Homicides — United States, 2007 and 2009

Joseph E. Logan, PhD

Jeffrey Hall, PhD

Dawn McDaniel, PhD

Mark R. Stevens, MSPH, MA

National Center for Injury Prevention and Control, CDC

Corresponding author: Joseph E. Logan, National Center for Injury Prevention and Control, CDC. Telephone: 770-488-1529; E-mail: ffa3@cdc.gov.

Introduction

According to 1981–2009 data, homicide accounts for 16,000–26,000 deaths annually in the United States and ranks within the top four leading causes of death among U.S. residents aged 1–40 years (1). Homicide can have profound long-term emotional consequences on families and friends of victims and on witnesses to the violence (2,3), as well as cause excessive economic costs to residents of affected communities (1,4). For years, homicide rates have been substantially higher among certain populations. Previous reports have found that homicides are higher among males (5–7), adolescents and young adults (6), and certain racial/ethnic groups, such as non-Hispanic blacks, non-Hispanic American Indian/Alaska Natives (AI/ANs), and Hispanics (6–9). The 2011 CDC Health Disparities and Inequalities Report (CHDIR) described similar findings for the year 2007 (10). For example, the 2011 report showed that the 2007 homicide rate was highest among non-Hispanic blacks (23.1 deaths per 100,000), followed by AI/ANs (7.8 deaths per 100,000), Hispanics (7.6 deaths per 100,000), non-Hispanic whites (2.7 deaths per 100,000), and Asian/Pacific Islanders (A/PIs) (2.4 deaths per 100,000) (10). In addition, non-Hispanic black men aged 20–24 years were at greatest risk for homicide in 2007, with a rate that exceeded 100 deaths per 100,000 population (10). Other studies have reported that community factors such as poverty and economic inequality and individual factors such as unemployment and involvement in criminal activities can play a substantial role in these persistent disparities in homicide rates (11). Public health strategies are needed in communities at high risk for homicide to prevent violence and save lives.

The homicide analysis and discussion that follow are part of the second CHDIR and update information presented in the first CHDIR (10). The 2011 CHDIR (12) was the first CDC report to take a broad view of disparities across a wide range of diseases, behavioral 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 (13). The purposes of this homicide report are to discuss and raise awareness of differences in homicide rates by some of these characteristics and to prompt actions to reduce these disparities.

Methods

To assess disparities in homicide rates in the United States, CDC analyzed data from the CDC National Vital Statistics System (NVSS), the same data source used for the 2011 CHDIR on homicides (10). In this report, NVSS data provided as of February 2012 were used. NVSS collects death certificate data filed in the 50 states and the District of Columbia (DC) (14). Death certificates provide information on the decedent's age, sex, race, ethnicity, and geographic region. They do not provide information on decedent income, sexual orientation, disability, or language spoken at home.

This report summarizes the homicide data by providing the number, proportion, and rates of homicides by age, sex, and race/ethnicity for the year 2009; providing the homicide rates by U.S. state for the year 2009; and comparing 2009 with 2007 data. Data in this report are based on homicides caused by any mechanism. More details on homicide rates by age, sex, and race/ethnicity for each state and census region can be accessed through the CDC Web-based Injury Statistics Query and Reporting System — Fatal (WISQARS Fatal) Injury Data module (available at http://www.cdc.gov/injury/wisqars/index.html). Data on individual and socioeconomic risk factors for homicide were unavailable for analysis. In addition, sufficient data were not available to assess disparities by certain racial/ethnic subgroups, household income, disability status, and sexual orientation. NVSS codes racial categories as white, black, AI/AN, and A/PI, and ethnicity is coded separately as Hispanic or non-Hispanic (14). In this report, references to whites, blacks, AI/ANs, and A/PIs refer to non-Hispanic persons. Hispanics might be of any race or combination of races. Crude homicide rates per 100,000 population were calculated by age, sex, and race/ethnicity, as well as by the combination of these three variables. Crude rates per 100,000 population by state in 2009 also are provided. Confidence intervals (CIs) of rates were calculated in two ways: 1) groupings of annual death counts of <100 were calculated by using a gamma estimation method (14), and 2) groupings of annual death counts of ≥100 were calculated by using a normal approximation approach. Rates calculated from <20 deaths were considered unreliable and are not reported.

