Recording of community violence by medical and police services.

OBJECTIVES To determine the extent to which community violence that results in injury treated in emergency departments appears in official police records and to identify age/gender groups at particular risk of under-recording by the police. METHODS Non-confidential data for patients with assault related injury treated in the emergency departments of two hospitals in one South Wales city (Swansea) during a six month period were compared with data relating to all recorded crimes in the category "Violence against the person" in the police area where the hospitals were located. RESULTS Over the six month period a total of 1513 assaults were recorded by Swansea emergency departments and the police (1019, 67.3% injured males and 494, 32.7% injured females). The majority of these assaults (993, 65.6%) were recorded exclusively by emergency departments; 357 (23.6%) were recorded only by the police and 163 (10.8%) were recorded by both emergency departments and the police. Equal proportions of males (67.3%) and females (67.5%) injured in assaults were recorded by both emergency departments and the police, but men were more likely to have their assault recorded exclusively in emergency departments (odds ratio (OR) 2.1, 95% confidence interval (CI) 1.7 to 2.7) while women were more likely to have their assault recorded exclusively by the police (OR 2.5, 95% CI 2.0 to 3.2). There were no significant relationships between exclusive emergency department recording and increasing age (OR 1.0, 95% CI 0.9 to 1.2), exclusive police recording and increasing age (OR 1.1, 95% CI 1.0 to 1.2), or between age and dual recording (OR 0.9, 95% CI 0.8 to 1.0). CONCLUSIONS Most assaults leading to emergency department treatment, particularly in which males were injured, were not recorded by the police. Assaults on the youngest group (0-10, particularly boys) were those least likely to be recorded by police and females over age 45, the most likely. Emergency department derived assault data provide unique perspectives of community violence and police detection.