Diagnostic Accuracy of Bayley III Screening Tool to Detect Developmental Delay in Moderate to High-Risk Indian

Sanmathi Suresh1, Abhinayaa Janakiraman2, Udayakumar Narasimhan3

  • 1Department of Pediatrics, Sri Ramachandra Institute of Higher Education and Research, Deemed to be University, Chennai, Tamil Nadu, India.

PubMed

Insights

The Bayley III screening tool shows good specificity for detecting developmental delay in Indian preterm infants. While effective in identifying potential issues, its sensitivity varies across different developmental domains.

Area of Science:

  • Pediatric developmental assessment
  • Neonatal screening tools
  • Developmental pediatrics

Background:

  • Early detection of developmental delay is crucial for preterm infants.
  • Accurate screening tools are needed for high-risk populations in diverse settings like India.
  • The Bayley Scales of Infant and Toddler Development, Third Edition (Bayley III) is a widely used developmental assessment.

Purpose of the Study:

  • To evaluate the diagnostic accuracy of the Bayley III screening tool.
  • To assess its ability to detect developmental delay in moderate to high-risk preterm children in India.
  • To compare its performance against the gold-standard Developmental Assessment Scale for Indian Infants (DASII).

Main Methods:

  • Cross-sectional study involving 94 preterm infants in India.
  • Administration of the Bayley III screener (index test) and the DASII (gold standard).
  • Blinded testing and comparison of diagnostic accuracy parameters (sensitivity, specificity).

Main Results:

  • With a DASII cut-off of <70, Bayley III showed 84.62% sensitivity and 85.18% specificity in the mental domain.
  • Motor domain sensitivity was 61.11% with 96.05% specificity (DASII cut-off <70).
  • Higher DASII cut-offs (e.g., 85) resulted in reduced sensitivity but maintained high specificity across domains.

Conclusions:

  • The Bayley III screener demonstrates good specificity in identifying developmental delay in Indian preterm infants.
  • Its sensitivity is more variable, particularly in the motor domain, suggesting potential for missed diagnoses.
  • The tool offers valuable predictive values but should be interpreted considering its performance characteristics.