The Clinical Severity of COVID-19 Variants of Concern: Retrospective Population-Based Analysis.

ABSTRACT

BACKGROUND: SARS-CoV-2 variants of concern (VOCs) emerged and rapidly replaced the original strain worldwide. The increased transmissibility of these new variants led to increases in infections, hospitalizations, and mortality. However, there is a scarcity of retrospective investigations examining the severity of all the main VOCs in presence of key public health measures and within various social determinants of health (SDOHs). OBJECTIVE: This study aims to provide a retrospective assessment of the clinical severity of COVID-19 VOCs in the context of heterogenous SDOHs and vaccination rollout. METHODS: We used a population-based retrospective cohort design with data from the British Columbia COVID-19 Cohort, a linked provincial surveillance platform. To assess the relative severity (hospitalizations, intensive care unit [ICU] admissions, and deaths) of Gamma, Delta, and Omicron infections during 2021 relative to Alpha, we used inverse probability treatment weighted Cox proportional hazard modeling. We also conducted a subanalysis among unvaccinated individuals, as assessed severity differed across VOCs and SDOHs. RESULTS: We included 91,964 individuals infected with a SARS-CoV-2 VOC (Alpha n=20,487, 22.28%; Gamma n=15,223, 16.55%; Delta n=49,161, 53.46%; and Omicron n=7093, 7.71%). Delta was associated with the most severe disease in terms of hospitalization, ICU admissions, and deaths (hospitalization adjusted hazard ratio [aHR] 2.00, 95% CI 1.92-2.08; ICU aHR 2.05, 95% CI 1.91-2.20; death aHR 3.70, 95% CI 3.23-4.25 relative to Alpha), followed generally by Gamma and then Omicron and Alpha. The relative severity by VOC remained similar in the unvaccinated individual subanalysis, although the proportion of individuals infected with Delta and Omicron who were hospitalized was 2 times higher in those unvaccinated than in those fully vaccinated. Regarding SDOHs, the proportion of hospitalized individuals was higher in areas with lower income across all VOCs, whereas among Alpha and Gamma infections, 2 VOCs that cocirculated, differential distributions of hospitalizations were found among racially minoritized groups. CONCLUSIONS: Our study provides robust severity estimates for all VOCs during the COVID-19 pandemic in British Columbia, Canada. Relative to Alpha, we found Delta to be the most severe, followed by Gamma and Omicron. This study highlights the importance of targeted testing and sequencing to ensure timely detection and accurate estimation of severity in emerging variants. It further sheds light on the importance of vaccination coverage and SDOHs in the context of pandemic preparedness to support the prioritization of allocation for resource-constrained or minoritized groups.