10.5 Interpersonal violence prevention: a recent public health mandate
Oxford Textbook of Public Health
Interpersonal violence prevention: a recent public health mandate
Deborah Prothrow-Stith, Howard Spivak, and Robert D. Sege
Definition and classification
The magnitude of the problem
School violence: more lethal
Rape and sexual assault
The characteristics of violence
Race and poverty
The contact that health professionals have with victims and perpetrators
The application of public health strategies
Identification of risk factors
Child neglect and abuse
Children and exposure to media violence
Psychological and behavioural factors
Approaches to violence prevention and control
Public health and criminal justice: interdisciplinary challenges
Primary, secondary, and tertiary prevention
Promising prevention programmes using public health strategies
This chapter on public health approaches to interpersonal violence is a response to the epidemic of adolescent and young adult homicide in the United States, and the growing international attention and concerns about this problem. The chapter provides a short history of the efforts within public health to address violence, a definition and description of the problem, and a discussion of examples of public health approaches to violence prevention. While several types of violence are briefly discussed, the focus of the chapter is youth violence and the increase in youth homicide in the United States. The 1987 United States homicide rate for 15- to 24-year-old men of 22 per 100 000 was the highest among industrialized countries not at war in the period 1986 to 1987 (Fig. 1). By 1991, it had increased to 37 per 100 000. While high homicide rates also plagued South Africa at the same time, the political instability and violent freedom struggle made it an exception.
Fig. 1 International comparisons of homicide rate per 100 000 population (males, aged 15–24 years) in the period 1986 to 1987.
The public’s demand for solutions to violence in the United States has generated increased multidisciplinary attention to the problem that goes beyond the traditional criminal justice responses of punishment and deterrence. In the past 25 years we have witnessed a dramatic effort led by public health professionals to confront violence in the United States. Leadership has emerged from the Centers for Disease Control (CDC), the Surgeon General’s office, and many state and local health departments. Many international meetings, initially geared towards discussion of peace and ending war, now include interpersonal violence on their agendas.
In 1983 the CDC established the Violence Epidemiology Branch for the study of homicide and suicide. Initial Morbidity and Mortality Weekly Reports revealed that homicide is the leading cause of death for black men between the ages of 15 to 24 years and 25 to 44 years, and is the second leading cause of death for all adolescents (CDC 1982a,b, 1983a,b). Additional information concerning the characteristics of homicides was published for public health audiences, and indicated that 58 per cent of the victims knew their assailants, 47 per cent were precipitated by an argument, and only 15 per cent were a result of another felony (burglary, rug trafficking, etc.) (CDC 1983a,b). The application of basic epidemiology and reporting techniques became the impetus for public health professionals across the country to confront the issue.
In October 1985, C. Everett Koop convened an invitational meeting, the Surgeon General’s Workshop on Violence and Public Health, in Leesburg, Virginia. The interdisciplinary meeting focused on assault and homicide, child abuse, rape and sexual assault, domestic violence, elder abuse, and suicide. The workshop and its published proceedings continue to fuel public health professionals’ efforts to frame violence as a mainstream public health problem.
Today, public health endeavours to understand and prevent violence continue to grow with increasingly more programmes, publications, and presentations. In 1994, the CDC established the National Center for Injury Prevention and Control and every Surgeon General following Dr Koop has encouraged the public health community to use its strategies to better understand and prevent violence.
In the United States, a country-wide effort to prevent violence utilizing standard epidemiology, community outreach, screening, community-based programmes, health education, behaviour modification, public awareness, and education campaigns continues, involving every aspect of the United States Public Health Service. The movement to prevent violence in the United States is based upon similar multidisciplinary efforts to prevent lung cancer deaths, heart disease, and fatal car crashes.
Although embryonic and consisting of thousands of insulated programmes scattered across the country, this movement has the potential for the same level of success that public health professionals have had with reducing smoking and drink-driving in the United States. The analogy between violence prevention and other public health problems is not flawless, yet two decades of experience employing comparable techniques and strategies indicates enough similarity for success.
Definition and classification
The National Center for Injury Prevention and Control at the CDC classifies both unintentional injuries (accidents) and intentional injuries (violence) as public health problems, as illustrated in Fig. 2. Intentional injuries are divided into self-directed violence (suicides and suicide attempts) and interpersonal violence (assaults and homicides) (Fig. 3). Violence is defined by the CDC as ‘the threatened or actual use of physical force or power against another person, against oneself, or against a group or community that either results or is likely to result in injury, death, or deprivation’.
Fig. 2 Unintentional and intentional injury.
Fig. 3 Intentional injury—violence.
Suicide, a more traditional problem for health and public health professionals, has several commonalities with interpersonal violence. Both often involve alcohol and other drugs, and the risk for both increases with the presence of a firearm. Media and entertainment values appear to have an impact on both. Adolescent suicide and homicide rates rose dramatically during the early 1980s. While suicide remains an important public health concern, recent efforts using public health strategies to address interpersonal violence have proliferated.
There are at least four reasons why interpersonal violence became an important concern for public health professionals in the United States: (a) the magnitude of the problem; (b) the characteristics of violence; (c) the contact that health professionals have with the victims and perpetrators of violence; (d) the application of public health strategies to both understanding and preventing violence.
Public health professionals have offered a unique approach to violence that has yielded significant contributions and offers further promise.
There are several sources of data on violence in America which are accessible to the public or through community partnerships with academic institutions. The National Center for Education Statistics and the Bureau of Justice Statistics are the primary federal entities for collecting, analysing, and reporting data related to education and crime respectively in the United States and other nations. The Department of the Treasury, Bureau of Alcohol, Tobacco and Firearms publication Commerce in Firearms in the United States is an annual report of activities relating to the regulation of firearms. The Youth Crime Interdiction Initiative Crime Gun Trace Analysis Reports: the Illegal Youth Firearms Market in 27 Communities brings together federal, state, and local law enforcement officials to improve information about the illegal sources of guns recovered from juveniles and adult criminals.
The Uniform Crime Reports, published by the United States Federal Bureau of Investigation (FBI), is the most frequently cited source of national information on violent crime. These annual reports, which date back to 1930, use police data that are submitted to the FBI and which are aggregated into a national data source. Homicides are mandatorily reported in these datasets, but other crimes are reported voluntarily and therefore inconsistently. The reports give cursory information on homicides and assaults, including victim and perpetrator relationship, weapons used, location of the violent episode, and race of victim and perpetrator.
The magnitude of the problem
The FBI estimates that 1.8 million Americans are victims of violence each year. Adolescents are more likely than any other age group to be victims, mostly from their peers. A complete representation of the magnitude of violence is not available because there are no reliable and consistent measures of non-fatal episodes of violence. Homicides, the tip of the iceberg, are more accurately measured as they are mandatorily reported by the United States’ local police departments to the FBI. Other countries have made their homicide rates available through the World Health Organization.
The magnitude of the problem of homicide in the United States is mind-boggling when compared with that of other industrialized nations not at war. Not only is the United States homicide rate 10 to 25 times higher than most industrialized nations, but the homicide rates rival those in some less developed countries facing war or considerable social, political, and economic turmoil (Wolfgang 1986).
The rate of youth violence in the United States is consistently higher than that of any other industrialized nation. In 1987, compared with 21 other developed countries, the United States had the highest homicide rate for males aged 15 to 24 years at 21.9 per 100 000 (Fingerhut and Kleinman 1990). Moreover, a study of 1985 homicide rates showed that the United States rate for both sexes, aged 1 to 19 years, was more than three times that of Canada which had the next highest rate (Williams and Kotch 1990).
Of the 16 914 murder victims in 1998, 94 per cent of the black victims were killed by black offenders and 87 per cent of the white victims were killed by white offenders. Firearms were used in six out of ten murders (FBI 1999). In 1997, the rate of homicide among males aged 15 to 19 years was 22.6 per 100 000—a decline of 12.4 per cent in one year. Homicide is the second leading cause of death for persons 15 to 24 years of age and is the leading cause of death for African Americans. Each month approximately 100 children die from gun violence (FBI 1997).
Non-fatal episodes of violence and assaults are not always reported to the police. Emergency departments captured a larger number of the assaults illustrated in the Northeast Ohio Trauma Study, where for every homicide there were 20 assaults reported to the police and 100 reported to the emergency rooms in one standard metropolitan statistical area in northeast Ohio. Many episodes of violence, particularly those occurring among friends and family, are not reported to the police or emergency departments.
The epidemic of youth violence in the United States is not limited to homicide. Police arrest data reveal an increase in non-fatal episodes of adolescent violence despite the limitations of the data set. The decade from 1980 to 1990 saw the juvenile violent-crime arrest rate for black adolescents increase by 19 per cent; for white adolescents it increased by 44 per cent, while the other-race category, despite a large increase in Asian youth, declined by 53 per cent (FBI 1992). Although minors make up less than 14 per cent of the United States population, they accounted for 37 per cent of homicide arrests, 28 per cent of rape arrests, and 51 per cent of robbery arrests in 1995. In 1982, 390 youths aged 13 to 15 years were arrested for homicide. By 1992, this figure had nearly doubled. Although these rates have been dropping since 1994, they are not dropping as quickly as adult violence rates (FBI 1997). The escalation of adolescent violent-crime rates in the last several decades cuts across race, class, and lifestyle, despite a common misconception that it is an urban black problem.
Violence involving youth has reached alarming levels in the United States. In 1991, the National Center for Health Statistics reported that homicide and legal intervention was the second leading cause of death for the group aged 15 to 24 years (22.4 deaths per 100 000, second only to motor vehicle accidents), at almost twice the rate for the overall United States population (11.7 deaths per 100 000) (CDC 1993a).
In 1997, 6146 people aged 15 to 24 years were victims of homicide. This amounts to an average of 17 youth homicide victims per day in the United States. Homicide is the second leading cause of death for persons aged 15 to 24 years, the leading cause of death for African Americans, and the second leading cause of death for Hispanic youths.
