Improving Autism and Developmental Screening for Practices Serving Latinos: The REAL-START Project
U.S. Latino children are diagnosed with Autism Spectrum Disorder [ASD] less often, at older ages, and at the point of greater severity. Improving ASD identification in the primary care setting may increase access to diagnosis and treatment and reduce disparities.
We conducted a 12-month intervention in 6 Oregon primary care practices serving Latino communities. The goal of the intervention was to raise rates of adherence to American Academy of Pediatrics [AAP] ASD and developmental screening guidelines at 18- and 24-month well child visits, and to increase Early Intervention [EI] referrals for children with high risk of ASD or other developmental conditions.
Forty-two medical providers from 6 practices attended a 1 ½ - hour initial in-person training on the ASQ-3 (Ages and Stages -3; general developmental screener) and MCHAT-R/F (Modified Checklist for Autism in Toddlers – R/F; autism screener), and on proper referral practices for children with screen failure. Three 1-hour follow-up trainings at 3, 6, and 9 months addressed topics specific to Latinos, such as communication about screening to families with low health literacy, the relationship between bilingualism and language delays, and access to autism services through Medicaid. Practice leaders met with a facilitator 4 times to review practice- and provider-level data on screening and referral rates, and conducted quality improvement activities to improve these rates. A medical record review captured screening and referral rates for all 18- and 24-month well visits at baseline and at 3, 6, 9, and 12 months after the initial intervention. A query of the Oregon Early Intervention database assessed EI eligibility rates. Estimates of screening rates accounted for clustering at the clinic and provider level and are adjusted for child age.
In total, 2357 18- and 24-month well child visits were captured during the intervention; 39% of visits were for Latino children. Practices improved age-adjusted rates of ASD screening with MCHAT-R/F at well visits, from 58% of well visits screened at baseline to 89% at the end of the intervention (p<0.001). Practices also improved rates of general developmental screening with ASQ-3, from 59% to 95% (p<0.001). Rates of complete adherence to AAP guidelines (i.e., conducting both screens) increased from 35% to 87% (p<0.001). The greatest improvements in screening rates occurred in the first 6 months of the intervention. Screening and referral rates were comparable for Latinos and non-Latinos throughout the intervention. While the proportion of children referred to EI due to screen failure stayed relatively constant, the absolute number of referrals increased. Additionally, the absolute number of children who qualified for EI services increased slightly (3.0 to 4.5 per study period; p = 0.28), but the proportion of children referred to EI who were found to be eligible decreased slightly.
This longitudinal intervention was effective in sustainably increasing screening for ASD and other developmental disabilities in primary care settings with many Latino patients. Increased screening rates were associated with increased Early Intervention referral rates, and a trend toward increased EI eligibility.