“Every day, approximately 22 American veterans commit suicide, totaling over 8,000 veteran suicides each year — I repeat, 8,000 veteran suicides each year.”
–Sen. John McCain (R-Ariz.), Senate hearing, Feb. 3, 2015
Both chambers unanimously passed the Clay Hunt Suicide Prevention for American Veterans Act, sending the bill to the president for his signature. The bill aims to improve mental-health and suicide-prevention services at the Department of Veterans Affairs. It is named after a former Marine sniper who committed suicide in March 2011 after struggling to receive mental-health care at the Houston Veterans Affairs medical center.
The statistic that there are 22 veteran suicides each day — or, more than 8,000 when multiplied by the number of days in a year — is a widely cited figure in reference to veteran suicides. It’s been used by Democratic and Republican lawmakers in both chambers, the VA, veteran groups and media outlets (including, in full disclosure, the author of this fact check).
Where does this figure come from, and what does it tell us about suicides among veterans? It is important to remember that suicide is already the tenth leading cause of death among Americans, so the question is whether the rate among veterans is significantly higher.
This statistic comes from the VA’s 2012 Suicide Data Report, which analyzed death certificates from 21 states, from 1999 to 2011. The report calculated a percentage of suicides identified with veterans out of all suicides in death certificates from the 21 states during the project period, which turned out to be 22 percent. (By point of reference, about 13 percent of U.S. adults are veterans, according to a 2012 Gallup poll.) Then the report applied that percentage against the number of suicides in the U.S. in a given year (approximately 38,000). Divided by number of days in a year, the report came up with 22 veteran suicides a day.
Researchers said they had reached a point of statistical significance because there was enough variety in the states studied (by region, population, veteran population, gun ownership, etc.) and the percentage of veteran suicides remained consistent. However, they acknowledged “significant limitations” in their available data, including people incorrectly identified as veterans in death certificates.
They cautioned against the use of the 22-deaths figure more than once in the study: “It is recommended that the estimated number of veterans be interpreted with caution due to the use of data from a sample of states and existing evidence of uncertainty in veteran identifiers on U.S. death certificates.”
To account for uncertainties, researchers gave a range of 18 to 22 veteran suicides a day, which is consistent with previous VA estimates using CDC data. The report does not include some states with the largest veteran population (including California, Texas, Georgia, Arizona and North Carolina), so it is unclear how this would affect the rate.
This was the first time the VA used death certificates from states to study the veteran population beyond those who receive services through the Veterans Health Administration. An updated report with data from at least 44 states is scheduled to be published this summer. (The VA is also working with the Defense Department and CDC for a study of 40 million veterans’ mortality records.) There is little reliable data on veteran suicides, so it makes sense that this figure is widely cited. However, it does not provide much context about veteran suicides.
While the figure is often cited in connection with recent wars in Iraq and Afghanistan (it also was used by Iraq and Afghanistan Veterans of America, the group that pushed for the Clay Hunt act), the study was more telling of suicides among older veterans. The average age of male veteran suicides was 59.6 years old — older than non-veteran male suicides. (The Gallup poll shows the percentage of veterans climbs dramatically as the population gets older, making up a majority of those older than 65.) “It is therefore possible that epidemiologic characteristic of suicide in the general population (i.e. higher rates of suicide among older adult males) may contribute to a comparatively high prevalence of Veterans among those who die from suicide,” the VA report noted.
In addition, the 22-a-day figure does not take into consideration the spike in suicides among veterans between 18 and 24 years old, which was published in a January 2014 update.
Robert Bossarte, director of the VA Office of Public Health’s epidemiology program who co-authored the report, said he is concerned the focus on the 22-a-day number is taking away from other findings that can help shape public policy.
“It’s not that I oppose the number 22 but it’s frequently not contextualized. So we focus on that number rather than what’s really happening within that number,” Bossarte said.
For example: Demographics of veterans who died from suicide were the same regardless of whether or not they received care through the Veterans Health Administration. More people were calling to the Veterans Crisis Line, the majority of them 50-59 years old. There are consistently higher rates of male and female veteran suicides compared to non-veteran suicides, but the difference is greater among males. Suicide among Americans in general has increased in recent years and veteran suicide trends mirror that, though risk factors vary.
A new study funded by the Army shows the suicide rate for veterans who served in recent wars is much lower than 22 a day. The study, published in the February 2015 Annals of Epidemiology, is the first large population-based study of post-service suicide risk among this population. Researchers used veteran records from two Defense Department databases, verified Social Security information and used the CDC’s National Death Index Plus. They studied 1.3 million veterans who were discharged between 2001 and 2007. Among deployed veterans in this report, 32.6 percent were born in 1978-1981 and 30 percent were born in 1982-1990.
Between 2001 and 2009, there were 1650 deployed veterans and 7703 non-deployed veteran deaths. Of those, 351 were suicides among deployed veterans and 1517 were suicides among non-deployed veterans. That means over nine years, there was not quite one veteran suicide a day.
This is not to say, however, that suicide is not a concern among that population. One in two veterans of Iraq and Afghanistan wars say they know a fellow service member who attempted or committed suicide, according to a Washington Post-Kaiser Health poll.
“This statistic points to an alarming and critical issue among veterans, and Sen. Isakson and the Senate VA Committee need better data from the VA. I am glad they are releasing an updated study that includes most of the states’ information; however, since 2012, we have repeatedly been given this figure by VA as the most current and accurate number, supposedly based on their own data,” said Amanda Maddox, Isakson’s spokeswoman.
The Pinocchio Test
Search this “22 veterans a day” number on Google and you will find endless use of this figure. What you will not find as easily, however, is an explanation of what it really means, or the various caveats that accompany it.
Suicide is a serious issue, and a difficult one to tackle regardless of the population. When it comes to veteran suicides, there is little reliable and comprehensive data.
The actual number of veteran suicides a day might be higher than 22 for a given population of veterans facing certain risk factors, and lower for another group. The repeated use of this number has been magnified by the lack of comprehensive research, but that does not make it acceptable to repeat an alarming figure with no context or caveats — especially one that researchers cautioned against repeatedly in the study.
The more important issue is whether the rate of suicides among veterans is higher than among the general population–and if so, by how much. That would be a better statistic to use than a raw number with little context or meaning.
(To contact the VA’s Crisis Line: Call 1-800-273-8255 and push “1” for Veteran services. Veterans Chat can be accessed at www.VeteransCrisisLine.net. Veterans Text is available at 838255.)
Send us facts to check by filling out this form