Deaths of Despair: Lessons from the Vietnam Draft Lottery


Understanding the causes of “deaths of despair”1—that is, mortality due to suicide, drugs, and alcohol—is a fundamental challenge for public health research and a serious concern for the nation.2 Studying that challenge, however, is difficult because the antecedents of such deaths are myriad and it is not possible to conduct a carefully controlled experiment to study the subject—or is it? In fact, 50 years ago this December 1, an experiment started that has helped illuminate the causes of deaths of despair and other public health challenges: the Vietnam Draft Lottery.

The Lottery arbitrarily assigned draft numbers to eligible men based on random draws of their birth dates. Thus, the individuals in the Draft’s “treatment condition”—those more likely to be inducted into service—exhibited the same traits prior to the “experiment” as those in its “control condition”—those not subject to the draft. Like an experiment to test medicines or weight loss plans, this feature of the Lottery allowed researchers to attribute differences between groups to the Draft’s intervention, just as scientists would were they studying a randomized experiment. However, unlike most experiments, the Vietnam draft did not affect a mere handful of individuals in a limited way. Instead, it affected an immense number of people and had the potential to influence those peoples’ life outcomes in drastic ways. Indeed, it ranks among the most significant and wide-ranging experiments in human history. And its first use was to understand something akin to deaths of despair.

In the 1970s and 1980s, stories of Vietnam veterans afflicted by alcoholism, drug use, and suicide appeared widely. By the mid-1980s, a team of researchers—Norman Hearst, Thomas B. Newman, and Stephen B. Hulley—asked a research question that probed whether these individual accounts signaled an epidemiological reality: Did induction for service in Vietnam make one more likely to suffer the most tragic of all consequences—death? That is, the trio looked squarely at the most severe and plausible effect of the draft—namely, its possible effect on men’s mortality. Analyzing vital records of California and Pennsylvania, Hearst, Newman, and Hulley earmarked all death certificates (from 1974 to 1983) of men born in years that made them eligible for the draft lotteries.3 Finding roughly 14,000 men with such records, they then looked within those records to see whether birth dates assigned lottery numbers that were called for induction appeared with greater frequency. In other words, Hearst, Newman, and Hulley used a procedure similar to a randomized medical trial. However, in this case, the treatment was not a drug, but rather a draft lottery number that increased one’s likelihood to serve in the military, while the control arm was represented by those who had numbers above the cut-off for induction.

Studying the draft in this manner, they noted a greater frequency of birth dates that called for induction in the mortality records they reviewed. As they reported in their article on the subject, the evidence indicated a 13% higher rate of suicide among draft-eligible men, as well as an 8% rise in the rate of motor-vehicle deaths, thus elevating total mortality by approximately 4%. Subsequent research4 re-examined the effect of lottery numbers on life expectancy and found that the effect of the draft on mortality seemed to dissipate over a longer timeframe.

In the years since that original study, other research used the lottery design to address basic questions that help illuminate the constituent elements of deaths of despair. For instance, clarifying how life experience and genetics shape addiction, researchers studied how lottery numbers—an exogenously imposed life experience—variably affected the smoking of individuals genetically disposed toward it, versus those genetically disinclined to start smoking.5 These researchers found that the draft tipped genetically inclined individuals toward smoking—and the risks of smoking related ailments—but it nary influenced men with a low genetic risk of smoking. Other research used the lottery design to understand drug use, violence, and employment—outcomes linked with deaths of despair.

Such scholarship using the draft lottery as a natural experiment has become sufficiently common that it might be viewed by some as a methodological gimmick, but we contend that such a perspective misses the point. Clean natural experiments that offer insight into major public health issues are rare occurrences and, thus, researchers ought to take full advantage of such opportunities.