Disparities were measured as deviations from a referent category rate or prevalence. The group with the largest population of the U.S. census data in each demographic category was used as the referent (e.g., females, non-Hispanic whites, or persons aged 30–49 years). Absolute difference was measured as the simple difference between a population subgroup estimate and the estimate for its respective reference group. The relative difference as a percentage was calculated by dividing the absolute difference by the value in the referent category and multiplying by 100. Relative differences in rates between each race/ethnic category were also stratified by sex and age. Rate comparisons were considered significantly different if they had nonoverlapping 95% CIs, which provide a conservative test for statistical significance.

Results

An estimated 18,361 homicides occurred in 2007 and 16,799 occurred in 2009 (Table 1). The relative rate difference reported for males was at least 250% higher than that of females in both data years. In addition, in each data year, the relative rate difference for non-Hispanic blacks was at least 650% higher than the rate reported for non-Hispanic whites. Non-Hispanic AI/ANs and Hispanics also had rates that far exceeded those of non-Hispanic whites in both years. Rates were highest among persons aged 15–29 years both in 2007 and 2009 and then decreased with each subsequent age group; however, the lowest rates reported in both years were among children aged 0–14 years.

The homicide rate for the U.S. population in 2009 was significantly lower than the U.S. homicide rate reported in 2007. Differences in rates also occurred among certain populations. Specifically, homicide rates were lower in 2009 than those reported in 2007 for males, non-Hispanic whites, non-Hispanic blacks, Hispanics, persons aged 15–29 years, and persons aged 30–49 years. None of the demographic groups had significantly higher rates in 2009 compared with 2007.

Among males, the risk for homicide was greatest among non-Hispanic blacks aged 15–29 years in both 2007 and 2009 (Table 2). Furthermore, for both years, the male homicide rate was significantly higher among non-Hispanic blacks than among those in other racial/ethnic groups in each age category assessed, except among men aged 50–64 years, for whom the 95% CIs overlapped with the rate for AI/ANs in 2009. Hispanic males had higher rates than non-Hispanic white males in every age group among males aged ≥15 years in both years as well.

Among females, the homicide rates also were generally higher among racial/ethnic minorities (Table 2). For example, in both years, female homicide rates were markedly higher among non-Hispanic blacks than among non-Hispanic whites in every age group <65 years. Female homicide rates also were higher among Hispanic than among non-Hispanic whites in every age group <30 years for both years. In 2009, the female homicide rate was higher among AI/ANs than among non-Hispanic whites aged 15–29 years as well.

Compared with 2007, homicide rates were significantly lower in 2009 in certain demographic populations (Table 2). Among non-Hispanic blacks, rates were significantly lower among males aged 15–29, 30–49, and 50–64 years and women aged 30–49 years. The homicide rates for each age category among Hispanic males aged 15–49 years also were lower in 2009 than in 2007.

State-specific homicide rates for 2009 ranged from 1.1 to 12.8 deaths per 100,000 population, and rates were generally higher in the southern states (Figure 1). Most states did not have any significant changes in homicide rates from 2007 to 2009; however, 10 states experienced significant decreases: Arizona, California, Florida, Georgia, Idaho, Maryland, New Jersey, North Carolina, Ohio, and Pennsylvania. Decreases in rates ranged from 12.4% in California to 55.8% in Idaho. The 2009 crude homicide rate for DC was an estimated 22.8 per 100,000 population.

