B80-4-38 Exploring the Relationship Between Indicators of Social Determinants of Health and Patient Outcomes Following Lung Surgery, Utilizing 2024 Sparcs Data in New York State

Abstract There is an inextricable link between social determinants of health (SDoH) and clinical outcomes in lung cancer patients 1. However, there remains a need for further exploration of these relationships in the context of surgical intervention for either benign or malignant pulmonary processes 2. In New York State, the Statewide Planning and Research Cooperative System (SPARCS) serves as a data repository for patient-level discharge information—capturing both clinical and sociodemographic variables aggregated annually. Here, we utilized select patient characteristics, as indicators of SDoH, as a means to investigate their relationship to clinical outcomes following lung surgery among patients admitted to New York State-funded Hospitals in 2024. We analyzed patients whose procedural diagnosis was classified as a lobectomy, wedge resection, or decortication (either open or thoracoscopic) for either benign or malignant disease (N = 3,127). Among this subset, we preselected relevant variables of interest including age, gender, race, insurance types, as outlined in Table 1, that have been well-established indicators of SDoH 3. Our primary outcome of interest was length of stay, defined as either greater than or less than the median (5 days). Our secondary outcome of interest was discharge disposition to either home/with services versus ‘other’ (discharge to short/long term nursing/rehabilitation facilities or those who died prior to discharge). Performing chi-squared analysis between age, gender, race, and insurance type and our outcomes of interest, we found statistically significant associations (p-value 0.001) between all SDoH variables and length of stay 5 days, as well as between all SDoH variables and discharge disposition to any place other than home. Given these associations, age, gender, race, and insurance type were included in two separate multivariate logistic regression models—one for each outcome of interest (both length of stay and discharge disposition). Multivariate LRM revealed that when controlling for the other covariates of interest, longer length of stays 5 days was statistically associated with advanced age (50years old), as well as having government/public insurance (p-value 0.001). These results were similarly echoed in the model with discharge disposition as the outcome of interest. This exploratory, descriptive analysis is illustrative of the potential correlations between indicators of SDoH and clinical outcomes following lung surgery, specifically in New York State. The findings we present here offer a starting point; emphasizing the critical need for further mixed-method statistical, geospatial, and even qualitative analyses that ultimately aim to improve post-operative outcomes in vulnerable populations. This abstract is funded by: None