068 - A Descriptive Study and Comparison of the Trends in Bayley Scales of Infant and Toddler Development - III Scores for the First 3 years of life of Infants Weighing Less Than And Above 1500 Grams

Publication Number: 68.3903

Mugil V. Shanmugam, Keck School of Medicine of the University of Southern California, Oak Park, CA, United States; Harini Thirumalai, Loma Linda University Children's Hospital, Loma Linda, CA, United States; Smeeta Sardesai, Keck School of Medicine of the University of Southern California, Los Angeles, CA, United States; Manoj Biniwale, Cedars Sinai Medical Center, Los Angeles, CA, United States

Background: Very low birth weight (VLBW) infants, and certain larger preterm and term infants are at risk of developmental delay. The Bayley Scales of Infant and Toddler Development-III is used at high-risk infant follow up (HRIF) clinics to identify delays in neurodevelopmental domains (language, motor and cognitive).

Objective: The objective of this study was to identify how the proportion of delays in the different domains evolves over time. We also compared proportional risk for developmental delay in all domains between VLBW infants < 1500 grams and larger infants >1500 grams.

Design/Methods: Charts of infants followed at HRIF clinic between June 2008 and August 2020, from time in the NICU to 3-year follow-up, were reviewed (Visit 1 at 6-12 months, Visit 2 at 18-24 months, Visit 3 at 36 months). Bayley score averages of 4 and above were classified as normal; anything below was abnormal. These overall scores were an average of the classification scores in language, cognitive, and motor domains; a score below 4 in a particular domain was considered delay in that domain. Z-tests of two proportions were utilized to identify any significant changes in the proportion of cognitive delay, motor delay, and language delay across age groups and sub-populations.

Results: For both groups of infants tested at the 3 visits, language was the most commonly delayed domain in the infants with abnormal Bayley scores, followed by motor and then cognitive (Table 1 and 2). The only exception to this trend was both VLBW and larger infants present more commonly with cognitive delay rather than motor delay specifically at 18-24 months. In larger infants, significant differences (p-value < 0.01) were identified between the relative proportion of motor delay at visit 1 vs 2, as well as visit 2 vs 3. A significant difference was also identified between the relative proportion of language delay at visit 1 vs 2. For VLBW infants, significant differences were identified between the relative proportions of motor delay at visit 1 vs 2, as well as at visit 2 vs 3. Significant differences were also observed between the relative proportions of language delay at visit 1 vs 2, as well as visit 1 vs 3.

Conclusion(s): A majority of infants with abnormal Bayley scores in any given age group and any weight can reliably be predicted to have delay in the language domain. The proportion of motor and language delay does change significantly across some age groups, compared to cognitive delay which does not. The similar trends across the weight groups as well as non-significant Z tests suggest the proportions of delays, regardless of weight, are about the same.

Table 1: Proportion of Abnormal Bayley Scores Involving Cognitive, Motor, and Language Delay in Low Birth Weight Infants Greater Than 1500 grams


Table 2: Proportion of Abnormal Bayley Scores Involving Cognitive, Motor, and Language Delay in Very Low Birth Weight Infants Less Than 1500 grams