Valuable information on the health of a population can be gained by looking at how many people die and what caused their death. Comparing causes of death between populations frames this information within the context of the wider community and can help to explain differences in the health of populations. Differences in patterns of death between populations can result from differences in behaviours, exposures to disease or injury, and social and environmental circumstances. This information can be used to guide and evaluate health policy and interventions.

This report examines leading causes of death among three Australian Defence Force (ADF) service status groups—personnel serving full time, personnel in the reserves, and ex-serving personnel. Comparisons are made with leading causes of death in the Australian population to determine if any groups of ADF personnel are at greater risk of death from particular causes. This analysis builds on findings from the 2018 AIHW study into the incidence of suicide among serving and ex-serving ADF personnel and is based on the same data as the comparative analysis undertaken for that study (AIHW 2018b).

Results are based on analysis of 1,939 deaths that occurred between 2002 and 2015 among serving and ex-serving personnel with at least 1 day of ADF service since 1 January 2001. As such, the maximum length of follow-up between discharge and death for ex-serving personnel in this study is 15 years, and less for those discharged after 2001.

Overseas deaths, including deaths that occur on operational deployments, are generally outside the scope of Australian death statistics, and are therefore largely not captured in these results. While the proportion of women in the ADF is increasing, the number of women serving in the ADF has historically been low. As such, analysis focuses on men as they made up the majority of both ADF personnel (around 85%) and deaths (92%) in the study. Some of the results in this study are based on small numbers, particularly those involving women. Rates produced using small numbers can be sensitive to small changes in counts of deaths over time. This should be taken into account when interpreting the findings from this study