Predictors of serious fall injury in hospitalized patients.

The purpose of this study was to identify predictors of falls that result in serious injury in hospitalized patients. The study involved secondary data analysis of 1,438 patient falls in a community hospital system between 2008 and 2010. The analysis included demographics, severity of illness, diagnosis-related group (surgical vs. medical), event type (bathroom, bed, chair, transfer, ambulating), risk factors identified by the Hendrich II fall risk assessment prior to the fall (confusion, depression, altered elimination, dizziness, antiepileptic or benzodiazepine medications), and contributing factors identified through an online event reporting system post-fall (incontinence, confusion, history of falls, alteration in mobility, and medication-related). Logistic regression results indicated that the overall model was a good fit and two predictors, age greater than 64 and male gender, were statistically reliable in predicting which patient falls would result in serious injury.