Development and Validation of a Pediatric Comorbidity Index

Area of Science:

• Pediatric Health Research
• Health Informatics
• Epidemiology

Background:

• Comorbidity scores are crucial for reducing bias in nonrandomized health-care database studies.
• Existing scores, designed for adults, may not accurately reflect pediatric patient complexity.
• There is a need for a validated comorbidity index specifically for pediatric populations.

Purpose of the Study:

• To develop and validate a novel pediatric comorbidity index.
• To assess the index's performance in predicting 1-year hospitalization risk in children.
• To compare the new index against existing adult comorbidity scores.

Main Methods:

• Utilized International Classification of Diseases, 10th Revision (ICD-10) data from the MarketScan database.
• Identified pediatric patients (≤18 years) continuously enrolled from October 1, 2015, to September 30, 2017.
• Employed logistic regression to predict 1-year hospitalization risk, incorporating predefined and empirically identified conditions.

Main Results:

• The developed pediatric comorbidity index includes 24 conditions.
• The index achieved a C statistic of 0.718 (95% CI: 0.714, 0.723) for predicting hospitalization risk.
• The pediatric index demonstrated superior performance compared to adult scores (C statistics 0.522–0.640).

Conclusions:

• The pediatric comorbidity index effectively summarizes disease burden in young patients.
• This index is suitable for risk adjustment in pediatric epidemiologic studies.
• The validated index offers a more appropriate measure for pediatric health research than adult-derived scores.