Furthermore, trends in adolescent mortality rates indicate that although overall death rates and death rates due to motor vehicle crashes both decreased for persons aged 10 to 24 years from 1979 to 1988, death rates for homicide increased by 6.7 per cent. The CDC estimates that by the year 2003 more Americans will die from firearms-related injuries than from motor vehicle accidents. This is already true in eight states (CDC 1994).
The 1990 National Crime Survey found that, even for non-fatal violent victimizations, age was one of the most important single predictors of an individual’s risk, which peaks at age 16 to 19 for both men (95 per 1000) and women (54 per 1000) (Bureau of Justice Statistics 1992, 1993). Clearly, our nation’s youth are at high risk for experiencing violence.
Dating violence is a form of adolescent and young-adult violence that is often overlooked, even within discussions of domestic violence. Very little research has been done and the general awareness of the problem of dating violence is relatively recent. A survey within a college population found that 21 per cent of the students admitted being in violent relationships, and 62 per cent knew personally of someone affected by a violent relationship (Makepeace 1981). Several studies indicate that between 12 and 19 per cent of high-school students are involved in dating violence, either as a victim or perpetrator. As with domestic violence, women are more often the victims.
School violence: more lethal
School suspension data offer another measure of the violence occurring, but there are several limitations. Suspension numbers may vary within a school depending upon the persons responsible for collecting the data. There are no standard criteria within or between school systems as to what behaviour will cause a student suspension.
Violence in schools is not new, but it is increasingly more severe and ethal. In the most recent Youth Risk Behaviour Survey conducted by the CDC, 18.3 per cent of students from the ninth to twelfth grades in the 50 states, including the District of Columbia and the Virgin Islands, reported carrying a weapon 30 days prior to the survey and 5.9 per cent reported that they had carried a gun. Seven per cent said they had been threatened or injured with a weapon while on school property in the 12 months before the survey. An estimated 74.2 separate incidents of weapon-carrying had occurred per 100 students 30 days prior to the survey. Among students nationwide, 36.6 per cent had been in a physical fight once or more during the 12 months preceding the survey.
The CDC survey also revealed that 4 per cent of the students said that they had missed at least one day of school in the previous month because they felt unsafe at school (no significant difference). In 1989, the United States Department of Justice found that 6 per cent of students report having to avoid certain places in school or on the way to or from school because they were afraid of being attacked. In a poll conducted by Metropolitan Life, 23 per cent of students said they never saw violent incidents at school. A Louis Harris Associates Poll of youth and guns conducted in 1993 of 2500 students in the sixth to twelfth grades showed that 15 per cent had carried a handgun to school in the past year, and 59 per cent said they could get a handgun if they wanted one. The addition of weapons to the typical school brawl has contributed significantly to the greater severity and mortality of school fights.
The CDC, the United States Department of Education, and the National School Safety Center are conducting a study of actual deaths in schools. Preliminary data show that 105 school-associated violent deaths (81 homicides, 19 suicides, and five unintentional firearm-related deaths) occurred in the school years 1992–1993 and 1993–1994. Sixty-six per cent of these occurred on school property and 75 per cent were committed with a firearm.
While the arrest rates for overall crime and violent crime are significantly lower for young women than for young men, recent increases in girls’ arrests have narrowed the gap. From 1981 to 1995, there was a 129 per cent increase in the violent-crime arrest rate for young women compared with a 56 per cent increase over the same time period for young men. Now, girls account for 25 per cent of the juvenile arrests for violent crime—unheard of two decades ago. More girls are entering the juvenile justice system and are doing so at younger ages; there has been a 10 per cent increase in the numbers of 13- and 14-year-olds coming into juvenile court.
Some experts, including Meda Chesney-Lind at the University of Hawaii, believe that there has not really been a significant increase in the proportion of young women committing crimes. Rather, ‘We’re criminalizing a lot of schoolyard scuffles where, in the past, we’d call it a cat fight, we’d giggle and keep walking. Now we’re calling the cops’.
Chesney-Lind and others also point to changes and biases within the juvenile justice system that they believe explain the increase in the number of girls being arrested. For example, girls are twice as likely as boys to be detained, with the detention period lasting five times longer for girls than for their male counterparts. Girls are more likely than boys to be charged with status offences, i.e. offences which, if committed by an adult, would not be considered a crime (e.g. running away from home, truancy, incorrigibility), and are more likely to be incarcerated for these offences than males.
Historically, suicide in the United States was viewed as primarily a problem of older adult white men with clinical depression or other mental disorders; thus suicide prevention involved identifying and treating mental illness. A dramatic rise in the adolescent and young adult (15–24 years old) suicide rates, from 4.5 per 100 000 in 1950 to 20.5 per 100 000 in 1997, created the need for new prevention strategies. The rise was alarming particularly as research indicated that only one out of three of those who committed suicide fit the criteria for clinical depression or other mental illness (Shaffer et al. 1988).
Race and gender disparities in suicide rates are striking. The rates for adolescent girls and women have remained relatively stable over the last 30 years and the rates for white men have levelled since 1988. However, there has been a dramatic rise in the suicide rates for young black males (15–24 years old) since 1986 (Shaffer et al. 1994). While the suicide rates for white males remain higher than those for black males at each age cohort, the rates among black males increased at a faster rate than any other group.
The CDC convened a panel of experts and conducted a study of youth suicide prevention programmes. The study reviewed the existing programmes and delineated eight suicide prevention strategies.
School gatekeeper training: this type of programme is directed at school staff to help them identify and defer students at risk of suicide and to organize the response in case of a suicide.
Community gatekeeper training: this type of programme provides the same service to community staff, clergy, police, merchants, etc.
General suicide education: these programmes are school-based education on suicide, often incorporating self-esteem building or social competency exercises.
Screening programmes: screening involves administering an instrument to identify high-risk youth in order to provide services.
Peer support programmes: school- or community-based programmes to help adolescents develop competency in relationships and to help each other.
Crisis centres and hotlines: these programmes provide 24-hour emergency counselling.
Means restriction: strategies to restrict access to firearms, drugs, or other means of committing suicide.
Intervention after a suicide: commonly called postvention, these programmes are designed to help survivors and prevent suicide clusters.
Domestic violence or partner violence is defined as violence between those involved in an intimate relationship. While it includes homosexual relationships and violence against men, it is usually the violence of a boyfriend, ex-boyfriend, husband, or ex-husband against a woman (Bureau of Justice Statistics 1994). It involves many kinds of attacks, physical, sexual, and verbal. Coercive control through degradation, malicious enforcement of petty rules, intermittent rewards, and isolations are examples of the methods employed to demonstrate and maintain power.
However, in family situations children under 12 years of age represent 62 per cent of all victims. Juveniles aged 12 to 17 comprise 30 per cent of the victims in overall offences and 23 per cent in family occurrences. Females are most frequently the victims of family and overall offences, comprising 74 per cent and 76 per cent of the victims of family and overall offences respectively (FBI 1998).
It is difficult to assess the amount of intimate violence that occurs, as most goes unreported. The Department of Justice regularly conducts the National Crime Victimization Survey showing that 2.5 million women annually report experiencing physical violence, two-thirds at the hands of an acquaintance or relation. The average annual victimization rates are 5 per 1000 by intimates, 1 per 1000 by other relatives, 8 per 1000 by acquaintances, and 5 per 1000 by strangers (Bureau of Justice Statistics 1994).
In 1985, the second large population-based study of domestic violence revealed that 3.4 per cent of adult women had been severely abused by an intimate partner, a prevalence of approximately 2 million. Minor acts of violence, pushing, shoving, or slapping, were reported by 11.6 per cent of the women.
In 1998, 27 per cent of those suffering family-related violence were reported to have been related to one or more of their offenders. A higher percentage of victims of family violence are over the age of 18 than are the victims of overall crimes of violence (80 per cent versus 76). Additionally, victims of family violence are overwhelmingly female (71 per cent for family violence and 58 per cent for overall violence) (FBI 1998).
Rape and sexual assault
In 1993, the FBI recorded 104 806 rapes for a national rate of 79 per 100 000 (FBI 1993). The National Crime Victimization Survey indicates that women report approximately 133 000 rapes each year, with half saying they reported them to the police and 55 per cent indicating that they knew the assailant.
Rape accounts for slightly less than 1 per cent of all violent offences. In particular, children under 12 comprise a larger portion of victims of family rape than all victims of rape (36 per cent versus 12) (FBI 1998).
Additional information is available from a national sample, the National Women’s Study (Kilpatrick et al. 1992), suggesting a higher incidence and prevalence. According to this study, an estimated 683 000 women are raped each year, 60 per cent at ages younger than 18. The perpetrator was a stranger in only 22 per cent of the cases. This study estimates that 12.1 million American women are raped at some point in their lives. Only 16 per cent of them report the rape to the police.
Each year United States citizens pay about $53.5 billion for criminal justice interventions for violence, and an additional $158 billion for cost of lifetime care for victims of violence (medical treatment, rehabilitation, and lost productivity). These figures reflect only the monetary costs of violence, not the pain, suffering, and lost quality of life for victims. They do not reflect the cost for safety measures—the inability of children and adults to walk or play in their own neighbourhood, the cost of guard dogs and guns for ‘protection’, and an immeasurable sense of fear of crime victimization. In considering the impact of violence on society, it is also important to note the costs of violent crimes.
A framework model developed by Miller et al. (1993) was used to quantify costs of violent crime; it incorporated direct losses other than property losses (medical, mental health, and emergency services, insurance administration), productivity losses (wages, fringe benefits, housework), and non-monetary losses (pain, suffering, lost quality of life). Costs to victims of crimes resulting in injury were estimated to be $60 000 for rape survivors, $22 000 for assault survivors, and almost $2.4 million per murder (in 1989 dollars). Moreover, the lifetime costs of criminal victimizations for persons aged 12 and older were estimated to be $10 billion for rape, $96 billion for assault, and $48 billion for murder (Miller et al. 1993).
