How Did an Autism Health Insurance Mandate Impact Academic and Behavioral Outcomes?

This study examines the impact of Massachusetts’ autism health insurance mandate (ARICA) on the academic and behavioral outcomes of students with autism spectrum disorder (ASD). Using administrative data on Massachusetts public school students, the researcher employs a triple-difference strategy to assess changes in special education inclusion, attendance, suspensions, and achievement. Findings indicate that ARICA increased full inclusion by 4 percentage points, decreased suspension days by 26%, and reduced the likelihood of suspension by 2 percentage points. While test scores did not significantly improve overall, middle school students saw math and reading gains of 0.09 and 0.13 standard deviations, respectively.

Full Citation and Link to Article

Coffey, Stephanie. "The Academic and Behavioral Impacts of an Autism Health Insurance Mandate: Evidence from Massachusetts." Education Finance and Policy, vol. 20, no. 1, 2023, pp. 110–145. DOI: 10.1162/edfp_a_00420.

Extended Summary

Central Research Question

This study investigates the impact of Massachusetts' autism health insurance mandate, known as ARICA (An Act Relative to Insurance Coverage for Autism), on the academic and behavioral outcomes of students with autism spectrum disorder (ASD). The central question guiding the research is: How did the implementation of ARICA affect the educational experiences of students with ASD, specifically in terms of their inclusion in general education settings, attendance, suspensions, and academic achievement? The research explores whether increased insurance coverage for ASD-related treatments led to measurable improvements in these areas.

Previous Literature

Prior studies have examined the broader effects of health insurance mandates on children’s health and educational outcomes. Research on Medicaid and the State Children’s Health Insurance Program (SCHIP) expansions of the 1980s and 1990s found evidence of long-term benefits for children, including improved test scores and higher educational attainment. Studies by Levine and Schanzenbach (2009) and Cohodes et al. (2016) showed that increased Medicaid coverage resulted in better school performance and higher rates of high school graduation and college attendance.

However, research on disability-specific health insurance mandates remains limited. One key study by Acton, Imberman, and Lovenheim (2021) examined the impact of Michigan’s ASD insurance mandate, finding that it led to a decrease in the use of special education services, as private insurance covered some ASD-related interventions outside of school. That study did not find significant effects on test scores. The current study builds on this literature by evaluating ARICA’s impact in Massachusetts, a state with a different regulatory approach and a high supply of ABA (Applied Behavioral Analysis) providers, which could influence the effectiveness of the mandate.

Data

The study uses administrative data on all Massachusetts public school students in grades K-8 from 2006 to 2014. The dataset includes student demographics, disability classification, special education placement, attendance records, suspension data, and standardized test scores in mathematics and English language arts (ELA). Special education setting data categorize students as being in full inclusion (less than 21% of the day outside general education), partial inclusion (21–60%), substantially separate placement (more than 60%), or a separate school.

The author also distinguishes between students likely to be affected by the mandate—those with private insurance—by using eligibility for free or reduced-price lunch (FRPL) as a proxy. FRPL-ineligible students are assumed to be more likely covered under private insurance, while FRPL-eligible students are more likely covered by Medicaid, which was not affected by ARICA during the study period.

Methods

The study employs a triple-difference (DDD) estimation strategy to measure the impact of ARICA. This approach compares changes in outcomes for students with ASD before and after ARICA, relative to students with other disabilities, while also comparing differences between students with private insurance (proxied by FRPL ineligibility) and those with public insurance (FRPL-eligible).

The model controls for various factors, including student demographics (race, gender, English language learner status), school characteristics, and fixed effects for grade and year. The key identifying assumption is that, in the absence of ARICA, trends in academic and behavioral outcomes for privately insured students with ASD would have followed the same trajectory as those for publicly insured students with ASD and for students with other disabilities.

Findings/Size Effects

The findings indicate that ARICA had a meaningful impact on several educational outcomes for students with ASD:

1. Special Education Inclusion

   • ARICA increased the likelihood that a student with ASD would be placed in a fully inclusive setting by 4 percentage points (a 6% increase relative to the pre-reform mean).

   • The likelihood of partial inclusion decreased by 3.6 percentage points, suggesting that some students transitioned from partial to full inclusion.

   • There was no significant effect on the probability of being placed in a substantially separate classroom or separate school.

2. Attendance and Suspensions

   • Attendance rates improved slightly, with an increase of 0.3 percentage points, equivalent to about half a day of additional school attendance per year.

   • Suspension rates declined significantly:

     • The number of days suspended dropped by 26% (0.038 days).

     • The likelihood of receiving any out-of-school suspension fell by 2 percentage points (a 28% reduction).

   • These reductions in suspension rates were most pronounced for boys and White/Asian students, with smaller effects for Black/Hispanic students.

3. Academic Achievement

   • Overall, there was no significant improvement in standardized test scores for students with ASD.

   • However, middle school students (grades 6–8) saw small but notable improvements:

     • Math scores increased by 0.09 standard deviations (SD).

     • ELA scores increased by 0.13 SD.

   • Elementary school students (grades K–5) did not experience statistically significant changes in test scores.

     
     

Conclusion

The findings suggest that ARICA had a positive impact on the educational experience of students with ASD in Massachusetts. The policy led to increased inclusion in general education classrooms and significant reductions in suspension rates. The fact that middle school students saw modest test score improvements suggests that the effects of increased ASD services may take time to manifest in academic performance.

However, the study also highlights disparities in impact. The benefits of ARICA were more pronounced for White and Asian students compared to Black and Hispanic students, possibly due to differences in access to ASD services. Additionally, while inclusion improved, there is little evidence that overall special education services within schools were significantly adjusted in response to the increased availability of private insurance coverage.

The study contributes to the literature on health and education policy by showing that insurance mandates for ASD treatment can have meaningful spillover effects on students’ schooling experiences. Future research could explore the long-term outcomes of these policies, particularly their effects on high school graduation and postsecondary success.