This study examines the association between SARS-CoV-2 genetic variants and the risk of hospitalization in 12,538 patients using data from a large, multisite observational cohort study. The authors used XGBoost with SHapley Additive exPlanations (SHAP) analysis and generalized linear mixed models to investigate the relationship between viral genomic variants and hospitalization, while controlling for clinical covariates such as COVID-19 vaccination status, outpatient monoclonal antibody treatment status, and underlying risk factors. The results indicate that several SARS-CoV-2 lineages, including B.1.1.7, AY.44, and AY.54, are associated with increased hospitalization risk. Omicron is associated with lower hospitalization risk compared to previous variants. Additionally, variants at specific amino acid locations within the Spike protein N-terminal domain and non-structural protein 14 are also associated with hospitalization risk. The study highlights the importance of considering both clinical and genomic data to better understand the impact of SARS-CoV-2 variants on health outcomes.This study examines the association between SARS-CoV-2 genetic variants and the risk of hospitalization in 12,538 patients using data from a large, multisite observational cohort study. The authors used XGBoost with SHapley Additive exPlanations (SHAP) analysis and generalized linear mixed models to investigate the relationship between viral genomic variants and hospitalization, while controlling for clinical covariates such as COVID-19 vaccination status, outpatient monoclonal antibody treatment status, and underlying risk factors. The results indicate that several SARS-CoV-2 lineages, including B.1.1.7, AY.44, and AY.54, are associated with increased hospitalization risk. Omicron is associated with lower hospitalization risk compared to previous variants. Additionally, variants at specific amino acid locations within the Spike protein N-terminal domain and non-structural protein 14 are also associated with hospitalization risk. The study highlights the importance of considering both clinical and genomic data to better understand the impact of SARS-CoV-2 variants on health outcomes.