The Health of Former Prisoners of War: Results from the Medical Examination Survey of Former POWs of World War II and the Korean Conflict (1992)

The Medical Follow-up Agency (MFUA) has been studying the health of former prisoners of war (POWs) since shortly after World Wax II (WW II). At that time, Cohen and Cooper (1954) assembled rosters of former WW II POWs and non-POW controls and characterized their mortality, morbidity, and disability after liberation. When it began, the study differed in two important aspects from other contemporary investigations of former WW II prisoners: it was designed to study a representative sample of all former prisoners, and it included control groups. The importance of controls was emphasized by Cohen and Cooper, who asserted that "whatever the consequences of imprisonment may be, they are not likely to be grossly different in kind from the effects of combat, disease, malnutrition, and other adverse experiences encountered outside prison camps. Rather, the important effects of imprisonment will manifest themselves as differences in severity or duration, or both, so that the bases of comparison which can be provided only by suitable control groups are indispensable" (Cohen and Cooper, 1954, p. 1). The original Cohen and Cooper follow-up was also characterized by its total dependence on records—death records, hospitalization records, disability records, and questionnaires—a characteristic shared by all subsequent MFUA follow-ups.

Starting with the rosters built by Cohen and Cooper for the original study, Nefzger (1970) added Korean conflict prisoners and controls and studied the mortality of all groups. A third study, by Beebe (1975), was a

20 year morbidity follow-up, collecting data from military and Department of Veterans Affairs (then Veterans Administration; VA) records and from questionnaires. A fourth study, undertaken by Keehn (1980), continued through 1976 the mortality follow-up begun earlier. A fifth study, directed by Page (1991), was a 1984–1985 mail questionnaire follow-up that focused on the persistence of depressive symptoms. The study presented here, which was designed to collect morbidity data from 1965 through approximately 1985, also includes data collected from these earlier follow-ups. Because deliberate efforts to collect complete and detailed cause-specific mortality data were not made, only brief anecdotal mortality data are reported.

Because this examination study followed up the MFUA cohort as assembled earlier, it is useful to review briefly the constitution of that cohort—both cases and controls. As originally constructed, the rosters of World War II ex-POWs and controls were expressly limited to white Army males, by far the largest demographically homogeneous segment of military personnel. This choice was wholly a matter of convenience "to avoid needless multiplication of variables" (Cohen and Cooper, 1954). A U.S. Army punchcard file of all prisoners liberated alive was the source of the study roster. Independent samples were drawn from the European theater of operations (which included both Mediterranean [MTO] and European [ETO] theaters); after the punchcard files were reduced to white males, sampling was carried out by selection of random digits in the service serial numbers. The samples were divided into a larger portion (group A) and a smaller (group B), with the more comprehensive morbidity follow-up performed only for group B and mortality follow-up performed for both groups A and B.

WW II controls were white Army males selected from units whose members were in combat in the theaters in which the prisoners were captured during the same time periods. Because the Pacific theater WW II prisoners were largely composed of the captured defenders of Bataan and Corregidor, and because virtually all survivors were captured, it was not possible to select controls who had contemporaneous combat service in the Pacific theater. Pacific combat controls therefore were identified from the payroll rosters of units that entered combat later. More details can be found in Cohen and Cooper (1954).

Nefzger (1970) added some Pacific prisoners to the sample, selecting them in the same way as those in the original sample except that a different combination of serial number digits was used. New controls were identified from the same payroll rosters used earlier for this study, but Nefzger matched individual controls to new prisoners on the basis of rank, arm or service,

age, and race (again, all subjects were white males). Nefzger also added a new group of malnourished prisoners from the European theater, which had been derived from a 20% sample of 1945 Army hospital admissions, and a group of Korean conflict prisoners and controls selected from a file of all known U.S. Army casualties of that war. POWs were selected from a list of all repatriated prisoners, and controls were selected from the group of men who were wounded and returned to action in Korea. Thus, the Korean conflict controls differ from all other control groups because each control is the victim of a combat injury. Prisoners and controls were matched on the first digit of their Army serial number, rank, date of casualty (capture or wounding), and Army unit at the time of casualty. Selection of Korean conflict prisoners and controls was not limited in terms of race, and they were not selected to match on race; nevertheless, 85% of the former and 87% of the latter were white. Further details are available in the study report (Nefzger, 1970). Approximately 2,000 POWs from all eras and an equal number of controls were eligible for this follow-up.

