Infant and toddler disease score was useful for risk of hospitalization based on data from administrative claims

Area of Science:

• Pediatric Health
• Epidemiology
• Health Services Research

Background:

• Administrative claims data are increasingly used in health research.
• Predictive tools are needed to assess pediatric morbidity and hospitalization risk.
• Existing tools may not adequately capture the specific needs of infants and toddlers.

Purpose of the Study:

• To develop and validate the Infant and Toddler Disease Score (IDS).
• To create a population-based predictive tool for morbidity status in children under two years old.
• To utilize administrative claims data for risk prediction in pediatric populations.

Main Methods:

• Prospective cohort study of 35,580 children (<2 years) from the French ERASME database.
• Logistic regression model developed on a construction sample (50%) to identify risk factors for incident hospitalization.
• Infant and Toddler Disease Score (IDS) derived from beta coefficients and validated on the remaining 50%.

Main Results:

• Key predictors of hospitalization included long-term disability, younger age, and prior hospitalizations.
• The IDS demonstrated concordant risk estimations between the construction and validation samples.
• The developed score effectively predicted hospitalization risk in the study population.

Conclusions:

• The Infant and Toddler Disease Score (IDS) is a valuable tool for assessing hospitalization risk in infants and toddlers.
• The IDS can be utilized to adjust for confounding in pharmacoepidemiologic studies using administrative claims data.
• This predictive index aids in understanding pediatric morbidity status based on administrative data.