Feed

Limited evidence exists on how multimorbidity combinations influence the risk of secondary bacterial infections. This study identified multimorbidity clusters among hospitalised COVID-19 patients in Victoria (2020-2023) and examined their association with secondary bacterial infections and admission outcomes, including ICU admission, hospital and ICU length of stay, and mortality.

We used population-wide linked hospital data and applied cluster analysis to ICD-10 coded chronic conditions to identify multimorbidity clusters. Risks of secondary bacterial infection across clusters were compared to patients with one or no chronic conditions. We used multivariate logistic regression, negative binomial regression, Kaplan-Meier curves, and Cox proportional hazards models to analyse associations with secondary bacterial infection and admission outcomes.

Among 179,688 COVID-19 hospital admissions, three multimorbidity clusters were identified: neuropsychiatric, cardiometabolic-multisystem, and neoplastic. Compared to no multimorbidity, each cluster was associated with significantly higher odds of secondary bacterial infection: neuropsychiatric (Odds Ratio (OR) 2.74, 95% CI 2.59-2.89), cardiometabolic-multisystem (OR 3.87, 95% CI 3.63-4.13), and neoplastic (OR 1.98, 95% CI 1.84-2.14; all p<0.001). For patients with secondary bacterial infection, cardiometabolic-multisystem multimorbidity had the highest increase in hospital length of stay (Incident Rate Ratio (IRR) 1.94, 95% CI 1.82-2.05) and ICU length of stay (IRR 2.75, 95% CI 2.36-3.20). Mortality was significantly elevated across all clusters and was highest in the cardiometabolic-multisystem group.

Multimorbidity was associated with increased risk of secondary bacterial infection and poorer clinical outcomes in hospitalised COVID-19 patients. Integrating multimorbidity profiles into clinical decision-making may enhance antimicrobial stewardship by identifying patients most likely to benefit from antibiotic therapy.