Now, in particular, is a crucial time to do so as we think about the health risks to our current veteran population from America’s longest running wars in Iraq and Afghanistan. According to the VA National Suicide Data Report 2005-2016, suicide rates from 2006 to 2015 grew faster among veterans than in the population at large. The website of the US Department of Veterans Affairs’ National Center for PTSD also reports that 10% of Iraq- and Afghanistan-war veterans obtaining services in the VA exhibit substance abuse problems. Figures quoted by researchers at the National Bureau of Economic Research suggest that war veterans overdose twice as frequently as nonveterans. Were these phenomena driven by military service or do they merely reflect an underlying psychological orientation of those who enter the armed forces? The experimental evidence of Hearst et al. would suggest that there is a real effect of service, net of selection, on mortality. More generally, results from studies using the lottery design suggest that the greatest threats to the health and wellbeing of veterans may occur right when they return home. Keeping in mind possible differences between our era and the Vietnam era (eg, levels of medical and psychological support for veterans, the volunteer nature of today’s forces, the higher quality of care from the Veterans Administration health care system, and so on), such discoveries show how the careful study of a natural experiment can illuminate key public health challenges.

Fortunately, research using natural experiments to illuminate public health issues continues today. Economists Resul Cesar, Joseph J. Sabia, and W. David Bradford recently analyzed a natural experiment in overseas military assignments and found that veterans who endured combat were much more likely than their nonexposed colleagues to use prescription painkillers or illicit heroin.6 Their research shows that it may be no coincidence that the current epidemic of “deaths of despair”—suicide, opioid overdose, and so on—is concentrated in regions of the country that send a high proportion of their populations to war.

These researchers carry the torch of Hearst et al. by recognizing that natural experiments afford unique opportunities to understand salient public health issues. Five decades ago this December, one of the first such experiments began; not only does it continue to illuminate public health, it also continues to inspire researchers to utilize other natural experiments as a means to understand societal challenges, such as deaths of despair, that plague our communities today.


  1. Case A, Deaton A. Mortality and morbidity in the 21st century. Brookings Pap. Econ. Activity. 2017;Spring:397-443.
  2. Koller CF. “Deaths of Despair” – Prevention for a Growing Crisis. Milbank Memorial Fund President’s Blog; January 28, 2019. Accessed August 25, 2019.
  3. Hearst N, Newman TB, Hulley SB. Delayed effects of the military draft on mortality. N Engl J Med. 1986;314:620-624.
  4. Conley D, Heerwig J. The long-term effects of military conscription on mortality: estimates from the Vietnam-era draft lottery. Demography. 2012;49(3):841-855.
  5. Schmitz L, Conley D. The long-term consequences of Vietnam-era conscription and genotype on smoking behavior and health. NBER Working Paper. 2015; No. 21348. Accessed August 26, 2019.
  6. Resul C, Sabia JJ, Bradford WB. Did the war on terror ignite an opioid epidemic? NBER Working Paper. 2019; No. 26264. Accessed September 19, 2019.

Published in 2019
DOI: 10.1111/1468-0009.12423

About the Authors

Dalton Conley is the Henry Putnam University Professor in Sociology at Princeton University and a faculty affiliate at the Office of Population Research and the Center for Health and Wellbeing. He is also a research associate at the National Bureau of Economic Research (NBER), and in a pro bono capacity he serves as dean of health sciences for the University of the People, a tuition-free, accredited, online college committed to expanding access to higher education. He earned an MPA in public policy (1992) and a PhD in sociology (1996) from Columbia University, and a PhD in Biology from New York University in 2014. He has been the recipient of Guggenheim, Robert Wood Johnson Foundation and Russell Sage Foundation fellowships as well as a CAREER Award and the Alan T. Waterman Award from the National Science Foundation. He is an elected fellow of the American Academy of Arts and Sciences and an elected member of the National Academy of Sciences.

See Full Bio

Tim Johnson is an associate professor of Public Management and Public Policy at Willamette University’s Atkinson Graduate School of Management. Before joining Willamette University’s faculty, Johnson completed his PhD studies at Stanford University and served as a predoctoral fellow at the Center for Adaptive Behavior and Cognition at the Max Planck Institute for Human Development in Berlin, Germany.

See Full Bio