BackHome
Refresh: OffView Personal Schedule
Individual Submission Summary
Share...

Direct link:

Neurobehavioral and Medical Predictors for Very Preterm Infants of M-CHAT Outcomes for Toddlers

Sat, March 25, 8:15 to 9:45am, Salt Palace Convention Center, Floor: 1, Meeting Room 151 A-C

Abstract

Introduction: The Modified Checklist for Autism in Toddlers (M-CHAT) is a screening tool for autism risk, but the accuracy of this measure is questionable (Carbone et al., 2020). Preterm infants and infants with medical morbidities are more likely to screen positive for autism (Kuban et al., 2009; Limperopoulos et al., 2018; Moore et al., 2012). The NICU Network Neurobehavioral Scale (NNNS; Lester & Tronick, 2004) is a well-validated assessment for newborns at risk for poor neurodevelopment that has been shown to predict poor outcomes at 2-years adjusted age on the Bayley-III and Child Behavior Checklist (CBCL; McGowan et al., 2022). Attention on the NNNS has been associated with social attention in infancy, a marker of autism risk (Bradshaw et al., 2020). Understanding the association between predictors of M-CHAT for very preterm (VPT) infants and the relationship between M-CHAT and developmental outcomes at 2-years may provide information about characteristics of children who screen positive on the M-CHAT.
Hypotheses: (1) VPT infants with dysregulated NNNS profiles and more medical morbidities will be more likely to screen positive on the M-CHAT Revised with Follow-up (R/F). (2) Those who screen positive on the M-CHAT R/F are more likely to have worse developmental and behavioral outcomes on the Bayley-III and CBCL.
Study Population: VPT infants (N=704) born <30 weeks PMA enrolled in The Neonatal Neurobehavior and Outcomes in Very Preterm Infants (NOVI) study.
Methods: Previous work used Latent Profile Analysis to determine mutually exclusive NNNS profiles with profiles 5 and 6 being the high neurobehavioral risk categories (Figure 1; McGowan et al., 2020). Data regarding four medical morbidities (dichotomized as 0-1; 2-4 diagnoses) and the results of NNNS (profiles 1-4; 5-6) were collected prior to NICU discharge. M-CHAT R/F, Bayley-III, and CBCL were collected at 2-years adjusted age. Logistic regressions predicted M-CHAT R/F screening with medical morbidities, NNNS profiles, and medical by NNNS interaction, while controlling for demographics. Unadjusted t-tests compared Bayley-III and CBCL scores between positive and negative M-CHAT R/F screens.
Results: M-CHAT data were available for 585 participants; 484 screened negative/91 screened positive. Covariates were based on significant association with M-CHAT R/F. Covariates controlled for in all regressions included socioeconomic status, maternal depression, education, partner status, minority race/ethnicity, infant birth weight, sex, and cerebral palsy diagnosis. VPT infants with 2 or more medical conditions had a 2.73 adjusted odds (aOR) of screening positive on the M-CHAT R/F, p=.001, 95% CI[1.47,5.03]. Being in profiles 5 & 6 had 2.20 aOR for screening positive, p=.005, 95% CI[1.26,3.81]. Specifically, being profile 5 alone was significant, aOR = 2.20, p=.001, 95% CI[1.46,4.63] and profile 6 was not, p=.912. Medical by NNNS interaction did not significantly predict M-CHAT R/F, p=.679. Positive screens on M-CHAT R/F had significantly higher scores on CBCL and significantly lower scores on Bayley-III (Table 1). Findings suggest there may be a subset of children who can be identified during NICU stay by medical morbidities and NNNS profile who are at risk of developmental delays possibly captured by the M-CHAT R/F at 2-years.

Authors