Surveillance for Violent Deaths --- National Violent Death Reporting System, 16 States, 2007

Debra L. Karch, PhD

Linda L. Dahlberg, PhD

Nimesh Patel, MS

Division of Violence Prevention, National Center for Injury Prevention and Control, CDC

Corresponding Author: Debra L. Karch, PhD, Division of Violence Prevention, National Center for Injury Prevention and Control, CDC, 4770 Buford Hwy, N.E., MS F-64, Atlanta, GA, 30341-3724. Telephone: 770-488-1307; Fax: 770-488-4222; E-mail: [email protected].

Abstract

Problem/Condition: An estimated 50,000 persons die annually in the United States as a result of violence-related injuries. This report summarizes data from CDC's National Violent Death Reporting System (NVDRS) regarding violent deaths from 16 states for 2007. Results are reported by sex, age group, race/ethnicity, marital status, location of injury, method of injury, circumstances of injury, and other selected characteristics.

Reporting Period Covered: 2007.

Description of System: NVDRS collects data regarding violent deaths obtained from death certificates, coroner/medical examiner reports, and law enforcement reports. NVDRS began operation in 2003 with seven states (Alaska, Maryland, Massachusetts, New Jersey, Oregon, South Carolina, and Virginia) participating; six states (Colorado, Georgia, North Carolina, Oklahoma, Rhode Island, and Wisconsin) joined in 2004, four (California, Kentucky, New Mexico, and Utah) in 2005, and two states (Ohio and Michigan) were funded to begin data collection in 2010, totaling 19 states. This report includes data from 16 states that collected statewide data in 2007. California data are not included in this report because NVDRS data are collected only in a limited number of California cities and counties rather than statewide. Ohio and Michigan are excluded because they did not begin data collection until 2010.

Results: For 2007, a total of 15,882 fatal incidents involving 16,319 deaths occurred in the 16 NVDRS states included in this report. The majority (56.6%) of deaths was suicides, followed by homicides and deaths involving legal intervention (i.e., deaths caused by police and other persons with legal authority to use deadly force, excluding legal executions) (28.0%), deaths of undetermined intent (14.7%), and unintentional firearm deaths (0.7%). Suicides occurred at higher rates among males, American Indians/Alaska Natives, non-Hispanic whites, and persons aged 45--54 years. Suicides occurred most often in a house or apartment and involved the use of firearms. Suicides were precipitated primarily by mental-health, intimate-partner, or physical-health problems, or by a crisis during the preceding 2 weeks. Homicides occurred at higher rates among males and persons aged 20--24 years; rates were highest among non-Hispanic black males. The majority of homicides involved the use of a firearm and occurred in a house or apartment or on a street/highway. Homicides were precipitated primarily by arguments and interpersonal conflicts or in conjunction with another crime. Other manners of death and special situations or populations also are highlighted in this report.

Interpretation: This report provides a detailed summary of data from NVDRS for 2007. The results indicate that violent deaths resulting from self-inflicted or interpersonal violence disproportionately affected adults aged <55 years, males, and certain minority populations. For homicides and suicides, relationship problems, interpersonal conflicts, mental-health problems, and recent crises were among the primary precipitating factors. Because additional information might be reported subsequently as participating states update their findings, the data provided in this report are preliminary.

Public Health Action: For the occurrence of violent deaths in the United States to be better understood and ultimately prevented, accurate, timely, and comprehensive surveillance data are necessary. NVDRS data can be used to monitor the occurrence of violence-related fatal injuries and assist public health authorities in the development, implementation, and evaluation of programs and policies to reduce and prevent violent deaths at the national, state, and local levels. The continued development and expansion of NVDRS is essential to CDC's efforts to reduce the personal, familial, and societal costs of violence. Further efforts are needed to increase the number of states participating in NVDRS, with an ultimate goal of full national representation.

Introduction

An estimated 50,000 persons die annually in the United States from violence-related injuries. Homicide is the second leading cause of death for persons aged 15--24 years, the third leading cause for persons aged 10--14 and 25--34 years, and the fourth leading cause for persons aged 1--9 years. Suicide is the second leading cause of death for persons aged 25--34 years, the third leading cause for persons aged 15--24 years, and the fourth leading cause for persons aged 10--14 and 35--44 years. Only unintentional injury in those aged 1--34 years and malignant neoplasms and congenital anomalies in children aged 1--14 years were more common (1).

Public health authorities require accurate, timely, and comprehensive surveillance data to better understand and ultimately prevent the occurrence of violent deaths in the United States (2). In 2000, CDC started planning to implement the National Violent Death Reporting System (NVDRS) (3,4). The goals of this system are to:

• collect and analyze timely, high-quality data that monitor the magnitude and characteristics of violent death at the national, state, and local levels;
• ensure that data are disseminated routinely and expeditiously to public health officials, law enforcement officials, policy makers, and the public;
• ensure that data are used to develop, implement, and evaluate programs and policies that are intended to reduce and prevent violent deaths and injuries at the national, state, and local levels; and
• build and strengthen partnerships among organizations and communities at the national, state, and local levels to ensure that data are collected and used to reduce and prevent violent deaths and injuries.

NVDRS was conceived as a state-based active surveillance system that would collect risk-factor data concerning all violence-related deaths, including homicides, suicides, and legal intervention deaths (i.e., deaths caused by police and other persons with legal authority to use deadly force, excluding legal executions) as well as unintentional firearm deaths and deaths of undetermined intent. NVDRS data are used to assist the development, implementation, and evaluation of programs and policies designed to reduce and prevent these deaths and injuries at the national, state, and local levels.

Before implementation of NVDRS, single data sources (e.g., death certificates or supplemental homicide reports) provided limited information and circumstances from which to understand patterns of deaths collected by this system. NVDRS fills this gap in national surveillance; it is the first system to provide detailed information on circumstances precipitating violent deaths, the first to link multiple source documents to enable researchers to understand each death more completely, and the first to link multiple deaths that are related to one another (e.g., multiple homicides, suicide pacts, and cases of homicide followed by the suicide of the suspected perpetrator).

NVDRS began operation in 2003 with seven states (Alaska, Maryland, Massachusetts, New Jersey, Oregon, South Carolina, and Virginia) participating; six states (Colorado, Georgia, North Carolina, Oklahoma, Rhode Island, and Wisconsin) joined in 2004, four (California, Kentucky, New Mexico, and Utah) in 2005, and two states (Ohio and Michigan) were funded to begin data collection in 2010, totaling 19 states (Figure). CDC provides funding for state participation and anticipates that NVDRS will expand to include all 50 states, the District of Columbia, and U.S. territories.

This report summarizes data for 2007 concerning deaths meeting NVDRS inclusion criteria from 16 states that collected statewide data (approximately 26% of the U.S. population). California data are not included in this report because NVDRS data are collected only in a limited number of California cities and counties rather than statewide. Ohio and Michigan were excluded because they did not begin data collection until 2010. Because additional information might be reported subsequently as participating states update their findings, the data provided in this report are preliminary. Annual updates of NVDRS data also are available through a web-based query system at http://wisqars.cdc.gov:8080/nvdrs/nvdrsDisplay.jsp.

Methods

NVDRS uses multiple, complementary data sources including death certificates, coroner/medical examiner (CME) records, and law enforcement reports. Some participating states use secondary sources (e.g., child fatality review team data, supplementary homicide reports, hospital data, and crime laboratory data). Bureau of Alcohol, Tobacco, Firearms, and Explosives traces information concerning firearms. NVDRS links together multiple documents for each violent death and also links multiple deaths that are related to each other (e.g., multiple homicides, a homicide followed by a suicide, or multiple suicides) into a single incident. The ability to analyze data linked in this way permits a comprehensive assessment of violent deaths.