Furthermore, these figures do not include property losses incurred during violent acts nor the mammoth costs incurred by collective society’s reactive response to violence, including law enforcement, adjudication, victim services, and correctional expenditures.
In 1993, the costs of direct medical services, emergency services, and claims processed for the victims of gun violence nationwide totalled approximately $3 billion. Average hospital charges for treating one child wounded by gunfire were more than $14 000.
Taxpayers pay for gun violence. The average cost of medical treatments for the hospital stay of one gunshot-wound patient (all age groups) is more than $33 000. Approximately 80 per cent of patients who suffer from violence are uninsured and/or eligible for government medical care assistance.
The characteristics of violence
Contrary to the stereotype of violence as predominantly stranger-related, or occurring in the context of criminal behaviour such as racial harassment, robbery, or drug-dealing, much of the violence experienced in the United States is intimate and occurs in the context of personal relationships (Spivak et al. 1988). A typical homicide involves two people who know each other, who are under the influence of alcohol, and who get into an argument that escalates in the presence of a gun. Only 15 per cent of homicides occur in the course of committing a crime, as compared with over 50 per cent that stem from arguments among acquaintances (CDC 1982a,b). This 50 per cent takes place in family relationships (e.g. child abuse, elder abuse, spouse abuse) or friends (interpersonal peer violence). In the remaining 35 per cent, the relationship between victim and perpetrator is unknown.
The perpetrator and victim of violence share many traits. They are likely to be young, male, and of the same race. They are likely to be poor and to have been exposed to violence in the past—especially family violence. They may be depressed and use alcohol and or other drugs (Prothrow-Stith and Weissman 1991). This incongruity between public perception and actual circumstances has resulted in demands for resources and solutions that address only part—possibly the smaller part—of the problem. While certainly not discarding established anticrime and antiviolence strategies, we must recognize the diversity of violent circumstances that exist and must build a broader base of efforts that not only responds to violent events, but also focuses on preventive services as well.
A closer look at the demographic characteristics reveal certain noteworthy factors contributing to a complex picture of adolescent violence. Breaking down the 10- to 24-year age group further, 1997 homicide rates (deaths per 100 000) were considerably higher among 20- to 24-year-olds (19.0) and 15- to 19-year-olds (11.7) than among 10- to 14-year-olds (1.7); however, it is still important to note that the rates increased among all three groups from 1979 to 1988. In terms of gender, males greatly exceed females in the number of violent victimizations, with the exception of sexual assault, and are also more likely to be violent offenders and witnesses to violence.
Race and poverty
There appear to be extremely large racial differences in violence rates among young Americans. In 1991, homicide was the leading cause of death for black youth aged between 15 and 24; the homicide rate for black youth (both sexes) was nine times the rate for white youth aged between 15 and 24 (90.0 per 100 000 versus 10.8 per 100 000). National statistics concerning other ethnic minority groups, such as Latin Americans, Asian Americans, and Native Americans, are scant.
The racial data are not indicative of any biological or genetic factor because they are confounded by socio-economic status, urban living, gun availability, and racism. Using family income as the primary indicator of socio-economic status, the National Crime Survey found an inverse relationship between income and the risk of violent victimization (Bureau of Justice Statistics 1992). In 1988, the risk of victimization was found to be 2.5 times higher for people in low income families (under $7500 per year), compared with high income families ($50 000 per year).
It is important to note, however, that the relationship between violence and social factors is complex and still unclear. For example, multivariate studies have shown a complicated interaction between race and socio-economic status: at low socio-economic levels black individuals have a higher risk of homicide than white individuals, but at higher socio-economic levels the difference disappears. William Julius Wilson’s work on neighbourhood poverty offers a possible explanation. Poor black people are much more likely than poor white people to live in neighbourhoods where the majority of the people are poor. Although it appears that race is a significant social predictor in certain studies, multivariate studies show a more complex situation.
Other studies have suggested that, in fact, socio-economic status is the major predictor and race is merely a marker. One study that used several markers for poverty, including number of people per square foot of housing, disaggregated the race and socio-economic variables. In this study overcrowded white people had the same high domestic homicide rates as did overcrowded black people. Less crowded members of both groups had the same lower rates (Centerwall 1984, 1993). In 1987, the homicide rate for young black men in the military was one-twelfth of their national rate, strongly indicating the influence of social, structural, cultural, and economic factors.
The contact that health professionals have with victims and perpetrators
The regular contact that physicians and nurses have with victims of violence, particularly in emergency departments, has caused many to begin to address this problem. The American College of Emergency Physicians has included violence prevention on the agenda of their annual meetings. The Journal of the American Medical Association has dedicated two special issues to the topic of violence, in parallel with the American Medical Association’s publication of manuals for health providers on domestic violence, child abuse, and rape and sexual assault.
The Northeast Ohio Trauma Study illustrates the need for greater data from emergency departments, in showing that five times the number of assaults reported to the police were reported to hospital emergency departments. Not only are non-fatal violent episodes inadequately measured with police data, but the greater contact that health providers have with victims provides an opportunity to offer public health prevention and intervention strategies in the emergency department. Such programmes have been started at Boston City Hospital, Cook County Hospital in Chicago, Harborview Hospital in Seattle, and Washington and Grady Memorial Hospital in Atlanta, Georgia.
There is a substantial body of research which suggests that crime rates reflect ‘community social disorganization’. Social disorganization theory was originally developed by the Chicago School researchers Clifford Shaw and Henry McKay in their classic work, Juvenile Delinquency and Urban Areas, published in 1942. Shaw and McKay demonstrated that the same socio-economically disadvantaged areas in 21 United States cities continued to exhibit high delinquency rates over a span of several decades despite changes in their racial and ethnic composition, indicating the persistent contextual effects of these communities on crime rates regardless of what populations experienced them. This observation led them to reject individualistic explanations of delinquency and to focus instead on community processes—such as disruption of local community organization and weak social controls—which led to the apparent transgenerational transmission of criminal behaviour. In general, social disorganization is defined as the ‘inability of a community structure to realise the common values of its residents and maintain effective social controls’. The social organizational approach views local communities and neighbourhoods as complex systems of friendship, kinship, and acquaintanceship networks, as well as formal and informal associational ties rooted in family life and ongoing socialization processes. From the perspective of crime control, a major dimension of social disorganization is the ability of a community to supervise and control teenage peer groups, especially gangs. Thus Shaw and McKay argued that residents of cohesive communities were better able to control the youth behaviours that set the context for gang violence. Examples of such controls include the supervision of leisure-time youth activities, intervention in street-corner congregation, and challenging youth ‘who seem to be up to no good’. Socially disorganized communities with extensive street-corner peer groups are also expected to have higher rates of adult violence, especially among younger adults who still have ties to youth gangs.
The application of public health strategies
Public health professionals have applied traditional public health strategies to violence prevention. They have brought a different perspective and orientation to bear on the problem. Applying public health techniques and strategies complements and strengthens the criminal justice approach.
Public health brings an analytic approach to problems that concentrates on identifying risk factors and important causes that could become the focus of preventive interventions. It also brings a record of accomplishment in controlling ‘accidental’ (unintentional) injuries through both environmental manipulations (e.g. seat belts and childproof caps on medicines), and behavioural change (e.g. laws and educational campaigns to reduce drink-driving).
Identification of risk factors
Major risk factors for youth violence have been identified. These factors can be broadly categorized as environmental or psychological. The major environmental risk factors include firearms, alcohol and other drugs, cultural factors, being a victim of child abuse, witnessing family violence, exposure to media violence, and exposure to high levels of peer and community violence. A consistent and strong environmental risk factor for homicide is the presence of poverty. The mechanism for this interaction is not completely understood, but may include the anger and frustration associated with not having money and essential commodities, the experience of classism, the probable absence of adult male role models, the scarcity of recreational, extracurricular, and after-school activities, and longer time spent watching television.
Corporal punishment is a controversial environmental factor that may be related to risk for violence. Certainly in its extreme form, abuse, there is evidence to suggest that it increases the risk of delinquency. Efforts to improve parenting and to reduce child abuse often focus on alternative disciplinary strategies. Other environmental risk factors for adolescents include peer pressure, the crack cocaine epidemic, and policing practices.
Child neglect and abuse
Child abuse and neglect are general terms used to encompass many harmful behaviours towards children. Verbal, emotional, and sexual abuse are included, as well as failure to meet a child’s needs and outright physical violence. Child sexual abuse is most often considered separately, yet each state has its own definition and guidelines for protective custody.
Because child abuse has been a reportable crime for many years, better statistics are available. An annual 50-state survey estimated 2 million reports of child abuse and neglect in 1986. Over the decade of the eighties, the number of reports increased by 184 per cent (Daro and Mitchel 1987). The number of child sexual abuse reports increased dramatically as well, a 12-fold increase within the decade. It is obvious that researchers have the same problem in documenting both child abuse and child sexual abuse as they do in documenting other forms of violence: unreliable data sources, under-reporting by victims, inconsistent definitions, and failure to recognize an event as precipitated by violence.
With child abuse, reporting biases work in both directions to inflate or deflate the numbers. Episodes of child abuse occurring in families of middle class and professional parents are less likely to be reported, even with mandatory reporting laws, which will diminish the prevalence estimates. Yet greater awareness and sensitivity to child abuse, and the advent of mandatory reporting, no doubt increase the numbers. There is a struggle among child health and human service professionals to determine the way to maintain mandatory reporting and improve the effectiveness of the state protective services. An over-reliance on foster care without adequate attention to family preservation seems to have been the rule in the past.
The cycle of violence
The relationship between child abuse, neglect, and witnessing violence to adolescent and adult violence has been demonstrated in several studies. Existing studies suggest that there is a greater likelihood of abuse by parents if they were abused as children. Estimates of the percentage of abusive parents who were abused as children range from 7 per cent (Gil 1973) to 70 per cent (Egeland and Jacobvitz 1984). Among adults who were abused, up to one-third abuse their children (Straus and Gelles 1990).
