Infection status associated with hypoxemia and death rates varying by virus type in a large US cohort.

Routine testing for SARS-CoV-2, influenza, and respiratory syncytial virus (RSV) was deployed in a large US healthcare system in 2022–2023. This policy allowed identification of a large cohort of co-infected patients and comparison of outcomes without confounding by testing indication. Patients “triple-tested” in the US Veterans Health Administration were classified by infection status in the first week of a positive test. Multivariable logistic regression was used to estimate associations of different infections with hypoxemia (SpO2 2 L/min) or death, separately, expressed as adjusted odds ratios (aOR) with 95% confidence intervals (CI). Among 835,987 triple-tested patients, 170,592 (20.4%) tested positive for SARS-CoV-2 alone, 30,454 (3.6%) influenza alone, 13,207 (1.6%) RSV alone, and 1,300 (0.2%) multiple viruses. Frequencies of hypoxemia and death were 8.0 and 1.9% with SARS-CoV-2, 7.7 and 0.8% with influenza, 9.3 and 1.1% with RSV, 8.7 and 1.5% with multiple viruses, and 8.9 and 2.0% with all-negative tests. After adjustment for age and immune-suppressive drugs, odds of hypoxemia were slightly higher with influenza (aOR = 1.12, CI 1.06–1.17), lower with RSV (aOR = 0.91, CI 0.85–0.97), and not significantly different with multiple viruses (aOR = 1.09, CI 0.89–1.34), relative to SARS-CoV-2 alone. Odds of death were lower with influenza (aOR = 0.52, CI 0.46–0.60) or RSV (aOR = 0.51, CI 0.43–0.60) and no different with multiple infections (aOR = 0.86, CI 0.54–1.36), relative to SARS-CoV-2 alone. Co-infection was rare (0.2% of tested cases), with incidences of hypoxemia and death similar to SARS-CoV-2 alone. Death was less frequent with influenza or RSV than SARS-CoV-2.