Association Between SARS-CoV-2 Mutations and Disease Severity Reveals Risk and Protective Effects Among Community-Sampled Patients in Israel

SARS-CoV-2 mutations play a key role in viral evolution, in immune escape, and potentially in disease severity. However, the clinical impact of most mutations remains poorly understood, particularly across different variants. A historical observational study was conducted using SARS-CoV-2 whole-genome sequencing data linked to clinical metadata from 175,503 COVID-19 cases in Israel. The dataset was stratified into four variant-specific periods: B.1.1.7, B.1.617.2, BA.1, and BA.2. Logistic regression models were applied separately within each period to assess the association between individual mutations and the need for hospitalization, adjusting for age, gender, and time since vaccination. False discovery rate correction was used to account for multiple testing. A total of 18 SARS-CoV-2 mutations were significantly associated with COVID-19 severity, of which eight remained statistically significant after false discovery rate correction. Among these, two were associated with increased risk and six with reduced risk. Most were non-synonymous mutations located in functionally relevant regions such as the spike protein and non-structural proteins. This study provides a variant-stratified assessment of SARS-CoV-2 mutations associated with clinical severity, revealing both known and novel associations. The findings highlight the importance of integrating genomic and clinical data in public health surveillance and may inform future outbreak preparedness by identifying mutations with potential clinical impact.