Suicides — United States, 2005–2009

Alex E. Crosby, MD1

LaVonne Ortega, MD2

Mark R. Stevens, MSPH, MA1

1National Center for Injury Prevention and Control, CDC

2Center for Surveillance, Epidemiology, and Laboratory Services, CDC

Corresponding author: Alex E. Crosby, Division of Violence Prevention, National Center for Injury Prevention and Control, CDC. Telephone: 770-488-4272; E-mail: aec1@cdc.gov.

Introduction

Injury from self-directed violence, which includes suicidal behavior and its consequences, is a leading cause of death and disability. In 2009, suicide was the 10th-leading cause of death in the United States and the cause of 36,909 deaths (1). In 2005, the estimated cost of self-directed violence (fatal and nonfatal treated) was $41.2 billion (including $38.9 billion in productivity losses and $2.2 billion in medical costs) (2). Suicide is a complex human behavior that results from an interaction of multiple biological, psychological, social, political, and economic factors (3). Although self-directed violence affects members of all racial/ethnic groups in the United States, it often is misperceived to be a problem affecting primarily non-Hispanic white males (4).

This report is part of the second CDC Health Disparities and Inequalities Report (CHDIR). The 2011 CHDIR (5) was the first CDC report to assess disparities across a wide range of diseases, behavior risk factors, environmental exposures, social determinants, and health-care access. The topic presented in this report is based on criteria that are described in the 2013 CHDIR Introduction (6). This report updates information that was presented in the 2011 CHDIR (7) by providing more current data on suicide in the United States. The purposes of this report are to discuss and raise awareness of differences in the characteristics of suicide decedents and to prompt actions to reduce these disparities.

Methods

To determine differences in the prevalence of suicide by sex, race/ethnicity, age, and educational attainment in the United States, CDC analyzed 2005–2009 data from the Web-based Injury Statistics Query and Reporting System — Fatal (WISQARS Fatal) (8) and the National Vital Statistics System (NVSS). In this report, NVSS data provided as of February 2012 were used. The 2009 data were used to describe the overall patterns in suicides. The aggregate 2005–2009 reporting period was used to describe patterns for the combined age group and race/ethnicity because sample sizes for any single year were limited. Mortality data were drawn from CDC's National Vital Statistics System (NVSS), which collects death certificate data filed in the 50 states and the District of Columbia (1). Data in this report include suicides from any cause during 2005–2009. The WISQARS database contains mortality data based on NVSS and population counts for all U.S. counties based on U.S. Census data. Counts and rates of death can be obtained by underlying cause of death, mechanism of injury, state, county, age, race, sex, year, injury cause of death (e.g., firearm, poisoning, or suffocation) and by manner of death (e.g., suicide, homicide, or unintentional injury) (8).

NVSS codes racial categories as non-Hispanic white, non-Hispanic black, American Indian/Alaska Native (AI/AN), and Asian/Pacific Islander (A/PI); ethnicity is coded separately as Hispanic or non-Hispanic (1). Persons of Hispanic ethnicity might be of any race or combination of races. Absolute differences in rates between two populations were compared using a test statistic, z, based on a normal approximation at a critical value of α = 0.05 (9).

Educational attainment is recorded by two methods on death certificates. In 28 states* and the District of Columbia (DC), the 2003 version of the standard certificate of death is used (which collects the highest degree completed), whereas 20 states† use the 1989 version of the certificate (which collects the number of years of education completed). For this reason, these two groups of states were analyzed separately. Death rates by educational attainment were based on population estimates from the U.S. Census Bureau's 2009 American Community Survey (ACS) (10). Data for Georgia and Rhode Island were excluded because educational attainment was not recorded on their death certificates. Rates are presented only for persons aged ≥25 years because persons aged <25 years might not have completed their formal education (9).

Unadjusted (crude) suicide rates were based on resident population data from the U.S. Census Bureau (10). Rates based on <20 deaths were considered unreliable and not included in the analysis. Confidence intervals were calculated in two ways: 1) groupings of <100 deaths were calculated by using the gamma method (9), and 2) groupings of ≥100 deaths were calculated by using a normal approximation (9).

