Abstract
This study investigated the relationship between Opioid Use Disorder (OUD) and in-hospital outcomes among COVID-19 patients, focusing on prolonged hospital stay, readmission, and mortality. Utilizing statistical methods like odds ratios (OR) and adjusted odds ratios (aOR), we analyzed a dataset of 39,884 COVID-19 patients, including 1.70% with OUD. The results indicated that patients with OUD had significantly higher risks of prolonged hospital stay (aOR = 6.32, CI = 5.23-7.63), readmission (aOR = 3.49, CI = 2.97-4.09), and mortality (aOR = 2.79, CI = 2.00-3.89). Additionally, the study evaluated the impact of vaccination, which substantially reduced the odds of extended hospitalization (aOR = 0.31, CI = 0.27-0.35) and mortality (aOR = 0.20, CI = 0.15-0.25), but did not significantly affect readmission rates (aOR = 1.00, CI = 0.95-1.05). Furthermore, this research employed multiple machine learning techniques to predict the length of stay (LOS) for 3,793 COVID-19 patients with Substance Use Disorder (SUD). These models demonstrated high predictive accuracy, with Balanced Random Forest (BRF) and Logistic Regression (LR) achieving 79 % and 80 % accuracy, respectively. This comprehensive approach provides valuable insights into the complexities of managing COVID-19 patients with concurrent SUD, highlighting the need for tailored healthcare strategies.