Impact of Social Determinants of Health on Predictive Models for Outcomes After Congenital Heart Surgery

Background: Despite documented associations between social determinants of health and outcomes post-congenital heart surgery, clinical risk models typically exclude these factors. Objectives: Characterize associations between social determinants and operative and longitudinal mortality; assess impacts on risk model performance. Methods: Demographic and clinical data were obtained for all congenital heart surgeries (2006–2021) from locally-held, CHS-COLOUR Society of Thoracic Surgeons-Congenital Heart Surgery Database (STS-CHSD) data. Neighborhood-level American Community Survey and composite sociodemographic measures were linked by ZIP code. Model prediction, discrimination, and impact on quality assessment were assessed before and after inclusion of social determinants in models based on the 2020 STS-CHSD Mortality Risk Model. Results: Of 14,173 total index operations across New York State, 12,321 cases, representing 10,271 patients at eight centers, had ZIP codes for linkage. 327 (2.7%) patients died in-hospital or before 30-days. 314 children died by 12/31/21 (total n=641, 6.2%). Multiple measures of social determinants of health explained as much or more variability in operative and longitudinal mortality than clinical comorbidities or prior cardiac surgery. Inclusion of social determinants minimally improved models’ predictive performance (operative: 0.834 to 0.844; longitudinal 0.808 to 0.811), but significantly improved model discrimination. 10.0% more survivors and 4.8% more mortalities changed risk quartile in appropriate directions. Wide variation in reclassification was observed by site, resulting in changes in center performance classification category for two of eight centers. Conclusions: While indiscriminate inclusion of social determinants in clinical risk modeling can conceal inequities, thoughtful consideration can help centers understand their performance across populations.