The Australian Vietnam Veterans Health Study: I. Study Design and Response Bias

BRIAN I O'TOOLE, RICHARD P MARSHALL, DAVID A GRAYSON, RALPH J SCHURECK, MATTHEW DOBSON, MARGOT FFRENCH, BELINDA PULVERTAFT, LENORE MELDRUM, JAMES BOLTON, JULIENNE VENNARD

The Australian Vietnam Veterans Health Study was set up to examine the post-war health of former soldiers 20 or more years after service and to examine the relation of combat exposure to physical and mental health. A prospective cohort study of a simple random sample of 1000 male Australian Army Vietnam veterans used information gathered from Army records, from personnel interview and questionnaires. Military records were used to examine response bias by determining the differences between 641 interviewed veterans, 50 known deceased veterans and 309 non-respondents (including 48 refusers and 213 non-traceable). Differences were evident between respondents and non-respondents, with logistic regression modelling pointing to pre-enlistment employment, antisocial behaviour, intelligence and post-Vietnam AWOL (absent without leave) as the most important discriminants with non-respondents performing worse. Compared to respondents, deceased left school earlier, had higher rank in Vietnam and at discharge, had a higher overall number of charges but not a higher rate overall, and were less likely to have gone AWOL. Deceased also received more casualty reports than respondents and non-respondents, were better behaved during service, and were better emotionally adjusted than non-respondents. Respondents compared with the Australian population had equivalent or better current socioeconomic status. There seems little bias due to non-response, but deceased tend to come from and older cohort than in the other two groups.