Understanding Pediatric Surgery Cancellation: Geospatial Analysis.

Area of Science:

• Health Services Research
• Geospatial Analysis
• Pediatric Surgery

Background:

• Day-of-surgery cancellations (DoSC) waste resources and inconvenience patients, impacting 2-20% of annual surgeries.
• Up to 85% of DoSCs may be preventable through behavioral modifications.
• Underlying factors and family barriers contributing to DoSC remain poorly understood.

Purpose of the Study:

• To conduct a geospatial analysis of patient-specific variables and socioeconomic factors to identify contributors to DoSC rate disparities.
• To understand neighborhood-level variations in DoSC rates using electronic health records and census tract data.

Main Methods:

• Geospatial analysis of electronic health records (EHRs) from two children's hospitals (CCHMC and TCH).
• Linked patient data with socioeconomic factors from the American Community Survey at the census tract level.
• Developed spatial models to analyze DoSC variations, comparing with nonspatial and deep learning models.

Main Results:

• Geospatial models identified key predictors of DoSC, including previous no-shows, proportion of African American inhabitants, and overcrowded households.
• L2-normalized generalized linear regression models demonstrated best performance in predicting DoSC rates at both CCHMC and TCH.
• DoSC rates varied significantly across census tracts, highlighting neighborhood-level influences.

Conclusions:

• Geospatial analysis can identify high-risk census tracts for targeted DoSC interventions.
• Home location, socioeconomic disadvantage, and racial minority status significantly impact pediatric surgical cancellations.
• Future efforts to reduce DoSC should consider social, economic, and cultural factors.