Prevalence and utilization of medical care services in persons with autism spectrum disorder using Medicare claims

There is a lack of information on how adults and children with autism use medical services covered by Medicare.  State Medicaid rules and birth cohort effects likely influence observable utilization.  Our purpose is to characterize the prevalence of autism among persons in Medicare claims and contrast with prevalence documented during disability determination in Social Security Administration (SSA) Supplemental Security Income (SSI) files.   

 CMS national 5% sample claims (2008-2010) were used to identify beneficiaries with autism diagnoses (299.xx) in the professional services (Carrier) file. Any single claim was designated as a case.  These were linked to other standard claims files. The nationally weighted estimate of persons with autism in Medicare files was compared with SSA/SSI data for beneficiaries eligible due to autistic disorder in 2010. 

 A national estimate of the number of expected persons with autism in the Medicare 100% claim file was 15,840.  National estimate 65+ was only 1,280.  90% of records were for persons less than 35.  Most (~66%) case claims had no autism diagnosis. SSA disability records indicate 151,260 persons <65 with autistic disorder in 2010. 

 Low estimates for older age cases are consistent with cohort effects.   Disabled SSI recipients <65 are automatically enrolled in Medicare Part A (hospital) insurance after 24 months.  Part B insurance requires premiums but most states automatically enroll SSI recipients into premium support programs (e.g., QMB’s).  Discrepancy might be explained by comprehensive state Medicaid coverage, low realized access to care, poor documentation of diagnosis in claims, or consequences of disability program rules.