OBJECTIVES: To estimate the incidence of ankle sprains and severe ankle sprains attending accident and emergency (A&E) units; to describe current practice for severe ankle sprains in A&E units in the United Kingdom. METHODS: Crude age and sex specific incidence rates were calculated for four health districts from cases ascertained from data on seven A&E clinical information systems. Case records of patients with ankle sprains at an A&E unit in another health district were audited and the proportion of severe ankle sprains calculated. UK A&E units were surveyed about their usual treatment of patients with severe ankle sprains. RESULTS: The estimate of the crude incidence rate of ankle sprains was a minimum of 52.7 per 10 000, rising to 60.9 (95% CI 59.4 to 62.4) when figures were adjusted for the proportion of patients without a diagnostic code (13.7%). There were important age-sex differences with unadjusted rates observed from 127.8 per 10 000 (CI 115.5 to 140.0) in girls aged 10-14 years to 8.2 (CI 4.2 to 12.3) in men aged 70-74 years. As 14% of ankle sprains attending A&E were classed as severe, this would equate to 42 000 severe ankle sprains per year in the UK. In the UK wide survey, there was a response rate of 79% (211 of 266). Among the responders, Tubigrip was used routinely in 55%, below knee casts in 3%, and braces in 2%. Boots were not used routinely in any unit. CONCLUSION: While there is considerable variation in severe ankle sprain management in UK A&E units, most are treated with the minimal mechanical support of Tubigrip.