Results

In 2009, a total of 36,909 suicides occurred in the United States, 83.5% of which were among non-Hispanic whites, 7.0% among Hispanics, 5.5% among non-Hispanic blacks, 2.5% among A/PIs, and 1.1% among AI/ANs (Table). Although AI/ANs represented the smallest proportion of suicides of all racial/ethnic groups, they shared the highest rates with whites. Overall, the crude suicide rate for males (19.2 per 100,000 population) was approximately four times higher than the rate for females (5.0 per 100,000 population). In each of the racial/ethnic groups, suicide rates were higher for males than for females, but the male-female ratio for suicide differs among these groups. Among non-Hispanic whites, the male-female ratio was 3.8:1; among Hispanics it was 4.5:1; among non-Hispanic blacks it was 4.7:1; among A/PIs it was 2.3:1; and among AI/ANs it was 2.8:1. These male-female ratios did not change significantly from those reported previously (7).

Overall, suicide rates varied by the level of educational attainment. Persons with the highest educational attainment had the lowest rates, those with the lowest educational attainment had intermediate rates, and those who had completed only the equivalent of high school (or 12 years of education) had the highest rates. This pattern was consistent for males, but the pattern of educational inequalities was different among females. Females with a lower educational level had the lowest suicide rates followed by those with the highest educational level, while those females with a high school education (12 years of education) had the highest suicide rates. For each version of the death certificate, whether overall or by sex, suicide rates differed significantly between levels of educational attainment, except that rates for females did not differ significantly between the lowest and highest educational attainment levels in the states on the basis of data from the 1989 death certificate version.

Suicide rates by race/ethnicity and age group demonstrated different patterns by racial/ethnic group, with the highest rates occurring among AI/AN adolescents and young adults aged 15–34 years (Figure). Rates among AI/ANs and non-Hispanic blacks were highest among adolescents and young adults, then declined or leveled off with increasing age, respectively. Among A/PIs and Hispanics, rates were highest among young adults in their early 20s, then leveled off among other adults but increased for those aged ≥65 years. In contrast, rates among non-Hispanic whites were highest among those aged 40–54 years. Although the 2009 overall rates for AI/ANs are similar to those of non-Hispanic whites, the 2005–2009 rates among adolescent and young adult AI/ANs aged 15–29 years were substantially higher.

Discussion

The burden of suicide among AI/AN youths is considerably higher than that among other racial/ethnic groups. In 2009, suicide ranked as the fourth leading cause of years of potential life lost (YPLL) for AI/ANs aged <75 years, accounting for 6.8% of all YPLL among AI/ANs (8). Studies examining the historical and cultural context of suicide among AI/AN populations have identified multiple contributors to the high rates such as individual-level factors (e.g., alcohol and substance misuse and mental illness), family- or peer-level factors (e.g., family disruption or suicidal behavior of others), and societal-level factors (e.g., poverty, unemployment, discrimination, and historic trauma [i.e., cumulative emotional and psychological wounding across generations]) (11). Although certain protective factors exist within AI/AN communities, including spirituality and cultural continuity, these factors often are overwhelmed by the magnitude of the risk factors (11). If the overall suicide rate among the AI/AN population (highest rate) could be decreased to that of non-Hispanic blacks (lowest rate), 271 (66.6%) of the total 407 AI/AN deaths during 2009 might have been prevented. This idea of achieving rates of the lowest group is similar to that proposed in the Healthy People 2010 objectives (12).

Prevention efforts and resources also should be directed toward adults aged 40–54 years because this age group has the highest (and increasing) suicide rate, but this age group often is overlooked as a group at which prevention efforts should focus (13). The National Strategy for Suicide Prevention has identified males in this age group as one of the populations at increased risk for suicide for whom additional surveillance, research, and prevention programs need to be focused (14).