Discussion

Homicide rates are still particularly high among non-Hispanic black, Hispanic, and non-Hispanic AI/AN populations and remain highest among young, non-Hispanic black males. The findings in this report estimate that 75 out of 100,000 non-Hispanic black males aged 15–29 years die from homicide in a given year. Moreover, 2009 data from the 16 U.S. states that report data on homicides to the National Violent Death Reporting System suggest that nearly half of homicides in this population were outcomes of escalated arguments and conflicts; one third were precipitated by another crime such as burglary, robbery, or assault; one fifth involved illicit drug activity; and approximately 16% were gang related (1,6).

Homicide remains less common among women; however, in 2009, homicide was the sixth leading cause of death among females aged 15–49 years (1). Similar to the findings from the 2011 CHDIR, data from this report indicate that non-Hispanic black and non-Hispanic AI/AN females experience death by homicide more frequently than women in other racial/ethnic populations (10). Female homicides are characteristically different from male homicides in that females are more likely to be killed by a family member during childhood or adolescence (15) and by an intimate partner during adulthood (16). Increased equality between men and women in regards to education, wages, and occupational status might increase women's access to services that prevent intimate partner homicide, such as protective orders, shelters, and advocacy services (17).

Although the findings in this report do not indicate whether a long-term decrease in homicide rates is occurring, rates were noticeably lower in 2009 than in 2007. The decrease in homicide rates between these 2 years was considerable, particularly among males aged 15–29 years, which is consistent with the long-term decreasing trend in homicide rate that has been observed among this demographic population since the early 1990s (1) (Figure 2). Possible explanations for this decreasing homicide rate among young males are reductions in drug trade and sales, increases in police response to youths who carry firearms, and increases in incarceration (17). Despite the decreases, the disparity in homicide rates between non-Hispanic black males and non-Hispanic white males is still pronounced. Although the rate among non-Hispanic black males aged 15–29 years in 2009 is half the rate reported in 1993 (75 vs. 158 per 100,000, respectively) (1), similar decreases have been reported for males of similar age and of other races/ethnicities.

Socioeconomic factors play a substantial role in homicide disparities by race/ethnicity, sex, age, and geographic area. For example, racial/ethnic minorities are more likely to live in disadvantaged neighborhoods (11). Residential areas with high levels of poverty, unemployment, and jobs with low wages can increase risk of income-generating crimes such as burglary and robbery, stress and conflict, and substance abuse among residents (18,19), all factors that increase risk for homicide and violence (11,20). One longitudinal study reported that after controlling for similar socioeconomic factors, such as living in a disadvantaged community, being on welfare, and having a young or single parent, race was not predictive of being a homicide offender (21). Similar risk factors might explain the differences in homicide rates by age and geographic area. Future studies controlling for socioeconomic factors might offer additional support for this conclusion.

Prevention strategies that can change the characteristics of communities, relationships, and persons that are associated with violence perpetration might reduce violence rates not only for all persons in the United States but also among groups with the highest rates of violence (22). Certain promising strategies have been developed that use multicomponent approaches, involve coordinated efforts by numerous relevant stakeholders, and include an appropriate mix of both universal interventions and interventions that address the needs of groups at highest risk for violence (22). Communities That Care (23,24), Promoting School-community-university Partnerships to Enhance Resilience (PROSPER) (25,26), Striving to Reduce Youth Violence Everywhere (STRYVE) (available at http://www.safeyouth.gov), and Urban Networks to Increase Thriving Youth (UNITY) (available at http://preventioninstitute.org/unity.html) are examples of coalition-based operating systems and violence prevention initiatives that can assist communities in developing the type of tailored, broad strategies described. Promising multicomponent programs such as CeaseFire (27) and Safe Streets (28), which work to change community norms regarding violence, cultivate skills for using alternatives to violence, and interrupt escalating tensions, also are promising strategies for preventing general violence, shootings, and shooting-related homicides.