A retrospective look at violent juvenile delinquents compared with non-violent juvenile delinquents showed a significantly higher rate of physical child abuse. Both interviews with the delinquents and medical chart reviews yielded evidence of greater victimization, skull fractures, emergency trauma visits, and other physical injuries.
A cohort study of abused or neglected children demonstrated a greater risk for delinquency, adult criminal behaviour, and violent criminal behaviour, even though the majority of such children do not demonstrate these behaviours. The abused children had a number of offences and began delinquent behaviour at earlier ages, regardless of race and gender (Widom 1989).
For black adolescents aged 11 to 19 years living in or around an urban housing project, the self-reported use of violence was associated with exposure to violence and personal victimization, hopelessness, depression, family conflict, and previous corporal punishments. Those with a higher sense of purpose in life and less depression were better able to handle the exposure to violence in the home and community (Durant et al. 1994).
Children and exposure to media violence
The association between childhood exposure to media violence and subsequent aggressive behaviour has been firmly established over the past four decades. The American Psychological Association collected these data and decided unequivocally to pronounce the negative influence of the entertainment media violence (American Psychological Association Commission on Violence and Youth 1993). The Association expected the report to have an impact on parents and policy-makers. Other reviews of the literature have been done with similar conclusions (Dietz and Strasburger 1991; Sege and Dietz 1994).
Preschool children exposed to violent activity in a controlled setting were observed to imitate and repeat the violent behaviours (Bandura et al. 1963a,b). An actor appearing on screen attacked a Bobo-the-clown doll. Following this attack, in three separate video sequences the actor was either praised, ignored, or punished. Those preschool-aged viewers who saw the violent behaviour rewarded on screen were more likely than the other two groups to repeat the violent actions when shown a Bobo doll themselves. This experiment demonstrated that children can learn violent behaviours from television, and are especially likely to do so when these activities are depicted as socially acceptable.
Older children’s behaviour is also heavily influenced by exposure to media violence. Meta-analysis of a series of experiences demonstrates conclusively that school-aged boys have more fights in the days following exposure to violent mainstream movies than they do in the days following exposure to less violent movies (Turner et al. 1986; Wood et al. 1991).
In a landmark cohort study involving children raised in Pennsylvania, Eron and Huesmann (1984) showed that preference for violent television programmes at age 8, as well as total hours of television viewing, predicted the severity of violent criminal convictions by age 30. However, this effect should be modified by parental interventions (Huesmann et al. 1983; Liebert 1988; Austin et al. 1990; Weaver and Barbour 1992; Sang et al. 1993).
Centerwall (1992) has shown that in three different countries (the United States, Canada, and South Africa), homicide rates doubled approximately 10 to 12 years after the introduction of English-language television. In the United States, homicide rates doubled first among those portions of the population exposed to television first (white urban dwellers), and only later among those segments of the population who received television later. He attributes approximately 10 000 deaths annually in the United States to the results of exposure to media violence.
Taken together, we believe that these studies satisfy most of the criteria for causality set forth in the Surgeon General’s report on smoking and health (US Department of Health, Education, and Welfare 1964), and establish that exposure to media violence places children at risk for subsequent violence.
Public debate flourishes concerning the roles of video games and violence-oriented musical lyrics in encouraging violence. Currently, however, no definitive data are available on these issues.
The United States has more firearms than any other industrialized nation not at war and the following facts extracted from the publication Not Even One (Carter Center 1994) are astounding.
A gun in the home is 43 times more likely to kill a family member or friend than it is to be used in self-defence.
In 1990, 4941 children in the United States under the age of 19 years died from gunshot wounds; 538 of these children were shot accidentally.
From 1973 to 1991, the number of licensed firearms dealers increased by 95 000—the total had reached more than 225 000 by 1994. In 1992, nearly 92 000 Americans applied to get or renew a federal firearms licence. Only 52 were denied.
In 1987, 1300 males under age 19 were murdered with guns in the United States. In the same year, fewer than 80 males under 19 were murdered with guns in Canada, Japan, France, West Germany, Australia, England, Wales, and Sweden combined.
None of the federal revenues on guns are designated for the medical care of victims of gun violence. In fact, all of the revenues from the firearm excise tax are required to go to hunting-related activities.
The firearm homicide rate for 15- to 19-year-olds increased 61 per cent from 1979 to 1989. The rate of homicide by all other methods remained stable or declined.
In 1991, the Bureau of Alcohol, Tobacco, and Firearms performed compliance inspections on fewer than 4 per cent of all existing gun dealerships; 5967 violations were found but only 17 dealers’ licences were revoked.
Firearms are used in more than 80 per cent of teenage homicides and about 68 per cent of homicides by all ages.
In 1990, 4.37 million guns were produced for the American market. That is 12 000 new guns every day. Although trigger lock loading indicators would save lives, few guns have them because no law requires them.
Suicide is the third leading cause of death for adolescents and young adults in the United States, after car crashes and homicides. In 1990, 3165 youths aged 15 to 24 killed themselves with guns.
Half of all Americans own guns. More than one-third of all male homicides are by firearms. Nearly half of all female homicides are by firearms.
From 1953 to 1978, the suicide rate of young people tripled; this rise paralleled an increase in the firearm-caused suicide rate.
From 1976 to 1987, more than twice as many American women were shot and killed by their husbands or boyfriends as by strangers using guns, knives, or any other means.
Guns are used in 60 per cent of all teenage suicides. The youth firearm-suicide rate in an American city with minimal restrictions on gun ownership is more than three times higher than in a Canadian city with strict gun-control laws.
Teenage and young adult homicide is a uniquely American problem. The high rate of youth homicide in the United States has been attributed to the country’s much higher rate of gun ownership. An international study of gun ownership and homicide found positive correlations between the rates of household gun ownership and the national rates and proportions of gun-related homicide (Lester 1988; Killias 1993).
Handgun availability appears to be playing an increasingly important role in youth homicides. As an example, the increasing trend seen in the total homicide rate among those aged 15 to 19 years between 1979 and 1989 is solely attributable to the increase in firearm homicides: the firearm-related homicide rate increased by 61 per cent (6.9 to 11.1 per 100 000), while at the same time the non-firearm-related homicide rates actually decreased by 29 per cent (3.4 to 2.4 per 100 000) (Fingerhut et al. 1992). From 1980 to 1989, over 65 per cent of the 11 000 homicides committed by high-school-aged youth were firearm related.
Handguns are widely accessible to adolescents in the United States. The national 1990 Youth Risk Behavior Survey found that about one in every 20 high-school students had carried a firearm at least once in the 30 days preceding the survey. The incidence was higher among males at 17 per cent. (CDC 1991). In another study of inner-city youths, as many as 35 per cent of males carried a gun outside of school (Sheley et al. 1992).
Firearms contribute to both the violent victimization of youth and the violent offences committed by youth. The presence of a weapon in the home is associated with a threefold increase in the likelihood of homicide compared with matched controls drawn from the neighbourhood surrounding the victim (Kellermann et al. 1993).
Comparisons of two cities (Seattle and Vancouver) which have similar demographic characteristics shows that Seattle’s excess homicide rate is entirely attributable to firearm homicides (Sloan et al. 1988) (Fig. 4). Another study, designed to look at the effect of implementation of gun control legislation, showed a positive effect of the enactment of tougher gun control laws in the District of Columbia compared with both neighbouring states (Loftin 1991).
Fig. 4 Weapons and violence. (Source: Sloan et al. 1988.)
All these studies demonstrate that availability of firearms is strongly correlated with increased homicide rates. Logically, this result coincides with the earlier observation that most homicides result from conflicts among people who know each other well, including friends and acquaintances, as well as relatives and spouses. In a situation of passionate conflict, handgun availability appears to increase the likelihood of serious injury or death.
Psychological and behavioural factors
In pioneering studies conducted in the late 1980s, Slaby and Guerra (1988) demonstrated that adolescents involved in violence have habits of thought which lead them into violent confrontations. They examined the responses of three groups of teenaged boys to a specific scenario. One group of boys were in custody for the commission of violent crimes, a second group were identified by their teachers as being violence prone, and a third group were identified by their teachers as not being violence prone. Each boy was presented with the same scenario: he was going after school to work on his batting so that he could make the baseball team. As he arrived, another boy took the last bat. In understanding this scenario, the violence-prone boys were more likely to assign malicious intent to the boy who took the bat than were the less violent boys. The violence-prone boys were able to imagine fewer alternative means to resolve this situation.
Slaby and Guerra (1988) concluded from this study that the more violence-prone boys got into more fights because they were more likely to see harmful intent in a given situation, and, having seen such intent, were less likely to come up with a peaceful way to resolve the situation. These results have been confirmed with a large-scale study conducted in New York City schools (CDC 1993b). Boys who reported having been in serious fights were more likely than the general school population to suggest that carrying a weapon or threatening to use a weapon were good ways to stay out of fights; they were also far less likely than the overall population to say that one could avoid fighting by apologizing.
In addition to the young people directly injured by violence, increasing attention is being given to the scores more who are affected indirectly, as witnesses to violent acts or by exposure to chronic violent environments (Groves et al. 1993). Pynoos et al. (1987) examined the appearance of post-traumatic stress disorder symptoms in children who experienced a fatal sniper attack on their elementary school and reported a correlation between the type and number of post-traumatic stress disorder symptoms and proximity to the violent incident, as well as more severe symptoms in children who knew the deceased child.
