Background
Various underlying medical conditions (UMCs) elevate the risk of influenza-associated hospitalization. We evaluated how these rates changed by type and number of UMCs.

Methods
Retrospective cohorts were constructed among adult members of two health systems aged ≥18 years with prior healthcare utilization. Across the 2016–17 to 2019–20 seasons, we estimated influenza-associated hospitalization rates by type and number of UMCs. Hospitalizations were defined using discharge diagnoses or laboratory confirmation. We calculated adjusted rate ratios (aRR) using Poisson regression controlling for site, season, and demographic characteristics. We used causal mediation to estimate the effect of UMCs on influenza-associated hospitalization accounting for influenza vaccination status.

Results
Among 870,888 cohort members, 1,403 were hospitalized with influenza at least once within a season across four seasons. Compared to those without, the aRR for influenza-associated hospitalization was highest for individuals with congestive heart failure (4.2, 95% CI: 3.6–4.9). The aRRs also increased with each additional UMCs compared to those with no UMCs. The effect of UMCs on influenza-associated hospitalizations was higher when not mediated by vaccination status; for those with ≥4 UMCs compared to no UMCs, rates were about 60% higher.

Conclusion
The burden of baseline medical conditions is associated with higher rates of influenza-associated hospitalization. Among those with varying types and number of UMCs, if vaccination prevalence had been lower than observed, influenza-associated hospitalization rates would have been higher. These findings highlight the importance of preventive medical care and annual influenza vaccination in reducing influenza-associated hospitalizations, particularly for individuals at high-risk.