COVID‐19 severity from Omicron and Delta SARS‐CoV‐2 variants

Abstract The Omicron variant of SARS‐CoV‐2 achieved worldwide dominance in late 2021. Early work suggests that infections caused by the Omicron variant may be less severe than those caused by the Delta variant. We sought to compare clinical outcomes of infections caused by these two strains, confirmed by whole genome sequencing, over a short period of time, from respiratory samples collected from SARS‐CoV‐2 positive patients at a large medical center. We found that infections caused by the Omicron variant caused significantly less morbidity, including admission to the hospital and requirement for oxygen supplementation, and significantly less mortality than those caused by the Delta variant.


| INTRODUCTION
The COVID-19 pandemic has caused significant morbidity and mortality worldwide. A new variant that partially escapes the existing vaccines, Omicron, was discovered by genomic surveillance teams in South Africa in November 2021 and was quickly identified as a variant of concern by the WHO. [1][2][3] Omicron was estimated to have accounted for approximately 95% of daily new infections in the United States by early January 2022. 4 Rapid exploration of the clinical severity of infections caused by the Omicron variant, as well as future variants, is critical to the targeted public health response to the evolution of the COVID-19 pandemic. The goal of this project was to compare the clinical severity of COVID-19 caused by SARS-CoV-2 whole genome sequence confirmed Omicron and Delta variants.

| MATERIALS AND METHODS
We compared the clinical severity of Omicron and Delta variants within a cohort of patients who tested positive for SARS-CoV-2 during a period of co-circulation of the Omicron and Delta variants. SARS-CoV-2 variants were identified with viral whole genome sequencing, and clinical outcomes were ascertained via systematic capture of clinical data from the electronic medical records. We examined a randomly selected subset (from a combined total-positive specimen bank) of To further explore the associations between viral variant and disease severity, we next performed logistic regression with the lrm function in the R package rms (version 6.2-0), 11 setting hospitalization and any oxygen use as the dependent variables. In addition to viral variant, age, sex, race, ethnicity, obesity (defined as a body mass index > 30), and SARS-CoV-2 vaccination status were a priori selected to be added to the model as independent variables due to their previously established associations with COVID-19 severity. Age was transformed with restricted cubic splines with five knots with the rms::rcs function. Due to the high number of missing data in race, ethnicity, and BMI categories, missing data were placed into a category called "Unknown," and results are presented for each predictor. Race was simplified to Black, White, Asian, Other, or Unknown. Interquartile odds ratios (OR) and their 95% CIs were found with rms::summary.rms, and P values for each independent variable were calculated with the rms::anova function.   (Table 1). Several indicators of high disease severity were more common in the Delta group than the Omicron group, including in-hospital care within 14 days of a positive test (13.1% vs. 5.3%), invasive mechanical ventilation (2% vs. 0%), and death (2% vs. 2%) ( Figure 1A,B). When controlling for age, sex, race, ethnicity, vaccination status, and obesity in logistic regression analyses, Omicron, com-

| DISCUSSION
We evaluated the severity of disease caused by Delta and Omicron variants over a brief period across the transition from Delta to Omicron dominance in our region. Potential history bias that may have obscured the relationship between variant and severity was minimized by the short period during which healthcare seeking behavior and methods of clinical testing were assumed to be largely unchanged. and death associated with Omicron infection. 12 They also agree with findings from other recent studies that suggest that severity of disease among inpatients is lower with Omicron compared to Alpha and Delta variants. 12,13 Our study had limitations. Our cohort was selected from patients seeking care in a single health system, limiting generalizability of results. Sample size was small compared to similar studies; however,

PEER REVIEW
The peer review history for this article is available at https://publons. com/publon/10.1111/irv.12982.

DATA AVAILABILITY STATEMENT
The Global Initiative on Sharing Avian Influenza Data (GISAID) accession numbers for the whole viral genomes are available in the supporting information.