Characteristics and Health Conditions of Publically-Insured Autistic Young Adults

Background:

As increasing numbers of autistic adolescents age into adulthood, a primary concern is whether adult healthcare needs are recognized and treated appropriately (Nicolaidis et al., 2012). Croen et al. (2015) found that adults with ASD were more likely to have psychiatric, immune, cardiovascular, metabolic, endocrine, neurologic, gastrointestinal, and other conditions than same-age adults without ASD. Public insurers may be used by adults with autism to meet their healthcare needs. Although Medicaid is the most well-known public payer, persons younger than 65 can obtain access to Medicare. Young adults with autism may be eligible for Medicare due to eligibility through SSDI or old-age and survivors benefits. Due to the increasing prevalence and public health burden of increasing care needs of aging adults with ASD, it is imperative to identify utilization among a publically insured sample of adults with ASD.

Objectives:

This presentation will describe differences in comorbid conditions among young adult Medicare beneficiaries aged 18-25 years with ASD and no intellectual disability (ASD-only), as compared to beneficiaries with ASD and ID (ASD+ID) and adults with ID only. Subgroup analyses by race/ethnicity and gender will be presented.

Methods:

We conducted a retrospective analysis of existing national Centers for Medicare and Medicaid (CMS) claims Limited Data Sets (LDS) for 2008-2010. Claims associated with beneficiaries with ICD-9 codes of 299.xx were flagged as ASD, and 317.xx, 318.xx or 319.xx represented beneficiaries with ID. Analyses were conducted on all extracted claim files for eligible beneficiaries (inpatient, outpatient, home health, hospice care, skilled nursing facility services and professional services). We used the Agency for Healthcare Research and Quality H-CUP Clinical Classification Software (CCS)® to create condition flags from each claim (AHRQ, 2015). CCS uses standard ICD-9-CM codes to flag claims. Additional confirmation of CCS condition flags with ICD-9 code classification is underway. Chi-square statistics were used to identify differences in rate of occurrence of each condition between groups. Bonferroni correction for multiple comparisons resulted in differences between groups with p less than .003 as considered statistically significant.

Results:

Due to space, only 2008 results are described below; presentation data will contain all three years. Table 1 displays rates and 95%CI for specific conditions between the three groups. Seizure disorder occurred more frequently among young adults with ASD+ID than the other two groups; this difference was observed in 2009 and 2010 claims. Parkinson’s disease was more prevalent among adults with ASD+ID than those with ID-only in 2008; this difference was not observed in other claim years. Adults with ID-only were significantly more likely than adults with ASD to experience 10 of the 17 conditions examined.

Conclusions:

Observed rates of psychiatric and medical conditions were significantly greater among young adults with ID than those with ASD for most conditions. Findings of greater rates of seizure disorder mirror those of other research. Additional analyses are ongoing to adjust for demographic characteristics that differ between groups. Practitioners, researchers and policymakers should attend to the impact that developmental disabilities, generally, have on health and health maintenance.