Autism Behavioral Phenotype and Health Across the Life Span

Background: Little is known about adults with autism spectrum disorders during the midlife and aging stages of the life course, as few in these age groups were correctly diagnosed with autism.  Yet the baby boom generation of individuals who warrant an autism diagnosis is now well into midlife and some are approaching old age, with a concomitant increase in the need for health care and services. 

Objectives: The goal of the present study is to describe the behavioral and health profile of autism in later life, drawing upon the available data from a 10-year longitudinal study of autism during adolescence and adulthood.  

Methods: The present sample included 406 individuals with autism spectrum disorders who were age 10-52 when the study began and are currently 10 years older.  Data have been collected six times over the 10 year period, making it possible to examine age-related changes in health, medications, functional abilities, behavioral problems and autism symptoms, and adult outcomes.  The present analysis includes three methodological approaches: (1) cross-sectional comparisons with adults who do not have disabilities, drawn from a nationally-representative sample, and adults with Down syndrome, (2) semi-longitudinal trajectories estimated by Friedman’s “supersmoother” (a non-parametric regression approach implemented in R, and (3) examination of mortality in adults during the 10-year study, including causes of death. 

Results: In comparison with age peers who do not have disabilities, adults with ASD were significantly more likely to be in poor health and less likely to be in excellent health.  Semi-longitudinal regressions revealed a pattern of slow age-related decline in health with more rapid decline after age 45, an increase in prescription medications (again more sharply after age 45), a steady decline in behavior problems during adulthood, and stability in functional abilities.  Eleven adults died during the 10-year study period, with four deaths caused by heart attacks and three as a result of accidents.  Adults with ASD who do not have an intellectual disability (ID) had better adult outcomes (residential independence, vocational independence, and social contact with friends) than adults with ASD and ID, or than adults with Down syndrome. 

Conclusions:  These findings suggest that adults with ASD have a long-term reliance on the public service system and the family, with limited profiles of independence in midlife.  The baby boom generation with ASD (who may or may not be diagnosed with ASD) is approaching old age and many will need long-term care services.  The need for planning for the future in terms of adult sibling involvement and other services is evident in the data. The heterogeneity of ASD is evident in the later years of the life course, just as it characterizes children with ASD.  Finally, the data offer implications for health surveillance and accident prevention in adults with ASD.