The association of state per capita income and military service deaths in the Vietnam and Iraq wars

Background In the United States, social burdens including war casualties are often distributed unequally across groups of individuals, communities, and states. The purpose of this report was to examine the association between war deaths and per capita income in the 50 states and District of Columbia during the Vietnam and Iraq wars. Methods The numbers of deaths by the home state of record for each conflict were obtained from Department of Defense records on the Internet as were key variables including age at death, gender, race, branch of service, rank, circumstances of death, home state of record and the ratio of wounded to dead. In addition, we obtained state per capita income and state population for the relevant times. Results Characteristics of decedents in the 2 conflicts were very similar with young, white enlisted men accounting for the majority of deaths. However, in the Iraq war, women accounted for a 2.4% of casualties. Also of note was the higher ratio of wounded to dead in Iraq. At the level of the state, the correlation between the ratio of deaths per 100,000 and per capita income was -0.51 (p < 0.0001) for Vietnam and -0.52 for Iraq (p < 0.0001). In both eras, states with lower per capita income tended to have higher ratios of deaths per population. Conclusion For military service members serving in the Vietnam and Iraq conflicts, there were many more women who died in the latter war. Whether war deaths resulted in lower per capita income cannot be determined from these cross sectional data; we simply note a strong association between per capita income and war casualty rates for both wars.


Background
In the United States, social burdens including war casualties are often distributed unequally across groups of individuals, communities, and states. In both Vietnam and Iraq the majority of deaths occurred in white enlisted men who served in the US Army or Marine Corps in hostile situations [1][2][3][4]. Less is known about the association between war casualties and income of states. The purpose of this report was to examine the association between war deaths and per capita income in the 50 states and District of Columbia during the Vietnam and Iraq wars. A secondary objective was to compare the characteristics of individuals who died in the 2 conflicts.

Methods
The numbers of deaths by the home state of record for each conflict were obtained from the World Wide Web [2,4] as were key variables including age at death, gender, race, branch of service, rank (enlisted versus officer or warrant officer), circumstances of death (hostile versus nonhostile), and the home state of record. Hostile deaths refer to those personnel who were killed in action, died of wounds, were missing in action, or died in captivity. Tabular counts of decedents (e.g., number of deaths by age) for the Vietnam war were obtained from the National Archives of the United States, which in turn received archival records from the Department of Defense [2]. Detailed individual level characteristics of decedents in the Iraq war were obtained from publicly available files provided by the Defense Manpower Data Center [4]. Death information for the Iraq war is changing as the war is ongoing. The ratio of wounded to dead was calculated for the conflicts in Vietnam and Iraq [5,6]. Race was defined differently in the 2 periods; in order to compare racial characteristics during the 2 time periods, race was categorized as white or non-white.

Results
The numbers of deaths in Vietnam exceeded those in Iraq by more than 10 times as deaths in Vietnam spanned the years 1956 through 1998, whereas those for the ongoing Iraq war covered the years 2003 to July 2008. Vietnam deaths after 1975 were for individuals whose status changed from missing to deceased. The characteristics of casualties in Vietnam and Iraq were similar, with the exception that the number of women increased significantly in Iraq (table 1). In Vietnam, there were only 8 recorded deaths in women, whereas in the Iraq there were 92 women who died. Another difference was that the ratio of wounded to dead was much higher in Iraq.
As seen in the table 2, the average per capita income in 1969 in 1999 dollars was $11.5 thousand compared to $20.9 thousand in 1999. The ratio of deaths per 100,000 was 29.8 for Vietnam and 1.6 for Iraq. The correlation between the ratio of deaths per 100,000 and per capita income was -0.51 (p < 0.0001) for Vietnam and -0.52 for Iraq (p < 0.0001). In both eras, states with lower per capita income tended to have higher ratios of deaths per population (figures 1 and 2).

Discussion
Much has been said about similarities and differences between the conflicts in Vietnam and Iraq. The former was fought by draftees and volunteers, whereas the current conflict includes only volunteers, as at this time there is no military draft in the United States. One striking difference was the numbers of deaths in Vietnam far exceeded those in the ongoing Iraq war. On the other hand, characteristics of decedents in the 2 conflicts were very similar with young, white enlisted men accounting for the majority of deaths. However, in the Iraq war, women accounted for a higher proportion of casualties. Also of note was the difference in the ratio of wounded to dead in the 2 conflicts. This may be an indication of improved medical care in the Iraq war, as personnel with severe injuries are surviving wounds that in previous wars were lethal [9].
It was also the case that states with fewer resources as measured by per capita income experienced higher casualty rates in both conflicts. The magnitude of the association was relatively strong and remarkably similar in both eras. For states and the armed forces, this meant the loss of young men and women; the loss of young women is particularly applicable to the Iraq conflict, where women currently account for 2.4% of deaths [2,4].
A major limitation of the study was that the state population was used as the denominator. A better denominator would have been the number of combatants from each state, as it is likely that individuals from poorer states were more likely to join the military than were those from wealthier states. Unfortunately, it was not possible to obtain counts of the number of combatants by state.
While there could have been selection bias in that men and women from poorer states were more likely to join the military, this does not change the fact that poorer states had higher casualty rates. This study was based on publicly available data over which the author had little control; information was used as provided. For example, 1999 was latest year for which state per capita income was  reported; this was 4 years prior to the start of the Iraq War. Furthermore, the information regarding deceased combatant's characteristics was taken "as is." The author was not able to assess the accuracy of the information; this applies particularly to the Vietnam war which started over 45 years ago.

Conclusion
In the Vietnam and Iraq conflicts, the burden of war casualties was shouldered primarily by young enlisted men, but as seen in Iraq, young women are beginning to share this burden. Less appreciated is the fact that populations may be affected by war casualties in military personnel. Whether war deaths cause states to have lower income cannot be determined from these cross sectional data; we simply note a strong association between per capita income and war casualty rates in the 2 conflicts.