Original Article
A comparative analysis of the outcomes of patients with influenza or COVID-19 in a tertiary hospital in Belgium

https://doi.org/10.1016/j.jiac.2022.07.012Get rights and content

Abstract

Introduction

The COVID-19 pandemic has emerged as a global health problem, associated with high morbidity and mortality rates. The aim of this study was to compare the outcomes of hospitalized patients with COVID-19 or with seasonal influenza in a teaching hospital in Belgium.

Methods

In this retrospective, single-center cohort study, 1384 patients with COVID-19 and 226 patients with influenza were matched using a propensity score with a ratio of 3:1. Primary outcomes included admission to intensive care unit (ICU), intubation rates, hospital length of stay, readmissions within 30 days and in-hospital mortality. Secondary outcomes included pulmonary bacterial superinfection, cardiovascular complications and ECMO.

Results

Based on the analysis of the matched sample, patients with influenza had an increased risk of readmission within 30 days (Risk Difference (RD): 0.07, 95% CI: 0.03 to 0.11) and admission to intensive care unit (RD: 0.09, 95% CI: 0.03 to 0.15) compared with those with COVID-19. Patients with influenza had also more pulmonary bacterial superinfections (46.2% vs 7.4%) and more cardiovascular complications (32% vs 3.9%) than patients with COVID-19.However, a two-fold increased risk of mortality (RD: −0.10, 95% CI: 0.15 to −0.05) was observed in COVID-19 compared to influenza. ECMO was also more required among the COVID-19 patients who died than among influenza patients (5% vs 0%).

Conclusions

COVID-19 is associated with a higher in-hospital mortality compared to influenza infection, despite a high rate of ICU admission in the influenza group. These findings highlighted that the severity of hospitalized patients with influenza should not be underestimated.

Keywords

Outcome comparison
Mortality
Influenza
COVID-19
Propensity score matched analysis

Cited by (0)

1

These authors contributed equally to the manuscript.

View Abstract