Aims
We aimed to investigate the degree to which co-occurring psychiatric conditions are associated with premature mortality in autistic people. We explored these associations across sex and co-occurring other neurodevelopmental conditions.
Method
We conducted a cohort study based on registry linkage. We identified 2 958 317 individuals born in Sweden from 1974 to 2004, including 70 546 (2.38%) individuals with an autism diagnosis. We identified diagnoses of 13 psychiatric conditions. The cohort was followed up from age 16 to age 46 at the oldest. We compared the risk of mortality (from all-causes, suicide, other external causes and natural causes) across autistic people with psychiatric conditions and (a) non-autistic people without psychiatric diagnoses; (b) autistic people without psychiatric diagnoses; and (c) non-autistic people with psychiatric diagnoses. We conducted separate analyses by sex and co-occurring neurodevelopmental conditions.
Results
The mortality rate in autistic people with psychiatric conditions was 3.2 deaths per 1000 person-years, which was higher than that observed in non-autistic (0.27 deaths per 1000 person-years) and autistic people without psychiatric conditions (0.96 deaths per 1000 person-years). The relative risk of premature mortality was higher in autistic people with psychiatric conditions compared with both non-autistic (hazard ratio 13.85, 12.86–14.91) and autistic people (hazard ratio 3.44, 2.94–4.02) without psychiatric conditions and also exceeded the risk of premature mortality in individuals with psychiatric conditions alone (hazard ratio 1.47, 1.37–1.58). Similar patterns emerged across sex, intellectual disability and attention-deficit hyperactivity disorder.
Conclusions
Co-occurring psychiatric conditions are a risk factor for premature mortality in autistic people. Timely detection and treatment of such conditions in autistic people may facilitate healthier, lengthier lives for autistic people.