Abstract
BACKGROUND: We examined the association between the coronavirus disease 2019 (COVID-19)-specific prefectural bed utilization rate and in-hospital mortality during the first 3 years of the pandemic in Japan. METHODS: This nationwide study included 58,175 COVID-19 patients from the COVID-19 Registry Japan, hospitalized between May 1, 2020 and November 30, 2022. Based on the weekly COVID-19-specific bed utilization rate in each prefecture at diagnosis, patients were categorized into four groups (<25%, 25% to <50%, 50% to <75%, and ≥75%). Odds ratios (ORs) were estimated by fitting a generalized linear mixed model with prefecture as a random intercept and adjusting for covariates (age, gender, body mass index, smoking and drinking status, and comorbidities). Additional analyses according to age group, gender, and wave of the pandemic were conducted. RESULTS: We observed 2,312 (4.0%) all-cause in-hospital deaths. All-cause in-hospital mortality increased with higher COVID-19 bed utilization rates at diagnosis (OR for multivariable model 1.35; 95% confidence interval [CI], 1.19-1.54 for 25% to <50%; OR 1.89; 95% CI, 1.66-2.16 for 50 to <75%; OR 2.16; 95% CI, 1.80-2.58 for ≥75%; P for trend <0.0001). Stronger associations were noted among the younger population (aged <70 years: OR 3.18; 95% CI, 1.96-5.19) and during the fourth (March 1-June 30, 2021: OR 3.81; 95% CI, 2.13-6.80) and sixth pandemic waves (January 1-June 30, 2022: OR 2.67; 95% CI, 1.68-4.23). CONCLUSION: Our results emphasize that preventing hospital bed shortages during outbreaks is an important public health strategy to reduce the associated mortality, particularly when new strains emerge and in younger people.