The Intercollegiate Academic Board of Sport and Exercise Medicine.

Objective-To determine injury patterns and characteristics specific to domestic violence in women who present to the accident and emergency (A&E) department. Design-A retrospective case note review of all female assaults over a one year period. The subjects were women who disclosed that their injuries were due to assaults by either a current or a previous male partner. Controls were female assault victims not injured by domestic violence. Setting-A medium sized urban A&E department. Results-There were 500 female assaults out of 48 169 new attendances. Domestic violence was disclosed in 103 cases. The following features were significantly associated with domestic violence in women: multiple injuries (p<0.001) (especially to the head and arms), fractures (p<0.05), loss of consciousness (p<0.05), abdominal injuries (p<0.05), pregnancy (p=0.01), injury occurring on "stairs" (p=0.01), and general practitioner referral (p<0.01). Conclusions-Women who have been assaulted are more likely to have been injured during domestic violence if they sustain multiple injuries (including fractures), abdominal injuries, have lost consciousness , or have been referred by their general practitioner. These markers may help medical staff to identify more cases of undisclosed domestic violence. The markers need to be tested further in a prospective study. In this study, domestic violence is defined as "the abuse of a women by a current or former male partner".' 2 A lifetime prevalence of 25%, in the UK, highlights domestic violence as a public health issue.' Between 5% and 11% of women who present acutely to the accident and emergency (A&E) department attend as a result of domestic violence."5 Many publications describe the epidemiology of domestic violence in patients attending an A&E department."' In comparison, few studies have attempted to identify patterns of injury and presentation that help distinguish between domestic violence and non-domestic violence."2-15 There is a lack of case-control studies.' '4 This is the first UK study with a control group. Identification of markers would be important in alerting clinicians to the possibility of domestic violence, as it is known that many cases are not disclosed,4 5 9 and that attacks tend to be repeated with both increasing frequency and severity.9 15-17 The study examines female assaults, comparing assaults due to domestic violence with a control group of women who were not injured by domestic violence. Methods One year of the A&E department medical and nursing case notes were hand searched by one author (RLS) to identify all female assaults. These were classified into …