Explainable Machine Learning for Public Health Informatics in HEDIS Childhood Immunization Status Combination 10

Area of Science:

• Public Health
• Healthcare Quality Improvement
• Pediatric Medicine

Background:

• The Healthcare Effectiveness Data and Information Set (HEDIS) Childhood Immunization Status Combination 10 (Combo 10) measure tracks vaccination rates in young children.
• Declining immunization rates pose a significant public health concern, potentially leading to outbreaks of preventable diseases.

Purpose of the Study:

• To analyze trends in HEDIS Combo 10 completion rates among US children aged 24-35 months.
• To identify key component-level drivers influencing Combo 10 completion using an explainable machine learning approach.

Main Methods:

• Analysis of survey-weighted data for US children aged 24-35 months.
• Application of an explainable machine learning model to identify significant factors in vaccination completion.

Main Results:

• A statistically significant decline in Combo 10 completion was observed, decreasing from 53.7% in 2021 to 44.6% in 2023.
• Influenza vaccination and completion of the rotavirus series were identified as the strongest predictors of overall Combo 10 completion.

Conclusions:

• Targeted interventions focusing on influenza and rotavirus vaccination can improve overall childhood immunization rates.
• Public health quality improvement initiatives should prioritize these key vaccination components to address declining trends.