ObjectiveTo describe the epidemiology of domestic swimming pool drowning and near drowning in Brisbane and to examine the efficacy of a broad range of preventive options, including pool fences.DesignA prospective, hospital‐based, injury surveillance system to describe the epidemiology of drowning and near‐drowning and a community survey to describe pool fencing.SettingThe surveillance questionnaire was completed at presentation in the Emergency Department by the parent, nurse and doctor. Personal interviews in households that were randomly selected by means of a stratified sampling scheme provided the pool fencing description.ParticipantsAll 139 children suffering from an immersion injury resulting in presentation at a hospital in the catchment area of The Mater Children's Hospital were included. There were 204 households with a swimming pool in the 1024 households interviewed in the community survey.ResultsThe 100 domestic pool drownings and near‐drownings were equivalent to 15.5 incidents per year per 100 000 children aged 0‐13 years and 64.9 per year per 100000 for the critical 1‐3 years age group. Of 72 children who gained unintended access to a domestic pool, 88.9% were less than 3 years of age and 52.8% were less than 2 years. All 10 of the children who drowned and five who were severely brain damaged (age range, 12‐32 months) were in this group. The risk of a drowning or near‐drowning involving unintended access to an unfenced pool is 3.76 times higher than the risk associated with a fenced pool (95% confidence limits for relative risk: 2.14, 6.62).ConclusionsPool fences are an effective method of preventing child drownings and near‐drownings. This effectiveness can be further improved if compliance with gate closure can be enhanced. This should be emphasised in health promotion accompanying the introduction of universal pool fencing.