Circadian rhythm of mortality among hospitalized cancer patients: A retrospective study.

340

Background: Circadian rhythms are intrinsic 24-hour biological cycles that regulate numerous physiological processes. Understanding the circadian pattern of death in cancer patients may inform the planning of palliative care services. This study aimed to investigate whether a circadian rhythm exists in the mortality of hospitalized cancer patients and to characterize its features. Methods: A retrospective study was conducted using medical records of 607 cancer patients who died during hospitalization at a tertiary cancer hospital in Shandong Province, China, from 2022 to 2024. The time of death was recorded to the minute. Cosinor analysis was applied to evaluate circadian rhythmicity. Subgroup analyses were performed by age, sex, cancer type, resuscitation status, and direct cause of death. Results: The cosinor analysis yielded a coefficient of determination (R²) of 0.3811, confirming a circadian pattern. The overall distribution of death times exhibited a bimodal pattern, with the primary peak occurring between 15:00 and 17:00 and a secondary peak between 09:00 and 11:00. Among patients who received cardiopulmonary resuscitation, the mortality peak occurred in the afternoon (15:00-17:00), whereas among those who did not, the peak occurred in the morning (09:00–11:00). The peak time differed by approximately two hours between males and females. Patients aged ≥60 years showed clearer rhythmicity (R²≈0.34), with a peak at 14:17. Peak timing also varied across cancer types and immediate causes of death, with hepatobiliary cancer peaking at 13:04. Deaths attributed to respiratory and circulatory failure peaked at 15:33, while those from multiple organ failure peaked at 12:24. Conclusions: Mortality among hospitalized cancer patients demonstrates a circadian rhythm with a bimodal distribution. The circadian rhythm of death timing in hospitalized cancer patients is likely the result of superposition between biological rhythms and medical system rhythms, with medical interventions, especially resuscitation and withdrawal processes, potentially being the dominant factor. This has implications for optimizing clinical resource allocation, improving withdrawal decision timing, and enhancing palliative care quality. Future prospective studies integrating biomarkers and process data are needed to provide a more solid foundation for evidence-based end-of-life care practice.