Los Angeles, California became a warzone of COVID-19 infections with up to one death every 10 minutes at the end of 2020. As resources thinned, and ICU beds and ventilators became scarce, physicians began agonizing over potentially rationing medical care. In this study, we conducted a retrospective cohort analysis of 7,429 confirmed COVID-19 positive patients from two community hospitals in Los Angeles, California between March 16, 2020 and June 9, 2021. We applied the Cox proportional hazards regression model to determine the risk factors most strongly associated with in-hospital mortality. Using the multivariable Cox proportional hazards model, there was a higher hazard ratio (HR) for mortality in patients who were older (age ≥60 years) [HR 2.189, 95% CI 1.991-2.407, p<0.001], had low triage oxygenation < 90% [HR 1.439, 95% CI 1.339-1.546, p<0.001], had chronic kidney disease (CKD) [HR 1.348, 95% CI 1.234-1.496, p = 0.001)], and who were obese (BMI ≥ 30 kg/m^2) [HR 1.221, 95% CI 1.155-1.340, p = 0.003)]. Overall, our study concluded that age ≥ 60 years, low triage oxygenation less than 90%, chronic kidney disease, and obesity were the top patient characteristics associated with increased mortality for both the univariate and multivariate Cox proportional hazards model analyses. Furthermore, by separating our data set into a development and validation set, we created a novel prediction tool to forecast in-hospital mortality and achieved 86% accuracy.