Evaluation of fall risk assessment score in a hospital

Abstract

As fall-prevention in hospitals is extremely important, assessment of the risk of falls using a fall assessment score on admission is commonly performed in hospitals. Although it is essential to validate a fall assessment system, no validation has been done in our hospital. Therefore, in the present study, we evaluated the use of a fall assessment score in our hospital. We enrolled all patients (15 years of age and over) who were admitted to the neurosurgery and neurology ward between April 2009 and March 2010. Among these 532 patients, 40 (7.5%) experienced a fall. Based on the fall assessment score, 2.7% of patients with a category 1 had falls while those with category 2 and category 3 had significantly more falls (5.3% and 15.3%, respectively). Univariate Cox proportional hazard modeling that considered the duration of hospitalization showed that there were no significant differences between patients who had experienced a fall and those who had not experienced a fall in terms of age, hearing and visual deficits, cognitive dysfunction, and patient's character. Multivariate Cox proportional hazard modeling that considered the duration of hospitalization revealed that inpatient falls were significantly associated with experience of a fall and use of medication related to falls. When dividing patients into 3 groups (patients with both experience of a fall and taking medication related to falls, patients with either experience of a fall or taking medication related to falls, and patients without either risk factor), the rate of falls in each group was found to be similar to the rates in the three classes defined based on 8 commonly used parameters. The current fall assessment score used in our hospital is considered to have good predictive validity. However, the score could potentially be simplified without losing predictive efficacy.