608 - Use of the Parent's Observations of Social Interactions (POSI) to assess Social Development in a High Risk Infant Follow-Up Program

Poster Number: 608
Publication Number: 608.310

Raye-Ann deRegnier, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, IL, United States; Sharon L. Veis, Northwestern University, Chicago, IL, United States; Robin Manus, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, IL, United States; Aarabhi Rajagopal, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, IL, United States

Background: The Parent's Observation of Social Interactions (POSI) is a newer screening test for autism.  In the cohorts studied thus far, it has shown similar or better sensitivity to the Modified Checklist for Autism, but somewhat lower specificity.  There is little data on the use of the POSI in a High Risk Infant Follow Up (HRIF) clinic with a case mix of children with normal development and those with neurodevelopmental impairments, delays and autism.  The features of the POSI (brevity, specific targeting of social and play behaviors) make it complementary to the use of more formal developmental testing. The association between POSI scores and cognitive, language and motor delays has not been studied.

Objective: The purpose of this study was to correlate results of the POSI with neonatal biomedical and demographic factors and developmental delays assessed using the Bayley Scales of Infant Toddler Development-III or IV (BSITD).

Design/Methods: In this retrospective study, we included toddlers tested using the POSI and BSITD at the same visit to our HRIF clinic at 18-35 months of age.  The POSI was completed by parents.  Per the scoring instructions, at-risk scores were those  > 3. The BSITD-III or IV was administered by an experienced therapist or psychologist.  Delays were defined as BSITD composite scores < 85. Correlations were performed using logistic regression with version of the BSITD (III vs IV) as a cofactor as pertinent.

Results: Table 1 shows the neonatal risk factors and demographic information for the 172 children studied.  At risk POSI scores were noted in 51 toddlers (29.7%). At risk POSI scores were not associated with sex, prematurity, neurologic abnormalities, cardiac surgery, congenital anomalies, or multilingual home language environment.  At risk POSI scores were associated with developmental delays (Table 2).  Toddlers with >1 domain of delay on the BSITD had the highest risk for at-risk POSI scores (OR 7.2; 95% CI: 2.8-17.8, P < 0.001) but those with isolated delays also had increased odds (OR 2.5; 95%CI: 1.02-6.042, p=0.045).Conclusion(s): Targeted screening of social development using the POSI was feasible in a HRIF program and a relatively high percentage of toddlers had at-risk scores. Toddlers with delayed development in any domain were at increased odds of an at-risk POSI score, particularly those with cognitive delays or multiple domains of delay.  Longitudinal follow up of this cohort is underway to assess the prediction of autism vs. delayed social development in this group of children who are at risk for multiple types of neurodevelopmental impairments.
Table 1. Neonatal Clinical and Demographic Characteristics*Mild-moderate, n=34; Severe, n=40. **Isolated major anomalies, n=46; multiple abnormalities/genetic syndromes, n=44.
Table 2. Association of BSITD Delays and At-Risk POSI scores*adjusted for Bayley version 3 vs. 4