Inhibited Toddlers and ASD Screening

Background: A child’s inhibited temperament greatly influences social engagement and exploration particularly in unfamiliar situations. Hence, inhibition can be confused with social-communication delays. Inhibition, defined by shyness, withdrawal, and under-activity, is rarely considered a mediator of ASD screening results.

Objectives: (1) Examine reliability and validity of inhibition observation. (2) Describe early ASD markers unique to those who were inhibited at one-year of age. (3) Determine whether inhibition increases false clinical referral for a developmental evaluation.

Methods: Parents of 84 12-month-old toddlers completed the First Year Inventory (FYI) an ASD screener. At 13 months toddlers participated in an in-home assessment including the Autism Observation Schedule for Infant (AOSI) and Mullen Scales of Early Learning (MSEL). The AOSI includes a reactivity item that captures both under- and over-reactivity. At 30 months parents completed the Infant Toddler Social Emotional Assessment (ITSEA) and toddlers participated in an in-home assessment including the MSEL and Autism Diagnostic Observation Schedule – Generic (ADOS-G). At 13 and 30 months toddlers were referred for a diagnostic evaluation if warranted by assessment results (n=18, 16, respectively). Inhibition was rated during the AOSI by experienced clinicians blind to the child’s risk status. Five children were excluded from the non-inhibited group due to their rating as over-active.

Results: Twelve infants were rated as inhibited, 7 of which were also captured by the AOSI reactivity item. Inter-rater agreement for inhibition observation was significant, PABAK = 0.79. Parents of all infants in this group stated that the toddler behaved as usual during the observation. Inhibited infants received a significantly higher risk on the FYI social-communication score but not on the sensory-regulatory score; 58.3% met FYI risk cutoff versus 10.4% in the non-inhibited group. Inhibited infants showed significantly higher risk rates on the following five AOSI items: visual tracking, social babbling, response to name, reciprocal social smile, and social interest and shared affect. The inhibited versus control group had significantly lower early learning composite scores at 13 months, and marginally lower at 30 months. At 30 months this group had a significantly higher clinical referral rate of 54.5% versus the non-inhibited children (16.7%). However, there was no difference in rates of false (positive or negative) referrals at 13 versus 30 months referral; or based on risk on the FYI relative to 13 month referral. At 30 months, the inhibited versus the non-inhibited group had significantly (p < .05) higher rates of meeting ITSEA 90^th percentile for inhibition to novelty (45.5%, 15.5% respectively) and depression/withdrawal scores (36.4%, 8.5% respectively).

Conclusions: Inhibition was reliably measured and persisted over time for about half of the children. An inhibited, withdrawn and difficult to engage infant has a higher likelihood of falling at risk on both parent and clinician based ASD screening measures. Nonetheless, inhibition did not inflate false screening results.