NVDRS defines a violent death as a death resulting either from the intentional use of physical force or power against oneself, another person, or a group or community. In addition, NVDRS collects information regarding unintentional firearm injury deaths (i.e., incidents in which the person causing the injury did not intend to discharge the firearm) and deaths of undetermined intent. NVDRS case definitions are coded on the basis of the International Classification of Diseases, 10th Revision (ICD-10) (5). Cases with selected ICD-10 codes are included in NVDRS (Box 1). ICD-10 case finding is completed by participating states.

Variables analyzed in NVDRS include the following:

• manner of death (i.e., the intent of the person inflicting a fatal injury);
• mechanism of injury (i.e., the method used to inflict a fatal injury);
• circumstances preceding injury (i.e., the precipitating events that led to the infliction of a fatal injury);
• whether the decedent was a victim (i.e., a person who died because of a violence-related injury);
• whether the decedent was a suspect (i.e., a person believed to have inflicted a fatal injury on a victim);
• whether the decedent was both a suspect and a victim (i.e., a person believed to have inflicted a fatal injury on a victim and then was fatally injured himself or herself);
• incident (i.e., an occurrence in which one or more persons sustained a fatal injury that was linked to a common event during a 24-hour period); and
• type of incident (i.e., a combination of the manner of death and the number of victims in an incident).

NVDRS is incident-based, and all decedents (both victims and alleged perpetrators [suspects]) associated with a given incident are grouped in one record. Decisions about whether two or more deaths belong to the same incident are made on the basis of the timing of the injuries rather than on that of the deaths. Examples of a violent death incident include 1) a single isolated violent death, 2) two or more related homicides (including legal interventions) when the fatal injuries were inflicted <24 hours apart, 3) two or more related suicides or deaths of undetermined intent when the fatal injuries were inflicted <24 hours apart, and 4) a homicide followed by a related suicide when both fatal injuries were inflicted <24 hours apart.

Data are obtained from individual information sources and entered into source-specific computerized data entry screens (i.e., police report data are entered into police report screens and death certificate data into death certificate screens). In addition to allowing independent entry of each source, this approach permits later review of what each source contributed and identification of missing sources. This permits comparisons of the quality and completeness of state-specific data sources and allows states to provide feedback to sources regarding the consistency of their data compared with data from other sources. In addition, the system permits automatic electronic importation of specific data sources without requiring manual entry.

Abstraction of identical variables across multiple source documents can result in data inconsistencies, which NVDRS resolves by assigning a primacy (i.e., hierarchical) rule for each variable. The primacy rules are applied to create a final analysis data set that uses data from all available sources. For each variable in NVDRS, primacy is established on the basis of a hierarchy of assumed reliability of all the sources for a single variable. For example, sex is collected in all three required documents (death certificate, CME record, and police report). The primacy for sex is expressed as death certificate/CME record/police report, which means the analysis file is constructed using the sex recorded in the death certificate; if this is left blank or is unknown, the sex recorded in the CME record is used; and if the CME record does not provide the sex or lists the sex as unknown, the police report is used.

Manner of Death

A manner (i.e., intent) of death for each decedent is assigned by a trained abstractor who takes into account information from all source documents. Typically, these documents are consistent regarding the manner of death, and the abstractor-assigned manner of death corresponds to that reported in all the source documents. On rare occasions, when a discrepancy exists among the source documents, the abstractor must assign a manner of death on the basis of the preponderance of evidence in the source documents. For example, if two sources classify a death as a suicide and a third classifies it as undetermined, the death will be coded as a suicide.

NVDRS classifies data using one of five abstractor-assigned manners of death:

• Suicide. Suicide is defined as a death resulting from the use of force against oneself when a preponderance of the evidence indicates that the use of force was intentional. This category includes deaths of persons who intended only to injure rather than kill themselves, deaths associated with risk taking behavior without clear intent to inflict fatal injury but associated with high risk of death (e.g., Russian roulette) and suicides involving only passive assistance to the decedent (e.g., supplying the means or information needed to complete the act). The category does not include deaths caused by chronic or acute substance abuse without the intent to die or deaths attributed to autoerotic behavior (e.g., self-strangulation during sexual activity). Corresponding ICD-10 codes included in NVDRS are X60--X84 and Y87.0.
• Homicide. Homicide is defined as a death resulting from the use of physical force or power, threatened or actual, against another person, group, or community when a preponderance of evidence indicates that the use of force was intentional. Two special scenarios that the National Center for Health Statistics (NCHS) regards as homicides are included in the NVDRS definition: 1) arson with no intent to injure a person and 2) a stabbing with intent unspecified. This category excludes vehicular homicide without intent to injure, unintentional firearm deaths (a separate category listed below), combat deaths or acts of war, and deaths of unborn fetuses. Corresponding ICD-10 codes included in NVDRS are X85--X99, Y00--Y09, and Y87.1.
• Unintentional firearm. The term "unintentional firearm death" is used when a death results from a penetrating injury or gunshot wound from a weapon that uses a powder charge to fire a projectile and for which a preponderance of evidence indicates that the shooting was not directed intentionally at the decedent. Examples of deaths included in this category include the death of a person as a result of celebratory firing that was not intended to frighten, control, or harm anyone; a soldier shot during a field exercise but not in a combat situation; and a person who received a self-inflicted wound while playing with a firearm. This category excludes firearm injuries caused by unintentionally striking a person with the firearm (e.g., hitting a person on the head with the firearm rather than firing a projectile) and unintentional injuries from nonpowder guns (e.g., BB, pellet, or other compressed air- or gas-powered guns). Corresponding ICD-10 codes included in NVDRS are W32--W34 and Y86 with a method of firearm.
• Undetermined intent. The term "undetermined intent" is used when a death results from the use of force or power against oneself or another person for which the evidence indicating one manner of death is no more compelling than evidence indicating another. This category includes CME rulings (e.g., accident or suicide, undetermined, jumped or fell, self-inflicted injuries) when records give no evidence or opinions in favor of either unintentional or intentional injury. Corresponding ICD-10 codes included in NVDRS are Y10--Y34, Y87.2, and Y89.9.
• Legal intervention. The term "legal intervention" is used when a decedent is killed by a police officer or other peace officer (a person with specified legal authority to use deadly force), including military police, acting in the line of duty. This category excludes legal executions. Corresponding ICD-10 codes included in NVDRS are Y35.0--Y35.4, Y35.6, Y35.7, and Y89.0.

Comparability of NVDRS Surveillance Summary Data Prior to 2006

Four changes were made to how variables were reported between 2005 and 2006 that affect their comparability; no variable changes were made between 2006 and 2007. The 2005 to 2006 changes involve race/ethnicity, location of injury, relationship of victim to suspect, and method of injury. In 2005, the race variable was reported in six categories (white, black, Asian Pacific Islander [API], American Indian/Alaska Native [AI/AN], other, and unknown). Ethnicity was categorized separately as persons of any race that reported Hispanic origin. When this methodology was used, Hispanics were reported both within their race category and then again separately by ethnicity. The 2006 and 2007 methodology classifies each person as non-Hispanic white, non-Hispanic black, API, AI/AN, Hispanic, other, and unknown. Race and ethnicity are combined in one variable. This change allows for better comparability with other violence-related data.

Location of injury is coded from a list of 31 location options in NVDRS. Because certain options are selected rarely, certain response categories have been combined. In 2006 and 2007, the category "bank" was included in "office building" rather than in "commercial/retail area" as it was in 2005. Also in 2006 and 2007, the category "synagogue/church/temple" was subsumed under "other" and not reported separately as in 2005.

Relationship of the victim to the suspect includes a new category, "other intimate-partner involvement," to refer to a death that is intimate-partner--related but that does not occur between the intimate partners themselves (e.g., when a child is killed by a parent's partner). In addition, the categories "rival gang member" and "victim was injured by a law enforcement officer" are reported in 2006 and 2007 as separate categories; in 2005, these categories were included in "other specified relationship." The categories "foster child" and "foster parent" also were moved from "other relative" to "child" and "parent," respectively.

