Two faller risk functions for geriatric assessment unit patients

Abstract

This study was done in order to construct a simple clinical failer predictor model, which quantifies risk for the individual elderly patient. Fifty patients sequentially admitted to a geriatric assessment unit were assessed systematically at admission for six potential failer risk factors: age, gait, speed, gait quality, gender, symptoms of urinary urgency, and symptoms of dizziness. Falls were systematically recorded during admission. Only age, abnormal gait, and self-selected gait speed were found to be associated with fallers. Multivariate analysis revealed that the two best models for predicting those most likely to be fallers were the factor subgroups: 1) age plus gait quality, and 2) age plus gait speed. The difference between the two models is the choice of gait measure incorporated. Gait speed is a continuous quantitative variable with no known limit of normal. Gait quality is a qualitative variable where abnormal gait is defined by clinical criteria. The predictability of each model was verified in another population (N=78).