In addition to acute incidents, other studies relate findings of correlations between exposure to chronic violence and distress symptoms (Fitzpatrick and Boldizar 1993; Freeman et al. 1993; Osofsky et al. 1993; Martinez and Richters 1993; Lorion and Saltzman 1993). In addition, Lorion and Saltzman (1993) described anecdotal reports from their research participants, including reports from teachers and administrators about children who lived in violent settings arriving at school in distress, who were unable to concentrate or maintain appropriate behaviour in class, and who hid in the classroom, afraid to return home or take the bus. Clearly, there is a need to address not only the physical threat of violence, but also the potential for psychopathological and/or emotional disturbances in both victims and bystanders (Emde 1993; Durant et al. 1994).
Approaches to violence prevention and control
Historically, society has relied almost exclusively on the criminal justice system both to respond to and to prevent violence. This tactic is rooted in the beliefs that violence is criminal, that those who commit violence should be punished, and that the threat of punishment is a potential deterrent to violent acts. A large, elaborate set of institutions has been developed to achieve these goals. That system includes police, prosecutors, public defenders, judges, probation officers, and prison guards. It is principally designed to respond to crimes after they have been committed by identifying, apprehending, prosecuting, punishing, and controlling the violent offender. It is guided not only by the practical goals of reducing crimes of all types (including violence), but also by the normative goal of assuring justice to victims and the accused.
The public health and criminal justice systems have been historically separate in their conceptualization of approaches to violence and the development of activities to reduce or prevent violence. The public health field has approached the issue through efforts to identify the risk factors related to violent behaviour. The field comes to this issue in reaction to the magnitude of intentional injuries that are present in health care settings. The criminal justice system has approached the issue through efforts to identify and assign blame for criminal behaviour, maintain public safety, and remove violent offenders from the community.
Viewed from the perspective of those interested in reducing violence, the criminal justice system’s responses have had only limited success. Part of the reason is inherent limitations in the overall approach of the criminal justice system. First, it is more reactive than preventive in its basic orientation. True, deterrence may produce some preventive results. True too, the criminal justice system has sought to rehabilitate offenders through special programmes in prisons, and to prevent children from becoming violent offenders through the development of the juvenile justice system whose most fundamental goal is to prevent future criminal activity by children. Nonetheless, the criminal justice system comes into play only after a crime episode occurs.
Second, the criminal justice system—and particularly the police—is focused primarily on the predatory violence that occurs among strangers on the street. The violence that emerges from nagging frustrations and festering disputes, and takes place in intimate settings, is far more difficult for the criminal justice system to deal with than stranger-inflicted violence that arises from greed or desperate need and takes place in the open. Robbery and burglary, and the violence that attends these, are more traditional and central to the criminal justice system’s business (and consciousness) than aggravated assaults that spring up among friends in bars, lovers in bedrooms, or teenagers at dances.
Public health and criminal justice: interdisciplinary challenges
Unfortunately, the collaboration of public health and criminal justice in the area of violence prevention has been fraught with tension. Some of this may stem from a basic failure in effectively reducing the problem of violence that has put both disciplines on the defensive—criminal justice for its failure to bring the problem under control and meet societal expectations, and public health for the slowness with which it has recognized and taken on the problem. However, much of this tension probably comes from the divergence of perspective of the two disciplines and the fact that there are inadequate resources directed to addressing violence, which has forced the disciplines to compete rather than collaborate.
Public health is primarily focused on identifying causality (or its approximation) and intervening to control or reduce the risk factors; it has little interest in assigning blame or meting out punishment, and does not discriminate between victim and offender. The public health community may agree that justice must be done, but is not professionally committed to the process. The criminal justice system, on the other hand, is deeply and morally rooted in ‘justice’, and the proper identification and punishment of criminal offenders. In this field there is less emphasis on the precursors or factors that may have led to the violent event. The criminal justice system is less likely to consider external factors that might have motivated the offender to engage in violence because it sees these issues as largely irrelevant to judgement of guilt and innocence. At worst, the claims that these other factors were causally important in the particular instance seems like a rationalization or an apology for what was a criminal deed. This rift is further exacerbated by the fact that the criminal justice profession continues to develop preventive agendas, such as first-offender programmes and community-policing initiatives, and probably feels that their ‘thunder’ and leadership are in jeopardy of being stolen by the entry of another professional player onto their turf.
This tension is clearly unproductive. It threatens effective collaboration and frustrates the opportunity to pool resources and expertise at a time when resources are seriously inadequate and the problem is increasing. Healing this rift requires a more collaborative spirit from both disciplines. The public health ‘purists’ must go beyond their science and recognize the invaluable contributions and practical experiences of the criminal justice professionals. The criminal justice ‘moralists’ must, in turn, recognize the limitations of a primary agenda of assigning blame and assuring justice is done.
If we are to get past these initial reactions and successfully exploit the complementary qualities of these two approaches to violence, it is essential to put aside professional jealousies. More important, we must better define the perspective, roles, and expertise both groups bring to the issue. This will not only lead to a more creative process but also to establishing productive working relationships.
Primary, secondary, and tertiary prevention
A conceptual framework that can alleviate interprofessional tension, facilitate definitions of roles in addressing the problem, and assist in developing a broader perspective on programmatic strategies, involves breaking the spectrum of violence into levels that reflect different points of intervention (Table 1). This framework, used frequently in public health circles, structures approaches to problems into three stages of prevention: primary, secondary (or early intervention), and tertiary (or treatment/rehabilitation). These distinctions have proved valuable in thinking about intervention efforts even though their boundaries are not discrete. In this discussion, it might be best to think of these distinctions in terms of concentric circles that widen out in space and time from a central point which is the occurrence of some violent event.
Table 1 Classification of preventive strategies
Tertiary prevention is distinguished from secondary and primary prevention in that it lies on the opposite side of the violent event from the other two. Its focus is on trying to reduce the negative consequences of a particular event after it has occurred, or on trying to find ways to use the event to reduce the likelihood of similar incidents occurring in the future. Thus one might think of improved trauma care, on the one hand, and increased efforts to rehabilitate or incapacitate violent offenders, on the other, as tertiary prevention instruments in the control of or the response to violence.
Primary prevention, which by definition addresses the broadest level of the general public, might seek to reduce the level of violence that is shown on television or to promote gun control. This would be an effort directed towards dealing with the public values and attitudes that may promote or encourage the use of violence.
Secondary prevention is distinguished from primary prevention in that it identifies relatively narrowly defined subgroups or circumstances that are at high risk of being involved in or occasioning violence, and focuses its attention on them. Thus secondary prevention efforts might focus on urban poor young men who are at particularly high risk of engaging in or being victimized by violence, and educating them in non-violent methods of resolving disputes or displaying competence and power.
The relative risk level of groups or circumstances is a continuum—with some people and circumstances at very high risk (a person who has been victimized by violence in his or her own home, also surrounded by violence in school, entering a bar in which members of a rival gang are drinking), and others at relatively low risk (say, a happily married professor, who owns no weapon more lethal than a screwdriver, writing on her computer at home). Moreover, it is generally true that the higher-risk groups are smaller than the lower-risk groups.
Primary prevention instruments are those that can affect larger and larger populations, ideally at relatively low cost. Indeed, the need to reach very large populations requires primary prevention efforts to be low cost per individual reached. Thus primary prevention instruments tend to be those providing information and education on the problem of violence through the popular media, for example the recruitment of Bill Cosby to the cause of using the media to prevent adolescent violence, or Sarah Brady’s efforts to advocate for gun control laws and educate the public about the risks of handguns, rather than providing non-violence training to the entire population. There are, of course, the ultimate long-term primary prevention goals that have to do with eliminating some of the root causes of violence such as social injustice and discrimination.
This public health model can be very useful when applied specifically to the issue of interpersonal violence. In the past, the criminal justice system has addressed each of the three points of intervention to varying degrees as represented, yet the bulk of the efforts have focused on the response to serious violent behaviour with moderate attention to early identification and intervention and limited efforts in the area of primary prevention.
The major activities of the criminal justice system have historically involved the roles of the police, the courts, and the prison system in responding to criminal or violent events. Most resources have been directed to investigating and punishing criminal behaviour. Tertiary prevention has generally involved incarceration. In the area of secondary prevention, the police have focused efforts on ‘situated’ crime prevention and the juvenile justice system has made attempts at early intervention with youthful offenders, although youths were frequently ignored by the courts and probation system until their criminal behaviour reached a relatively high level of concern. Primary prevention efforts have focused on elementary-school drug and violence prevention education by the police, as well as on controlling ‘criminogenic’ commodities such as drugs, guns, and alcohol.
With the more recent involvement of the public health system, attention has been broadened with enhanced efforts in the prevention arena. The public health agenda has focused primarily on prevention and early intervention, playing only a small role in the treatment of individuals with serious violence-related problems. The role and activities of the public health system are newer, less extensive, and therefore less evolved than that of the criminal justice system.
Traditionally, public health responded by treating the violence-related injury in the emergency setting. Today a new generation of committed health practitioners, community violence-prevention practitioners, social workers, and community activists have devised numerous intervention programmes to serve medium- to high-risk adolescents. At the primary prevention level, efforts have focused on gun control and safety, and enhanced public awareness of risk factors and the true characteristics of most violence to dispel myths and modify societal values around the use of violence. Additionally, some educational interventions (e.g. violence-prevention curricula) have been applied to broader, less high-risk settings. Again, much of this work is relatively recent and therefore has not yet established a long track record to assess fully its effects. Finally, public health has applied its analytical expertise to enhance greatly the understanding of risk factors, allowing for a broader vision in the planning and development of preventive approaches (Spivak et al. 1988; Prothrow-Stith and Weissman 1991).
In the area of secondary prevention, public health has been involved in the development of educational interventions specifically focused on behaviour modification of high-risk individuals, particularly children and youth. A number of curricula are currently in use addressing both the risks of violence in solving problems and conflict resolution techniques (Spivak et al. 1988; Prothrow-Stith and Weissman 1991).
