Abstract
The specific quantitative relationship between the calculated mean total nursing hours per patient-day (derived from National Health Insurance Form 9) and in-hospital mortality in Japanese acute-care settings remains largely unexamined. This study aims to evaluate the clinical relevance and patient safety impact of actual nursing care hours by identifying practical thresholds below which patient mortality risk substantially increases. This retrospective, multicentre study included 75,708 patients admitted to general hospital beds between April 1, 2019, and March 31, 2020, across nine acute-care hospitals in Japan. Patients from Intensive Care Units and High Care Units were excluded to focus on general ward care. The main outcome measure was in-hospital mortality (dichotomous outcome). The mean daily nursing hours per patient was 3.88 h (median: 3.75 h). Mixed-effects logistic regression analysis showed that while higher nursing hours generally correlated with lower mortality, this risk reduction plateaued beyond 4.5 h. Compared to the 4.5–5.0 h care category (reference), patients receiving less than 4.0 h of care experienced a substantial increase in the odds of in-hospital mortality, with the risk elevating by 34%–52% (adjusted odds ratios ranging from 1.34 to 1.52). Our study provides evidence of a significant inverse association between actual nursing care hours and in-hospital mortality in Japanese acute care hospitals, suggesting that a minimum threshold of 4.0 h of actual daily nursing care may serve as a practical safety boundary for patient survival. These findings suggest that current staffing standards in Japan may warrant careful reconsideration from the perspective of patient safety.