352. Socioeconomic Factors Affect Late Re-Admissions and Midterm Mortality Following Isolated CABG
Garrett N. Coyan, Amber Okoye, Ayesha Shah, Floyd Thomas, *Christopher M. Sciortino, Arman Kilic, *Thomas G. Gleason, *Danny Chu
University of Pittsburgh, Pittsburgh, PA
Invited Discussant: *Scott A. LeMaire
Objective: To determine the effect of socioeconomic status (SES) and social determinants of health on late re-admission rates and mid-term mortality following isolated coronary artery bypass grafting (CABG).
Methods: Consecutive patients undergoing isolated CABG from July 2011 to December 2017 at an academic health system were retrospectively reviewed. Data was prospectively recorded in local institutional STS database and augmented with median household income (HI) and percent high school graduation in those over 25 years old (HS) (both by zip code tabulation area) obtained via the American Factfinder Database (U.S. Census Burau). HI and HS were divided into quartiles to facilitate comparison. Re-admissions data and mortality data were obtained via longitudinal follow-up and Social Security Death Index. Primary endpoints were 1-year re-admission and 5-year mortality rates following isolated CABG by SES. Analysis was conducted with chi-square, Kruskal-Wallis test, Kaplan-Meier Analysis, and multivariate logistic regression.
Results: There were 5320 patients who underwent isolated CABG during the study period. Median HI for the cohort was $48,061, and median HS was 91.5%. When stratified by HI and HS quartiles, lower quartiles were associated with increased pre-operative co-morbidities including congestive heart failure, previous myocardial infarction, diabetes, peripheral vascular disease, depression, anticoagulation use, iv drug and tobacco use, and chronic lung disease (all p<0.05). Lower quartiles of HI and HS were associated with higher presentation rates of cardiogenic shock and heart failure, and more often required blood products and a prolonged ICU stay post-operatively. Figure 1 demonstrates a Kaplan-Meier curve showing significantly decreased actuarial survival at 5 years in unadjusted lower income quartiles (p=0.006). Average interval to re-admission was longer in those with lower quartiles of both SES variables (p<0.05 in both cases), and the total number of 1-year re-admissions per patient was higher in those with higher HI and HS (p<0.05). Multivariable logistic regression analysis demonstrated 30-day, 90-day, and 1-year re-admission rates lower among the lowest quartile of HS (p<0.05 in each) after adjusting for patient-level covariates.
Conclusions: Lower SES is associated with increases in pre-operative co-morbidities and more acute presentation in CABG patients. Higher SES is associated with more frequent and earlier re-admissions in our cohort, where lower SES is associated with higher mid-term mortality.