Associations between socioeconomic characteristics and autism treatment: A longitudinal analysis

research objective:  Early childhood (ages 3-5 years) often marks a shift from a focus on early detection and diagnosis of autism spectrum disorder (ASD) to a focus on treatment.  A longitudinal understanding of treatment patterns for young children with ASD is lacking.  The objective of this study is to analyze subject-specific trends in the receipt of educational therapies for ASD and determine whether socioeconomic disparities in trends exist. 

study design:   A longitudinal survey of children participating in preschool special education across the US was conducted between the 2003-04 and 2008-09 school years.  The primary outcome variables included the receipt of speech, occupational, or physical therapy, behavior therapy, and mental health services either inside or outside of school over the first three survey years.  Explanatory variables included race, maternal education, and family income.  Severity of disability was used as a control variable.  Generalized mixed-effects models with random effects for individual intercepts and time trends were used to analyze the data. 

population studied: The study sample included 250 children with parent or teacher identified ASD. 

principal findings:   There was a significant amount of child-specific variation in the patterns of treatment over time.  In models estimating the log-odds of receiving any of the five services, the intra-class correlation consistently found between 54-62% of the variability in rates attributed to child-specific variability.  Significant individual variability was also evident in the slope of the time trend (=475.89, p<0.001).  There were no significant fixed effects in the models examining the effect of explanatory variables on service use independent of service setting.  There was a significant effect of race over time in models examining the effect on service use outside of school.  Conditional on random intercept and trend, children of non-white race were significantly less likely to receive service outside of school than white children (p<0.033).

conclusions:   Significant variation over time in the receipt of treatment services for ASD exists but is often independent of socioeconomic characteristics.  The presence of variation and the direction of the time trend can help support the need for policy to unify a service system for the treatment of ASD.