Background: It is known that more people die during the nighttime compared to daytime in the community setting. Many health care providers are required to care for patients in a critical condition which includes around the time of death, and which may be at night. However, there have been few studies with respect to the factors associated with death at night. We identified factors associated with death at night.
Methods: For consecutive patients who died over the period of one year from April 2012 and March 2013, we retrospectively analyzed the factors associated with nighttime (17:00-8:59) death compared to daytime (9:00-16:59) death. As independent variables, age, gender, marital status, residence, activity of daily living, Charlson comorbidity index, seasons of the year at admission, weekday or weekend admission, daytime or nighttime admission, means of transfer to the hospital, department of admission care, and sequential organ failure assessment score were obtained. Logistic regression model was used for multivariable adjusted comparison.
Results: Of 851 patients (mean age, 78 years; 355 (42%) woman), 284 (33.4%) patients died during the daytime and 567 (66.6%) patients died during the night. The median time from their initial visit to death was 15 days. 270 (31.7%) patients died during a weekend or a holiday. Univariate analysis showed that patients admitted during the weekend for admission (p=0.038) and the department of admission for their care (p=0.041) were significant factors. Multivariate analysis revealed that the being over 75 years of age (adjusted odds ratio 1.68, 95% CI 1.03-2.73) was the only independent predictor of death at night.
Conclusions: Our study revealed that dying at night occurs more commonly in patients over the age of 75 years compared to other patient factors. There is a need to provide effective medical care during the nighttime in acute care hospitals, especially in Japanese facilities with an increasingly aging population.