This purpose of this study was to determine whether an identification bracelet is effective in preventing falls among high-risk patients who are undergoing in-patient physical rehabilitation. A stratified, randomized, balanced controlled clinical trial was conducted; participants were blinded as to the outcome and the study hypothesis. All patients having one or more risk factors that predisposed them to falls were randomized to receive either a blue identification bracelet or no bracelet. The identification bracelet was intended to increase patients' vigilance about falling. Two risk strata were specified. The high risk stratum consisted of patients with stroke or ataxia, urinary incontinence, or a history of falls. The low risk stratum comprised patients older than 80 years and those on one or more medications that had been identified as contributing to an individual's risk of falling. This report presents the effect of the identification bracelet only among persons in the high-risk stratum. Over 1 year, 65 high-risk subjects were randomized to receive the blue identification bracelet and 69 high-risk subjects were controls. In the intervention group, 27 persons (41%) fell at least once, whereas in the control group 21 persons (30%) fell at least once yielding a hazard ratio of 1.3 (95% confidence interval: 0.8 to 2.4). These results suggest that the identification system was of no benefit in preventing falls among high-risk persons.