Abstract
General Overview: This paper reports on the findings from a study of data from Emergency Departments of participating hospitals in Queensland, to compare data on injuries from spousal and non-spousal assault.

Objective: The study aimed to assess (1) who are the most vulnerable to spousal assault; (2) whether spousal assault injuries are more or less severe than injuries from non-spousal assault or accidental injuries; and (3) whether the recorded figures for assault injuries on women understate the number of assault injuries.

Methods: Based on data from the participating hospitals’ emergency departments, comparisons were made between the nature of injuries resulting from spousal assault with injuries from non-spousal assault and accidental injuries. Econometric tests were applied to the data.

Discussion: The paper provides a review of the literature. This includes a discussion of profiles of victims of spousal assault, the under–reporting of spousal abuse, the disguising of spousal abuse injuries as accidents and the cost of such abuse to Australia.

Results: The study found that those most likely to suffer spousal abuse were female, Indigenous, unemployed or a home-maker and aged 22 to 45 years of age. In grading injury in terms of very urgent, fairly urgent and not urgent, the study found that spousal assault injuries were more likely to be assessed as needing very urgent treatment, compared to non-spousal assault injuries. However, they were not significantly different from accidental injuries. The study used Bayes’ theorem to estimate the ‘true’ number of women who are injured by being struck by or colliding with a person (i.e. spousal abuse). It found that the reported level underestimated the true level of spousal assaults by almost one-fourth.

Conclusions: The paper concludes that injuries from spousal assault are significantly likely to be more severe than non-spousal assault injuries. In addition, those suffering spousal injuries are more likely to refuse further attention through medically advised hospital admission than persons with other injuries. The authors suggest that Bayes’ theorem provides a more accurate estimate of the number of injuries due to spousal assault. The study argues that while gender, Indigenous status and age are personal factors, employment can be modified through policy. That is, increasing the labour market participation of women could reduce their probability of being injured through spousal assault.