A large Israeli cohort study of more than 500,000 infants found that routine milestone checks during well-child visits could identify which early developmental delays persisted into the second year of life. Machine-learning models and simple milestone-count rules both predicted persistent delay with moderate accuracy, supporting the use of standardized surveillance data, such as the Tipat Halav Israeli Surveillance (THIS) developmental scale, to strengthen early identification and intervention in pediatric care.

Source: JAMA Network Open