The high homicide rates among youths in late adolescence and young adulthood suggest that the school years are an important developmental point for intervention. Creating a positive school environment is an example of one way to improve youths' access to a safe, stable, nurturing setting; promote norms of nonviolence; facilitate the formation of supportive and positive social relationships; and maximize the development of social and problem-solving skills (29). The influences, experiences, and socialization provided by such school environments could help youths become more adept at navigating problematic interactions and adapting to challenges and difficulties that could lead to serious violence (29). Such school environments might particularly be important for youths who lack other positive influences (29).

Limitations

These findings in this report are subject to at least three limitations. First, small numbers of homicides precluded stable rate estimations among some populations. Second, data on individual and environmental risk factors for homicide were unavailable, which precluded closer examination of possible sources of disparities by age, sex, race/ethnicity, and geography. Third, racial misclassification might result in overestimated homicide rates for non-Hispanic blacks and non-Hispanic whites and underestimated rates for AI/ANs, A/PIs, and Hispanics (8).

Conclusion

Effective evidence-based strategies to reduce violence are available (30); however, additional work is needed to build organizational and community capacity to make best use of these programs, policies, and strategies. Many health-related disparities can be reduced by altering influential socially embedded conditions such as 1) neighborhood living conditions, 2) opportunities for learning and capacity for development, and 3) employment opportunities and community development (31,32). Because these outcomes mediate the effects of social determinants of health, they might be viable mechanisms for changing or eliminating social influences that create or increase disparities in homicide rates. Promising strategies such as implementing business or community improvement districts might help decrease levels of violent crimes by increasing employment opportunities for local residents and creating physical or cultural environments that are more aesthetically and economically attractive (33,34). These community-level strategies might reduce or offset the effects of poverty, improve the social environments of communities, and implement safety measures (33,34). To eliminate homicide disparities, more research is needed to understand the scope of the problem and the risk and protective factors implicated in these violent events, evaluate programs that prevent and reduce violence, and better understand how to adapt, disseminate, and implement these strategies in the communities and populations in greatest need.

Acknowledgment

The findings in this report are based, in part, on contributions by Nimeshkumar Patel.

