Linking Public Safety And Public Health Data For Firearm Suicide Prevention In Utah

In 2013–17 a total of 220,306 Americans died by suicide, and half of them used a firearm.¹ Suicide rates vary widely across US states, ranging in 2017 from lows of 8–9 per 100,000 people in Massachusetts, New Jersey, and New York to highs of 20–26 per 100,000 in Alaska, Idaho, Montana, New Mexico, and Wyoming. In 2017 Utah had the eighth-highest suicide rate in the nation, and the rate had increased since 2000 at a faster clip (from 13.3 to 21.4 per 100,000) than had the national rate (from 10.4 to 14.5 per 100,000). Of particular concern to Utah policy makers, suicide is the leading cause of death in that state among residents ages 15–24 and the fourth-leading cause among those younger than age 65.²

In H.B. 440 in 2016, the Utah State Legislature directed the Utah Division of Substance Abuse and Mental Health to conduct a study of firearms and suicide in the state to improve data on suicide.³ A spirit of bipartisan support and collaboration among multiple state agencies, cultivated during the then-recent formation of Utah’s Firearm Safety Workgroup, created the opportunity for the study to link sensitive public health and public safety databases. The legislature specified that the study track patterns of suicide in the state and identify opportunities for prevention by documenting how decedents obtained guns used in suicide, whether they could possess guns legally, and whether they had a history of mental illness. To obtain additional information on gun ownership rates and storage habits, the state also added three questions to the survey instrument used in 2017 for its telephone-based survey, the Behavioral Risk Factor Surveillance System.

This article describes the Utah experience as a case study for how policy makers, researchers, firearm stakeholders, and others can reach across ideological lines to study their local firearm suicide problem and identify actionable prevention strategies. We describe how legislation that called for the study gained bipartisan support, briefly describe how the study was carried out, provide examples of policy-relevant findings and policy implications derived from those findings, and discuss how state leaders and others are applying the insights gained. In doing so, we hope to spur policy makers and suicide prevention leaders in other states to study their own firearm suicide problems and involve people with a variety of perspectives—including those who value guns—in formulating the study questions and interpreting the findings. For context, we begin by briefly describing why firearms are an important focus for suicide prevention efforts.

The Role Of Firearms In Suicide

In Utah, as in the US overall, most firearm deaths are suicides (85 percent in Utah and 60 percent in the US in 2017), and firearms are the leading method of suicide.¹ Suicide has multiple determinants, and a host of interventions seek to reduce the risk that someone will attempt suicide. Reducing access to firearms at times of risk is intended primarily to reduce the odds that an attempt, should it occur, will prove fatal. This has been a growing public policy and clinical priority, with organizations as diverse as the American Foundation for Suicide Prevention and the National Shooting Sports Foundation calling for household guns to be stored in a way that makes them inaccessible to people at risk of suicide.³

The rationale underlying these calls is based on four well-established observations. First, among commonly used methods of suicide in the US, firearms are by far the most lethal.^4,5 Second, the interval between thinking about attempting suicide and making the attempt is often quite brief—ten minutes or less in about half of suicide attempts, according to one study.⁶ Moreover, predicting when someone, even a person at very high risk, will make a suicide attempt is nearly impossible.⁷ Third, the method people use in suicidal acts often depends in part on the method’s ready availability.^8,9 Fourth, if a person survives a suicide attempt, the prognosis for continued survival is good. Nine out of ten people who make a suicide attempt—even those making medically serious attempts—do not go on to die by suicide.¹⁰

Cultivating Broad Support For Firearm Suicide Prevention

In 2014, in response to the growing toll of firearm suicide in Utah, a state legislator, Rep. Steve Eliason (R-45), contacted the director of Means Matter (Catherine Barber, one of the authors of this article) to discuss prevention strategies. Means Matter is a project based at Harvard University that disseminates research and training on reducing a suicidal person’s access to highly lethal suicide methods. Representative Eliason was heartened to hear about an approach that Means Matter and the New Hampshire Firearm Safety Coalition had applied to firearm suicide.³ That approach brings opinion leaders in the gun-owning community together with public health and suicide prevention professionals to study the local suicide problem and develop largely nonlegislative interventions, such as a “friends don’t let friends drive drunk” approach to reducing the use of guns in suicide attempts. As a result, he filed H.B. 134, legislation (which passed in 2014 with strong support) that directed the Utah Division of Substance Abuse and Mental Health to create materials that encourage gun owners to routinely store their guns locked and to take further steps (such as storing guns away from home) if someone in the home (the gun owner or someone else) is at risk for suicide.

