Alex D Michaels1, Max O Meneveau2, Robert B Hawkins2, Eric J Charles2, J Hunter Mehaffey3. 1. Department of Surgery, Wake Forest School of Medicine, Winston-Salem, NC. 2. Department of Surgery, University of Virginia Health, Charlottesville, VA. 3. Department of Surgery, University of Virginia Health, Charlottesville, VA. Electronic address: jhm9t@virginia.edu.
Abstract
BACKGROUND: There is a strong association between socioeconomic status and surgical outcomes; however, the optimal method for socioeconomic risk-stratification remains elusive. We aimed to compare 2 metrics of socioeconomic ranking by ZIP code, the Distressed Communities Index, and the Area Deprivation Index and their association with surgical outcomes. METHODS: This retrospective study included all general surgery cases performed at a single institution from 2005 to 2015. Each patient was assigned Distressed Communities Index and Area Deprivation Index scores based on ZIP code. Both indices are normalized composite measures of socioeconomic status derived from census data. Primary outcome was 30-day morbidity; secondary outcomes included long-term mortality and cost, stratified by socioeconomic status. The utility of the addition of each metric to the American College of Surgeons National Surgical Quality Improvement Program risk calculator was assessed. RESULTS: The 9,843 patients had normally distributed Distressed Communities Index (47.3 ± 22.4) and Area Deprivation Index (35.4 ± 19.0). Patients who experienced any complication or readmission had significantly higher Distressed Communities Index (48.6 vs 47.1, P = .04) and Area Deprivation Index (37.2 vs 35.1, P = .002). Risk-adjusted models demonstrated that only Area Deprivation Index independently predicted postoperative complications (odds ratio 1.11, P = .02), improved the discrimination of risk-stratification when added to the American College of Surgeons National Surgical Quality Improvement Program risk calculator (area under curve 0.758-0.790, P = .02), and was associated with hospitalization cost ($1,811 ± 856/quartile, P = .03). CONCLUSION: Area Deprivation Index provides improved socioeconomic risk-adjustment in this surgical population. The addition of Area Deprivation Index to risk-stratification tools would allow us to better inform our patients of their expected postoperative courses, more accurately account for the increased cost of providing their care, and identify patients and regions that are most in need of improvements in health and healthcare.
BACKGROUND: There is a strong association between socioeconomic status and surgical outcomes; however, the optimal method for socioeconomic risk-stratification remains elusive. We aimed to compare 2 metrics of socioeconomic ranking by ZIP code, the Distressed Communities Index, and the Area Deprivation Index and their association with surgical outcomes. METHODS: This retrospective study included all general surgery cases performed at a single institution from 2005 to 2015. Each patient was assigned Distressed Communities Index and Area Deprivation Index scores based on ZIP code. Both indices are normalized composite measures of socioeconomic status derived from census data. Primary outcome was 30-day morbidity; secondary outcomes included long-term mortality and cost, stratified by socioeconomic status. The utility of the addition of each metric to the American College of Surgeons National Surgical Quality Improvement Program risk calculator was assessed. RESULTS: The 9,843 patients had normally distributed Distressed Communities Index (47.3 ± 22.4) and Area Deprivation Index (35.4 ± 19.0). Patients who experienced any complication or readmission had significantly higher Distressed Communities Index (48.6 vs 47.1, P = .04) and Area Deprivation Index (37.2 vs 35.1, P = .002). Risk-adjusted models demonstrated that only Area Deprivation Index independently predicted postoperative complications (odds ratio 1.11, P = .02), improved the discrimination of risk-stratification when added to the American College of Surgeons National Surgical Quality Improvement Program risk calculator (area under curve 0.758-0.790, P = .02), and was associated with hospitalization cost ($1,811 ± 856/quartile, P = .03). CONCLUSION: Area Deprivation Index provides improved socioeconomic risk-adjustment in this surgical population. The addition of Area Deprivation Index to risk-stratification tools would allow us to better inform our patients of their expected postoperative courses, more accurately account for the increased cost of providing their care, and identify patients and regions that are most in need of improvements in health and healthcare.