Logo image
Back
Association Between Medicaid Waivers and Medicaid Disenrollment Among Autistic Adolescents During the Transition to Adulthood
Journal article   Open access   Peer reviewed

Association Between Medicaid Waivers and Medicaid Disenrollment Among Autistic Adolescents During the Transition to Adulthood

Meghan E. Carey, Sha Tao, Kaitlin H. Koffer Miller, Steven C. Marcus, David S. Mandell, Andrew J. Epstein and Lindsay L. Shea
JAMA network open, v 6(3), 232768
01 Mar 2023
DOI: https://doi.org/10.1001/jamanetworkopen.2023.2768
PMID: 36912840
Featured in Collection :   UN Sustainable Development Goals @ Drexel
url
https://doi.org/10.1001/jamanetworkopen.2023.2768View
Published, Version of Record (VoR)CC BY-NC-ND V4.0,  Open

Abstract

Health Policy Online Only Research Letter
Introduction Autistic youths are more likely than their nonautistic peers to disenroll from Medicaid as they enter adulthood due to eligibility changes and lack of appropriate Medicaid services.1 States can provide services through waivers that target eligibility criteria based on age and/or diagnosis for specific populations.2 We examined whether waivers were associated with reduced risk of Medicaid disenrollment among autistic adolescents entering adulthood. Methods This cohort study was approved by the Drexel University Institutional Review Board, which waived the need for informed consent owing to the secondary data analysis. The study followed the STROBE reporting guideline. We extracted data from Medicaid for the period 2008 to 2016 from 47 states and Washington, DC (Arizona, Rhode Island, and Vermont utilize section 1115 of the Social Security Act vs traditional waivers and were therefore excluded). Individuals were included if they: (1) had 12 or more consecutive months of Medicaid enrollment, (2) had at least 1 inpatient or at least 2 other claims with an autism spectrum disorder (ASD) diagnosis code,3 and (3) were aged 14 to 26 years during the study period. We constructed longitudinal analytic data sets at the person-month level, combining individual-month–level information on Medicaid eligibility and service use with state-level information on Medicaid programs and procedures. The binary outcome was first disenrollment while enrolled in Medicaid.1 The exposure was a time-varying categorical measure of waiver availability constructed as residence in a state with an active ASD-specific 1915(c) waiver (hereinafter ASD-specific), a non–ASD-specific 1915(c) waiver (hereinafter other), or no waiver. The exposure was interacted with age in a given month to assess variation in the waiver availability–disenrollment association by age. Beneficiary characteristics served as covariates, along with state of residence and calendar year. Characteristics were compared between continuously enrolled and disenrolled individuals using χ2 tests with a 2-sided α of .05. Risk of disenrollment was assessed using marginal predicted probabilities as generated from an adjusted person-month discrete-time proportional hazards model. Additional details are provided in the eMethods in Supplement 1. Results This study included 133 955 autistic individuals: 14 739 who had disenrolled in Medicaid and 119 216 who were continuously enrolled; the mean (SD) age for the total cohort was 16.5 (3.4) years. A higher proportion of disenrolled vs continuously enrolled individuals were Medicaid-eligible due to poverty (32.8% vs 5.3%, respectively); a lower proportion were Medicaid-eligible because of disability (43.2% vs 75.3%, respectively) (Table). A smaller proportion of disenrolled vs continuously enrolled individuals were dually enrolled in Medicare (5.2% vs 14.3%, respectively), and similar proportions of individuals had no waivers available in their state (3.3% vs 3.5%, respectively). The annual probability of disenrollment (Wald P < .001) (Figure) was similar among individuals in states with an ASD-specific waiver, other waiver, or no waiver through age 21 years. Disenrollment increased at age 19 years among those in states with waivers but was less pronounced in states without waivers. At age 22 years, the probability of disenrollment in states without waivers increased markedly (>13%) and remained elevated through age 26 years. In states with a waiver, probability of disenrollment remained stable (approximately 2%) for the same age range. Discussion Residence in a state with Medicaid waivers was associated with a 6-fold or greater decrease in probability of disenrollment among autistic adolescents entering adulthood. These findings are important for state policy, as being insured is associated with fewer unmet health care needs4,5 and may reduce inpatient and long-term care services and expenses.6 States without waivers available for autistic young adults should consider implementing or extending existing waivers to maintain insurance coverage and subsequently improve health outcomes. Interestingly, ASD-specific waivers were not associated with greater decreases in disenrollment compared with waivers for other diagnoses or functional eligibility criteria, suggesting that ASD-specific waivers may not mitigate disenrollment. Our findings suggest that requirements for remaining continuously enrolled in Medicaid when poverty-eligible may be challenging, as small changes in income may result in disenrollment. Study limitations include the potential for misclassification due to incomplete or missing claims data. Policy makers should consider bolstering continuity of waiver availability to keep autistic youths insured and connected to necessary health care services or explore other eligibility mechanisms for Medicaid.2

Metrics

23 Record Views
7 citations in Web of Science
6 citations in Scopus

Details

UN Sustainable Development Goals (SDGs)

This publication has contributed to the advancement of the following goals:

#3 Good Health and Well-Being

InCites Highlights

Data related to this publication, from InCites Benchmarking & Analytics tool:

Collaboration types
Domestic collaboration
Web of Science research areas
Medicine, General & Internal
Logo image