The division’s suicide prevention coordinator (Kim Myers, one of the authors) chaired the Utah Suicide Prevention Coalition and established Utah’s Firearm Safety Workgroup to guide the development of these materials. The members of the workgroup represent a broad base, including the Utah Departments of Public Safety and Human Services; the state’s largest gun lobbying group (the Utah Shooting Sports Council, or USSC); health care organizations; the Office of the Medical Examiner; firearm instructors and retailers; suicide researchers; and suicide survivors. Barber proposed at a workgroup meeting in 2015 that the state adopt a brief, voluntary suicide prevention module that firearm instructors could opt to include in the classes that people seeking a Utah concealed firearm permit must take. Clark Aposhian, a committee member and chair of the USSC, frowned while reading through the proportion of firearm deaths that are suicides in the state (85 percent), shook his head, and looked up. “If the data looked any different, I’d agree,” he said. “But given the numbers, I think this needs to be mandatory.” Means Matter drafted the slide set for the module, committee members edited it, it was pilot-tested with a small group of firearm instructors, and a draft was shared with over a thousand instructors in an online survey for comment. Two-thirds of the instructors said that they would like to teach it. In 2016 the Department of Public Safety released the new curriculum.

The process of creating prevention materials together led members of the Firearm Safety Workgroup, including USSC leaders, to gain a greater, shared commitment to the role that gun owners can play in reframing suicide awareness as a basic tenet of firearm safety. Aposhian convinced the council’s board that progun leaders should become more knowledgeable about firearm suicide and take a lead in preventing these deaths. As he frequently stated in workgroup meetings, “We can’t be afraid of the data.”

Passing Legislation To Fund A Study Of Firearms And Suicide

In 2016 another state legislator, Rep. Brian King (D–28), sponsored new language in H.B. 440 that directed the Utah Division of Substance Abuse and Mental Health to conduct a study on firearms and suicide in Utah. Historically, a Democratic call for a firearm study would be unlikely to pass the legislature, where Republicans had a supermajority. However, Representative Eliason and the Firearm Safety Workgroup had laid the groundwork for bipartisan support. The message in the workgroup’s existing materials was not that Utahns should not have guns, but rather that Utahns with guns should consider storing them away from home or otherwise inaccessibly if the gun owner or another household member is at risk for suicide. This reframing turned what was previously a political lost cause (focusing on firearm suicide in a gun-friendly state) into a topic on which common ground could be found. Earlier in 2016 Republican House Speaker Greg Hughes (R-51) urged legislators to use data to drive policy. Representative King pointed to the need for data to better understand firearm suicide and coordinated with the state’s lead firearm lobbyist to support the bill in committee hearings. H.B. 440 passed the Utah House on March 7, 2016, by a nearly unanimous vote (there was one nay vote and one abstention), and it passed the Senate by a two-thirds majority shortly thereafter.

Study Methods, Data Sources, And Linkage

The Utah Division of Substance Abuse and Mental Health contracted with the Harvard Injury Control Research Center (the study team at the center consisted of Barber, John Berrigan, David Hemenway, Deborah Azrael, and Matthew Miller, authors still affiliated with the center; and Morissa Henn, an author who was then a doctoral student there) to conduct the study. As noted above, H.B. 440 specified that the study determine whether firearm suicide decedents were legally entitled to possess a gun, how they obtained their guns, and whether they had a history of mental illness—information that is available only by linking sources of public health and public safety data. The psychological autopsy examiner at the Office of the Medical Examiner (Michael Staley, an author) served as the conduit for the disparate data sources and played a key role in preparing the necessary interagency data-sharing agreements. The office sent lists of people who had died by suicide in Utah in 2014 or 2015 to the state’s Departments of Public Safety and Human Services, which in turn assembled data on decedents’ history of hospital inpatient and emergency department (ED) visits stretching back to 2011, whether they could have passed a background check for legal firearm possession on their date of death, and whether they had ever had a concealed carry permit. (An additional set of data elements based on federal Bureau of Alcohol, Tobacco, Firearms and Explosives records is still being assembled at the time of this writing, to learn who purchased the guns used in the suicides and how long before the suicide.) Online appendix exhibit A summarizes the data linked in the study.¹¹ Details on the data sources are available in the report Suicide and Firearm Injury in Utah.¹²

