A multifactorial approach to understanding fall risk in older people

Objective: To identify the interrelationships and discriminatory value of a broad range of objectively measured explanatory risk factors for falls.

Design: Prospective cohort study with 12-month follow-up period.

Setting: Community sample.

Participants: Five hundred community-dwelling people aged 70 to 90.

Measurements: All participants underwent assessments on medical, disability, physical, cognitive, and psychological measures. Fallers were defined as people who had at least one injurious fall or at least two noninjurious falls during a 12-month follow-up period.

Results: Univariate regression analyses identified the following fall risk factors: disability, poor performance on physical tests, depressive symptoms, poor executive function, concern about falling, and previous falls. Classification and regression tree analysis revealed that balance-related impairments were critical predictors of falls. In those with good balance, disability and exercise levels influenced future fall risk-people in the lowest and the highest exercise tertiles were at greater risk. In those with impaired balance, different risk factors predicted greater fall risk-poor executive function, poor dynamic balance, and low exercise levels. Absolute risks for falls ranged from 11% in those with no risk factors to 54% in the highest-risk group.

Conclusions: A classification and regression tree approach highlighted interrelationships and discriminatory value of important explanatory fall risk factors. The information may prove useful in clinical settings to assist in tailoring interventions to maximize the potential benefit of falls prevention strategies.