A literature review carried out as an earlier stage of this project concluded that there is a dearth of information regarding alcohol-affected pedestrian crashes. The present study reports a detailed analysis of crashes involving intoxicated pedestrians in the ATSB Coronial Database. Records for 240 cases for the years 1999, 2001 and 2002 were examined in the study. A researcher read over each report, then attempted to use the coronial record to answer a standard set of questions about each case. Although the general pattern of findings relating to intoxicated pedestrian crashes was replicated (ie male victims, high BACs, night and weekend occurrence, and over-representation of indigenous persons),. there were new findings relating to behavioural patterns. The most frequent drinking venues were hotels and other people?s homes. Victims tended to have been drinking for long periods, many were identified as drinking in company, and were almost entirely local residents. In 14% of cases, there was an argument which precipitated the victim?s leaving the drinking venue. At the time of the crash, the victim was walking along the road in approximately 30% of cases, lying on the road in another 20% of cases, sitting in the road in another 2%, and standing on or near the edge of the road in a further 5%. Almost all pedestrians were struck by the vehicle within two hours of finishing drinking. Surprisingly, there was no association between BAC level and pedestrian movement at the time of the crash. Recommendations arising from the study include strengthening Responsible Serving of Alcohol programs, publicity aimed fellow patrons and drivers, programs targeted at disadvantaged groups in the community, traffic engineering treatments outside drinking venues, and continuing monitoring of alcohol affected traffic casualties.