Research Letter
May 15, 2024
Mortality in Patients Hospitalized for COVID-19 vs Influenza in Fall-Winter 2023-2024
Yan Xie, Taeyoung Choi, Ziyad Al-Aly
JAMA. Published online May 15, 2024. doi:10.1001/jama.2024.7395
In the first year of the COVID-19 pandemic, risk of death in people hospitalized for COVID-19 was substantially higher than in people hospitalized for seasonal influenza.1,2 The risk of death due to COVID-19 has since declined. In fall-winter 2022-2023, people hospitalized for COVID-19 had a 60% higher risk of death compared with those hospitalized for seasonal influenza.3 New variants of SARS-CoV-2 have continued to appear, including the emergence of JN.1, the predominant variant in the US since December 24, 2023.4 This study evaluated the risk of death in a cohort of people hospitalized for COVID-19 or seasonal influenza in fall-winter 2023-2024.
Methods
Based on US Department of Veterans Affairs electronic health records from all 50 states, we identified people who were admitted to the hospital with a diagnosis of COVID-19 or seasonal influenza between October 1, 2023, and March 27, 2024, and within 2 days before and 10 days after a positive test result for SARS-CoV-2 or influenza. Patients with either infection hospitalized for another reason or those hospitalized for both COVID-19 and seasonal influenza were excluded. The cohort was followed up for 30 days, until death, or until March 31, 2024. Baseline characteristics between patients hospitalized for COVID-19 vs influenza were compared using absolute standardized differences; a standardized difference less than .01 suggests good balance.
We adjusted for differences in baseline characteristics between the groups using inverse probability weighting. Logistic regression was used to calculate a propensity score (probability of being assigned to the COVID-19 group) that was then applied to balance the 2 groups; covariates are listed in Supplement 1. Weighted Cox survival models were used to estimate the difference in risk of death between COVID-19 and seasonal influenza groups. Results were reported as adjusted death rates and hazard ratios (HRs) with 95% CIs in the COVID-19 group compared with the seasonal influenza group.
We also examined the difference in risk of death between people hospitalized for COVID-19 before and during the JN.1-predominant era (before vs on or after December 24, 2023). Analyses were performed with SAS Enterprise Guide version 8.3 (SAS Institute Inc). We defined statistical significance as a 95% CI that did not cross 1.00. The study was approved with a waiver of informed consent by the VA St Louis Health Care System Institutional Review Board.
Results
The cohort included 8625 participants hospitalized for COVID-19 (unadjusted death rate, 5.70% at 30 days) and 2647 participants hospitalized for seasonal influenza (unadjusted death rate, 3.04% at 30 days). The COVID-19 and seasonal influenza groups were balanced after propensity score weighting (Table 1).
Patients hospitalized for COVID-19 had a higher risk of death compared with those hospitalized for seasonal influenza (adjusted death rate, 5.70% vs 4.24% at 30 days; adjusted HR, 1.35 [95% CI, 1.10-1.66]). There was no statistically significant difference in the risk of death among people hospitalized for COVID-19 before and during the JN.1-predominant era (adjusted death rate, 5.46% vs 5.82% at 30 days; adjusted HR, 1.07 [95% CI, 0.89-1.28]) (Table 2).


Discussion
The study found that in fall-winter 2023-2024, the risk of death in patients hospitalized for COVID-19 was greater than the risk of death in patients hospitalized for seasonal influenza. Compared with a study using the same database and methods,3 the death rate at 30 days was 5.97% in 2022-2023 vs 5.70% in 2023-2024 for COVID-19 and 3.75% in 2022-2023 vs 4.24% in 2023-2024 for influenza. Both adjusted HRs were statistically significant, with an HR of 1.61 in 2022-2023 and 1.35 in 2023-2024, with overlapping 95% CIs. Changes in either the SARS-CoV-2 or influenza viruses or in their care (eg, use of vaccines or antivirals) may influence the comparative risk of death each season. The findings should be interpreted in the context of nearly twice as many hospitalizations for COVID-19 compared with seasonal influenza during 2023-2024.5,6
The results also showed that at the level of statistical power available in this study, there was no significant difference in risk of death among those hospitalized for COVID-19 before and during the JN.1-predominant era—suggesting that JN.1 may not have a materially different severity profile than the variants that immediately preceded it.
Study limitations include that the Veterans Affairs population (older age and predominantly male) may not represent the general population and causes of death were not examined.