Four new categories were added to method of injury in 2006 and 2007: "firearm and poisoning," "firearm and other method type," "poisoning and other method type," and "other combination of methods." All deaths in these new categories involved more than one method, and the evidence did not indicate which method caused the fatal injury. For example, a homicide victim might have injuries from both a firearm and a sharp instrument, but the method that actually caused the fatal injury might be unclear. In this case, the method of injury would be categorized as "firearm and other method."

Variables Analyzed

NVDRS collects approximately 250 unique variables (available at http://www.cdc.gov/violenceprevention/nvdrs). The number of variables recorded for each incident depends on the content and completeness of the source documents. Variables include manner of death, demographics, ICD-10 and underlying cause-of-death codes and text, location and date/time of injury and death, toxicology results, bodily injuries, precipitating circumstances, decedent-suspect relationship, and method of injury (Boxes 2 and 3).

Circumstances Preceding Death

The circumstances preceding death are defined as the precipitating events that led to the infliction of a fatal injury (Box 3). The circumstances that preceded a fatal injury are reported on the basis of the content of the CME record and police reports. Different sets of circumstances are coded for suicide/undetermined deaths, homicide/legal-intervention deaths, and unintentional firearm deaths. The variable "circumstances known" is a gateway variable to a list of potential circumstances. Each incident requires the data abstractor to code all circumstances in cases for which the circumstances are known. If circumstances are not known (e.g., for a body found in the woods with no other detail) the data abstractor leaves the gateway variable blank, and these cases are excluded from the denominator for circumstance values. If either the CME record or the police report indicates that the circumstance is reported to be true, then the abstractor enters data as confirmed (e.g., if the police report indicated that a decedent had disclosed an intent to commit suicide, then suicidal intent is accepted to be true).

Coding Training and Quality Control

Coding training is held annually for all participating states. Ongoing coding support is provided through an e-mail help desk, monthly conference calls with all states, and regular conference calls with individual states. A coding manual is provided. Software features enhance coding reliability, including automated validation rules and a hover-over feature containing variable-specific information. Details regarding NVDRS procedures and coding are available at http://www.cdc.gov/violenceprevention/nvdrs/publications.htm.

States are requested to perform blind reabstraction of cases using multiple abstractors to identify inconsistencies. CDC also runs a quality-control analysis in which multiple variables are reviewed for their appropriateness, with special focus on abstractor-assigned variables (e.g., method selection and manner of death). If CDC questions any variable, CDC notifies the state for a response or correction.

Time Frame

States are required to report all deaths within 6 months of the end of each calendar year for the preceding January--December time frame. States then have an additional 12 months to complete each incident record. Although states typically meet these timelines, additional details sometimes arrive after a deadline has passed. New incidents also might be identified after the deadline (e.g., if a death certificate is revised, new evidence is obtained that changes a manner of death, or a miscoded ICD-10 is corrected to meet NVDRS inclusion criteria). These additional data are incorporated into NVDRS. Analysis files are updated monthly at CDC. On the basis of previous experience, CDC estimates that case counts might increase 1%--2% after the initial 18-month data collection period.

Fatal Injuries During 2007

This report provides preliminary data concerning fatal injuries meeting the NVDRS case definition in 2007 for 16 participating states that were received by CDC as of August 31, 2009. Data from California were not included in this report because NVDRS was implemented only in a limited number of cities and counties rather than statewide. Participating states used vital statistics death certificate files to identify deaths meeting NVDRS case definitions. Each state reported all deaths of their residents that occurred within the state and deaths of state residents that occurred elsewhere. Once a death was identified, NVDRS data abstractors linked source documents, linked deaths within each incident, coded data elements, and wrote a short narrative of the incident. These narratives were reviewed for all incidents in which coded data were unclear or incomplete. State-level data then were consolidated and analyzed for this aggregate report. Numbers, percentages, and crude rates are presented in aggregate for all deaths by abstractor-assigned manner of death and for special situations and populations (e.g., homicide followed by suicide, suicides of former or current military personnel, and intimate-partner--related homicides). Rates for cells with a frequency of <20 are not reported because of the instability of those rates. In addition, rates could not be calculated for variables (e.g., marital status, precipitating circumstances) because denominators were unknown. Bridged-race 2007 population estimates were used as denominators in the rate calculations (6). For compatible numerators for rate calculations to be derived, person records listing multiple races were recoded to a single race when possible, using a bridging algorithm provided by NCHS (available at http://www.cdc.gov/nchs/nvss/bridged_race.htm).

Results

All Deaths

Deaths by Manner, Method, and Location

The 16 NVDRS states included in this report collected data concerning 15,882 incidents and 16,319 deaths that occurred during 2007. The crude death rate was 19.9 deaths per 100,000 population. Suicides (n = 9,245) accounted for the highest rate of violent death (11.6 per 100,000 population) followed by homicide/legal-intervention (n = 4,563) deaths (5.7 per 100,000 population). Deaths of undetermined intent (n = 2,403) and unintentional firearm deaths (n = 107) occurred at lower rates (3.0 and 0.1 per 100,000 population, respectively). Of all incidents occurring in 2007 in the 16 states included in this report, 2.2% were known to have multiple victims. Firearms accounted for 48.2% of injury deaths, poisoning for 20.4%, and hanging/strangulation/suffocation for 14.2% (rates: 9.9, 4.2, and 2.9 per 100,000 population, respectively); rates for other methods were lower. For all deaths, a house or apartment was the most common location (69.9%). The next-most-common location of injury (8.2%) was a street or highway (Table 1).

Toxicology Results of Decedent

Tests for alcohol were conducted for 73.8% of decedents, and drug tests for amphetamines, antidepressants, cocaine, marijuana, and opiates were conducted for 52.5%, 43.2%, 55.8%, 36.5%, and 54.1% of decedents, respectively. Among decedents who tested positive for alcohol (33.4%), 59.1% had a blood alcohol concentration (BAC) of >0.08 g/dL (the legal limit in the majority of states). Opiates, including heroin and prescription pain killers, were identified in 26.2% of cases tested for these substances (antidepressants [23.5%], cocaine [13.5%], marijuana [11.1%], and amphetamines [4.6%]) (Table 2).

Suicides

Sex, Race/Ethnicity, Age Group, and Marital Status

The 16 NVDRS states included in this report collected data concerning 9,233 fatal suicide incidents and 9,245 suicides that occurred during 2007. Rates of suicide by month showed little variation throughout the year (range: 0.8--1.1 per 100,000 population) (Table 3). Overall, the crude suicide rate was 11.6 per 100,000 population. The rate for males was more than three times that for females (18.4 and 5.0 per 100,000 population, respectively). Non-Hispanic whites accounted for the largest number of suicide deaths, and AI/ANs and non-Hispanic whites had the highest rates of suicide (18.2 and 14.0 per 100,000 population, respectively). The highest rates of suicide by age group occurred among persons aged 45--54 years, 75--84 years, and 35--44 years (17.6, 16.4, and 16.3 per 100,000 population, respectively). Children aged 10--14 years had the lowest rates of suicide among all age groups (0.8 per 100,000 population). Rates of suicide among adolescents aged 15--19 years (6.9 per 100,000 population) were approximately half of those for persons aged ≥30 years (Table 4).

Decedents aged 35--64 years accounted for 55.0% of suicide deaths among males. Rates among males were highest for those aged ≥85 years followed by those aged 75--84 years (43.9 and 35.8 per 100,000 population, respectively). AI/AN males had the highest rates of any racial/ethnic population and had rates that were more than four times the rate for API males. Among females, decedents aged 35--64 years accounted for 65.0% of suicides. Rates for females peaked at 9.0 per 100,000 among those aged 45--54 years. As with males, female suicide rates were highest among AI/ANs (7.3) followed closely by non-Hispanic whites (6.2). Among females, the lowest rates of suicide were among non-Hispanic blacks (1.5) and Hispanics (1.7). Of all decedents aged ≥18 years for which marital status was known, 38.7% were married, 29.3% had never married, and 22.1% were divorced at the time of death (Table 4).

