163 - Diagnostic Accuracy of the Ages and Stages Questionnaires for Neurodevelopmental Impairment Screening in Children Born Very Preterm

Poster Number: 163
Publication Number: 163.363

Emily M. Youngblood, University of Alabama at Birmingham, Birmingham, AL, United States; Vivek Shukla, University of Alabama at Birmingham, Birmingham, AL, United States; Seabrook H. Jeffcoat, University of Alabama at Birmingham, Hoover, AL, United States; Rachel L. Benz, University of Alabama at Birmingham, Birmingham, AL, United States; Jan Wallander, University of California Merced, Merced, CA, United States; Waldemar A. Carlo, University of Alabama School of Medicine, Birmingham, AL, United States; Colm P. Travers, University of Alabama School of Medicine, Birmingham, AL, United States

Background: The Ages and Stages Questionnaires, Third Edition (ASQ-3) are accessible developmental screening tools. There are limited data on the accuracy of the ASQ in children born very preterm when compared with the standard diagnostic test, the Bayley Scales of Infant and Toddler Development, Third Edition (BSID-III).

Objective: Test the hypothesis that the ASQ-3 can accurately screen for neurodevelopmental impairment (NDI) in children born very preterm.

Design/Methods: This study included children born < 29 weeks’ gestational age (GA) attending the Newborn Follow-Up Clinic of the University of Alabama at Birmingham from 2014 to 2019 who completed the ASQ-3 and the BSID-III. Moderate-severe NDI was defined as cognitive or motor composite score < 85 (1 standard deviation (SD) below the mean) on the BSID-III, a Gross Motor Function Classification System (GMFCS) scale score > 2, bilateral blindness, or bilateral hearing loss needing amplification. ASQ-3 screenings were classified as abnormal if one or more domains were > 2 standard deviations (SD) below the age-specific mean. Our primary outcome was the accuracy of the ASQ-3 for detecting moderate-severe NDI using the area under the receiver-operating curve (AUROC). Akaike information criterion (AIC) determined which BSID-III cutoff, < 85 versus < 70 (2 SD), provided higher predictive utility for the ASQ-3.

Results: We included 303 children with a mean (SD) age of testing for the ASQ-3 of 18 (9) months corrected GA and a mean (SD) age of testing for the BSID-III of 23 (3) months corrected GA. According to BSID-III scores, 111 (37%) had NDI at score < 85 and 49 (16%) at score < 70. Additionally, 58 (19%) had cerebral palsy with 5 (1.6%) classified as severe at GMFCS level > 2. AUROC analyses determined there was an 82% and 87% chance of distinguishing normal from abnormal results using BSID-III score cutoff of < 85 [Figure 1] and < 70 [Figure 2] respectively. Analyses comparing ASQ-3 to BSID-III score < 85 yielded the following: sensitivity 86%, specificity 56%, positive predictive value (PPV) 52%, and negative predictive value (NPV) 87%. This was compared to BSID-III score < 70, which yielded a sensitivity 96%, specificity 48%, PPV 26%, and NPV 98%. The predictive utility of the ASQ-3 was higher at BSID-III cutoff score < 70 (AIC 184) compared with score < 85 (AIC 300). Individual ASQ-3 and BSID-III domains had moderate positive correlation (all p < 0.01) [Table 1].

Conclusion(s): The ASQ-3 may be a useful neurodevelopmental screening tool but does not eliminate the need for formal diagnostic testing in children born very preterm.