Although the current survey continues the follow-up of a sample of POWs and controls originally assembled some 35 years ago, it differs from Beebe's morbidity follow-up in its inclusion of the malnourished European POWs. Thus, the survey reports data from seven study groups, shown below with the abbreviated titles that will be used in this report:

PWP (Prisoners of War, Pacific theater, WW II)

WP (War veterans, Pacific theater, WW II)

PWE (Prisoners of War, European theater, WW II)

PWEM (Prisoners of War, European theater, WW II, Malnourished)

WE (War veterans, European theater, WW II)

PWK (Prisoners of War, Korean conflict)

WK (War veterans, Korean conflict).

A brief discussion of the results of earlier studies provides a context for this follow-up. The earliest study, by Cohen and Cooper, found excess mortality in Pacific theater prisoners—mainly accidents and tuberculosis—but no excess mortality in European prisoners. It also identified tuberculosis, residual effects of malnutrition, psychological problems, ophthalmic changes, gastrointestinal disorders, and cardiovascular disease as frequent conditions worthy of continued study. Nefzger's mortality follow-up showed that the early excess mortality had begun to diminish by the mid-1960s among WW II POWs, while Korean conflict former POWs continued to have higher mortality throughout the entire 12 years of follow-up in that study. Beebe, in a 1965 questionnaire and record follow-up, noted that

Pacific WW II prisoners were experiencing both somatic and psychological aftereffects, whereas European prisoners apparently were experiencing only psychological aftereffects. Higher rates of illness were manifested in higher hospitalization rates, higher VA disability compensation rates, and more self-reported health problems. Keehn's mortality follow-up through the mid-1970s documented the fact that the increased early risk of dying among WW II Pacific and Korean conflict prisoners had lessened over time. In fact, for these POWs, as well as for the WW II European POWs, overall death rates in the most recent follow-up period were indistinguishable from those of the comparably aged U.S. population.

Particularly important to the current study are the results of the latest completed follow-up, a 1984–1985 mail questionnaire survey. This follow-up found that detrimental psychological effects of military captivity persisted for as long as 40 years after repatriation. (Beebe had shown earlier that 20 years after repatriation, the most striking and persistent aftereffects of the POW experience were psychological.) The 1984–1985 follow-up collected data on depressive symptoms, as measured by the Center for Epidemiologic Studies depression (CES-D) scale developed at the National Institute of Mental Health; it found rates of depressive symptoms among POWs that were from three to five times higher than expected from comparable studies of the general population. Equally important, the follow-up showed that depressive symptoms were more severe among POWs who had been treated more harshly during captivity, although POWs who were older at capture and who had more education had less depressive symptomatology than younger, less well-educated POWs. It was speculated, however, that these findings on depressive symptoms were actually due to a higher prevalence of some underlying diagnosis, probably posttraumatic stress disorder.

The design of the current study drew on these earlier follow-ups in deciding on the basic kinds of data that should be collected. First, it was clearly time for a more general morbidity survey; using VA medical practitioners to collect data as part of a comprehensive medical examination would provide the first such examination data since repatriation, as well as the most complete morbidity information since Beebe's follow-up study in 1967. Second, the 1984–1985 results pointed out the need to obtain diagnostic data from physicians as well as self-reported symptom data. Thus, the current study consisted of a comprehensive medical examination, to achieve the first objective, plus a structured clinical interview and comprehensive psychiatric evaluation, to achieve the second.

Beebe GW. 1975. Follow-up studies of World War II and Korean war prisoners: II. Morbidity, disability, and maladjustments. Am. J. Epidemiol. 101:400–422.

Cohen BM, Cooper MZ. 1954. A Follow-up Study of World War II Prisoners of War. Veterans Administration Medical Monograph. Government Printing Office, Washington, DC.

Keehn RJ. 1980. Follow-up studies of World War II and Korean war prisoners. III. Mortality to January 1, 1976. Am. J. Epidemiol. 111:194–211.

Nefzger MD. 1970. Follow-up studies of World War II and Korean war prisoners. I. Study plan and mortality findings. Am. J. Epidemiol. 91:123–138.

Page WF, Engdahl BE, Eberly RE. 1991. Prevalence and correlates of depressive symptoms among former prisoners of war. J. Nerv. Ment. Dis. 179:670–677.