Method and Location of Injury

Firearms were used in the majority (50.7%) of suicide deaths, followed by hanging/strangulation/suffocation (23.1%) and poisoning (18.8%) (Table 5). The most common method used by male suicide decedents was a firearm (56.0%) followed by hanging/strangulation/suffocation (24.4%). Among females, poisons were used most often (40.8%) followed by firearms (31.9%). The most common place of self-inflicted injury was a house or apartment (77.2%) followed by natural areas (4.4%), and streets or highways (3.0%). A total of 120 (1.3%) suicides occurred in a jail or prison setting (115 males and 5 females) (Table 5).

Toxicology Results of Decedent and Precipitating Circumstances

Tests for alcohol were conducted for 68.9% of suicide decedents, and drug tests for amphetamines, antidepressants, cocaine, marijuana, and opiates were conducted for 44.2%, 39.8%, 47.2%, 35.7%, and 47.1% of suicide decedents, respectively. Among suicide decedents who tested positive for alcohol (34.5%), 62.0% had a BAC of >0.08 g/dL. Opiates (e.g., heroin, prescription pain killers) were identified in 22.8% of cases tested for these substances; cocaine and marijuana were identified in 8.9% and 7.5% of tested cases, respectively. Of suicide decedents who were tested for antidepressants, 30.0% were positive at the time of their death (Table 6).

Precipitating circumstances were known for approximately 86% of suicide decedents. Overall, mental-health problems were the most commonly noted circumstance for suicide decedents with 41.5% described as experiencing a depressed mood at the time of their deaths. Nearly 45% were described as having a mental-health problem although only 33.8% were receiving treatment (Table 7). Among those with a current mental-health problem, 74.9% had received a diagnosis of depression/dysthymia, 14.5% had been diagnosed with bipolar disorder, and 8.1% with an anxiety disorder (Table 8). Among suicide decedents with known circumstance information, 19.9% had a history of previous suicide attempts, 28.0% disclosed their intent prior to dying, and 33.3% left a suicide note (Table 7). Other than mental health conditions, circumstances noted most often were a crisis in the preceding 2 weeks or intimate-partner problems, each indicated in approximately 30% of suicides with known circumstance information. Physical-health problems also were noted in 21.4% of cases with circumstance information.

Similar percentages of male and female suicide decedents were observed to have a depressed mood at the time of death; however, a higher percentage of females than males had a current mental-health problem (62.9% and 39.6%, respectively) or were being treated for a mental-health problem (50.6% and 28.8%, respectively). Approximately the same percentage of male and female suicide decedents experienced physical-health problems in the period before their deaths, although a higher percentage of males than females had job, financial, or criminal problems in the period preceding their deaths. Intimate-partner problems also were cited as a precipitating factor in a higher percentage of male suicides than female suicides (31.7% and 26.4%, respectively). Although occurring in only a limited percentage of cases, being a perpetrator of interpersonal violence in the month before death was more common among male suicide decedents (5.6%) than being a victim of such violence (0.2%) whereas the proportions were similar for females (1.6% and 1.1%, respectively) (Table 7).

Homicides

Sex, Race/Ethnicity, Age Group, Marital Status

The 16 NVDRS states included in this report collected data concerning 4,324 homicide incidents and 4,563 homicides that occurred during 2007. Overall, the crude homicide rate was 5.7 deaths per 100,000 population in 2007. Rates of homicide by month showed little variation throughout the year (range: 0.4--0.6 per 100,000 population) (Table 9).

The majority (53.0%) of homicide decedents aged ≥18 years for which marital status was known had never been married, and 22.7% were married at the time of their death. In 41.2% of homicides, the relation of the victim to the suspect was not known. When a suspect was identified, the suspect most often was an acquaintance or friend (15.7%), a spouse or intimate partner (10.5%), or a stranger (8.4%). Perpetrators were other relatives of the decedent in <10% of cases with known information about the relation of the victim to the suspect (Table 10).

The homicide rate for males was approximately 3.6 times that for females (9.0 and 2.5 per 100,000 population, respectively). Non-Hispanic blacks accounted for the majority (52.0%) of homicide deaths and had the highest rate (19.3 deaths per 100,000 population) followed by AI/ANs (10.5) and Hispanics (7.2). Age-specific homicide rates were highest (14.7 deaths per 100,000 population) among those aged 20--24 years followed by those aged 25--29 years (12.4 deaths per 100,000 population). The rate for infants aged <1 year was more than four times that for children aged 1--4 years (9.7 and 2.2 per 100,000 population, respectively) and similar to that for adolescents aged 15--19 years (9.5 per 100,000 population). Rates were lowest among children aged 5--14 years and persons aged ≥65 years. The majority (62.3%) of male homicide decedents were aged 20--44 years; males aged 20--24 years had the highest rates of homicide (24.5 per 100,000 population). For females, homicide rates were highest (8.2 deaths per 100,000 population) among infants aged <1 year (Table 11).

Method and Location of Injury

Firearms were used in 66.1% of homicides, followed by sharp instruments (12.1%) and blunt instruments (5.8%). No other single method was used in more than 3.4% of homicides (Table 9). Firearms were the most common method used in homicides of males (71.7%) and females (46.4%). Hanging/strangulation/suffocation was over six times more common among female homicide decedents than among males (8.6% and 1.4%, respectively). A house or apartment was the most common location of homicide for both males and females (44.7% and 72.6%, respectively). The next-most common location of homicide for males was a street or highway (26.0%), a parking lot or public garage (5.4%), and a motor vehicle (4.0%); for females, the next-most common locations were a street or highway (6.8%), or a commercial/retail or natural area (2.7% each) (Table 12).

Toxicology Results of Decedent and Precipitating Circumstances

Tests for alcohol were conducted for 77.9% of homicide decedents, and drug tests for amphetamines, antidepressants, cocaine, marijuana, and opiates were conducted for 54.4%, 34.8%, 58.8%, 33.8%, and 52.4% of homicide decedents, respectively. Among homicide decedents who tested positive for alcohol (34.6%), 55.4% had a BAC of >0.08 g/dL. Marijuana, cocaine, and opiates were identified in 20.0%, 15.7%, and 6.8% of homicide decedents tested, respectively (Table 13).

Precipitating circumstances were identified for 64.9% of homicide deaths. Approximately one third of those homicides were precipitated by another crime. In 76.0% of cases precipitated by another crime, the crime was in progress at the time of the incident (Table 14). The crime was most often robbery (37.4%), followed by assault (23.3%), burglary (9.3%), drug-related (8.2%), rape/sexual assault (3.3%), or motor-vehicle theft (3.0%) (Table 15). Other common precipitating circumstances were an argument, abuse, or conflict over something other than money or property (37.5%); drug-related circumstances (13.3%); justifiable self defense (8.0%); or an argument about money or property (5.4%). In 20.4% of cases with known circumstance information, intimate-partner violence was identified as a contributing factor. In approximately 1% of the cases, the decedent was a police officer killed in the line of duty or an intervening person assisting a crime victim (Table 14).

An argument, abuse, or a conflict unrelated to money or property was a factor in more homicides among males than females (41.9% and 24.5%, respectively). Drug-related homicides accounted for 15.0% of male homicides and 8.2% of female homicides. Intimate-partner violence was a precipitating factor in 52.5% of female homicides but only 9.5% of male homicides. In 13.7% of male homicides with known circumstance information, the decedent also used a weapon during the altercation, compared with 1.9% of female homicides (Table 14).

