OBJECTIVES: To determine the effectiveness of multifactorial intervention to prevent falls in cognitively intact older persons with recurrent falls. DESIGN: Randomised controlled trial of multifactorial (medical, physiotherapy and occupational therapy) post-fall assessment and intervention compared with conventional care. SETTING: Accident & Emergency departments in a university teaching hospital and associated district general hospital. SUBJECTS: 313 cognitively intact men and women aged over 65 years presenting to Accident & Emergency with a fall or fall-related injury and at least one additional fall in the preceding year; 159 randomised to assessment and intervention and 154 to conventional care. Outcome measures: primary outcome was the number of falls and fallers in 1 year after recruitment. Secondary outcomes included injury rates, fall-related hospital admissions, mortality and fear of falling. RESULTS: There were 36% fewer falls in the intervention group (relative risk 0.64, 95% confidence interval 0.46-0.90). The proportion of subjects continuing to fall (65% (94/144) compared with 68% (102/149) relative risk 0.95, 95% confidence interval 0.81-1.12), and the number of fall-related attendances and hospital admissions was not different between groups. Duration of hospital admission was reduced (mean difference admission duration 3.6 days, 95% confidence interval 0.1-7.6) and falls efficacy was better in the intervention group (mean difference in Activities Specific Balance Confidence Score of 7.5, 95% confidence interval 0.72-14.2). CONCLUSION: Multifactorial intervention is effective at reducing the fall burden in cognitively intact older persons with recurrent falls attending Accident & Emergency, but does not reduce the proportion of subjects still falling.