All linkages using personally identifiable data were done in Utah by the agencies that held the data, and data were deidentified (with the exception of National Violent Death Reporting System [NVDRS] case identification numbers), before the Office of the Medical Examiner transferred the information to the Harvard study team. The NVDRS is a state-based reporting system funded and coordinated by the Centers for Disease Control and Prevention. It joins detailed data from death certificates, police reports, and coroner’s or medical examiner’s reports on suicides and other violent deaths.¹³ The study team linked information from the various data sets using NVDRS identification numbers. Data from that system formed the backbone of the linked data set.

To provide a full census of fatal and nonfatal events, the Utah Department of Health also provided the study team with a set of deidentified data about hospital visits by patients who were discharged alive from ED or inpatient care at a Utah hospital in 2014 or 2015 following a suicide attempt or act of deliberate self-harm. Details on the hospital data source have been presented elsewhere.¹⁴ The hospital and suicide data were used to calculate population-based incidence rates for fatal and nonfatal events and case fatality rates (the percentages of self-harm acts that were fatal).

The study was approved by Institutional Review Boards at the Harvard T. H. Chan School of Public Health and the Utah Department of Health.

Examples Of Policy- And Program-Relevant Findings

Findings from the linked data were presented to the legislature in late 2018 in a report titled Suicide and Firearm Injury in Utah.¹² Here we highlight examples of findings that are instructive because of their policy relevance and because they spurred state actions in the public or private sectors.

Background Checks For Legal Firearm Possession

Data Finding:

Ninety-two percent of firearm suicide decedents could have passed a background check on the day they died. In other words, only 8 percent of those using a firearm, compared with 19 percent of those using other methods, were prohibited possessors.

Policy Implications:

The extent to which universal background check legislation could help reduce suicide rates would necessarily be modest, given that nine out of ten decedents could have passed the check. Friends, family members, and others therefore play an important role in urging those at risk to store their guns away from home or otherwise inaccessibly until the situation improves.

State Actions:

In 2019 the legislature established a dollar-for-dollar public-private matching fund of up to $2 million aimed at preventing firearm suicide. The money will be used for a campaign to promote social norms related to routinely storing firearms locked and taking further steps (such as storing them away from home) when a household member is struggling with a mental health or substance abuse problem or painful life crisis. Intermountain Healthcare, Utah’s largest provider of inpatient and outpatient services, announced in early 2019 that it is committing the lead gift toward the $2 million fund. The Church of Jesus Christ of Latter-day Saints (the state’s largest religious group) and the Utah Shooting Sports Council have also committed major gifts.

Concealed Firearm Permit Status

Data Findings:

Twenty-three percent of male firearm decedents had a concealed firearm permit. Among all male suicide decedents who had such a permit, 84 percent used a gun in their suicide.

Policy Implications:

The concealed firearm permit process provides an avenue to reach Utah gun owners with education about firearms and suicide. As of 2017 the course that Utah permit seekers must take briefly covers suicide prevention. In the 2016 survey of Utah concealed carry permit instructors, several commented that people in their classes were not the sort who would take their own lives. Learning that a quarter of male firearm decedents were permit holders, and that such people who did take their own lives almost always used a gun, could strengthen instructors’ resolve in covering the material.

State Actions:

The state passed legislation in 2019 that provided resources for relevant state agencies to supply brochures to concealed carry permit applicants about the role of safe firearm storage in suicide prevention, offer discounts on gun safes for permit applicants, and provide a short online video that people renewing their permits must view.

History Of Hospital Care

Data Findings:

Twenty-five percent of firearm suicide decedents were seen in the hospital with a behavioral health diagnosis in the year before their death, but only 6 percent were seen for a suicide attempt or self-harm. Most of the previous suicide attempts were with drugs or sharp instruments.

Policy Implications:

Hospitals are a useful venue for suicide prevention. However, focusing only on patients who are treated for a suicide attempt would miss 94 percent of firearm suicides. Focusing on people who visit the hospital with a behavioral health issue could reach a quarter of would-be suicides. One message for clinicians to convey to patients is the potential safety advantage of storing guns away from home or otherwise inaccessibly to the patient until they recover.