Deaths of Undetermined Intent

Sex, Race/Ethnicity, Age Group, Education, and Marital Status

The 16 NVDRS states included in this report collected data concerning 2,392 incidents involving 2,403 deaths during 2007 for which a determination of intent could not be made. Rates of undetermined death by month were at 0.2 or 0.3 per 100,000 population throughout the year (Table 16). Overall, the crude rate for undetermined deaths was 3.0 per 100,000 population. Rates of undetermined death were higher among males than females (3.8 and 2.3 per 100,000 population, respectively). Although non-Hispanic whites accounted for 71.5% of undetermined deaths, rates were highest among AI/ANs (6.1 per 100,000 population). Nearly half (49.1%) of decedents for whom the manner of death was undetermined were aged 35--54 years. Rates were highest (22.0 per 100,000 population) among infants aged <1 year. Among decedents aged >18 years with an undetermined manner of death for which marital status was known, 39.3% had never been married, 27.9% were married, and 24.3% were divorced at the time of death. AI/AN males had the highest rates (7.5 per 100,000 population) of undetermined death compared with males or females of any other racial/ethnic population (Table 17).

Method and Location of Injury

The most common method of injury was poisoning (65.0%). No other known single method accounted for >2.7% of undetermined deaths. Among both males and females for which the method of injury was known, poisoning was reported for 63.6% and 67.4% of deaths, respectively. The majority of undetermined deaths occurred in a house or apartment, making it the most common place of injury for both males and females (75.1% and 82.6%, respectively). A natural area was the second most common setting, accounting for 3.6% of deaths among males and 2.5% among females (Table 18).

Toxicology Results of Decedent and Precipitating Circumstances

Tests for alcohol were conducted for 85.4% of decedents of undetermined intent, and drug tests for amphetamines, antidepressants, cocaine, marijuana, and opiates were conducted for 80.8%, 72.4%, 83.4%, 44.8%, and 84.9% of decedents, respectively. Among decedents who tested positive for alcohol (28.2%), 55.2% had a BAC of >0.08 g/dL. Among decedents tested for opiates, 56.6% were positive; of those tested for cocaine, 21.1% were positive; of those tested for marijuana, 9.4% were positive; and of those tested for antidepressants, 27.8% were positive (Table 19).

Precipitating circumstances were known in approximately 73% of deaths of undetermined intent. Of those, 28.7% of decedents had a problem with alcohol, and 62.4% had other substance-abuse problems (e.g., illicit drug or prescription abuse). Although a current depressed mood was reported for only 13.0% of decedents, 39.0% of decedents with known circumstance information had a current mental-health problem, 31.7% were receiving treatment at the time of their death, 10.5% had a history of suicide attempts, 7.2% had disclosed an intent to commit suicide, and 1.6% left a suicide note. Other circumstances noted most often were physical-health problems (29.0%), a crisis during the preceding 2 weeks (15.0%), or an intimate-partner problem (10.8%) (Table 20). Of those with a current mental-health problem, 61.2% had received a diagnosis of depression/dysthymia, 16.0% of bipolar disorder, and 12.5% of an anxiety disorder (Table 21).

A greater percentage of male than female decedents were reported to have an alcohol problem (33.1% and 21.6%, respectively) or other substance-abuse problems (65.1% and 57.9%, respectively) at the time of death. Mental-health problems were reported in a higher percentage of undetermined deaths of females than of males (52.0% and 30.9%, respectively), and a higher percentage of females were receiving treatment for a mental-health problem than males (44.6% and 23.7%, respectively) and had a history of suicide attempts (13.8% and 8.4%, respectively) (Table 20).

Unintentional Firearm Deaths

Sex, Race/Ethnicity, Age Group, and Seasonality

The 16 NVDRS states included in this report collected data concerning 107 unintentional firearm deaths during 2007. Males accounted for 90.7% of decedents. The majority (68.2%) were non-Hispanic whites, followed by non-Hispanic blacks (22.4%). Approximately 20% of unintentional firearm fatalities occurred among persons aged 15--19 years. November had the highest percentage of unintentional firearm deaths (17.8%) at nearly twice the frequency of any other month (Table 22).

Location of Injury

Approximately 64.5% of all unintentional firearm fatalities took place in a house or apartment, making it the most common place of injury for both males and females, followed by natural areas (16.8%) (Table 22).

Context of the Injury and Associated Circumstances

Overall, unintentional firearm injury deaths occurred more commonly while victims were playing with a gun (29.9%), hunting (24.7%), showing a gun to others (14.3%), or loading or unloading a gun (10.4%). The circumstances of injury included thinking that a gun was unloaded, unintentionally pulling the trigger, and experiencing a gun malfunction (26.0%, 19.5%, and 5.2%, respectively) (Table 23).

Special Topics

Violent Deaths with Multiple Decedents

The 16 NVDRS states included in this report collected data on 347 incidents that resulted in multiple decedents. Firearms were the most common method (73.7%) used in incidents with multiple decedents, followed by poisonings (5.7%), sharp instruments or hanging/strangulation/suffocation (5.2% each), and blunt instruments (3.3%) (Table 24). Of a total of 784 victims, 461 (58.8%) were males; 334 (90.8%) of 368 suspects also were males. Non-Hispanic whites accounted for the highest percentage of decedents (57.0%), followed by non-Hispanic blacks (26.5%) and Hispanics (10.5%). Rates for decedents were highest for non-Hispanic blacks and persons aged 15--54 years. Suspects most commonly were aged 20--54 years (Table 25).

Homicide Followed by Suicide

The 16 NVDRS states included in this report collected data concerning 172 violent incidents that occurred during 2007 in which a homicide was followed by the suicide of the suspect. Of 240 homicide decedents, 174 (72.5%) were female and 160 (93.0%) suspects (suicide decedents) were male. Homicide rates were similar (0.3 per 100,000) among racial/ethnic groups which had results reported; 70.4% of homicide decedents were non-Hispanic whites. Among suspects who killed themselves after committing homicide, 66.9% were non-Hispanic whites and 15.7% were non-Hispanic blacks. The highest percentages of both homicide and suicide decedents were aged 35--54 years (31.7% and 49.4%, respectively) (Table 26).

The majority of homicide decedents and suspects (34.7% and 32.4%, respectively) were married at the time of death (not necessarily to each other) (Table 26). With respect to location, 75.4% of the homicides occurred in a house or apartment and 2.1% each in a street/highway or commercial/retail area. Firearms were the most common (approximately 80%) method used by suspects both in committing the homicide and in subsequently killing themselves (Table 27).

Tests for alcohol were conducted for 79.6% of homicide decedents and 72.7% of suicide decedents. Among decedents who tested positive for alcohol (18.3% of homicide victims; 32.8% of suicide decedents), 48.6% of homicide decedents and 53.7% of suicide decedents had a BAC of >0.08 g/dL at the time of death. Suspects who killed themselves following a homicide and who were tested subsequently for drugs had higher percentages of positive tests for antidepressants, cocaine, marijuana, and opiates than homicide victims (Table 28).

Although 8.3% of persons who killed themselves following a homicide had a current depressed mood, only 3.6% were receiving mental-health treatment at the time of the fatal incident. Intimate-partner--relationship problems preceded homicide followed by suicide in 81.0% of suspected suicides. Other nonintimate-partner--relationship problems contributed to 13.7% of suspected suicides. Of suspects who killed themselves, 91.1% had had a personal crisis within the preceding 2 weeks. Previous criminal legal problems were noted in 19.1% of suspected suicides and noncriminal problems in 3.0%; physical health or financial problems were contributing circumstances in 6.6% and 4.2% of suspected suicides, respectively; 6.0% of suicide decedents had disclosed their intent to kill themselves; and 1.8% had a history of suicide attempts (Table 29).