References

1. CDC. Web-based Injury Statistics Query and Reporting System (WISQARS). Atlanta, GA: US Department of Health and Human Services, CDC; 2012. Available at http://www.cdc.gov/injury/wisqars/index.html.
2. Rheingold AA, Zinzow H, Hawkins A, Saunders BE, Kilpatrick DG. Prevalence and mental health outcomes of homicide survivors in a representative U.S. sample of adolescents: data from the 2005 National Survey of Adolescents. J Child Psychol Psychiatry 2012;53:687–94.
3. Schiavone DB. The effects of exposure to community violence on aspects of adolescent identity development. J Child Adolesc Psychiatr Nurs 2009;22:99–105.
4. Hemenway D. Measuring the cost of injury: underestimating the costs of street violence. Inj Prev 2011;17:289–90.
5. Shahpar C, Li G. Homicide mortality in the United States, 1935–1994: age, period, and cohort effects. Am J Epidemiol 1999;150:1213–22.
6. CDC. Surveillance for violent deaths—National Violent Death Reporting System, 16 states, 2007. MMWR 2010;59(No. SS-4).
7. Miniño AM. Mortality among teenagers aged 12–19 years: United States, 1999–2006 Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics; 2010. NCHS data brief no. 37. Available at http://www.cdc.gov/nchs/data/databriefs/db37.htm.
8. CDC. Fatal injuries among children by race and ethnicity—United States, 1999–2002. MMWR 2007;56(No. SS-5).
9. Najem GR, Aslam S, Davidow AL, Elliot N. Youth homicide racial disparities: gender, years, and cause. J Natl Med Assoc 2004;96:558–66.
10. CDC. Homicides—United States, 1999–2007. In: CDC health disparities and inequalities report—United States, 2011. MMWR 2011;60(Suppl; January 14, 2011).
11. Krueger PM, Bond Huie SA, Rogers RG, Hummer RA. Neighbourhoods and homicide mortality: an analysis of race/ethnic differences. J Epidemiol Community Health 2004;58:223–30.
12. CDC. CDC health disparities and inequalities report—United States, 2011. MMWR 2011; 60(Suppl; January 14, 2011).
13. CDC. Introduction. In: CDC health disparities and inequalities report—United States, 2013. MMWR 2013;62(No. Suppl 3).
14. Xu J, Kockanek KD, Murphy SL, Tejada-Vera B. Deaths: final data for 2007. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics. Natl Vital Stat Rep 2010;58.
15. Muftić LR, Moreno RD. Juvenile homicide victimization: differences and similarities by gender. Youth Violence Juv Justice 2010;8:386–98.
16. Vieraitis LM, Britto S, Morris RG. Assessing the impact of changes in gender equality on female homicide victimization: 1980–2000. Crime Delinq. Epub November 8, 2011. Available at http://cad.sagepub.com/content/early/2011/09/10/0011128711420100.
17. Blumstein A, Rivara FP, Rosenfeld R. The rise and decline of homicide—and why. Annu Rev Public Health 2000;21:505–41.
18. Gould ED, Weinberg BA, Mustard DB. Crime rates and local labor market opportunities in the United States: 1979–1997. Rev Econ Stat 2002;84:45–61.
19. Hill TD, Angel RJ. Neighborhood disorder, psychological distress, and heavy drinking. Soc Sci Med 2005;61:965–75.
20. Loeber R. The prediction of violence and homicide in young men. J Consult Clin Psychol 2005;73:1074.
21. Farrington DP, Loeber R, Berg MT. Young men who kill. Homicide Stud 2012;16:99–128.
22. Krug EG, Dahlberg LL, Mercy JA, Zwi AB, Lozano R, eds. World report on violence and health. Geneva, Switzerland: World Health Organization; 2002.
23. Hawkins JD. Preventing crime and violence through communities that care. Eur J Crim Policy Res 1999;7:443–58.
24. Hawkins JD, Catalano RF, Arthur MW. Promoting science-based prevention in communities. Addict Behav 2002;27:951–76.
25. Spoth R, Greenberg M, Bierman K, Redmond C. PROSPER community—university partnership model for public education systems: capacity-building for evidence-based, competence-building prevention. Prev Sci 2004;5:31–9.
26. Spoth RL, Greenberg MT. Toward a comprehensive strategy for effective practitioner–scientist partnerships and larger-scale community health and well-being. Am J Community Psychol 2005;35:107–26.
27. Skogan WG, Hartnett SM, Bump N, Dubois J. Evaluation of CeaseFire—Chicago. Chicago, IL: Northwestern University; 2008.
28. Webster DW, Vernick JS, Mendel J. Interim evaluation of Baltimore's Safe Streets Program. Baltimore, MD: Johns Hopkins Bloomberg School of Public Health, Center for the Prevention of Youth Violence; 2009.
29. Hahn R, Fuqua-Whitley D, Wethington H, et al. Effectiveness of universal school-based programs to prevent violent and aggressive behavior: a systematic review. Am J Prev Med 2007;33(Suppl):S114–29.
30. Mihalic S, Irwin K, Elliot D, Fagan A, Hansen D. Blueprints for violence prevention. Washington, DC: US Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention; 2001.
31. Anderson LM, Scrimshaw SC, Fullilove MT, Fielding JE. The Community Guide's model for linking the social environment to health. Am J Prev Med 2004;24:12–20.
32. Williams DR, Costa MV, Odunlami AO, Mohammed SA. Moving upstream: how interventions that address the social determinants of health can improve health and reduce disparities. J Public Health Manag Pract 2008;14:S8–17.
33. MacDonald JM, Blumenthal RN, Golineli D, et al. Neighborhood effects on crime and youth violence: the role of business improvement districts in Los Angeles. Santa Monica, CA: Rand; 2009.
34. MacDonald J, Golinelli D, Stokes RJ, Bluthenthal R. The effect of business improvement districts on the incidence of violent crimes. Inj Prev 2010;16:327–32.