The aim of this study was to examine the incidence of cancer experienced by the 17,381 Australian male veterans of the Korean War compared with that experienced in the Australian community for the period 1982-1999. These cancer patterns were examined for all Korean War veterans and for each of the Services-Navy, Army and RAAF.

Work was commissioned by the Department of Veterans' Affairs (DVA) and was undertaken by AIHW under the supervision of the Study Consultative Committee and a Scientific Advisory Committee. The DVA and AIHW Ethics Committees provided approval for the work.

The outcome of the study showed that Korean veterans experienced a significantly greater overall cancer risk than the Australian community, with an excess of between 13% and 23%, and varying significantly in the pattern across Services.

At the commencement of the study period (1982), there were 15,041 veterans identified as being alive and eligible for the study. An additional 884 veterans (5%) could not be found after extensive searches of data systems and are referred to as 'veterans whose status is unknown'.

There were 3,543 cancers identified in the veteran population in the period 1982-1999 using the DVA Korean nominal roll and the AIHW National Cancer Statistics Clearing House. The most common cancers found among the veteran population were cancer of the prostate (21% of the total cancers), lung cancer (19%), colon (8%), melanoma (7%), rectum (6%) and head & neck cancers (5%).

To ascertain whether veterans experienced cancer at a similar rate to the Australian community, a series of steps were required to calculate the population 'at risk' for each year of the study (i.e. remove deaths). Once this was established the Australian community rate was applied to the 'at risk' population to determine how many cancers would be expected amongst the veterans.

Deaths of veterans were identified both here in Australia using the AIHW's National Death Index and the state and territory Registrars of Births, Deaths and Marriages databases, and overseas using the New Zealand Registrar of Births, Deaths and Marriages database.

The 5% of veterans whose status is unknown posed a problem for the study in that they needed to be taken account of, as including them or excluding from the population 'at risk' significantly influenced how many cancers might be expected amongst veterans.

Two 'at risk' population Scenarios were derived to manage this issue. Scenario 1 excludes veterans whose status was unknown from the at-risk population and Scenario 2 includes this group in the at-risk population. The report presents findings under both Scenarios.

The study compared cancers in veterans with those expected based on Australian community rates under the two Scenarios, by producing standardised cancer incidence ratios. Where the ratio exceeds one, the actual cases of cancer among veterans are higher than the expected number, and vice versa. However, these results also need to be considered along with the confidence intervals around each ratio, which help indicate statistical significance. The findings reported below only highlight those results that have been found to be statistically significant.