It is important to note that the criminal justice system has increased its involvement with primary and secondary prevention efforts. For example, some criminal justice professionals have become increasingly involved in gun control initiatives. In 1974, the Juvenile Justice and Delinquency Prevention Act was passed and gave the Justice Department primary responsibility for delinquency prevention programmes. The Office of Juvenile Justice and Delinquency Prevention was designed in part to encourage development of model delinquency prevention programmes. One such programme is the Boys Clubs of America Targeting Programs for Delinquency Intervention. Other community groups refer at-risk boys to the programme, who are then recruited. Early evaluations of these programmes seem promising. Data indicate that 39 per cent of the boys did better at school and 93 per cent who completed the programme have not been reinvolved with the juvenile system (Boys Clubs of America 1986). These types of interventions reflect an important interface between the criminal justice and public health professions. With further attention, the dedication of resources of the public health system to this issue, and the broadening vision of criminal justice, a more reasonable balance between prevention and treatment can be achieved in the future. Efforts can be broadened to reflect more fully the range of efforts needed to both reduce the extent of violent behaviour and to respond to the violence that does occur. The emphasis of the public health system will be on prevention, with the criminal justice system prioritizing the response to violence, but with both disciplines working together across the spectrum.
Cigarette smoking reduction: a model for intervention
To illustrate the advantages of this approach, it is useful to review how it has worked successfully in other areas. One example, which on the surface appears to be a considerable stretch from violence, is the multidisciplinary approach that has been developed to deal with tobacco use. It is important to note that, while this example illustrates a collaboration between public health and the medical care system, it represents a useful analogy to the possible collaboration between public health and criminal justice.
Smoking is a major contributing factor to death and disability in this country. Significant inroads have been made in turning the tide on this major health threat. What was once a valued, sexy, and socially acceptable behaviour is now viewed as disgusting, unhealthy, and socially unacceptable. Heroes in the media used to smoke all the time; now they rarely do. Nationally, the number of people who smoke has declined dramatically, although smoking was and still is a learned behaviour—one that can be unpleasant or distasteful to start but is extremely difficult to stop.
The strategy to deal with smoking involved a three-pronged approach: primary prevention for those not yet smoking to teach the reasons for not starting and to support the decision not to start, secondary prevention to encourage stopping or reducing use for those who had already started smoking (often this involves helping individuals to identify alternative behaviours to replace the smoking), and treatment in the form of surgery, chemotherapy, and other medical interventions for those smokers who have developed cancer or other health consequences of their behaviour. Broad public initiatives to alter the societal values that encouraged smoking were also established to support the above efforts. This has been done through legislation (package labelling, advertising constraints, restrictions on sales to minors, establishment of smoke-free environments), public education, and pressure on media to change images and role models. Although, as stated earlier, this is an example of a public health–medical care interface, it represents an important success that has value when looking at the possibilities of a public health and criminal justice collaboration in addressing violence.
A similar approach could and should be taken with respect to violence. Primary prevention strategies and more targeted secondary prevention efforts need to be applied that proactively value and teach non-violent behaviours in response to anger and conflict. This is particularly important given the growing evidence that violence is a learned behaviour (Bandura et al. 1963a,b; Liebert et al. 1973; Slaby and Quarfoth 1980; Allen 1981; Eron and Huesmann 1984; Prothrow-Stith and Weissman 1991; Straus 1991; Vissing et al. 1991). Well-child health visits in neighbourhood health centres provide an ideal window of opportunity for early intervention. Peter Stringham, a paediatrician at the East Boston Neighborhood Health Center, incorporates a violence-prevention protocol for families, from the new-born visit to the teenage years. Social skills are as important to teach our children as the academic subjects we now emphasize in our society. This will in no way eliminate the underlying societal stresses that influence violent behaviour, but can affect and direct responses to these stresses towards a prosocial and productive outcome. Curricula that place emphasis on decision-making, non-violent conflict resolution, and self-esteem development currently exist but are extremely under-utilized and viewed as an ‘add on’ in academic settings rather than a basic component. A move to place more emphasis on the use of such curricula, with enhanced investment in social and support services for families and youth, will be an important step in countering the learned use of violence by our youth. Such a move would also require that the education, human service, and public health institutions play major roles in effecting these changes in our communities.
Indeed, the recognition that education designed to teach non-violent behaviours might be an important part of a combined public health–criminal justice response to the problem of violence helps to remind us that the modern view of how the law operates on behaviour in the society has become far more narrow that it once was. In our modern conceptions of the law, we imagine it operating on individual behaviour primarily through its incentive effects—the promise of punishment for misconduct made concrete and credible through individual prosecutions.
In the classic writings on laws, however, a great deal of attention was devoted not only to the passage of laws and to their application to individual cases, but also to their promulgation throughout the society (Friedman 1975). Extensive efforts to explain and educate citizens as to why the laws were necessary helped to ensure both their justice and their efficacy. Unless citizens knew about the law—its spirit as well as its letter—they could not reasonably be held accountable for failures to obey it. If the purpose of the law were not made clear, voluntary compliance—which was crucial to the law’s effect—could not be counted upon.
The public health community’s interest in non-violence education can be viewed as the modern rediscovery of the importance of explaining to and educating the public about violence, as well as simply having laws and applying them. It also incorporates an important modern discovery about the promulgation of obligations: it is often far easier to persuade people to comply with an important obligation when one can show individuals that it is in their best interests to do so, and when one can help them comply with the law. Persuasion and assistance are often more effective tools than accusation and blame. Still, it often helps in persuading and assisting if there is a broad social rule against violence that becomes part of the context for the education. Thus behavioural change may depend on a combination of education and laws that used to be called promulgation.
Gun control legislation efforts represent an important example of the interconnection between education and laws. Although there is growing support for increased handgun ownership restrictions as a primary prevention strategy, legislation alone is unlikely to create great change in violent injury rates in the foreseeable future. With over 60 million handguns in circulation in the United States (Bureau of Alcohol, Tobacco and Firearms 1991), an understanding and acceptance of the risks of handgun ownership and carrying is as important as legislative restrictions to reducing intentional handgun injuries.
Secondary level strategies require a more targeted effort. It requires early identification of individuals at high risk for, or already beginning to exhibit, violent behaviour and the development of treatment services for these individuals. This area represents an important interface between the human service and the criminal justice systems because the early identification of individuals at high risk for violence requires considerable collaboration. Points of early identification occur in schools, health facilities, police departments, courts, and a variety of other community institutions. Professional training in early identification and appropriate evaluation and treatment is necessary. This is not an easy process. Professional definitions and institutional boundaries have been established that encourage limited one-dimensional approaches.
Treatment interventions (tertiary prevention) for the most seriously affected individuals represent a key focal point for the criminal justice system. Violent behaviour cannot be condoned; punishment is an appropriate response to violent crimes or episodes and some individuals with serious pathology are not able to live in the general society. While it is essential that we understand how violent behaviour evolves, we must deal with it firmly to maintain safety within communities.
Although tertiary prevention falls most extensively into the criminal justice realm, with incarceration as the major strategy, public health needs to work along with the prison system in the area of rehabilitation. Without increased attention to rehabilitative efforts, including supportive services for those returning from prison to the community, most will continue to leave the prison system without the skills to avoid violence in the future. Public health must advocate for and support drug and alcohol treatment services, job training efforts, and conflict resolution and violence prevention skills, as well as the development of more extensive behaviour-change interventions. To date, successful rehabilitative efforts are limited, further reinforcing the need for more attention to be focused on this area as well.
Promising prevention programmes using public health strategies
Researchers Joy G. Dryfoos and Lisbeth Schorr both agree that the most effective programmes must be comprehensive, family and community oriented, and collaborative in nature. Some schools, communities, social agencies, and politicians around the country have incorporated this formula for success and have developed strategies to help children and their families prevent or cope with violence. These programmes offer the opportunity to learn from their successes and failures.
The Boston Violence Prevention Program
The Boston Violence Prevention Program, an intensive community-based outreach and education effort run by Boston’s Department of Health and Hospitals, was launched in 1986 as part of its Health Promotion Program for Urban Youth. Much of its early work, supported by foundations, was focused on training teachers and youth services staff how to teach adolescents about the risks of violence, and discussing measures that can be employed to avoid fights using the Violence Prevention Curriculum for Adolescents. Regular training is offered for teachers, health workers, street outreach workers, and peer workers. The programme’s peer leaders group is trained and actively participates in training sessions offered by the programme. A mass media campaign, Friends for Life Don’t let Friends Fight, was developed to create a new community ethos in support of violence prevention. The programme has been part of the city budget for the last 5 years and has generated several spin-off activities.
Resolving Conflict Creatively Program—school-based conflict resolution programme in New York, California, and Alaska
The Resolving Conflict Creatively Program, a holistic school-based conflict resolution programme, works with the entire school structure to create safe schools. The K-12 curriculum, developed and refined since 1986, requires support from school administration although it does not mandate that every teacher be trained. Each of the 32 school districts in New York City that use the curriculum maintain a certain autonomy, leaving decision-making to the school community. After an intense 40-hour training on the curriculum, teachers incorporate the methods into their classrooms.
The progamme invites and encourages the participation of parents in training. Both teachers and administrators in these schools document fewer fights and a sense of peace. Students become peer mediators and receive special training to negotiate and mediate arguments that break out during school hours.
The programme provides a win–win situation. Not only do students develop leadership skills and a sense of responsibility for peace and respect, but teachers and administrators find practical use in their lives for conflict resolution skills. The programme also offers regular advanced training sessions in such topics as helping students deal with death and grief.
A walk through the hallways of the Satellite Academy High School in the Bronx gives the sense that, despite unsafe surroundings, the school offers an oasis of safety, a place of mutual respect—a place where learning takes place. Programme evaluations show that both teachers and students notice a positive change in the schools. A more comprehensive programme evaluation was recently funded through a grant from the CDC.