The findings regarding the association of suicide rates and educational attainment are mixed in this study and in others. Certain studies (15) have found an inverse relationship between educational status and suicide among males (i.e., suicide rates decrease as educational attainment increases), whereas other studies (16) have not found this pattern. Patterns among females identified in other studies seem more consistent (i.e., the lowest rates occur among those with the lowest educational attainment) or find no association, but the underlying explanation is unclear (17,18). It has been suggested that studies on the association between education and suicide should perform more specific analysis (e.g., by examining combinations of age, ethnicity, culture, and sex variables to assess the true association) (19).

As a result of multiple challenges (e.g., narrow theoretical focus, lack of longitudinal studies to provide a range of modifiable risk and protective factors, and insufficient study designs), the evidence for the proven effectiveness of suicide prevention programs is sparse (20). Suicide prevention efforts often focus on counseling, education, and clinical intervention strategies for persons at high risk for suicide, neglecting a broader population-based approach (20). Although these efforts might assist those persons at the highest risk for adverse outcomes, they also require high levels of effort and commitment and might have a limited population-level impact, a critical goal of public health (21). In contrast, strategies that seek to address societal-level factors demonstrated to be associated with suicide (e.g., economic strain, poverty, and misuse of alcohol and other psychoactive substances) and improving the health-care system infrastructure in impoverished and underserved communities to address this problem might have a greater population impact but need additional development and testing (22).

Limitations

The findings presented in this report are subject to at least four limitations. First, suicides often are undercounted on death certificates, and studies have indicated that they are differentially undercounted for females and racial/ethnic minorities (23); therefore, the suicide rates in this analysis are likely to be underestimated. Second, injury mortality data likely underestimate by 25%–35% the actual numbers of deaths for AI/ANs and certain other racial/ethnic populations (e.g., Hispanics) because of the misclassification of race/ethnicity of decedents on death certificates (24). Third, data on educational attainment must be interpreted with caution because of misclassification of the decedent's years of education, which has been shown with comparisons between educational attainment as recorded on the death certificate versus that in census surveys (9). Finally, certain variables that have been associated with suicidal behavior (e.g., psychiatric illness, sexual orientation, and social isolation) are not collected in U.S. mortality data, and therefore patterns of suicide based on these factors cannot be described. Other data sources (e.g., the National Violent Death Reporting System) that collect a broader array of information about the circumstances surrounding suicides and other violent deaths can provide additional insight (25).

Conclusion

Comprehensive suicide prevention programs focus on risk and protective factors, including coping skills, access to mental health treatment, substance misuse, and social support. However, only a limited number of programs have been developed specifically for higher risk or racial/ethnic minority populations (3). An example of a comprehensive prevention program that has been reported to reduce suicidal behavior within an AI/AN community is the Natural Helpers Program (26). This program includes health education and outreach activities to the community and at-risk persons, training for community members in identification of at-risk persons, and support for local behavioral health efforts like alcohol and substance abuse programs.

Strategies that address the health and well-being of persons at risk and that support the widespread implementation of culturally relevant and effective programs are needed to reduce the rates of suicide among groups that are disproportionately affected. To address some of these issues, CDC has focused on studying and promoting individual and organizational connectedness as a way to prevent suicide (27).

