P3.P317: Understanding the Associations Between Sociodemographic Factors and Infant Bayley IV Performance

Background: The Bayley Scale of Infants and Toddler Development IV (BSID-IV) is an established tool utilized in NICU follow-up clinics to evaluate high-risk infants for neurodevelopmental delays. Numerous studies suggest that lower scores on the BSID-IV are associated with male sex, maternal education, maternal employment, and residing in high-risk neighborhoods. Our research aims to expand on this knowledge and investigate whether other sociodemographic factors are also linked to BSID-IV performance.

Methods: We identified infants who were scheduled for NICU follow-up clinic appointments between Jan 1, 2021, and Dec 31, 2022. Infants were excluded if: they attended ≤1 appointment, the BSID-IV was not administered, there were missing domains of the conducted BSID-IV, or the adjusted age for preterm infants and chronological age for term infants was ≤ 17 months and 15 days. Retrospective clinical and sociodemographic information was collected for each infant. For the statistical analysis, chi-squared and nonparametric tests were performed to determine independent sociodemographic factors associated with poor performance on the BSID-IV.

Results: After exclusion, 103 infants remained in our study for analysis. The majority of our infants identified as Black/Hispanic (65%); race/ethnicity was not significant with any component of the BSID-IV. We found that male sex, higher discharge weight, and longer length of stay were significantly associated with lower scores in the cognitive (see Table 1), language, and motor domains. Public insurance, compared with private, was associated with lower scores in the cognitive (mean 87.0 [95% CI 83.4-90.6] vs 94.7 [91.3-98.1]) and language domains (mean 82.3 [95% CI 78.6-86.0] vs 91.6 [87.5-95.8]). Infants of single mothers were associated with lower performance in the cognitive domain (mean 80 [95% CI 74.5-85.9] vs 91.83 [86.7-97.0]). Infants who resided in English-speaking households, compared to Spanish-speaking, scored higher in the language domain (mean 87.93 [95% CI 84.7-91.2] vs 80.12 [74.9-85.3]). Overall, lower opportunity neighborhoods, as defined by the Childhood Opportunity Index, were associated with lower scores in the language domain (Fig. 1).

Conclusion: In our majority non-white population, the cognitive and language domains compared to the motor domain of the BSID were associated with various clinical and sociodemographic factors, especially the impact of language spoken at home and level of neighborhood opportunity in the language domain. Developmental professionals should keep this in mind when interpreting the BSID-IV as a tool for clinical prognosis and severity of neurodevelopmental outcomes. Structural and policy interventions can be designed in the future with the intent of improving early language proficiency for this vulnerable population of children.