Risk factors for elderly falls in a rural community of central Taiwan

Objectives: A one-year prospective study was conducted to determine the incidence rate and risk factors for falls among 368 rural elderly people aged 65 years or older in Shin-Sher Township, Taichung County, Taiwan. Methods: Information on demographics, health-related behaviors, previous fall history, activities of daily living (ADL), health-related quality of life, vision, hearing, cognitive function, and number of chronic diseases was collected through personal interviews at each subject's home. Furthermore, timed Up & Go test was examined in the field. The elderly were asked to send a postcard to inform researchers when a fall occurred, and researchers also called them every two months to ascertain whether they experienced a fall in the past two months. We validated the self-reported falls using their next of kin and medical records. The PWP (Prentice, Williams, Peterson) proportional hazards model was employed to investigate the relationship between the potential risk factors and falls. Results: The incidence rate of falls among the rural elderly was 226 per 1000 person years. Furthermore, previous fall history (relative hazard (RH)= 1.76, 95% confidence interval (CI)= 1.08 to 2.88), impaired number of instrumental activity of daily living (IADL)(RH= 1.22, 95% CI= 1.08 to 1.39), and longer duration of timed Up & Go (compared with the elderly with ≦ 10 second, RH11-20= 1.99, 95% CI= 1.00 to 3.97; RH ≧ 21= 3.01, 95% CI= 1.22 to 7.41) were positively associated with the risk of falls. Seventy-five percent of the self-reported falls were identified by their next of kin, and 69.6% of those who reported to seek medical care due to a fall were corroborated by medical records. Conclusions: A high-risk group of falling among rural elderly people can be identified using a set of simple measures such as Fall history, IADL, and timed Up & Go.