Prevalence, Incidence, and Prodromal Presentation of Dementia in Medicaid Beneficiaries with Autism Spectrum Disorder

Background/Purpose: Autistic adults now have the opportunity to live long lives in their communities, yet it is unknown whether autism spectrum disorder (ASD) confers increased risk for dementia. Known dementia risk factors—including depression, hypertension, hyperlipidemia, and diabetes—are elevated in autistic adults, and these factors may contribute to the emergence of dementia. Population-based epidemiological work is needed to clarify the extent of clinical dementia in autistic adults and assess prodromal risk factors. Our objective was to describe prevalence and incidence of dementia in autistic adults compared to adults with intellectual disability (ID) and to assess the impact of prodromal dementia risk factors (depression, diabetes, hypertension, hyperlipidemia) on incident dementia.

Methods: We assessed Medicaid claims for adults (≥21 years) who ever had two claims for ASD or ID on two different days during their Medicaid enrollment. Our sample included 3,336 autistic adults without ID, 2,524 autistic adults with ID, and 13,482 adults with ID. We excluded adults with Down syndrome because of the known link between Down syndrome and dementia. We analyzed claims from January 1, 2008 to December 31, 2018. Dementia claims were extracted from codes for any dementia from the Center for Medicaid Services Chronic Conditions Data Warehouse Condition Categories. We required ≥3 years Medicaid enrollment to ensure validity of dementia claims; therefore, beneficiaries entered our cohort at any point between 2008 and 2015. We categorized age at first claims by age group (<40, 40-54, ≥55) to account for confounding by age and compared dementia prevalence and incidence by diagnostic group using an age category-adjusted Poisson regression. Finally, we compared prevalence of prodromal risk factors (depression, diabetes, hypertension, hyperlipidemia) as identified in claims prior to incident dementia by group.

Results: In adults aged 55 and older, 19.09% (N=61) of autistic adults without ID, 21.6% (N=132) of autistic adults with ID, and 24.9% (N=1,623) of adults with ID had claims for dementia. In the ASD groups, dementia was 0.48 times as prevalent (95% Confidence interval: 0.4, 0.6) in the non-ID compared to the ID group in an age-adjusted Poisson regression. Incidence did not differ between ASD without ID, ASD with ID, and ID groups in the 40-54 (log rank test P=0.06) or ≥55 age group (P=0.3). Prevalence of prodromal depression (45.7%) was highest in the ASD only group while prevalence of prodromal diabetes (26.9%), hypertension (49.1%), and hyperlipidemia (41.5%) were highest in the ID group.

Conclusions and Implications: While our interpretation is limited by reliance on claims rather than direct assessment and a lack of non-developmental disability control group, our findings from a state-wide health service system indicate that dementia is highly prevalent in autistic adults after the age of 55. High dementia prevalence in Medicaid beneficiaries with ASD underscores the need to develop dementia services and supports for autistic adults as they age and reside in Medicaid-funded assisted living and/or skilled nursing facilities. Findings also stress the need for exploration into how dementia manifests in autistic adults.