Cohort Differences in the Social Participation of Youth on the Autism Spectrum Receiving Special Education Services

Background: Rates of intellectual disability and mean impairment severity among children identified in surveillance studies of autism have been steadily declining for over a decade. No studies have investigated whether there have been historical changes in rates of social participation (measured here as extracurricular activities and friendship with peers) among youth on the autism spectrum.

Objectives: Compare rates across two cohorts of social participation in youth on the autism spectrum receiving special education services. Hypothesis 1: Social participation rates would be higher in the more-recent cohort. Hypothesis 2: Low parental education, low household income, low independent functioning and severely impaired communication ability would all be associated with lower rates of social participation in the more-recent cohort.

Methods: We used secondary data from two related cohort studies that were funded by the U.S. Department of Education and conducted 10 years apart: Wave 1 (2001-2002) of the National Longitudinal Transition Study-2 and the National Longitudinal Transition Study-2012. Both studies were designed to yield nationally representative estimates of the characteristics and experiences of youth with special needs who received special education services. The goal of valid cross-cohort comparison was designed into the 2012 study by using similar sampling methods and identical question wording for many items. We limited analyses to youth who were ages 13 to 17 years. Our outcomes were dichotomous indicators of a) any participation in any kind of extracurricular school activities (e.g., sports teams, band or chorus, school clubs, or student government) during the prior 12 months, and b) never saw friends at all during the prior 12 months. We used a cohort dummy variable in logistic regression to estimate the differences between cohorts adjusting for covariates (youth and household demographics, indicators of impairment severity and disability characteristics), weights and sampling method.

Results: The rates of any extracurricular school activities were 27% (95% CI: 22,34) for the 2001 cohort and 60% (95% CI: 56,65) for the 2012 cohort. The rates of never seeing any friends in the past 12 months were 41% (95% CI: 33,49) for the 2001 cohort and 36% (95% CI: 32,41) for the 2012 cohort. Extracurricular participation was significantly higher in the 2012 cohort after adjusting for covariates (odds ratio: 4.6; 95% CI: 3.0,6.9). There was no significant adjusted difference between cohorts on the rate of never seeing friends (OR: 0.8; 95% CI: 0.5,1.3).

Conclusions: We found higher rates of participation in structured extracurricular activities but no significant differences in the rate of unstructured meeting with friends. Structured extracurricular activity participation is associated with better young adult outcomes in typically developing samples. Whether such participation enhances postsecondary outcomes for youth on the autism spectrum should be investigated. Schools need to prepare for growing numbers of youth on the autism spectrum participating in structured activities. Research should also focus on ways to support friendship development and reduce social exclusion.