State Actions:

The state is working to ensure that its mental health workforce has training in lethal means counseling. Intermountain Healthcare collaborated with Means Matter and the developers of Counseling on Access to Lethal Means (CALM) to create CALM-Utah, an online course available free of charge to any Utah clinician. The course was launched in early 2019 and encourages clinicians to discuss making household guns and dangerous quantities of household medications inaccessible not only to patients who have disclosed suicidal thoughts, but also to patients further upstream—those in serious mental distress.

Role Of Arguments And Youth Gun Suicide

Data Findings:

According to the NVDRS data, 52 percent of people who took their life with a gun were reportedly going through relationship problems such as divorce or family conflict. Arguments were noted as playing a role in a quarter of all suicides, across all methods. In addition, one in three argument-related gun suicides took place during the argument, which was often with an intimate partner. Suicides by other methods rarely occurred in the midst of an argument.

Among rural youth younger than age twenty-one, 62 percent of firearm suicides were by rifle or shotgun—often the youth’s own gun.

Policy Implications:

Professionals such as social workers, clergy, and divorce attorneys could advise clients and parishioners who are experiencing relationship conflicts to store firearms away from home or inaccessibly during crisis periods. Regarding youth suicide, urging parents to lock their guns would inadequately address the problem if their teens know where the keys are (or what the combination is) or if their teens are the gun owners and control gun storage. A more useful message to parents might be to lock all guns and ensure that youth have no unsupervised access to the keys (or do not know the combination).

State Actions:

LDS Family Services—a service arm of the Church of Jesus Christ of Latter-day Saints—has encouraged its staff members to complete the CALM-Utah training. In response to the finding on youth suicide, the state passed legislation in 2019 that directed retailers to distribute a locking device (supplied free by the state) with all sales of rifles and shotguns. Federal law already requires dealers to supply locks for handguns.

Fatal And Nonfatal Events: Drivers Of Higher Suicide Rates In Rural Counties

Data Findings:

According to NVDRS data, the suicide rate was higher in Utah’s rural counties than its metropolitan counties (23.3 versus 18.9 per 100,000 people). However, the suicide attempt rate (based on hospital data) was not higher in rural areas, nor was the suicide rate by nongun methods (see appendix exhibit B).¹¹ Statewide, the case fatality rate was higher for guns (87 percent) than for all other methods (1–44 percent). Method-specific case fatality rates were similar in rural and metropolitan counties. Suicidal behavior was not more prevalent in rural areas, nor were method-specific attempts more deadly there. What drove the higher rural suicide rate was the higher proportion of attempts involving a firearm. The responses to Utah’s 2017 Behavioral Risk Factor Surveillance System Survey indicated that gun ownership is higher in Utah’s rural counties and that more rural owners store a gun loaded and unlocked than do owners in metropolitan areas.

Policy Implications:

Given the high case fatality rate for firearms, if a proportion of Utahns who would otherwise attempt suicide with a firearm were prevented from using a gun, there would likely be fewer suicide deaths—even if the person substituted another method.¹⁵ Gun owner venues such as gun shops and shooting clubs could serve as messengers that storing firearms away from home or inaccessibly when a household member is struggling with a behavioral health problem or serious life crisis is a basic tenet of firearm safety.

State Actions:

The state’s mental health agency has provided grants to seven rural communities to support outreach to gun owners on suicide prevention and safe firearm storage.

A Swift Policy Response

Within only six months of the formal release of Suicide and Firearm Injury in Utah¹² in late 2018, stakeholders had responded to the study findings and the policy opportunities identified. The legislature had established the $2 million matching fund, Intermountain Healthcare and LDS Family Services had announced new initiatives, and the state had developed new mechanisms to educate gun owners about suicide prevention and to distribute locking devices free of charge or at reduced prices.

What explains the speedy response? The study team apprised the study’s key legislative sponsors of their preliminary findings in late summer 2018—before the report’s formal release. This enabled legislators to consider legislation for the 2019 session, which began in late January. Utah’s forty-five-day annual legislative session is one of the shortest in the nation, which ensures quick turnaround on legislative proposals.

In addition, in recent years leaders in the Utah legislature have encouraged data-driven policy making. Since 2011 the state’s suicide prevention efforts have routinely included policy solutions as part of a comprehensive approach. With a psychological autopsy examiner at the medical examiner’s office, suicide prevention coordinators in key state agencies, and committed and competent staff members in charge of relevant public safety and public health databases, the state had the internal infrastructure to bring data to bear on the suicide problem. Utah’s suicide prevention plan was revamped in 2017 and now highlights reducing access to highly lethal means of suicide as one of three core priorities. In addition, the Utah Health Improvement Plan identified mental health promotion and suicide prevention—including reducing access to lethal means—as one of three statewide health priorities. Together, these initiatives and infrastructure provided a foundation for a swift response from the public sector.

With respect to the private sector, the report to the legislature was released as Intermountain Healthcare was finalizing a suicide prevention strategy for its system of 170 clinics and 23 hospitals. The recently named director of community health in charge of Intermountain’s “Zero Suicide” initiative (Henn) ensured that data from the report were woven into the discussion of and planning for the initiative at relevant meetings of clinical and administrative leaders. The findings helped persuade them to focus on lethal means reduction. What had previously been seen as politically untenable in a gun-friendly state was now seen as a crucial area of health care engagement. Firearm suicide prevention became a centerpiece of Intermountain’s clinical and community health approaches.¹⁶ The report also provided a rationale to engage in collaborative conversations with other major Utah health care providers, including University of Utah Health and LDS Family Services.

Conclusion

The data provided epidemiologic evidence to suggest that firearm access is one of the drivers of geographic variation in suicide rates.

The collaboration reported on here—in which firearm stakeholders, state agency representatives, legislators, and suicide researchers came together to find common ground on preventing suicide—provides an example for other states seeking to tackle firearm suicide. The linked data helped establish realistic parameters for the proportion of suicide decedents whom interventions such as background checks and hospital-based counseling could reach. The data also provided epidemiologic evidence to suggest that firearm access is one of the drivers of geographic variation in suicide rates. And they delivered insights into how to tailor safe firearm storage messaging and highlighted novel venues—such as the concealed firearm permitting process—for reaching gun owners with locking devices and gun owner–friendly suicide prevention materials.

All states in the US now participate in the National Violent Death Reporting System and gather rich data on suicide. Researchers cannot use the deidentified national data set to link to other data sources. However, individual state NVDRS offices can facilitate these linkages using state-specific data, and it is state-specific numbers that are of the greatest utility and motivation to state leaders.^17,18

Collaboration and buy-in by Utah firearm stakeholders was vital in building support for the state-mandated study.

Engaging thoughtful opinion leaders in the gun-owning community can lead to bipartisan support for studying the problem.

Collaboration and buy-in by Utah firearm stakeholders was vital in building support for the state-mandated study, interpreting its findings, and tailoring gun owner–friendly interventions. As stated in Suicide and Firearm Injury in Utah, “Firearms are a valued part of the fabric of many homes in Utah. This report was assembled to…help stakeholders craft [suicide] prevention strategies that are responsive to the local problem and that build on the strengths and values of the communities and individuals most affected by firearm suicides: gun owners and their families.”¹² The experience in Utah indicates that engaging thoughtful opinion leaders in the gun-owning community on how to reduce firearm suicide can lead to bipartisan support for studying the problem. Linking public health and public safety data delivers actionable findings that can guide prevention efforts on the part of all stakeholders.

ACKNOWLEDGMENTS

A portion of the time that Catherine Barber, Deborah Azrael, and David Hemenway spent on this study was covered by a contract with the Utah Division of Substance Abuse and Mental Health. A portion of Matthew Miller's time was supported by a grant from the Joyce Foundation. Barber has received funding from the Suicide Prevention Resource Center and Intermountain Healthcare in the past year. The authors gratefully acknowledge the careful work of the Utah state agency personnel who collected and linked the data presented in this article.

NOTES

• 1 Centers for Disease Control and Prevention. CDC WONDER: underlying cause of death, 1999–2017 [Internet]. Atlanta (GA): CDC: [cited 2019 Aug 13]. Available from: http://wonder.cdc.gov/ucd-icd10.html Google Scholar
• 2 Centers for Disease Control and Prevention. Injury prevention and control: welcome to WISQARS^TM [Internet]. Atlanta (GA): CDC; [last reviewed 2019 Aug 6; cited 2019 Aug 13]. Available from: http://www.cdc.gov/injury/wisqars/index.html Google Scholar
• 3 Barber C, Frank E, Demicco R. Reducing suicides through partnerships between health professionals and gun owner groups—beyond docs vs Glocks. JAMA Intern Med. 2017;177(1):5–6. Crossref, Medline, Google Scholar
• 4 Spicer RS, Miller TR. Suicide acts in 8 states: incidence and case fatality rates by demographics and method. Am J Public Health. 2000;90(12):1885–91. Crossref, Medline, Google Scholar
• 5 Vyrostek SB, Annest JL, Ryan GW. Surveillance for fatal and nonfatal injuries—United States, 2001. MMWR Surveill Summ. 2004;53(7):1–57. Medline, Google Scholar
• 6 Deisenhammer EA, Ing CM, Strauss R, Kemmler G, Hinterhuber H, Weiss EM. The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? J Clin Psychiatry. 2009;70(1):19–24. Crossref, Medline, Google Scholar
• 7 Belsher BE, Smolenski DJ, Pruitt LD, Bush NE, Beech EH, Workman DEet al. Prediction models for suicide attempts and deaths: a systematic review and simulation. JAMA Psychiatry. 2019;76(6):642–51. Crossref, Medline, Google Scholar
• 8 Eddleston M, Karunaratne A, Weerakoon M, Kumarasinghe S, Rajapakshe M, Sheriff MHet al. Choice of poison for intentional self-poisoning in rural Sri Lanka. Clin Toxicol (Phila). 2006;44(3):283–6. Crossref, Medline, Google Scholar
• 9 Skopek MA, Perkins R. Deliberate exposure to motor vehicle exhaust gas: the psychosocial profile of attempted suicide. Aust N Z J Psychiatry. 1998;32(6):830–8. Crossref, Medline, Google Scholar
• 10 Carroll R, Metcalfe C, Gunnell D. Hospital presenting self-harm and risk of fatal and non-fatal repetition: systematic review and meta-analysis. PLoS One. 2014;9(2):e89944. Crossref, Medline, Google Scholar
• 11 To access the appendix, click on the Details tab of the article online.
• 12 Barber C, Azrael D, Berrigan J, Miller M, Sobelson M, Hemenway D. Suicide and firearm injury in Utah: linking data to save lives [Internet]. Boston (MA): Harvard T. H. Chan School of Public Health; 2018 Oct [cited 2019 Aug 13]. Available from: https://dsamh.utah.gov/pdf/suicide/Suicide%20and%20Firearm%20Injury%20in%20Utah%20-%20Final%20Report.pdf Google Scholar
• 13 Paulozzi LJ, Mercy J Jr, Frazier L, Annest JL. CDC’s National Violent Death Reporting System: background and methodology. Inj Prev. 2004;10(1):47–52. Crossref, Medline, Google Scholar
• 14 Hedegaard H, Schoenbaum M, Claassen C, Crosby A, Holland K, Proescholdbell S. Issues in developing a surveillance case definition for nonfatal suicide attempt and intentional self-harm using International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) coded data. Natl Health Stat Report. 2018;(108):1–19. Google Scholar
• 15 Miller M, Azrael D, Barber C. Suicide mortality in the United States: the importance of attending to method in understanding population-level disparities in the burden of suicide. Annu Rev Public Health. 2012;33:393–408. Crossref, Medline, Google Scholar
• 16 Cheney C. Health system targets gun safety to prevent suicide. Health Leaders [serial on the Internet]. 2019 Jun 14 [cited 2019 Aug 14]. Available from: https://www.healthleadersmedia.com/clinical-care/health-system-targets-gun-safety-prevent-suicide Google Scholar
• 17 Hernandez-Meier JL, Akert B, Zheng C, Guse CE, Layde PM, Hargarten S. Status of legal firearm possession and violent deaths: methods and protocol for a retrospective case-control study. Inj Prev. 2019 Jan 31. [Epub ahead of print]. Crossref, Medline, Google Scholar
• 18 Weis MA, Bradberry C, Carter LP, Ferguson J, Kozareva D. An exploration of human services system contacts prior to suicide in South Carolina: an expansion of the South Carolina Violent Death Reporting System. Inj Prev. 2006;12(Suppl 2):ii17–21. Medline, Google Scholar