Temporal Trends in the Chronobiology and Epidemiology of Sepsis Admissions over 21 Years: A Nationwide Study in Northwestern Spain

Background: Sepsis remains one of the leading causes of hospital morbidity and mortality worldwide. While its epidemiology has been extensively investigated, the chronobiology of sepsis—the temporal patterns underlying its occurrence and outcomes—has received comparatively little attention, despite its potential relevance for anticipating clinical demand and optimizing healthcare resource allocation. Objective: To analyze trends in the epidemiology and chronobiology of hospital admissions for sepsis across a large healthcare region in Spain over a 21-year period. Design: Retrospective observational study based on clinical–administrative records. We included all patients ≥ 18 years who had a principal diagnosis of sepsis or septic shock admitted between 2001 and 2021, identified through ICD-9-CM and ICD-10-CM coding in the Spanish National Hospital Discharge Records Database. Annual incidence, in-hospital mortality, and temporal distribution of admissions and deaths were assessed. Joinpoint regression was used to evaluate trends, Fourier spectral analysis to identify dominant rhythms, and multi-harmonic cosinor models to test for circannual rhythmicity. Results: We identified 39,622 sepsis admissions. Hospital incidence increased significantly over time (annual percent change [APC]: +9.6%; p < 0.001), with two inflection points (2008 and 2014). In-hospital mortality decreased linearly from 42.1% in 2001 to 31.9% in 2021 (p < 0.001), despite a progressive increase in patient age. Chronobiological analyses revealed no significant circannual rhythm in incidence or mortality (cosinor p = 0.14), although mortality was disproportionately clustered in winter months (p < 0.001). In multivariable analyses, clinical and epidemiological variables were independent predictors of in-hospital mortality. Conclusions: Over the last two decades, sepsis incidence has risen steadily, whereas hospital mortality has declined. Although no regular biological rhythm was demonstrated, excess winter mortality suggests extrinsic seasonal influences. This study provides novel evidence by jointly examining the epidemiology and chronobiology of sepsis, and supports their integration into healthcare planning strategies.