Intimate-Partner Homicide

The 16 NVDRS states included in this report collected data concerning 562 incidents comprising 612 deaths of intimate-partner--related homicides that occurred during 2007. Of 612 homicide victims, 394 (64.4%) were female. Although 51.3% of homicide victims were non-Hispanic whites, rates were higher for AI/ANs and non-Hispanic blacks (2.1 and 1.6 per 100,000 population, respectively). Of 580 suspects, 451 (77.8%) were male, 252 (43.5%) were non-Hispanic whites and 186 (32.1%) non-Hispanic blacks. The highest percentages of victims and suspects (26.1% and 23.5%, respectively) were persons aged 35--44 years. The highest percentage (37.8%) of victims were married at the time of death (Table 30). Tests for alcohol were conducted for 81.5% of the victims. Of the 33.9% of decedents who tested positive for alcohol, 60.4% had a BAC of >0.08 g/dL. The percentage of victims tested for substances other than alcohol varied (range: 37.8%--56.5%) for various drugs; cocaine was evident in approximately 13% of victims tested for this substance (Table 31).

Suicide of Former or Current Military Personnel

The 16 NVDRS states included in this report collected data concerning 1,774 suicides by former or current military personnel that occurred during 2007. Of these decedents, 1,713 (96.6%) were male, and 1,627 (91.7%) were non-Hispanic whites. The greatest percentage of decedents were persons aged ≥35 years. The most common method (67.2%) used was a firearm followed by hanging/strangulation/suffocation (14.5%) and poisoning (12.3%) (Table 32). Among the 64.2% former or current military personnel suicide decedents who were tested for alcohol, 31.6% tested positive; 64.2% of these decedents had a BAC of >0.08 g/dL (Table 33). Although 43.2% were depressed at the time of death, and 36.8% had a mental-health problem, only 27.4% were receiving mental-health treatment. With respect to substance abuse, 16.4% had an alcohol problem, and 7.9% had a problem with other substances. Among those with known circumstance information, 25.2% had experienced a problem with an intimate partner, 37.2% had a physical-health problem, and 32.4% had experienced an acute crisis during the preceding 2 weeks. With respect to life stressors, 11.0% had experienced a job problem, 10.0% a financial problem, and 7.5% a criminal legal problem. Approximately one third (34.7%) left a suicide note, 13.8% had made a previous suicide attempt, and 26.9% had disclosed an intent to commit suicide (Table 34).

Legal Intervention

The 16 NVDRS states included in this report collected data on 145 legal-intervention incidents in 2007 resulting in 141 single-victim deaths and four deaths where the legal-intervention victim had recently committed a homicide. Of the 145 legal-intervention decedents, 45.5% were non-Hispanic whites and 35.2% were non-Hispanic blacks. With respect to location, 37.9% of legal-intervention deaths occurred in a house or apartment, 31.7% on a street or highway, and 9.0% in a parking lot or public garage (Table 35). The majority of decedents were aged 20--54 years (Table 36). Of the 87.6% of legal-intervention decedents tested for alcohol, 38.6% were positive for alcohol and 67.4% of these decedents had a BAC of >0.08 g/dL. The percentage of victims tested for other substances varied (range: 49.7%--75.9%). The presence of other drugs for which tests were positive also varied: 30.6% of those tested for marijuana, 24.6% of decedents tested for cocaine, 20.4% of those tested for amphetamines, 9.3% of those tested for opiates, and 5.4% of those tested for antidepressants were positive for these substances (Table 37).

Suicide Among Persons Aged ≥50 Years

In 2007, NVDRS collected data for 3,677 persons aged ≥50 years who died by suicide. Of those, rates of suicide were highest among those aged 50--59 years (16.7 per 100,000 population), followed by those aged ≥80 years (15.8 per 100,000). Beginning with those aged 60--69 years, rates increase with age from 13.4 per 100,000 to 14.4 for those aged 70--79 years, and 15.8 for those aged ≥80 years. Among persons aged ≥50 years, rates were five times higher among males than among females (26.1 and 5.1 per 100,000 population, respectively). Rates were highest among non-Hispanic whites (17.7 per 100,000 population), followed by AI/ANs (11.7 per 100,000 population), Hispanics (8.6), APIs (5.6), and non-Hispanic blacks (3.9). At the time of death, persons aged 50--69 years most often were either married or divorced. Those aged 70--79 years and those aged ≥80 years most often were either married or widowed (Table 38).

The majority (81.9%) of suicide decedents aged ≥50 years died in a house or apartment. The second-most-common location for those aged 50--59 and 60--69 years was a natural area (4.7% and 3.5%, respectively). The second-most-common location for those aged 70--79 years was a street/highway (2.2%) and for those aged ≥80 natural areas, street/highway and motor vehicle were equally common at 1.2% each. As to method used by suicide decedents aged ≥50 years, firearms accounted for 60.6% of deaths (rate: 9.3 per 100,000 population), poisoning for 19.6% (3.0 per 100,000 population), and hanging/strangulation/suffocation for 12.6% (1.9 per 100,000 population). Rates of firearm suicide were highest among persons aged ≥80 years (12.1 per 100,000) and those aged 70--79 years (11.1 per 100,000 population) (Table 38).

Precipitating circumstances were identified for approximately 86% of older adult suicides. Current depressed mood (39.1%), current mental-health problem (39.1%), and physical-health problems (33.0%) were the most commonly identified circumstances; 30.8% left a suicide note, and 24.7% disclosed their intent to commit suicide (Table 39).

Discussion

The findings in this report indicate clear variations in patterns of death from violence-related injuries, unintentional firearm-related injuries, and deaths of undetermined intent reported from the 16 states included in this report. Rates for these deaths were disproportionately higher among males, adults aged <55 years, and minority populations. A residence (house or apartment) was the most common location for all deaths. Of all incidents meeting NVDRS inclusionary criteria in 2007 in the 16 states included in this report, approximately 98% involved a single victim.

Suicide Patterns

Suicide rates were higher among males than among females, AI/ANs, and non-Hispanic whites than among non-Hispanic blacks, and highest among persons aged 45--54 years. These findings are similar to those that have been documented in other reports (7--9). Persons aged ≥80 years have typically had the highest rates of suicide in the United States (1). However, in 2006, rates of suicide among persons aged 45--54 years in the United States surpassed those for persons aged ≥80 years (1). This pattern also was noted in the findings from NVDRS states in 2006 (10) and has continued in 2007. Problems related to mental health, jobs, finances, or relationships might have contributed to the high rates of suicide in this age group. Mental health and/or substance-abuse problems, relationship problems and losses, and recent crises were frequent precipitants for suicide. These factors have been documented in other studies as important risk factors for suicide (8,11).

Alcohol was a factor in approximately one third of the reported suicides, and 62% of these decedents had a BAC of >0.08 g/dL at the time of death. Alcohol and drug abuse are second only to depression and other mood disorders as the most frequent risk factors for suicidal behavior (8,12,13). However, the relation between these factors is complex. Previous research indicates that alcohol intoxication might act as an important proximal (i.e., an experience often occurs immediately prior to a suicide) risk factor for suicidal behavior (14). In cases where there is alcohol dependency, the alcohol abuse might lead directly to depression or indirectly through a sense of decline or failure that is experienced by many persons who are dependent on alcohol. Alcohol also might be a form of self-medication to alleviate depression. Both depression and alcohol abuse also might be the result of specific stresses in a person's life (15). The co-occurrence of mood disorders with substance abuse, including alcohol abuse, greatly increases the risk for suicidal behavior (16).

Many of the circumstances surrounding suicides (e.g., mental health problems, alcohol problems, and recent crises) also were found for certain special populations examined in this report. However, other factors also were important. Physical health problems were more frequently noted for former or current military personnel and in adults aged ≥70 years. Physical illness is believed to be a common antecedent to suicide among the elderly, although prevalence estimates vary widely, and it appears to be a stronger contributing factor when mood disorders, depressive symptoms or other factors also are present (8,15). For example, untreated or undertreated pain, anxiety about the progression of an illness, fear of dependence, and fear of burdening family members are major contributing factors in suicidal behavior among elderly persons with physical illnesses (8). Many of these factors also are evident in military populations with disabling conditions. The proportion of former or current military personnel reported to be experiencing health problems also might reflect a difference in reporting and contact with health-care professionals.

