Five Questions With: Arizona’s EMS Suicide Investigators
We have to understand it before we can solve it, and to a large extent we’re still getting our arms around the issue of suicide in the public safety professions. Helping define the problem is a new study from investigators in Arizona, “Death by Suicide: The EMS Profession Compared to the General Public,” published in Prehospital Emergency Care.
The goal of this study was to determine the statewide mortality odds ratio (MOR) of suicide completion among EMTs compared to non-EMTs in Arizona. Its authors analyzed seven years’ worth of adult deaths from the state’s death registry, then manually reviewed everyone’s occupations to identify EMTs. Their discomfiting results: After all statistical adjustments, EMTs’ suicide MOR was “significantly higher” than that of non-EMTs [aMOR: 1.39; 95% CI (1.06–1.82)].
Two of the piece’s authors, University of Arizona med student Neil Vigil and Ben Bobrow, MD, FACEP, a distinguished professor of emergency medicine at UA and medical director for the state EMS bureau, elaborate in the latest Five Questions With.
EMS World: There’s been a lot of recent concern about suicide risk among emergency providers. Why did your team undertake this study?
Vigil/Bobrow: In 2014 a large North American survey was published by NREMT suggesting EMTs have a tenfold higher incidence of both suicidal ideation and suicide attempts. Around the same time we also experienced several EMT suicide deaths within our Arizona EMS community. While we were aware of the significant rise in U.S. suicide rates in the general public, we know survey data can be strongly biased. We started by asking the basic question, “What is the suicide rate of EMTs in Arizona?”
As our research team at the University of Arizona Emergency Medicine Research Center began to try to answer this question, we realized that while simple, it was not an easy question to answer because there is no sole-source EMT suicide database. So we attempted to answer the question utilizing a combination of Arizona death records data (searching for occupation fields in the death registry that corresponded to EMT). It was extremely laborious and imperfect, but it was a way we could at least estimate the true incidence of this problem.
We found EMTs in Arizona have a 39% relative increase in their odds of dying by suicide compared to the state’s general population (whose average is significantly higher than the average for the country). We did not calculate a rate but instead an odds of death by this specific cause.
That suicide rates are higher among emergency providers has long been suspected. What’s some of the previous data or existing literature that’s suggested this?
Surveys suggest a tenfold higher incidence of suicidal thoughts and attempts among EMTs, which is different than suicide completions. The two are different, but one of the strongest predictors of suicide completion is suicide attempts, so they are likely related. Remarkably, there is precious little data on actual suicides among EMTs. We hope our paper helps highlight this fact and the desperate need for better data.
Can you describe that process of identifying EMTs in the death data? Do you feel it captured everyone (EMT-trained firefighters, volunteers, etc.)?
We had to use a case-by-case search process of the death registry, which was very time-consuming. We looked for search terms we believed represented occupations where the individual was most likely an EMT, such as paramedic, firefighter, flight medic, etc.
We know there’s a strong bias against categorizing unclear deaths as suicides, and thus we believe we most likely underestimated how many EMT suicides there are.
Were you surprised by what you found, or the strength of the association? How would you characterize your team’s reaction to the results?
We were surprised at how many actual EMT suicides we found every year in Arizona, which was on average 9 deaths. Each one is a tragedy. They made this study very personal for us and motivated us to publish this paper and to work toward improving this horrible situation.
While you didn’t investigate causes or contributing factors to EMT suicide, were you able to develop any sense of any of those? Anything we can infer from the data?
As you point out, our study did not address the cause(s) of the problem; as a starting point we simply attempted quantify it. We go by the adage, “If you can’t measure it, you can’t improve it.” So this is a very preliminary paper we hope will spur more much-needed work to identify the cause(s) and try to figure out effective interventions.
Our hypothesis is that the causes are multifactorial and likely related to things like repetitive work stress related to EMS, shift work (working at night), a culture of not wanting to ask for help, repeat exposure to seeing suicide and other tragedies, and other factors. Our sincere hope is to help highlight this major problem and catalyze interventions to dramatically decrease EMT suicide.