40. Interpersonal Violence

The Nature, Burden, and Causes of Interpersonal Violence

WHO (WHO Global Consultation on Violence and Health 1996, 2-3) defines violence as follows: "The intentional use of physical force or power, threatened or actual, against oneself, another person, or against a group or community that either results in or has a high likelihood of resulting in injury, death, psychological harm, mal-development, or deprivation." This definition emphasizes that, for the act to be classified as violence, a person or group must intend to use force or power against another person. Thus, violence is distinguished from unintended incidents that result in injury or harm.

The nature or mode of violence may be physical, sexual, or psychological, or it may involve deprivation and neglect. Given the difficulties of measuring deprivation and neglect, this chapter concentrates on the physical, sexual, and psychological modes.

Acts of interpersonal violence are classified as family violence or community violence. Family violence is further categorized by victim: child, intimate partner, or elder. Child abuse, as defined by WHO (1999, 15), is "physical and/or emotional ill-treatment, sexual abuse, neglect or negligent treatment or commercial or other exploitation, resulting in actual or potential harm to the child's health, survival, development, or dignity in the context of a relationship of responsibility, trust, or power." Behavior within an intimate relationship that causes physical, psychological, or sexual harm is typically labeled intimate partner violence or domestic violence. Elder abuse is mistreatment of older people, generally those older than age 60 or 65, in the home or in an institutional setting.

Community violence is categorized by two types of perpetrators: acquaintances and strangers. It includes sexual assault by strangers and violence in institutional environments, such as residential care facilities, jails, workplaces, and schools. Youth violence, with perpetrators and victims typically 10 to 29 years of age, is also a form of community violence.

Outcomes of Interpersonal Violence

Identifying the outcomes of interpersonal violence helps to determine the magnitude of the problem.

Data

As noted earlier, a fundamental aspect of the public health approach is the collection of accurate information, such as demographic characteristics of victims and perpetrators, weapon involvement, settings in which violence occurs, situational determinants, and nature and severity of resultant injuries and other harm. Data sources include death certificates, vital statistics records, medical examiners' reports, hospital and other medical records, police and judiciary records, and self-reported information from victim surveys and special studies. Multiple data sources, with their inherent strengths and limitations, are essential.

The most widely encountered sources of information are from the health and criminal justice sectors. Reliable data on violent deaths are not routinely collected in most countries. Where data collection systems are in place, coroner and mortuary reports, death certificates, and vital statistics records usually provide additional data about the victim. The health sector typically documents characteristics of the decedent and the cause, location, circumstances, and time of death. The criminal justice sector documents deaths or arrests resulting from interpersonal violence, including sometimes recording information about the relationship between the victim and the offender, the circumstances surrounding the violence, and the demographics of the perpetrator.

Theoretically, health and criminal justice sector data include information about nonfatal violence at all levels of severity, including threats of violence and instances of psychological violence, deprivation, and neglect. In practice, however, only data about violence-related injuries presenting at hospital emergency departments are collected. Studies from a variety of countries show that for every victim reporting violence to the police, at least two more present only at health agencies (Houry and others 1999; Kruger and others 1998; Sutherland, Sivarajasingam, and Shepherd 2002; additional sources online). Victims of nonfatal violence treated by the health sector may provide information about the perpetrator-victim relationship, about the circumstances surrounding the attack, and about contextual and developmental risk factors. However, the health sector is frequently restricted in recording information about perpetrators.

In LMICs, population-based surveys are a more useful source of information about violence-related injuries at all severity levels (Sethi, Habibula, and others 2004). Such surveys have been conducted in Bangladesh (Rahman, Andersson, and Svanstrom 1998); Colombia (Duque, Klevens, and Ramirez 2003); Iraq (Roberts and others 2004); Pakistan (Ghaffar 2001); South Africa (Butchart, Kruger, and Lekoba 2000; additional sources online); and Uganda (Kobusingye, Guwatudde, and Lett 2001). Demographic and health surveys with questions about violent victimization also collect information about the relationship between violence and other health conditions, but they can provide only limited insight into the perpetrators.

Hospital emergency departments have been used in some postconflict settings to monitor weapons-related injuries and evaluate the relative contributions of collective and interpersonal violence to the caseload (Meddings and O'Connor 1999; Michael and others 1999). Some developing countries, such as Bangladesh, Kenya, and Uganda, also use violence and injury surveillance systems based in health facilities to monitor hospitalizations resulting from violence and other causes of injury (Kobusingye and Lett 2000; Odero and Kibosia 1995; Rahman and others 2001). Where emergency and forensic medical services are reasonably well developed and where access to such services is equitable, violence and injury surveillance tools have been integrated into hospital emergency departments (Hasbrouck and others 2002; additional sources online), prenatal clinics (Dunkle and others 2004), forensic service centers for rape victims (Swart and others 2000), and mortuaries (Butchart and others 2001). Those efforts have proven effective in obtaining victim-based, descriptive epidemiological information and insights into the relationships between victims and perpetrators.

Deaths Resulting from Interpersonal Violence

Global burden of disease estimates indicate that, in 2001, approximately 1.6 million people died as a result of violence. Of those deaths, 34 percent were due to interpersonal violence (table 40.1).