Save Our Sons and Daughters—a community-based programme working in Detroit public schools
Clementine Barfield Chism, Executive Director of the community-based programme Save Our Sons and Daughters (SOSAD), works daily to restore a sense of safety in a community that regularly experiences violence. She is one of the first in this country to recognize the feelings of children after a violent episode: ‘They are scared. They see violence is a way of life so they act out. It is normal’. She cringes when she hears people talk about inner-city children experiencing post-traumatic stress disorder. To her, it is not a disorder, but a natural response to the chaotic violence around them.
After losing her son to violence, she founded SOSAD to provide grief counselling for her other sons. Over the past years, SOSAD activities have grown to include a K-12 curriculum—the Peace Program. The Peace Program was implemented in some Detroit schools in the communities most devastated by violence. The Peace Program teaches age- and culture-appropriate conflict resolution, and helps children and adolescents find a sense of empowerment in establishing and maintaining peace in their school by teaching them to become peacemakers. It also deals with the grief and pain that comes with losing someone to violence.
Peace Program schools proudly display a peace chain. Each child signs a peace pledge, and the chain links them all together. Early assessments indicate that teachers report improved school climates and that children are excited with their roles as peacemakers and peacekeepers.
The teachers of one particularly tough group of freshman girls, known for their physical fighting at one of Detroit’s high schools, affirm the success of this programme. These girls have made a pledge not to fight. They are fully engaged in learning and practising the skills of peace.
School-based management—Comer method schools
More than 300 schools from around the country have adopted an educational model designed by Yale University’s James P. Comer. A psychiatrist, Comer began his school-based work in one of New Haven’s most troubled schools in 1968. He designed a multidisciplinary approach to school management. The programme stresses that a partnership between educators and parents is critical to a school’s success. Its philosophy is based on the premise that each child is special and that schools should be places of learning.
Comer’s model is also based on the premise that the initial relationship between schools and disadvantaged parents is too often wrought with mistrust and alienation. Comer schools require parental representation on all levels, including the school’s management team. These parents also work to increase the involvement of other parents in the school’s mission to educate all children.
Powerful Schools of Seattle, Washington
I’ve been wanting to go to night school and learn how to use a computer, but when I get home after work, I just don’t have the energy to drive half way across town, hunt for a parking place, and take a class at a community college . . . I really like the idea that the school buildings are being used full time. (A parent from Seattle’s Powerful Schools programme)
Seattle’s Powerful Schools used a collaborative community-based model to improve students’ performance, ‘to serve as a neighbourhood resource and empower parents, students, teachers and community’. This innovative coalition of four Rainier Valley elementary schools is a national leader in community-based school reform. These schools offer a wide variety of after-school programmes and classes for children and adults. Businesses and community groups help by planning field trips, providing support, and serving as a work-world contact for parents and students. The 30 to 40 classes offer reduced tuition fees and include topics such as gymnastics, blues harmonica, people’s law school, and foreign languages.
A programme for students who exhibit emotional disturbances—Montgomery County, Maryland, public schools
A 20-member committee that included principals, teachers, and support services representatives in Montgomery County, Maryland, developed an interagency plan to serve students who exhibit violent behaviour. The co-operating agencies include public schools, department of social services, juvenile justice system, police, family resources services, recreation department, drug, alcohol, and mental health services, the State Attorney’s Office, and the Office of Management and Budget.
The school district, mandated to serve students under the age of 16, has traditionally placed violent students in home instruction and provided 6 hours of instruction per week. The new plan includes creating centres that have school days longer than 8 hours and a half-day session. The longer school days coupled with the mandated Saturday sessions are meant to encourage students to return to public school.
The programme employs a case management approach. First, an assessment team evaluates the student to determine individual and family needs. The intense interagency programme is then implemented according to each student’s identified needs. The school instructional day includes science and mathematics, life-skills building, physical education, group counselling, and English and social studies. The day also includes meetings with students, teachers, case managers, and counsellors. By co-ordinating existing resources and redirecting existing funds, the programme requires only moderate budget increase.
Overcoming Obstacles, based in Los Angeles, is an education, jobs, and entrepreneurship programme that teaches young people the skills needed to succeed in education, find jobs, develop entrepreneurial skills, and involve themselves in the community. It is an example of business partnering with a community programme to give students a chance to succeed. This three-phased programme includes course work to:
develop a sense of personal responsibility
instil a sense of pride in community
set realistic goals
develop communications and conflict resolution skills
gain employment-seeking and retention skills.
An evaluation demonstrated positive behaviour change in successful students during the school year and after graduation.
After the first educational phase, high-school students move to a job placement phase. Through a network, part-time and summer employment is secured for students, while full-time employment is offered for programme graduates. One example is a special programme with ARCO Product Company/Prestige Stations Inc. Students working for this company receive a salary subject to increases and bonuses based on performance, and have the opportunity to attend college, paid for by ARCO.
Phase three encourages students to learn to become managers and owners of a business. Presently, funding has been provided for several businesses designed and managed by graduates of Overcoming Obstacles.
The Health Department of Contra Costa County, California, is widely known for its efforts to develop comprehensive programmes, harnessing existing resources, to alleviate poor health outcomes. Rather than designing and implementing their own stand-alone projects, the Contra Costa County Program co-ordinates and develops existing programmes to meet identified community needs. It serves as a lead agency for a number of health-related issues, including violence prevention. Some of its premises are that coalitions:
offer more resources for less money
can reach more people than a single organization
provide greater credibility than single organizations
offer more political clout
serve a community networking function
offer more diverse opinions and talents.
The Contra Costa County Program defines eight steps to building effective coalitions. An initial planning group should:
analyse programme objectives and decide if a coalition is needed;
recruit relevant and effective organizations/community representatives;
develop preliminary objectives and activities;
convene the group.
By using the input from the planning group, the coalitions should:
develop a budget and structure;
maintain coalition vitality (communications, public relations);
based on evaluation, offer recommendations to improve programmes.
A guide further defining the eight-step process is available from the Contra Costa County Health Department.
A school-based planning model: Pittsburgh Safe Schools Project
The Safe Schools Project of the Pittsburgh Public Schools is a model for multidisciplinary violence prevention coalition. Members of the Pittsburgh Public Schools, the Jewish Healthier Foundation, the Western Psychiatric Institute and Clinic, the Center for Injury Research and Control at the University of Pittsburgh, and the Boys and Girls Club of Western Pennsylvania formed a working group that produced a Blueprint for Violence Reduction in Pittsburgh Public Schools. It is an action plan based on a sound theoretical framework, data collection and analysis, a commitment to understanding the causes of violence, and an analysis of state-of-the-art school violence prevention programmes.
The Blueprint contains a review of the project’s components and discusses each step in a thorough manner, which lends itself to replication in other school districts. It also includes a set of valuable guiding principles for school-based program implementation:
violence prevention must be a long-term priority for the school district
adequate resources should be focused on very young children, particularly those at risk of developing aggressive lifestyles
developmentally appropriate programmes should be integrated in a comprehensive approach for all grade levels
students, teachers, and parents should participate in planning and assessing violence prevention activities
activities should be culturally and racially appropriate
prevention efforts should include home, school, and community co-ordination
programme evaluation measures should be integrated into the programme design
new programmes should be built upon successful existing programmes.
The contributions made by public health professionals towards efforts to prevent violence have been tremendous. The continued application of public health strategies to the understanding and prevention of violence assures success. The public health campaign to reduce smoking took 30 years after the first Surgeon General’s report to have an effect. Violence reduction can be expected to take at least as long and require as many, if not more, diverse strategies.
Allen, N.H. (1981). Homicide prevention and intervention. Suicide and Life Threatening Behavior, 11, 167–79.
American Psychological Association Commission on Violence and Youth (1993). Violence and youth: psychology’s response. American Psychological Association, Washington, DC.
Austin, E.W., Roberts D.F., and Nass, C.I. (1990). Influences of family communication on children’s television-interpretation processes. Communication Research, 4, 545–65.
Bandura, A., Ross, D., and Ross, S.A. (1963a). Imitation of film-mediated aggressive models. Journal of Abnormal Psychology, 63, 3–11.
Bandura, A., Ross, D., and Ross, S.A. (1963b). Vicarious reinforcement and imitative learning. Journal of Abnormal Psychology, 63, 601–7.
Boys Clubs of America (1986). Targeted Outreach Newsletter, Vol. II-1.
Bureau of Alcohol, Tobacco and Firearms (1991). Firearm census report. US Treasury Department, Washington, DC.
Bureau of Justice Statistics (1992). Criminal victimization in the United States, 1991. US Department of Justice, Washington, DC.
Bureau of Justice Statistics (1993). Highlights from 20 years of surveying crime victims: the national crime victimization survey, 1972–1992. US Department of Justice, Washington, DC.
Bureau of Justice Statistics (1994). Violence between inmates. Report NCJ-149259, Office of Justice Programs, US Department of Justice, Washington, DC.
Carter Center (1994). Not even one: a report on the crisis of children and firearms. Carter Center, Atlanta, GA.
CDC (Centers for Disease Control) (1982a). Homicide. Morbidity and Mortality Weekly Report, 31, 594.
CDC (Centers for Disease Control) (1982b). Homicide—United States. Morbidity and Mortality Weekly Report, 31, 599–602.
CDC (Centers for Disease Control) (1983a). Violent deaths among persons 15–24 years of age—United States, 1970–78. Morbidity and Mortality Weekly Report, 32, 453.
CDC (Centers for Disease Control) (1983b). Homicide surveillance, high risk racial and ethnic groups—Blacks and Hispanics, 1970–83. Public Health Service, US Department of Health and Human Services, Washington, DC.
CDC (Centers for Disease Control) (1990). Homicide among black males—United States, 1978–1987. Morbidity and Mortality Weekly Report, 39, 869–72.
CDC (Centers for Disease Control) (1991). Weapon carrying among high school students—United States, 1990. Morbidity and Mortality Weekly Report, 40, 681–4.
CDC (Centers for Disease Control) (1992). Youth suicide prevention program: a resource guide. National Center for Injury Prevention and Control, CDC, Atlanta, GA.
