Objective: To systematize results of prospective cohort studies on multiple risk factors for falls in the elderly and to assess their methodological quality.
Methods: Systematic review of epidemiological studies from Medline, SciELO and Lilacs database. We included prospective cohort studies with samples of more than 100 subjects of both sexes, older than 64 years, and living either in the community or a nursing home.
Results: Of 726 studied identified, 15 met the inclusion criteria of being published between 1988 and 2005. The methodology of the studies varied. The main factors associated with increased risk of falls include: previous falls, altered gait, functional impairment, cognitive impairment, psychotropic medication use and excessive physical activity. Despite contradictory findings, being a woman at an advanced age may also be a predictor of falls.
Conclusions: Methodological limitations were identified in prospective cohort studies on falls. There is a need for further studies on extrinsic determinants, including evaluator blinding and closer monitoring during follow-up with reduced time of recall.