Predictors of Early Intervention Referral after a Positive Developmental Screen in Community Primary Care Clinics.

Abstract

Background: Primary care physicians (PCPs) may rely upon factors other than screening test scores in making referral decisions to developmental services. This study investigated which patient, provider, and screening test factors predict a PCP's IDEA Part C Early Intervention (EI) referral after a positive screening test result.

Methods: Child demographics, developmental screening test results and EI referral decisions were collected via medical record review of 2,756 15-, 18-, 24- and 30-month well-child checks conducted at 7 community primary care clinics in 4 Oregon counties, in 2020-2021. A provider survey collected PCP demographic and professional characteristics. We tested the association of receipt of EI referral with screening test (Ages & Stages-3 [ASQ-3] and Modified Checklist for Autism in Toddlers - Revised [MCHAT-R]) scores, provider demographic information, child demographic data (sex, language, race/ ethnicity), using multivariable logistic regression.

Results: 54.1% of children with positive MCHAT-R screens, and 42% of children with positive ASQ-3 screens received EI referrals. Multivariable analyses showed that MCHAT-R score, ASQ-3 Communication and Gross Motor scores were associated with referral after a positive screen. Child sex, race, ethnicity, and language, and provider demographics had no multivariable association with referral. Referral rates varied substantially by site and individual provider.

Conclusion: The majority of toddlers with positive developmental screens were not receiving EI referrals from their PCP during the COVID-19 pandemic. Screening test thresholds and clinical thresholds for EI referral differ substantially, and some portions of the ASQ-3 do not seem to impact provider referral decisions. These findings may help inform physician training on developmental screening in primary care, specifically during times of unprecedented healthcare challenges.