1B-03 - Risk of mortality of COVID-19 & influenza co-infection within 30-days: A clone-censor-weight study

Background: In the transition to the post-COVID-19 era, it is crucial to reassess the clinical severity of COVID-19 and influenza co-infection to update the currently available evidence and provide better guidance for medical practice.

Objectives: To compare the risk of 30-day all-cause mortality between patients with co-infection of COVID-19 and influenza, to those with single infection.

Methods: We retrieved data from the Hong Kong Clinical Data Analysis and Reporting System. We identified adult patients (aged >18) who received test for COVID-19 and influenza using reverse transcription polymerase chain reaction (RT-PCR) between 2023-01-01 – 2024-01-31. RT-PCR-confirmed COVID-19 and influenza within 7 days were considered as co-infection. Clone-censor-weight method was used to reduce immortal time bias. We first created clones, assigned into 2 arms (co-infection / single infection of COVID-19 or influenza). Then, we censored the person-time when their actual record violated the assigned treatment. However, this will create a selection bias due to informative censoring. Lastly, an inverse probability of censoring weight (IPCW) was computed using time-fixed variables (sex, age group, prior morbidities) and time-varying variables (dexamethasone use and antiviral use). The risk of 30-day mortality was compared between patients with co-infection and (1) COVID-19 infection or (2) influenza only. Pooled logistic regression models were used to estimate the hazard ratio between co-infection patients and single infection patients. Robust variance estimation was applied to compute 95% confidence intervals.

Results: Out of 47,956 adult patients tested for both COVID-19 and influenza, 32,523 (68%) were confirmed positive only for COVID-19 via RT-PCR, while 15,103 (31%) tested positive only for influenza. A total of 330 (0.69%) patients had co-infection. According to the pooled logistic regression analysis, patients with co-infection had a twofold increased risk of mortality within 30 days [HR (95% CI) = 2.11 (1.58, 2.81)] compared to those with COVID-19 alone. This risk further escalated to threefold [HR (95% CI) = 3.45 (2.19, 5.42)] when compared to patients with influenza alone.

Conclusions: Our data covered a period when pandemic restrictions started to relax and addressed the recent strains of COVID-19. The risk of 30-day mortality for patients with co-infection was found to be significantly higher those with single infection, warranting vigilance on COVID-19 and influenza co-circulation. Further investigation is needed, particularly in subgroups defined by age and co-morbidities.