Finally, approximately 30% of suicide victims had disclosed their intent to commit suicide, and approximately 20% had made a previous suicide attempt. A previous suicide attempt is an important predictor of subsequent fatal suicidal behavior (8,15). Disclosure of intent also is an important warning sign of suicidal intentions, although persons in close contact with potential victims of suicide often are unaware of the significance of these warnings or unsure how to act on them (17).

Homicide Patterns

Homicide rates were higher among males than among females, among non-Hispanic blacks compared with members of other racial/ethnic populations, and among persons aged 20--24 years compared with persons in other age groups. These findings also are consistent with patterns documented in other reports. Homicide is the second leading cause of death in the United States among persons aged 15--24 years, and rates among non-Hispanic blacks in this age group exceed those of other racial/ethnic populations by approximately fourfold to sevenfold (1). Males also are disproportionately represented among victims of homicide in the United States and elsewhere (1,18).

The majority of homicides involved a single victim. Multiple decedent homicides and homicide-suicide incidents accounted for <3% of violent deaths. The majority of homicides were related to interpersonal conflicts. Crime was a factor in approximately one third of all homicide/legal-intervention deaths, with robbery being the primary circumstance. These findings are consistent with other research on homicide. Arguments and conflicts are immediate motivations for the majority of both male and female homicides in the United States (19). One factor that distinguishes male from female homicides is the relationship between the victim and the perpetrator. In the United States, approximately one in three homicides of females is committed by a current or former spouse or partner (20). Among male homicide victims, approximately 5% are killed by intimate partners. The findings of this report indicate that male homicide decedents were more frequently killed following arguments or conflicts with persons other than an intimate partner or for other reasons (e.g., crime or drug-related) whereas more than half of homicides involving a female victim involved intimate-partner--related violence.

As with suicide decedents, alcohol was present in approximately one third of homicide decedents; more than half of these decedents (55.4%) had BACs of >0.08 g/dL. Alcohol is an important situational factor in interpersonal violence. In the case of interpersonal violence among youths, excessive alcohol consumption might increase impulsivity and make some drinkers more likely to resort to violence in a confrontation or argument (21,22). Reduced physical control and the ability to assess risks in potentially dangerous situations also can make some drinkers more vulnerable to victimization (21,22). In the case of intimate-partner violence, excessive alcohol consumption by one or both partners might exacerbate financial or child care problems or other stressors and increase tension and conflict in the relationship (23). Alcohol also can be a form of self-medication to cope with previous or current experiences of abuse (23).

Method of Injury

Approximately two thirds of all homicides and approximately one half of all suicides in the United States are committed with a firearm (1). In the 16 states included in this report, firearms were the most common method used in homicides, incidents involving multiple victims, and incidents of homicide followed by suicide. Previous research indicates that interpersonal disputes can escalate and cause serious violent injury or death, especially when weapons of lethal means (e.g., firearms) are involved in the dispute (24,25). Firearms also were the most common method used in suicides; however, methods differed by sex. Firearms were the most common method used by males to complete suicide followed by hanging/strangulation/suffocation. Poisoning was the most common method used by females, followed by firearms.

Deaths where the intent was undetermined were primarily the result of poisonings or had an unknown cause. Poisoning was the most common method for both males and females. Toxicology results documented a high prevalence of alcohol and other substances at the time of death. For example, approximately 85% of decedents with undetermined intent were tested for opiates, and nearly 57% tested positive for these substances. Whether these deaths were related to unintentional drug poisonings, (which have increased substantially in recent years, particularly among adults aged 35--54 years [26,27]) or were suicides is unknown. The majority (49%) of decedents in the 16 states were aged 35--54 years. Substance-abuse problems involving drugs other than alcohol were the most commonly noted circumstance; approximately 40% of such decedents had a mental-health problem, and 10% had a history of suicide attempts.

Prevention Opportunities

Information concerning the precipitating circumstances in violent deaths described in this report provides important clues regarding where to focus prevention efforts. For example, relationship problems, interpersonal conflicts, and recent crises were important precipitating factors for both homicide and suicide. Intimate-partner--related problems, in particular, were a factor in many types of violent death. Primary prevention programs designed to enhance social problem-solving and coping skills to deal with stressful life events, health and financial problems, or other problems that occur within interpersonal relationships can potentially reduce violence (28). In addition to demonstrating the need to address situational stressors, the findings in this report underscore the importance of changing cultural and social norms (e.g., attitudes condoning the use of violence as a means of resolving conflict) addressing the social and economic conditions within communities that often give rise to violence (e.g., inequities with regard to the distribution of and access to resources and opportunities, social isolation, lack of connectedness among persons, families, and communities) and intervening much earlier by teaching young persons the skills to develop and promote respectful, nonviolent interpersonal relationships. Some of the strategies that offer the strongest evidence of effectiveness with respect to the latter are primary prevention strategies that focus on family environments, school environments, and building individual social, emotional, and behavioral competencies (28--31).

The findings of this report also highlight the importance of addressing mental health problems. Mental-health problems were highly prevalent among suicide decedents, yet many were not receiving treatment at the time of death. Reasons why persons do not seek care for mental health problems include the belief that the problem will resolve on its own, financial barriers, a lack of awareness of available services, fear of hospitalization, embarrassment, fear of what others might think, and the belief that acknowledgment or discovery of a mental health problem could damage one's career or relationships (8). Despite public education efforts focused on the nature, causes, and treatment of mental illness, the stigma of mental illness is one of the most significant barriers deterring persons from seeking treatment (8). Stigma leads persons to fear, reject, and distance themselves from persons with mental health problems. Findings from a national study indicate that approximately 47%--63% of persons in the United States want to distance themselves from persons with depression and schizophrenia (32). The consequences of stigma for persons suffering from these problems include diminished opportunities, lowered self-esteem, shame and concealment of symptoms, and lower help-seeking behavior (8). These results underscore the need for prevention measures aimed at changing cultural attitudes and norms surrounding mental health problems so that persons in need of treatment can seek the care of professionals and reach out to family, friends, and others without hesitation.

Limitations

The findings provided in this report are subject to at least seven limitations. First, the availability, completeness, and timeliness of data are dependent on the sharing of data among state health department NVDRS teams, CMEs, and law enforcement personnel in their states. This is particularly challenging when states have independent county coroner systems rather than a centralized CME system and large numbers of law enforcement jurisdictions. NVDRS incident data might be limited or incomplete for areas in which these data-sharing relations are not developed fully.

Second, toxicology data are not collected consistently across all states or for all alcohol and drug categories. The percentage of decedents testing positive might be affected by selective testing biases in medical examiner or coroner offices (33). Third, abstractors are limited to the data included in the reports they receive. Reports might not fully reflect all information known about an incident, particularly in the case of homicides, when data are less readily available until after prosecutions are complete. Fourth, case definitions present challenges when a single death is classified differently in different documents (e.g., "unintentional" in a police report, "homicide" in a CME report, and "undetermined" on the death certificate). NVDRS abstractors reconcile these cases using standardized NVDRS case definitions and select a single manner of death on the basis of all source documents. Fifth, NVDRS data are available only from a limited number of states and therefore are not nationally representative. Sixth, although extensive coding training is conducted and help desk support is available daily, variations in coding might occur depending on the abstractor's level of experience. For this reason, states regularly conduct blinded reabstraction of cases to test consistency and identify training needs. Finally, protective factor data (i.e., characteristics or circumstances that reduce the risk for violent death) are not collected by NVDRS because of the nature of death certificate, CME record, and police reports, which typically contain only circumstances associated with risk factors.

