Statistical Analysis
Continuous data was expressed in terms of mean +/- standard deviation or median (IQR) and compared with ANOVA or Mann-Whitney U test. Categorical data was evaluated using chi-squared test. Multivariate logistic regression for weighted outcomes was utilized to compute odds ratio for in-hospital mortality and complications. We considered non-emergency CABG and emergency CABG separately and performed the analysis for all centers and transplant centers for each. Total hospital charges and LOS were evaluated with linear regression testing. In the linear and logistic regression testing, the general population (no history of transplant or immunosuppression, or Non-Tx) was set as the reference category. The multivariate analysis was adjusted for potential confounding factors. In order to attain robust statistical conclusions, total hospital change and LOS values were transformed to normal distribution via log-transformation procedure. Additionally, we adjusted the total charges based on consumer inflation index (CPI) 2020 to be able to compare the charges present. Statistical analysis was performed using IBM’s Statistical Package for Social Sciences (SPSS) software version 25 (IBM Corp, Armonk, NY).
Results: