The American Psychiatric Association (APA) has updated its Privacy Policy and Terms of Use, including with new information specifically addressed to individuals in the European Economic Area. As described in the Privacy Policy and Terms of Use, this website utilizes cookies, including for the purpose of offering an optimal online experience and services tailored to your preferences.

Please read the entire Privacy Policy and Terms of Use. By closing this message, browsing this website, continuing the navigation, or otherwise continuing to use the APA's websites, you confirm that you understand and accept the terms of the Privacy Policy and Terms of Use, including the utilization of cookies.

×
ArticlesFull Access

Changes in Suicide Mortality for Veterans and Nonveterans by Gender and History of VHA Service Use, 2000–2010

Abstract

Objective:

Veterans are believed to be at high risk of suicide. However, research comparing suicide rates between veterans and nonveterans is limited, and even less is known regarding differences by history of Veterans Health Administration (VHA) service use. This study directly compared veteran and nonveteran suicide risk while for the first time differentiating veterans by VHA service use.

Methods:

The cross-sectional study analyzed data from 173,969 adult suicide decedents from 23 states (2000–2010) included in the U.S. Department of Veterans Affairs suicide data archive. Annual standardized mortality ratios (SMRs) were computed for veterans compared with nonveterans and for veterans who used VHA services compared with veterans who did not, overall and separately for males and females.

Results:

After the analysis controlled for age and gender differences, the number of observed veteran suicides was approximately 20% higher than expected in 2000 (SMR=1.19, 95% confidence interval [CI]=1.10–1.28), and this increased to 60% higher by 2010 (SMR=1.63, CI=1.58–1.68). The elevated risk for female veterans (2010 SMR=5.89) was higher than that observed for male veterans (2010 SMR=1.54). Trends for non–VHA-utilizing veterans mirrored those of the veteran population as a whole, and the SMR for VHA-utilizing veterans declined. Since 2003, the number of suicides among VHA-utilizing veterans was less than expected when compared directly with the suicide rate among non–VHA-utilizing veterans.

Conclusions:

Veterans are members of the community and, as such, are an important part of observed increases in U.S. suicide rates. Not all veterans are at equal or increasing risk of suicide, however. VHA-utilizing veterans appear to have declining absolute and relative suicide rates.

In 2010, suicide was the tenth leading cause of death in the United States (1). Although preventing suicide is a complicated problem, the National Strategy for Suicide Prevention asserts that suicide is indeed preventable (2). In 2012, the National Action Alliance for Suicide Prevention made suicide a top public health priority, setting the goal to save 20,000 lives in the next five years (2). To achieve this goal, the strategy called for the development of an agenda prioritizing research on individuals at increased risk of suicide. Members of the Armed Forces and veterans were among 11 high-risk populations of interest named in this call to action.

An increase in suicide among active military personnel has been documented (3,4). Further research is needed, however, to better understand the distribution of suicide risk among veterans after active-duty service and to understand population-based differences between those with and without a history of military service, after accounting for key demographic differences (57). The demographic composition of the veteran population is changing as the proportions of younger, female, and nonwhite veterans increase (8). There are also early indicators of changes in the prevalence of service-related injuries, such as traumatic brain injury (9), and exposures, such as military sexual trauma (10), that may precipitate changes in suicide risk among post–9/11 veterans (11,12). A more complete picture of the absolute and relative risk of suicide for all veterans and within meaningful subgroups is necessary to appropriately target and evaluate treatment and prevention programs, allocate suicide prevention resources, and track changes in suicide rates over time.

A small number of studies limited to veterans who used services provided by the Veterans Health Administration (VHA) observed a 42% to 66% increase between 2000 and 2007 in relative risk of suicide compared with the general U.S. population (13,14), but these studies were unable to determine the absolute or relative mortality for veterans not using VHA services. Little is known about differences in suicide rates between veterans who use VHA services and those who do not. One study estimated a 50% increased risk among male veterans using VHA services (15). Although these findings are important, they may not be generalizable to the broader veteran population.

Thus, as an initial step toward improving our understanding of suicide risk among veterans, this study aimed to characterize veteran suicide risk compared with risk among nonveterans, characterize differences in relative suicide risk between veterans with and without a history of VHA service use, and examine the role gender plays in these differences.

Methods

Study Population

Data for this project were obtained from the U.S. Department of Veterans Affairs (VA) archive of state suicide data. Data collection is ongoing and includes suicide mortality information (ICD-10 codes X60–X84 and Y87.0) directly from state death certificate records beginning with calendar year 1999 (16). Information was analyzed for all suicides of persons age 18 years and older reported by 23 states with available, validated data for 2000–2010. Four Northeastern states (New Jersey, New York, Pennsylvania, and Rhode Island), five Midwestern states (Iowa, Kansas, Michigan, Minnesota, and Nebraska), eight Southern states (Alabama, Arkansas, Florida, Louisiana, North Carolina, Tennessee, Texas, and West Virginia), and six Western states (Alaska, Idaho, Montana, Oregon, Utah, and Washington) were included. The 2010 suicide rates for these states ranged from 8.0 to 23.1 per 100,000; ten states were at or below the national average of 12.4 per 100,000, and 13 were above it (17).

Variable Classifications

Variables for age at death, gender, and cause of death were used as reported on the death certificate. Veteran status was validated through an extensive linking procedure using VA and Department of Defense data (16). Veteran status of all suicide decedents is validated by using this process, regardless of an indication of military history on the death certificate. “VHA-utilizing veterans” were defined as validated veterans with any record of VHA service use in the fiscal year prior to or of death. Service utilization included any inpatient, outpatient, extended, or fee-basis care recorded in the National Patient Care Database. Receipt of pharmacy benefits only was not considered service utilization. [Details of the suicide decedent validation process are illustrated in a figure included in an online supplement to this report.]

Statistical Analysis

Age and gender distributions were compared over time for four groups: all veterans, VHA-utilizing veterans, non–VHA-utilizing veterans, and nonveterans. Age at death was categorized to capture meaningful age-related differences in suicide rates and to permit direct comparison to population estimates (18–29, 30–39, 40–49, 50–59, 60–69, 70–79, and ≥80). Annual crude suicide rates were computed overall for each study group as the number of suicide deaths divided by the corresponding population at risk and reported per 100,000 lives. Direct age-adjusted suicide rates (18) were also computed for veterans and nonveterans overall and by gender but not by VHA utilization history because of sample size limitations. Estimates for the population at risk in each group within age- and gender-defined strata were computed on the basis of data for the 23 states of interest from the American Community Survey (ACS), National Center for Veterans Analysis and Statistics (VetPop), and VHA Support Services Center. For each calendar year, the total veteran population was determined by using VetPop estimates, and the nonveteran population was computed by subtracting VetPop estimates from ACS figures. The population of VHA-utilizing veterans was estimated from VHA support service figures, and the non–VHA-utilizing veteran population was computed by subtracting VHA figures from VetPop estimates. VetPop figures are reported by fiscal year (October 1–September 30) but were assumed to closely estimate the calendar year veteran population.

To compare suicide-specific mortality between veterans and nonveterans, annual standardized mortality ratios (SMRs) were computed for all veterans and by VHA utilization history. Overall age- and gender-standardized SMRs were computed as well as gender-stratified, age-standardized SMRs. To evaluate precision, 95% likelihood ratio-based confidence intervals were computed (19). SMRs were computed as the ratio of observed to expected suicides. The expected number of suicides for a given year and for a veteran subgroup were computed by multiplying the population at risk by the corresponding calendar year-, age-, and gender-specific nonveteran suicide rates. To enhance comparability of findings for VHA-utilizing and non–VHA-utilizing veterans, overall and gender-stratified SMRs for VHA-utilizing veterans were also computed by using non–VHA-utilizing veteran suicide rates as the standard. A sensitivity analysis computing gender-specific direct age-adjusted rate ratios was conducted for veterans compared with nonveterans. All procedures were approved by the Syracuse (New York) VA Medical Center Institutional Review Board.

Results

Of the 173,969 suicide decedents included in the data, 25% were veterans; 20% did not have a history of VHA service use, and 5% were VHA-utilizing veterans. Veteran suicide decedents were older than nonveteran suicide decedents, and a larger proportion was male. From 2000 to 2010, the proportion of veteran suicide decedents with a recent history of VHA service use increased marginally (17% to 19%), and the proportion of all veterans using VHA services nearly doubled (from 14% to 25%) (Table 1). Due primarily to an influx of younger veterans and an increase in VHA service use among veterans over age 60, the number of VHA-utilizing veterans increased by approximately one million during the study period despite a decrease in the overall veteran population of nearly two million.

TABLE 1. Changes in the distribution (%) of demographic characteristics among persons in 23 states, by veteran status and use of Veterans Health Administration (VHA) services, 2000–2010a

CharacteristicNonveteransAll veteransVHA-utilizing veteransNon–VHA-utilizing veterans
20002010200020102000201020002010
Age
 18–2924.423.64.04.62.84.64.24.6
 30–3921.717.810.28.26.35.810.89.0
 40–4921.118.814.213.913.59.414.315.4
 50–5913.718.023.917.122.216.824.217.3
 60–698.611.419.425.718.429.919.524.3
 70–796.36.320.717.025.617.719.816.8
 ≥804.24.17.713.411.215.97.212.6
Gender
 Male41.543.894.192.190.591.394.792.4
 Female58.556.25.97.99.58.75.37.6
Use of VHA services14.125.2
Population total92,360,633107,236,97513,278,66911,299,1331,877,4072,844,52911,401,2628,454,604

aPercentages represent the population of veterans and nonveterans; thus the population total indicates the total population at risk of suicide in a given group and year in 23 states (Alabama, Alaska, Arkansas, Florida, Idaho, Iowa, Kansas, Louisiana, Michigan, Minnesota, Montana, Nebraska, New Jersey, New York, North Carolina, Oregon, Pennsylvania, Rhode Island, Tennessee, Texas, Utah, Washington, and West Virginia).

TABLE 1. Changes in the distribution (%) of demographic characteristics among persons in 23 states, by veteran status and use of Veterans Health Administration (VHA) services, 2000–2010a

Enlarge table

From 2000 to 2010, both the crude and age-adjusted veteran suicide rates increased by approximately 25% while the comparable nonveteran rates increased by approximately 12% (Table 2 and Figure 1). Of note, a 13% increase was observed for nonveteran females (4.8 to 5.4 per 100,000 lives) while a 40% increase was observed for veteran females (24.7 to 34.6 per 100,000). Although VHA-utilizing veterans experienced a higher suicide rate in 2000 than non–VHA-utilizing veterans (34.5 versus 27.6 per 100,000), the suicide rate among VHA-utilizing veterans declined over the study period, and by 2010, the rate among VHA-utilizing veterans was lower than the rate among non–VHA-utilizing veterans (27.6 versus 38.7 per 100,000).

TABLE 2. Crude suicide rates (deaths per 100,000 lives at risk) among veterans and nonveterans and by use of Veterans Health Administration (VHA) services, 2000–2010

YearVeterans
Nonveterans
TotalMaleFemaleTotalMaleFemaleVHA-utilizingNon–VHA utilizing
200011.120.14.828.628.824.734.527.6
200111.420.54.929.429.626.432.428.8
200211.821.05.130.130.524.332.129.6
200311.520.45.031.331.429.329.231.8
200411.820.55.430.831.126.630.131.0
200511.820.45.331.832.028.529.932.3
200611.920.95.231.531.926.929.832.0
200712.221.05.533.133.230.728.834.3
200812.722.05.633.934.230.430.934.8
200912.321.45.335.736.031.730.137.4
201012.421.45.435.936.034.627.638.7

TABLE 2. Crude suicide rates (deaths per 100,000 lives at risk) among veterans and nonveterans and by use of Veterans Health Administration (VHA) services, 2000–2010

Enlarge table
FIGURE 1.

FIGURE 1. Direct age-adjusted suicide rates (per 100,000 lives) among adult veterans and nonveterans, overall and by gender, 2000–2010

After accounting for age and gender differences using SMRs, the number of observed suicides among veterans was significantly higher than expected had suicide rates been the same as those observed for nonveterans. Specifically, in 2000 the number of observed veteran suicides was approximately 20% higher than expected, and the number increased to over 60% higher than expected in 2010. Findings for males followed the same pattern. Observed suicides among female veterans, however, were 390% higher than expected in 2000, increasing to nearly 490% higher than expected in 2010 (Figure 2).

FIGURE 2.

FIGURE 2. Standardized veteran suicide mortality ratios, 2000 –2010a

a All lines in panel 1 and the total line in panel 2 are age and gender standardized. All other SMRs are gender stratified and age standardized. An SMR of 1 indicates that the number of observed deaths equals the number of expected cases.

In addition to gender differences, meaningful differences emerged when VHA service utilization was examined. Although the magnitudes and time trends for non–VHA-utilizing veterans mirrored those of the veteran population as a whole, this was not the case for VHA-utilizing veterans (Figure 2). In 2000, the age- and gender-adjusted SMR for VHA-utilizing veterans compared with nonveterans was greater than that for non–VHA-utilizing veterans, but it declined considerably over the next three years. From 2003 to 2010, SMRs for VHA-utilizing veterans compared directly with non–VHA-utilizing veterans were significantly less than 1 (Figure 2), indicating that the observed number of suicides for VHA-utilizing veterans was less than expected had their suicide rates been the same as for non–VHA-utilizing veterans. The sensitivity analysis [see online supplement] suggests that these SMRs provide conservative estimates of the relative risks of interest, which may be slightly higher.

Finally, considering gender and VHA service utilization simultaneously, the decline in SMRs among VHA veterans was primarily observed for males (Figure 2). In 2000, male VHA-utilizing veterans experienced 36% more suicides than expected compared with their nonveteran peers; this declined to a low of 16% in 2003 and increased slightly to 23% by 2010. Conversely, no clear period of change emerged for female VHA-utilizing veterans, who experienced approximately one-and-a-half to two times the number of suicides expected compared with nonveteran females. Estimates for female VHA-utilizing veterans were less precise due to the small number of annual suicides in some age groups. When female VHA-utilizing veterans were compared directly with female non–VHA-utilizing veterans, however, SMRs were consistently and significantly less than 1, indicating that female VHA-utilizing veterans experienced fewer suicides than expected if their rates had been similar to veterans who did not use VHA services (Figure 2 and online supplement). Female veterans who did not use VHA services experienced the greatest number of excess suicides compared with their nonveteran peers throughout the study period (Figure 2).

Discussion

Evidence of increasing suicide risk among select U.S. subgroups is mounting (20). Most notably, significant increases have been observed among middle-aged men and women since 1999 (21). The question of how veterans contribute to this change has received considerable attention in recent years (57). Prior SMR estimates using indirect estimation methods for VHA-utilizing veterans compared with the general population have been published (13,14). Suicide SMRs for VHA patients by fiscal year were previously reported to have decreased from 1.7 in 2000 to 1.4 in 2007 (14). To the best of our knowledge, however, ours is the first study to directly compare veteran and nonveteran suicide risk using a validated variable for veteran status while documenting differences in suicide rates between veterans who were VHA service users and those who were not. On the basis of these analyses, veterans appear to be another select U.S. subpopulation experiencing greater than average increases in suicide rates. Importantly, however, such increases were not observed uniformly across the veteran population. Crude suicide rates declined among veterans using VHA services from 2000 to 2010, and, even after the analysis accounted for age and gender, the excess risk among VHA-utilizing veterans compared with their nonveteran peers decreased as well. Conversely, increases in suicide rates among veterans without a history of VHA service use were occurring alongside, and more rapidly than, increases among U.S. adults without a history of military service. It is reasonable to assert, therefore, that factors within the veteran population may make this group particularly susceptible to stressors and subsequent mental health conditions associated with increased risk of suicide (22), which may be partially mitigated by use of VHA services.

Combat-related exposures have been suggested as one factor contributing to increases in veteran suicide risk compared with nonveterans; however, recent research indicates that this may not fully explain observed differences (3). Alternatively, increased suicide risk among some veteran groups may be related to characteristics that are not unique to veterans but that may be associated with selection into military service. For example, adverse childhood experiences have been shown to increase the risk of a suicide attempt (23) and may serve as motivating factors for selecting military service (24)—essentially providing an escape route from suboptimal family environments. Additional research is needed on factors differentially associated with selection of U.S. military service that may explain differences in suicide risk. It is also possible that veterans are overrepresented in populations that are particularly vulnerable to stressors related to suicide, such as the recent economic downturn (25), that affect the U.S. population as a whole. These may include marginally employed populations or those with greater economic vulnerability (26) and homeless populations (27).

Study findings clearly suggest that veterans without a history of VHA service use are at particularly high risk of suicide. However, it is likely that this group is quite heterogeneous, and national data on clinical characteristics and risk exposures are not readily available. Although some veterans who do not use VHA services may be healthy or may have private health insurance, others may choose not to use VHA services for reasons related to difficulties accessing care, misconceptions regarding quality of care, a lack of knowledge about eligibility for services, or perceived stigma surrounding seeking mental health care. Research is warranted to better understand suicide risk among veterans who do not use VHA services and whether characteristics related to VHA service eligibility and use are related to suicide risk.

In addition, despite overall improvements, clear gender differences emerged for veterans who used VHA services. SMRs for VHA-utilizing and non–VHA-utilizing female veterans were substantially higher than those observed for males. Although female veterans who used VHA services consistently experienced fewer suicides than expected relative to female veterans who did not use VHA services, there was no meaningful change in this estimate over time. Conversely, SMRs decreased over time among male VHA-utilizing veterans. Research on risk and protective factors unique to female veterans is needed.

Finally, although most of the decline in SMRs among VHA-utilizing veterans in this study occurred prior to the onset of activities in support of current VHA suicide prevention programs (28), this does not suggest that prevention programs have been ineffective. Rather, the plateau observed among VHA-utilizing veterans from 2004 to 2009 occurred in contrast to a rising suicide rate among non–VHA-utilizing veterans. One possible reason for this relative stability could be that VA’s Mental Health Enhancement Initiative and Suicide Prevention Program are successfully countering rising veteran suicide rates. That said, gender differences discussed above suggest possible gender-based differences in the effectiveness of VA suicide prevention initiatives. There may also be differences in health and demographic factors among male and female veterans who use VHA services, or there may be a difference in baseline risk of suicide related to eligibility for or self-selection into VHA services that may vary by gender. These results further support continued efforts to better understand health disparities, including suicide, among women veterans and the most effective clinical and public health strategies to prevent suicide among this fastest-growing subgroup of veterans.

Although this study is a novel and important contribution to the literature, it had limitations. The data source was limited to suicides. Because information was not available on other causes of death, particularly undetermined causes and accidental overdoses, it was not possible to account for potential misclassification of suicides as other causes of death. Such misclassification likely differs by age and gender (29). Misclassification may also differ by veteran status given that the availability of information to classify cause of death might vary by veteran status and veterans are more likely to use firearms for suicide, a mechanism that is less likely to lead to misclassification. At this time, however, death certificates are the gold standard for documenting the cause of death and were used as such in this study. In addition, although this study reported only on the first 23 states that provided complete data to the VA’s suicide mortality database, it appears that the first 23 states are representative of the United States as a whole in terms of geographic distribution and the range of state-specific suicide rates. As the database expands, analyses will continue to extend information and improve precision for female veterans who use VHA services. Finally, for this initial analysis, SMRs were chosen to estimate the true standardized rate ratios. The utility of this measure for dealing with sparse data while providing valid estimates to compare mortality between populations makes this approach preferable to other methods such as direct standardization (30). Specifically, a relatively small number of suicides for a single year within certain age groups of VHA-utilizing female veterans made direct age-adjustment of suicide rates in this group infeasible at this time. However, given that SMRs only estimate the true rate ratio of interest, we conducted a sensitivity analysis computing gender-specific direct age-adjusted rate ratios for all veterans compared with nonveterans; the results demonstrated that the SMR truly does provide a conservative estimate, and time trends observed were very similar [see online supplement]. Likewise, it is safe to assume that the SMR is an appropriate, conservative estimate to compare veterans who do and do not use VHA services.

In summary, these novel findings indicate a decline in the relative risk of suicide among VHA-utilizing veterans compared with their nonveteran peers despite continued increases among veterans who did not use VHA services. The VA Mental Health Enhancement Initiative launched in 2005 and the Suicide Prevention Program launched in 2007 are designed to provide readily available, integrated care through awareness and education campaigns, a 24-hour Veterans Crisis Line, and placement of suicide prevention coordinators to track and follow up with high-risk patients in each VA medical center, with evidence-based treatments available for high-risk patients. Research is needed to directly evaluate the impact of these programs overall on veteran suicide rates in recent years and to identify which components of the program are or are not effective. For example, a recent evaluation of Denmark’s suicide prevention clinics demonstrates a possible approach to evaluating national VA suicide prevention initiatives (31). In addition, parallel research is needed to determine how to best engage eligible, at-risk veterans not currently utilizing VHA services.

Conclusions

In 2009, suicide became the most common cause of violent (32) and injury-related (33) deaths in the United States, which has drawn considerable and renewed attention to the need for enhanced prevention efforts. This report offers new, valuable insight into changes in suicide among veterans during this period of national increase. Consistent with previous analyses, our results indicate that suicide is a complex outcome, with risk mediated by individual and service-related characteristics, such as gender, age, and VHA service use. Combating veteran suicide is certainly a high priority within the VHA, and veterans should continue to be recognized as a high-priority suicide research population. On the basis of the 2010 veteran suicide rate observed in this study (35.9 per 100,000 lives) and the fact that there are approximately 22 million living veterans in the United States (11), the potential exists to save approximately 800 lives annually by reducing veteran suicide rates by 10%. Such a reduction would be a considerable step toward reaching our nation’s overall suicide reduction goal of 4,000 lives annually over the next five years.

Dr. Hoffmire and Dr. Kemp are with the VISN2 Center of Excellence for Suicide Prevention, U.S. Department of Veterans Affairs, Canandaigua, New York (e-mail: ). Dr. Bossarte is with the Office of Public Health, U.S. Department of Veterans Affairs, Washington, D.C. Dr. Hoffmire and Dr. Bossarte are also with the Department of Psychiatry, University of Rochester, Rochester, New York.

The authors thank Brady Stephens, M.S., for careful data management of the U.S. Department of Veterans Affairs archive of state suicide data. They also thank the departments of health and related agencies in the 23 participating states: Alabama, Alaska, Arkansas, Florida, Idaho, Iowa, Kansas, Louisiana, Michigan, Minnesota, Montana, Nebraska, New Jersey, New York, North Carolina, Oregon, Pennsylvania, Rhode Island, Tennessee, Texas, Utah, Washington, and West Virginia. Interpretations and conclusions are those of the authors and not those of agencies in the participating states. The views expressed are those of the authors and do not necessarily represent the views of the U.S. Department of Veterans Affairs or the institutional affiliations listed.

The authors report no financial relationships with commercial interests.

References

1 Hoyert DL, Xu J: Deaths: preliminary data for 2011. National Vital Statistics Reports 61:1–51, 2012Google Scholar

2 National Strategy for Suicide Prevention: Goals and Objectives for Action: A Report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. Washington, DC, Office of the Surgeon General, 2012Google Scholar

3 Armed Forces Health Surveillance Center: Deaths by suicide while on active duty, active and reserve components, US Armed Forces, 1998–2011. Medical Surveillance Monthly Report 19:7–10, 2012Google Scholar

4 Black SA, Gallaway MS, Bell MR, et al.: Prevalence and risk factors associated with suicides of army soldiers 2001–2009. Military Psychology 23:433–451, 2011CrossrefGoogle Scholar

5 Kaplan MS, McFarland BH, Huguet N, et al.: Estimating the risk of suicide among US veterans: how should we proceed from here? American Journal of Public Health 102(suppl 1):S21–S23, 2012Crossref, MedlineGoogle Scholar

6 Miller M, Azrael D, Barber C, et al.: A call to link data to answer pressing questions about suicide risk among veterans. American Journal of Public Health 102(suppl 1):S20–S22, 2012Crossref, MedlineGoogle Scholar

7 Bossarte RM, Claassen CA, Knox KL: Evaluating evidence of risk for suicide among veterans. Military Medicine 175:703–704, 2010Crossref, MedlineGoogle Scholar

8 Data and Statistics. Washington, DC, US Department of Veterans Affairs, National Center for Veterans Analysis and Statistics, Nov 21, 2013. Available at www.va.gov/vetdata. Accessed Sept 2013Google Scholar

9 Wojcik BE, Stein CR, Bagg K, et al.: Traumatic brain injury hospitalizations of US army soldiers deployed to Afghanistan and Iraq. American Journal of Preventive Medicine 38(suppl):S108–S116, 2010Crossref, MedlineGoogle Scholar

10 Department of Defense Annual Report on Sexual Assault in the Military Fiscal Year 2012. Volume 1. Washington, DC, US Department of Defense, Sexual Assault Prevention and Response, April 15, 2013Google Scholar

11 Brenner LA, Ignacio RV, Blow FC: Suicide and traumatic brain injury among individuals seeking Veterans Health Administration services. Journal of Head Trauma Rehabilitation 26:257–264, 2011Crossref, MedlineGoogle Scholar

12 Kimerling R, Street AE, Pavao J, et al.: Military-related sexual trauma among Veterans Health Administration patients returning from Afghanistan and Iraq. American Journal of Public Health 100:1409–1412, 2010Crossref, MedlineGoogle Scholar

13 McCarthy JF, Valenstein M, Kim HM, et al.: Suicide mortality among patients receiving care in the Veterans Health Administration health system. American Journal of Epidemiology 169:1033–1038, 2009Crossref, MedlineGoogle Scholar

14 Blow FC, Bohnert AS, Ilgen MA, et al.: Suicide mortality among patients treated by the Veterans Health Administration from 2000 to 2007. American Journal of Public Health 102(suppl 1):S98–S104, 2012Crossref, MedlineGoogle Scholar

15 Katz IR, McCarthy JF, Ignacio RV, et al.: Suicide among veterans in 16 states, 2005 to 2008: comparisons between utilizers and nonutilizers of Veterans Health Administration (VHA) services based on data from the National Death Index, the National Violent Death Reporting System, and VHA administrative records. American Journal of Public Health 102(suppl 1):S105–S110, 2012Crossref, MedlineGoogle Scholar

16 Hoffmire CA, Piegari RI, Bossarte RM: Misclassification of veteran status on Washington State death certificates for suicides from 1999 to 2008. Annals of Epidemiology 23:298–300, 2013Crossref, MedlineGoogle Scholar

17 McIntosh JL, Drapeau CW: USA Suicide 2010: Official Final Data. Washington, DC: American Association of Suicidology, Sept 20, 2012. Available at www.suicidology.orgGoogle Scholar

18 Klein RJ, Schoenborn CA: Age Adjustment Using the 2000 Projected US Population. Healthy People Statistical Notes, no 20. Hyattsville, Md, National Center for Health Statistics, Jan 2001Google Scholar

19 Venzon DJ, Moolgavkar SH: A method for computing profile-likelihood-based confidence intervals. Applied Statistics 37:87–94, 1998CrossrefGoogle Scholar

20 Baker SP, Hu G, Wilcox HC, et al.: Increase in suicide by hanging/suffocation in the US, 2000–2010. American Journal of Preventive Medicine 44:146–149, 2013Crossref, MedlineGoogle Scholar

21 Centers for Disease Control and Prevention (CDC): Suicide among adults aged 35–64 years: United States, 1999–2010. Morbidity and Mortality Weekly Report 62:321–325, 2013MedlineGoogle Scholar

22 Kang HK, Bullman TA: Risk of suicide among US veterans after returning from the Iraq or Afghanistan war zones. JAMA 300:652–653, 2008Crossref, MedlineGoogle Scholar

23 Felitti VJ, Anda RF, Nordenberg D, et al.: Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The Adverse Childhood Experiences (ACE) Study. American Journal of Preventive Medicine 14:245–258, 1998Crossref, MedlineGoogle Scholar

24 Blosnich JR, Dichter ME, Cerulli C, et al.: Disparities in adverse childhood experiences among individuals with a history of military service. JAMA Psychiatry 71:1041–1048, 2014Crossref, MedlineGoogle Scholar

25 Reeves A, Stuckler D, McKee M, et al.: Increase in state suicide rates in the USA during economic recession. Lancet 380:1813–1814, 2012Crossref, MedlineGoogle Scholar

26 Yip PS, Caine ED: Employment status and suicide: the complex relationships between changing unemployment rates and death rates. Journal of Epidemiology and Community Health 65:733–736, 2011Crossref, MedlineGoogle Scholar

27 Bossarte RM, Piegari R, Hill L, et al.: Age and suicide among veterans with a history of homelessness. Psychiatric Services 64:713–714, 2013LinkGoogle Scholar

28 VHA Handbook 1160.01: Uniform Mental Health Services in VA Medical Centers and Clinics. Washington, DC, US Department of Veterans Affairs, Sept 11, 2008Google Scholar

29 Rockett IR, Hobbs G, De Leo D, et al.: Suicide and unintentional poisoning mortality trends in the United States, 1987–2006: two unrelated phenomena? BMC Public Health 10:705, 2010Crossref, MedlineGoogle Scholar

30 Goldman DA, Brender JD: Are standardized mortality ratios valid for public health data analysis? Statistics in Medicine 19:1081–1088, 2000Crossref, MedlineGoogle Scholar

31 Erlangsen A, Dam Lind B, Stuart EA, et al.: Short-term and long-term effects of psychosocial therapy for people after deliberate self-harm: a register-based, nationwide multicenter study using propensity score matching. Lancet Psychiatry (Epub ahead of print, Nov 24, 2014)MedlineGoogle Scholar

32 Karch DL, Logan J, McDaniel D, et al.: Surveillance for violent deaths—National Violent Death Reporting System, 16 states, 2009. MMWR Surveillance Summaries 61:1–43, 2012MedlineGoogle Scholar

33 Rockett IR, Regier MD, Kapusta ND, et al.: Leading causes of unintentional and intentional injury mortality: United States, 2000–2009. American Journal of Public Health 102:e84–e92, 2012Crossref, MedlineGoogle Scholar