References

1. CDC. Deaths: final data for 2009. Atlanta, GA: US Department of Health and Human Services, CDC; 2011. Available at http://www.cdc.gov/nchs/data/dvs/deaths_2009_release.pdf.
2. CDC. Web-based Injury Statistics Query and Reporting System (WISQARS): cost of injury reports. Atlanta, GA: US Department of Health and Human Services, CDC; 2012. Available at http://wisqars.cdc.gov:8080/costT.
3. Goldsmith SK, Pellmar TC, Kleinman AM, Bunney WE, eds. Reducing suicide: a national imperative. Washington, DC: National Academies Press; 2002.
4. Earls F, Escobar JI, Manson SM. Suicide in minority groups: epidemiologic and cultural perspectives. In: Blumenthal SJ, Kupfer DJ, eds. Suicide over the life cycle: risk factors, assessment, and treatment of suicidal patients. Washington, DC: American Psychiatric Publishing; 1990:571–98.
5. CDC. CDC health disparities and inequalities report—United States, 2011. MMWR 2011;60(Suppl; January 14, 2011).
6. CDC. Introduction. In: CDC health disparities and inequalities report—United States, 2013. MMWR 2013;62(No. Suppl 3).
7. CDC. Suicides—United States, 1999–2007. In: CDC health disparities and inequalities report—United States, 2011. MMWR 2011;60(Suppl; January 14, 2011).
8. CDC. Web-based Injury Statistics Query and Reporting System (WISQARS): fatal injury data. Atlanta, GA: US Department of Health and Human Services, CDC. Available at http://www.cdc.gov/injury/wisqars/fatal.html.
9. Xu J, Kochanek KD, Murphy SL, Tejada-Vera B. Deaths: final data for 2007. National Vital Statistics Report 2010;58(19).
10. US Bureau of the Census. 2009 American Community Survey 1-Year Estimates, prepared by the US Census Bureau. Available at http://factfinder2.census.gov.
11. Substance Abuse and Mental Health Services Administration. To live to see the great day that dawns: preventing suicide by American Indian and Alaska Native youth and young adults. Rockville, MD: US Department of Health Human Services, Substance Abuse and Mental Health Services Administration; 2010. Publication no. SMA 10-4480.
12. US Department of Health and Human Services. Healthy people 2010. 2nd ed. With Understanding and Improving Health and Objectives for Improving Health. 2 vols. Washington, DC: U.S. Government Printing Office; 2000.
13. Phillips JA, Robin AV, Nugent CN, Idler EL. Understanding recent changes in suicide rates among the middle-aged: period or cohort effects? Public Health Rep 2010;125:680–8.
14. US Department of Health and Human Services, Office of the Surgeon General and National Action Alliance for Suicide Prevention. 2012 national strategy for suicide prevention: goals and objectives for action. Washington, DC: US Department of Health and Human Services; 2012.
15. Abel EL, Kruger ML. Educational attainment and suicide rates in the United States. Psychol Rep 2005;97:25–8.
16. Fernquist RM. Education, race/ethnicity, age, sex, and suicide: individual-level data in the United States, 1991–1994. Current Research in Social Psychology 2001;6:277–90. Available at http://www.uiowa.edu/~grpproc.
17. Kposowa A J. Marital status and suicide in the National Longitudinal Mortality Study. J Epidemiol Community Health 2000;54:254–61.
18. Lorant V, Kunst AE, Huisman M, Costa G, Mackenbach JP. Socio-economic inequalities in suicide: a European comparative study. British Journal of Psychiatry 2005;187:49–54.
19. Stack S. Education and risk of suicide: an analysis of African Americans. Sociol Focus 1998;31:295–302.
20. De Leo D. Why are we not getting any closer to preventing suicide? Br J Psychiatry 2002;181:372–4.
21. Frieden TR. A framework for public health action: the health impact pyramid. Am J Public Health 2010;100:590–5.
22. Knox KL, Conwell Y, Caine ED. If suicide is a public health problem, what are we doing to prevent it? Am J Public Health 2004;94:37–45.
23. Phillips DP, Ruth TE. Adequacy of official suicide statistics for scientific research and public policy. Suicide Life Threat Behav 1993;23:307–19.
24. Arias E, Schauman WS, Eschbach K, Sorlie PD, Backlund E. The validity of race and Hispanic origin reporting on death certificates in the United States. Vital Health Stat 2 2008;148:1–23.
25. Steenkamp M, Frazier L, Lipskiy N, et al. The National Violent Death Reporting System: an exciting new tool for public health surveillance. Inj Prev 2006;12(Suppl 2):ii3–5.
26. May PA, Serna P, Hurt L, DeBruyn LM. Outcome evaluation of a public health approach to suicide prevention in an American Indian tribal nation. Am J Public Health 2005;95:1238–44.
27. Kaminski JW, Puddy RW, Hall DM, Cashman SY, Crosby AE, Ortega LAG. The relative influence of different domains of social connectedness on self-directed violence in adolescence. J Youth Adolesc 2010;39:460–73.