LAUSR

Introduction In our former study, we had validated the previously developed predictive model for in-hospital falls (Saga fall risk model) using eight simple factors (age, sex, emergency admission, department of admission, use of hypnotic medications, history of falls, independence of eating, and Bedriddenness ranks [BRs]), proving its high reliability. We found that only admission to the neurosurgery department, history of falls, and BRs had significant relationships with falls. In the present study, we aimed to clarify whether each of these three items had a significant relationship with falls in a different group of patients. Methods This was a single-center based, retrospective study in an acute care hospital in a rural city of Japan. We enrolled all inpatients aged 20 years or older admitted from April 2015 to March 2018. We randomly selected patients to fulfill the required sample size. We performed multivariable logistic regression analysis using forced entry on the association between falls and each of the eight items in the Saga fall risk model 2. Results A total of 2932 patients were randomly selected, of whom 95 (3.2%) fell. The median age was 79 years, and 49.9% were men. Multivariable analysis showed that female sex (odds ratio [OR] 0.6, 95% confidence interval [CI] 0.39–0.93, p = 0.022), having a history of falls (OR 1.9, 95% CI 1.16–2.99, p = 0.010), requiring help with eating (OR 1.9, 95% CI 1.12–3.35, p = 0.019), BR of A (OR 6.6, 95% CI 2.82–15.30, p < 0.001), BR of B (OR 7.5, 95% CI 2.95–19.06, p < 0.001), and BR of C (OR 4.1, 95% CI 1.53–11.04, p = 0.005) were significantly associated with falls. Conclusion History of falls and BRs were independently associated with in-hospital falls.