Introduction: This is the first prospective longitudinal study carried out in a Chinese elderly population with the objective of identifying the incidence and predictors of falls.
Materials and methods: This is a population-based cohort study in Hong Kong with 1517 ambulatory elderly Chinese recruited using a multi-stage sampling method. Baseline data on demographic, comorbid diseases, drugs, Activities of Daily Living (ADL) [Barthel Index and Lawton's Instrumental Activities of Daily Living (IADL)], Geriatric Depression Scale (GDS-15), cognitive assessment by the Abbreviated Mental Test (AMT), fear of falling, self-perceived mobility problem, hand grip strength, lower limb power, balance and gait tests were performed. Every subject was followed up for 1 year.
Results: Four hundred and one falls occurred in 294 fallers (19.3%) over 1 year of follow-up. The prevalence of falls and recurrent falls were 19.3% and 4.75%, respectively. The incidences of falls (i.e., the fall events) were 220, 324 and 270 per 1000 person-years for men, women and both gender, respectively. The independent predictors of falls were previous history of falls, advancing age, Parkinson's disease, knee extension power and gait speed. The independent predictors of recurrent falls were previous history of falls, self-perceived mobility problem, the knee extension strength and the Total Mobility Score of the Tinetti Balance and Gait Evaluation.
Conclusions: The incidence of falls in the Chinese elderly was 270 per 1000 person-years. History of falls, old age, Parkinson's disease, decreased lower limb power and impairment in balance and gait function were important independent predictors of falls or recurrent falls in the Chinese elderly. Effective fall prevention programmes targeted at improving these risk factors for falls should be developed for the Chinese elderly in Hong Kong and Asia.