Conclusion

Accurate, timely, and comprehensive surveillance data can be used to monitor the occurrence of violence-related fatal injuries and assist public health and other authorities in the development, implementation, and evaluation of programs and policies that reduce and prevent violent deaths and injuries at the national, state, and local levels (34,35). Continued development and expansion of NVDRS is critical to the public health and criminal justice communities at the federal, state, and local levels that work to reduce the personal, familial, and societal costs of violence. Further efforts are needed to increase the number of states participating in NVDRS, with the ultimate goal of full national representation, including all 50 states, the District of Columbia, and U.S. territories.

Acknowledgments

Contributors to this report included participating state Violent Death Reporting Systems; participating state agencies, including state health departments, vital registrars' offices, coroners' and medical examiners' offices, crime laboratories, and local and state law enforcement agencies; partner organizations, including the State and Territorial Injury Prevention Directors' Association, National Violence Prevention Network, National Association of Medical Examiners, National Association for Public Health Statistics and Information Systems, Council of State and Territorial Epidemiologists, the International Association of Chiefs of Police, and Association of State and Territorial Health Officials; federal agencies, including the Department of Justice (Bureau of Justice Statistics and the Federal Bureau of Investigation), the Department of the Treasury (Bureau of Alcohol, Tobacco, and Firearms); other stakeholders, researchers, and foundations, including Harvard University School of Public Health, the Joyce Foundation, and Fenton Communications; and the National Institute for Occupational Safety and Health, National Center for Health Statistics, CDC.

References

1. CDC. Web-based Injury Statistics Query and Reporting System (WISQARS(tm)). Atlanta, GA: US Department of Health and Human Services, CDC; 2008. Available at http://www.cdc.gov/ncipc/wisqars/default.htm.
2. Doll L, Bonzo S, Mercy J, et al, eds. Handbook of injury and violence. New York, NY: Springer; 2007.
3. Paulozzi LJ, Mercy J, Frazier L, et al. CDC's National Violent Death Reporting System: background and methodology. Inj Prev 2004;10:47--52.
4. CDC. Surveillance for violent death---National Violent Death Reporting System, 16 states, 2005. In: CDC Surveillance Summaries, April 11, 2008. MMWR 2008;57(No. SS-3).
5. World Health Organization. International classification of diseases, version 10. Geneva, Switzerland: World Health Organization; 2007. Available at http://www.who.int/classifications/icd/en/index.html.
6. CDC. U.S. census populations with bridged race categories. Hyattsville, MD: US Department of Health and Human Services, CDC; 2007. Available at http://www.cdc.gov/nchs/about/major/dvs/popbridge/popbridge.htm.
7. Kung HC, Hoyert DL, Xu JQ, et al. Deaths: final data for 2005. Natl Vital Stat Rep 2008;56:1--124.
8. Institute of Medicine. Reducing suicide: a national imperative. Washington, DC: National Academies of Science; 2002.
9. CDC. Surveillance for fatal and nonfatal injuries---United States, 2001. In: CDC Surveillance Summaries, September 3, 2004. MMWR 2004;53(No. SS-7).
10. Karch DL, Dahlberg LL, Patel N, et al. Surveillance for violent deaths--- National Violent Death Reporting System, 16 States, 2006. CDC Surveillance Summaries, March 20, 2009, MMWR 2009;58(No. SS-1).
11. US Department of Health and Human Services. National strategy for suicide prevention: goals and objectives for action. Rockville, MD: United States Department of Health and Human Services, Public Health Service; 2001.
12. Borges G, Walters EE, Kessler RC. Associations of substance use, abuse, and dependence with subsequent suicidal behavior. Am J Epidemiol 2000;15:781--9.
13. Tondo L, Baldessarini RJ, Hennen J, et al. Suicide attempts in major affective disorder patients with comorbid substance use disorders. J Clin Psychiatry 1999;60(Suppl 2):S63--9.
14. Hufford MR. Alcohol and suicidal behavior. Clin Psychol Rev 2001;21:797--811.
15. DeLeo D, Bertolote J, Lester D. Self-directed violence. In: Krug EG, Dahlberg LL, Mercy JA, et al., eds. World report on violence and health. Geneva, Switzerland: World Health Organization; 2002:183--212.
16. Moscicki EK Epidemiology of completed and attempted suicide: Toward a framework for prevention. Clinical Neuroscience Research 2001;1:310--23.
17. American Association of Suicidology. Understanding and helping the suicidal individual. Fact sheet. Available at http://www.suicidology.org/c/document_library/get_file?folderId=232&name=DLFE-30.pdf.
18. Krug EG, Dahlberg LL, Mercy JA, et al. World report on violence and health. Geneva, Switzerland: World Health Organization; 2002.
19. Schwartz, J. Gender differences in homicide offending. In: DeLisi M, Conis P, eds. Violent offenders: theory, research, public policy, and practice, Boston, MA: Jones & Bartlett; 2007:119--40.
20. Federal Bureau of Investigation. Crime in the United States, 2007. Washington, DC: US Department of Justice, Federal Bureau of Investigation; 2007. Available at http://www.fbi.gov/ucr/cius2007/offenses/expanded_information/homicide.html.
21. Parker RN. Alcohol and violence: connections, evidence and possibilities for prevention. J Psychoactive Drugs 2004 (Supp2):S157--63.
22. Graham K. Social drinking and aggression. In Mattson MP, ed. Neurobiology of aggression: understanding and preventing violence. 1st ed. Totowa, NJ: Humana Press; 2003:253--74.
23. World Health Organization. Alcohol and intimate partner violence. Fact sheet, 2005. Geneva, Switzerland: World Health Organization; 2005. Available at http://www.who.int/violence_injury_prevention/violence/world_report/factsheets/ft_intimate.pdf.
24. Fagan J, Wilkinson DL. Guns, youth violence, and social identity in inner cities. Crime and Justice 1998;24:105--88.
25. Wilkinson DL. Guns, violence and identity among African-American and Latino youth. New York, NY: LFB Scholarly Publishing; 2003.
26. Paulozzi LJ, Budnitz DS, Xi Y. Increasing deaths from opioid analgesics in the United States. Pharmacoepidemiol Drug Saf 2006;15:618--27.
27. CDC. Unintentional poisoning deaths---United States, 1999--2004. MMWR 2007;56:93--6.
28. Lutzker JR, ed. Preventing violence: research and evidence-based intervention strategies. Washington, DC: American Psychological Association; 2006.
29. Hahn R, Fuqua-Whitley D, Lowry J, et al. The effectiveness of universal school-based programs for the prevention of violence: a report on recommendations of the Task Force on Community Preventive Services. Am J Prev Med 2008;33:S114--29.
30. Wilson SJ, Lipsey MW, Derzon JH. The effects of school-based intervention programs on aggressive behavior: a meta-analysis. J Consult Clin Psychol 2003;71:136--49.
31. Henggeler SW, Clingempeel WG, Brondino MJ, et al. Four-year follow-up of multisystemic therapy with substance-abusing and substance-dependent juvenile offenders. J Am Acad Child Adolesc Psychiatry 2002;41:868--74.
32. Link BG, Phelan JC, Bresnahan M, et al. Public conceptions of mental illness: Labels, causes, dangerousness, and social distance. Am J Public Health 1999;89:1328--33.
33. CDC. Toxicology testing and results for suicide victims---13 states, 2004. MMWR 2006;55:1245--8.
34. Karch, D, Logan, J. Data consistency in multiple source document:findings from homicide incidents in the National Violent Death Reporting System, 2003--2004. Homicide Studies, 2008;12:264--76.
35. Logan J, Karch D, Crosby A. Reducing unknown data in violent death surveillance: a study of death certificates, coroner/medical examiner and police reports from the National Violent Death Reporting System, 2003--2005. Homicide Studies 2009;13:385--97.

All MMWR HTML versions of articles are electronic conversions from typeset documents. This conversion might result in character translation or format errors in the HTML version. Users are referred to the electronic PDF version (http://www.cdc.gov/mmwr) and/or the original MMWR paper copy for printable versions of official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.

**Questions or messages regarding errors in formatting should be addressed to [email protected].