Rates and patterns of violent death vary by country and region (figure 40.1). Homicide rates were highest in developing countries in Sub-Saharan Africa and Latin America and the Caribbean and lowest in East Asia, the western Pacific, and some countries in northern Africa. Studies show a strong, inverse relationship between homicide rates and both economic development and economic equality (Butchart and Engstrom 2002; Fajnzylber, Lederman, and Loayza 2000). Poorer countries, especially those with large gaps between the rich and the poor, tend to have higher rates of homicide than wealthier countries.
[Figure 40.1]

Homicide rates differ markedly by age and sex (table 40.2). Gender differences were least marked for children. For the 15 to 29 age group, male rates were nearly six times those for female rates; for the remaining age groups, male rates were from two to four times those for females. Female homicide rates doubled after age 14 and gradually but steadily increased with age, and male rates increased more than 14 times after age 14, peaked in the 15 to 29 age group, and then gradually decreased with age. Overall, homicides resulted in the deaths of 3.4 males per female.

Violence-related Burden of Disease

The sum of years of potential life lost because of premature mortality and years of productive life lost because of disability is not a particularly useful measure of the burden of violence. Disability-adjusted life years rely, in part, on estimates of nonfatal events. In the case of violence, those estimates are restricted to injuries and physical disabilities, both markedly underreported. In addition, given that psychological and other noninjury health consequences of violence are substantial, failure to include them in the measurement of disability-adjusted life years means that estimates of the nonfatal burden of violence may be grossly underestimated.

Violence-related morbidity can be analyzed as four distinct, but often co-occurring, outcome clusters: injuries and disabilities, mental health and behavioral consequences, reproductive health consequences, and other health consequences.

Studies in a number of countries show that, for every homicide among young people age 10 to 24, 20 to 40 other young people receive hospital treatment for a violent injury (Mercy and others 2002). Injuries range from minor, which can be self-treated, to severe. Severe injuries are those that may require resource-intensive emergency medical treatment and inpatient care and may result in lifelong disabilities, such as amputations, brain damage, or paraplegia. Few countries have information systems for monitoring nonfatal violent injuries, and existing systems typically record only data on violent injuries presenting at hospital emergency departments. Data from those sites cannot be directly compared, given the marked differences between and within countries in the availability and accessibility of emergency medical services.

The mental health consequences of violence are far reaching. Child abuse has well-documented sequelae of psychiatric disorders and suicidal behaviors (Runyan and others 2002). Both short- and long-term sequelae have been demonstrated (Mercy and others 2002, Heise and Garcia-Moreno 2002), including depression, anxiety disorders, substance abuse disorders, aggression, cognitive problems, sleep disorders, and post-traumatic stress disorder. The severity and duration of those consequences vary with the child's age and the length of time the child suffers the abuse, as well as the duration and intensity of the abuse, the child's relationship to the abuser, and the treatment received (Runyan and others 2002).

Intimate partner violence results in an increased incidence of suicide and suicide attempts, as well as in depression, anxiety, and phobias (Heise and Garcia-Moreno 2002). Additional consequences include substance abuse, eating and sleep disorders, poor self-esteem, posttraumatic stress disorder, psychosomatic disorders, and risky sexual behaviors. Sexual assault results in consequences that can be long lasting and severe, including posttraumatic stress disorder, depression, and conduct disorders, as well as sleep and eating disorders (Jewkes, Sen, and Garcia-Moreno 2002).

According to Jewkes, Sen, and Garcia-Moreno (2002), among adolescents and women age 12 to 45, the frequency of pregnancy as a result of rape varies from 5 to 18 percent. In addition, younger rape victims often have an increased rate of later, unintended pregnancies. Rape frequently results in gynecological problems, problems of sexual functioning, and sexually transmitted diseases, including HIV infection. HIV infection and the stigma it carries put both female and male victims of sexual assault at increased risk of further violence. A similar range of reproductive health consequences may also follow intimate partner violence.

A strong, graded relationship exists between the breadth of exposure to abuse or household dysfunction during childhood and the presence of adult diseases, including ischemic heart disease, cancer, chronic lung disease, skeletal fractures, and liver disease (Felitti and others 1998). In developed countries, abuse and other violent events of childhood have been associated with a 4- to 12-fold increased risk for alcoholism, drug abuse, depression, and suicide attempt; a 2- to 4-fold increased risk for smoking, poor self-rated health, 50 or more sexual intercourse partners, and sexually transmitted disease; and a 1.4- to 1.6-fold increased risk for physical inactivity and severe obesity (Anda and others 1999; Dietz and others 1999; Dube and others 2001, 2002; Hillis and others 2000, 2001; Williamson and others 2002). Similar exposures to violence in developing countries may have different, yet equally wide-ranging, impacts beyond direct physical and psychological injuries.

Data on Violence in Developing Countries

Studies documenting the human and economic toll of violence in LMICs are strikingly scarce. In addition to disparate levels of economic development, other differences between countries strongly influence levels and patterns of interpersonal violence and the toll that such violence takes on society. Countries with weak governments and institutions are at considerably higher risk for interpersonal violence than countries with developed institutions, and countries at war are likewise at higher risk than countries at peace. The same factors that lead to high levels of interpersonal violence—lack of economic development; weak social, political, and judicial institutions; social disturbances; and warfare—also adversely affect nations' ability to collect data and to address the causes or consequences of this violence.

Risk Factors for Understanding Violence

Risk factors for violence are conditions that increase the possibility of becoming a victim or perpetrator of violence. No single factor explains why a person or group is at a high or low risk of violence. Rather, violence is an outcome of a complex interaction among many factors. This relationship is captured in an ecological model that classifies risk factors for violence by four levels: individual, relationship, community, and societal (Dahlberg and Krug 2002). Although some risk factors may be unique to a particular type of violence, the various types of violence more commonly share a number of risk factors (table 40.3).