CDC (Centers for Disease Control) (1993a). Advance report of final mortality statistics, 1991. Monthly Vital Statistics Report, August. CDC, Atlanta, GA.
CDC (Centers for Disease Control) (1993b). Violence-related attitudes and behaviors of high school students—New York City, 1991. Morbidity and Mortality Weekly Report, 40, 773–7.
CDC (Centers for Disease Control) (1994). Deaths resulting from firearm- and motor-vehicle-related injuries—United States, 1968–1991. Morbidity and Mortality Weekly Report, 3, 37–42.
Centerwall, B.S. (1984). Race, socio-economic status, and domestic homicide, Atlanta, 1971–72. American Journal of Public Health, 8, 813–15.
Centerwall, B.S. (1992). Television and violence: the scale of the problem and where to go from here. Journal of the American Medical Association, 267, 3059–63.
Centerwall, B.S. (1993). Race, socio-economic status, and domestic homicide in New Orleans. Presented at the 2nd World Conference on Injury Control.
Daro, D. and Mitchel, L. (1987). Deaths due to maltreatment soar: the results of the 1986 Annual Fifty State Survey. National Center on Child Abuse Prevention Research, National Committee for Prevention of Child Abuse, Chicago, IL.
Dietz, W.H. and Strasburger, V.C. (1991). Children, adolescents and television. Current Problems in Pediatrics, 21, 8–31.
Durant, R., Pendergast, R., and Cadenhead, C. (1994). Exposure to violence victimization and fighting behavior by urban black adolescents. Journal of Adolescent Health, 15, 311–18.
Egeland, B. and Jacobvitz, D. (1984). Intergenerational continuity of parental abuse: cases and consequences. Presented at the Conference on the Bio Social Perspectives on Abuse and Neglect, York, ME.
Emde, R.N. (1993). The horror! The horror! Reflection on our culture of violence and its implication for early development and morality. Psychiatry, 56, 119–23.
Eron, L. and Huesmann, L.R. (1984). Television violence and aggressive behavior. In Advances in clinical child psychology (ed. B. Lahey and A. Kardin). Plenum Press, New York.
FBI (Federal Bureau of Investigation) (1991). Uniform crime report: crime in the United States. US Department of Justice, Washington, DC.
FBI (Federal Bureau of Investigation) (1992). Uniform crime report: crime in the United States. US Department of Justice, Washington, DC.
FBI (Federal Bureau of Investigation) (1993). Uniform crime report: crime in the United States. US Department of Justice, Washington, DC.
FBI (Federal Bureau of Investigation) (1997). Uniform crime report: crime in the United States. US Department of Justice, Washington, DC.
FBI (Federal Bureau of Investigation) (1998). Uniform crime report: crime in the United States. US Department of Justice, Washington, DC.
FBI (Federal Bureau of Investigation) (1999). Uniform crime report: crime in the United States. US Department of Justice, Washington, DC.
Fingerhut, L.A. and Kleinman, J.C. (1990). International and interstate comparisons of homicide among young males. Journal of the American Medical Association, 24, 3292–4.
Fingerhut, L.A., Ingram, D.D., and Feldman, J.J. (1992). Firearm and non-firearm homicide among persons 15 through 19 years of age. Journal of the American Medical Association, 22, 3048–53.
Fitzpatrick, K.M. and Boldizar, J.P. (1993). The prevalence and consequences of exposure to violence among African American youth. Journal of the American Academy of Child and Adolescent Psychiatry, 32, 424–30.
Freeman, L., Mokros, H., and Poznanski, E. (1993). Violent events reported by normal urban school-aged children: characteristics and depression correlates. Journal of the American Academy of Child and Adolescent Psychiatry, 32, 419–23.
Friedman, L.M. (1975). The legal system: a social science perspective, pp. 56–66. Russell Sage Foundation, New York.
Gil, D. (1973). Violence against children: physical child abuse in the United States. Harvard University Press, Cambridge, MA.
Groves, B.M., Zuckerman, B., Marans, S., and Cohen, D.J. (1993). Silent victims: children who witness violence. Journal of the American Medical Association, 269, 262–4.
Huesmann, L.R., Eron, L.D., Klein, R., Brice, P., and Fischer, P. (1983). Mitigating the imitation of aggressive behaviors by changing children’s attitudes about media violence. Journal of Personality and Social Psychology, 44, 899–910.
Kellermann, A.L., Rivara, F.P., Rushforth, N.B., et al. (1993). Gun ownership as a risk factor for homicide in the home. New England Journal of Medicine, 15, 1084–91.
Killias, M. (1993). International correlations between gun ownership and rates of homicide and suicide. Canadian Medical Association Journal, 148, 1721–5.
Kilpatrick, D.G., Edmunds, C.N., and Seymour, A.K. (1992). Rape in America: a report to the nation. National Victim Center, Arlington, VA.
Lester, D. (1988). Firearm availability and the incidence of suicide and homicide. Acta Psychiatrica Belgica, 387–93.
Liebert, R.M. (1988). Early window: the effects of television on children and youth (6th edn). Allyn and Bacon, Needham, MA.
Liebert, R., Neale, J., and Davidson, E. (1973). Early window: the effects of television on children and youth. Pergamon Press, New York.
Loftin, C., McDowall, D., Wiersema, B., and Cottey, T.J. (1991). Effects of restrictive licensing of handguns on homicide and suicide in the District of Columbia. New England Journal of Medicine, 325, 1615–20.
Lorion, R.P. and Saltzman, W. (1993). Children’s exposure to community violence: following a path from concern to research to action. Psychiatry, 56, 55–65.
Makepeace, J.M. (1981). Courtship violence among college students. Family Relations, 30, 97–102.
Martinez, P. and Richters, J. (1993). The NIMH community violence project. II: Children’s distress symptoms associated with violence exposure. Psychiatry, 56, 22–35.
Miller, T.R., Cohen, M.A., and Rossman, S.B. (1993). Datawatch: victim cost of violent crime resulting injuries. Health Affairs (Millwood), 12, 186–97.
Osofsky, J., Wewers, S., Hann, D.M., and Fick, A.C.. (1993). Chronic community violence: what is happening to our children? Psychiatry, 56, 36–45.
Prothrow-Stith, D. and Weissman, M. (1991). Deadly consequences: how violence is destroying our teenage population and a plan to begin solving the problem, pp. 1–203. Harper Collins, New York.
Pynoos, R.S., Frederick, C., Nader, K., et al. (1987). Life threat and post-traumatic stress in school-age children. Archives of General Psychiatry, 44, 1057–63.
Sang, F., Schmitz, B., and Tasche, K. (1993). Developmental trends in television coviewing of parent–child dyads. Journal of Youth and Adolescence, 5, 531–43.
Satcher, D. (1985). The public health approach to violence. presented at the National Education Association National Conference, Los Angeles, California.
Sege, R. and Dietz, W. (1994). Television viewing and violence in children: the pediatrician as agent for change. Pediatrics, 94, 600–7.
Shaffer, D., Garland, A., Gould, M., Fisher, P., and Trautman, P. (1988). Preventing teenage suicide: a critical review. Journal of the American Academy of Child and Adolescent Psychiatry, 27, 673–87.
Shaffer, D., Garland, M., and Hicks, R. (1994). Worsening suicide rate in black teenagers. American Journal of Psychiatry, 151, 12.
Sheley, J., McGee, Z., and Wright, J. (1992). Gun-related violence in and around inner-city schools. American Journal of Diseases of Childhood, 146, 677–82.
Slaby, R.G. and Guerra, N.G. (1988). Cognitive mediators of aggression in adolescent offenders: 1. Assessment. Developmental Psychology, 4, 580–8.
Slaby, R. and Quarfoth, G. (1980). Effects of television on the developing child. Advanced Behavioral Pediatrics, 1, 225–66.
Sloan, J.H., Kellerman, A.L., Reay, D.T., et al. (1988). Handgun regulations, crime, assaults, and homicide: a tale of two cities. New England Journal of Medicine, 319, 1256–62.
Spivak, H., Prothrow-Stith, D., and Hausman, A. (1988). Dying is no accident: adolescents, violence, and intentional injury. Pediatric Clinics of North America, 35, 1339–47.
Straus, M. (1991). Discipline and deviance: physical punishment of children and violence and other crime in adulthood. Social Problems, 38, 137–54.
Straus, M. and Gelles, R. (1990). How violent are American families? Estimates from the National Family Violence Resurvey and other studies. In Physical violence in American families: risk factors and adaptations to violence in American families: risk factors and adaptations to violence in 8145 families (ed. M. Straus and R. Gelles). Transaction, New Brunswick, NJ.
Turner, C.W., Hesse, B.W., and Peterson-Lewis, S. (1986). Naturalistic studies of the long-term effects of television violence. Journal of Social Issues, 51–73.
United States Department of Health, Education, and Welfare (1964). Smoking and health. Report of the Advisory Committee to the Surgeon General. Public Health Service, Washington, DC.
Vissing, Y., Straus, M., Gelles, R., and Harrop, J. (1991). Verbal aggression by parents and psychological problems of children. Child Abuse and Neglect, 15, 223–38.
Weaver, B. and Barbour, N. (1992). Mediation of children’s televiewing: families in society. Journal of Contemporary Human Services, 4, 236–43.
Widom, C.S. (1989). The cycle of violence. Science, 244, 160–6.
Williams, B.C. and Kotch, B.J. (1990). Excess injury mortality among children in the United States: comparison of recent international statistics. Pediatrics, 86, 1067–73.
Wolfgang, M. (1986). Homicide in other industrialized countries. Bulletin of the New York Academy of Medicine, 62, 400–12.
Wood, W., Wong, F.Y., and Chachere, J.G. (1991). Effects of media violence on viewer’s aggression in unconstrained social interaction. Psychological Bulletin, 109, 371–83.