A cross-sectional study was conducted to determine the prevalence of domestic violence victims among patients using emergency services at Sydney's Royal North Shore Hospital, in an affluent urban area of New South Wales. This study used a self-administered questionnaire (used in a similar study at the Royal Brisbane Hospital) to investigate the history of domestic violence among patients attending the emergency department during 64 randomly selected nursing shifts in October-November 1994. Adult domestic violence was reported by 19.3 per cent of females and 8.5 per cent of males, confirming the results of the Brisbane study. Evidence for underreporting was found: 4 per cent of females and 6.3 per cent of males who did not report being victims revealed experiences of abuse on nine measures of types of violence, including six taken from the Conflict Tactics Scale. Results supported evidence from other studies suggesting that experience of abuse as a child is a risk factor for being in abusive relationships as an adult. In the past, comparison of results has been limited because of variation in definitions of domestic violence; this has been overcome by intentional replication of the Brisbane study. The study was enhanced by inclusion of patients from non-English-speaking backgrounds and a cohort of parents of children attending. Similar prevalence estimates were found in these groups. Results have implications for the detection and treatment of victims of domestic violence across all strata of society and have potential to raise awareness and affect attitudes towards this significant community problem.