---

Press the print button on your browser.
Click here to return to the previous page.

Featured Report:
AMA Data on Violence Between Intimates (I-00)
Full Text

Scope of Report  Definitional Issues Methods  Previous Council Reports and AMA Policy
Intimate Partner Violence Violence Directed at Women  Violence Directed at Men Injuries and Reports to Authorities Perpetrators Violence Between Same-Sex Partners Elder Abuse Immigrant and Ethnic Groups Child Maltreatment  Perpetrators Child Maltreatment and Intimate Partner Violence Aftereffects of Violence	Alcohol and Family or Intimate Partner Violence  Data From Victim and Perpetrator Reports Alcohol and Violence in the Home Alcohol and Child Maltreatment How Alcohol Affects Intimate Partner Violence CONCLUSION RECOMMENDATIONS (Adopted AMA Policy) References

---

NOTE: This report, written in response to  Resolution 410 (I-99), represents the medical/scientific literature on this subject as of December 2000. It was presented as CSA Report 7 at the 2000 AMA Interim Meeting.

---

Scope of Report: The primary purpose of this report is to assess the foundation of American Medical Association (AMA) policy on family violence and to gauge the need for possible changes in those policy statements. The report examines current literature on family violence, providing an overview of this vast literature. The report is not intended to be comprehensive but rather a concise summary of data on violence between intimate partners, including former partners, as well as issues related to child maltreatment and elder abuse. The role of alcohol is also examined. Wherever possible, data sources will be limited to key studies that provide national estimates on the incidence and prevalence of intimate partner violence and family violence for both men and women. New studies continue to appear in the literature, but many are limited to one gender (for example, Coker et al¹) or particular locales (for example, Bensley et al²).

Included will be recent statistics and a sketch of current strategies for dealing with violence between male and female intimate partners as well as same sex partners in intimate relationships. Comparable material on child maltreatment, and elder abuse will be covered as well. However, the literature across these fields is uneven in both quantity and quality, making some issues difficult to study. The medical literature on immigrant domestic violence, for example, is limited to a few groups in particular locales, meaning no firm conclusions can be drawn, and the literature on violence against women, which is the most common, generally addresses the health effects of violence only in a cursory fashion.³ Compounding the problem is the fact that studies of violence within families tend to focus on particular elements, and few links between types of violence are made.⁴ For example, child maltreatment researchers tend not to deal with spouse abuse (and vice versa), and research on elder abuse only rarely deals with other types of family violence.

Definitional issues: For this report, the phrase "intimate partner violence" will refer to violent behavior between partners, regardless of their genders. Indeed, "intimate partner violence" is now the preferred term for this issue, having replaced "domestic violence." The term includes violence between former partners as in cases involving former spouses or former boyfriends or girlfriends. Intimate partners need not be cohabiting nor is sexual activity necessarily involved,⁵ and in general, intimate partner violence consists of a continuing pattern of behavior rather than a single violent act.⁶

Child maltreatment includes all types of maltreatment of children by responsible parties. In most cases, these parties are adults, either parents or guardians, but may include older youth such as adolescent babysitters. Both physical abuse and sexual abuse are included as are psychological abuse and neglect. In the case of adolescents, the youths may be either victims or perpetrators, and the interested reader is referred to the Council on Scientific Affairs report on this topic (CSA Report I. "Family Violence: Adolescents as Victims and Perpetrators," [A-92]), which although now eight years old thoroughly examines the issues. In this report, the term "child maltreatment" is preferred over "child abuse and neglect."

Elder abuse may include physical, psychological or financial abuse or neglect, usually in a deliberate attempt to inflict harm or injury. Unintentional mistreatment can occur when an act stemming from ignorance, inability or inexperience results in harm to an elderly person.⁷ The extent of elder abuse in the United States is poorly documented.
Back to Top

Methods

The literature reviewed for this report came from MEDLINE searches using relevant sets of key words for literature published in English within the past 5 years. For the types of violence reviewed the following key words were used in MEDLINE:

• Intimate partner violence—"Domestic Violence"
• Violence between same-sex partners—"Domestic Violence and Homosexuality"
• Elder abuse—"Elder Abuse"
• Immigrant and ethnic groups—"Domestic Violence and Emigration and Immigration"
• Child maltreatment—"Child Abuse"

In addition, knowledgeable experts were consulted to recommend studies providing national statistics. These experts included researchers at the Centers for Disease Control and Prevention National Center for Injury Prevention and Control as well as authors of the AMA’s series of diagnostic and treatment guidelines dealing with various aspects of family violence. National studies were preferred over smaller scale studies in particular locales.
Back to Top

Previous Council Reports and Existing AMA Policy

The CSA has prepared a number of reports on violence between intimates and within families over the past decade, and considerable policy (AMA Policy Database) has been developed. Key reports in chronological order include:

1. Violence Against Women (I-91), Policy H-515.984
2. Family Violence: Adolescents as Victims and Perpetrators (A-92), Policy H-515.981
3. Mental Health Consequences of Interpersonal and Family Violence (A-93), Policy H-515.976
4. Alcohol, Drugs, and Family Violence (A-93), Policy H-515.975
5. Violence Toward Men: Fact or Fiction? (full text) (I-94), Policy H-515.972

Additional policy has emerged through reports from the Board of Trustees and the Council on Ethical and Judicial Affairs as well as numerous resolutions. 
Back to Top

Intimate Partner Violence

This is the issue most commonly thought of when the subject of violence between intimates is raised. Oftentimes, the reference is to "wife-battering" or "domestic violence," and a common depiction would be a woman beaten by her husband. However, as noted above, the preferred term is broader, including a wider range of cases and not limiting the issues to marital relationships or to male on female violence.

Intimate partner violence "is characterized as a pattern of coercive behaviors that may include repeated battering and injury, psychological abuse, sexual assault, progressive social isolation, deprivation, and intimidation."^{8 p4} Heise and colleagues^{6 p5} note that "partner abuse can [include] physical assaults such as hits, slaps, kicks, and beatings; psychological abuse such as constant belittling, intimidation, and humiliation; and coercive sex. It frequently includes controlling behaviors such as isolating a woman from family and friends, monitoring her movements, and restricting her access to resources." It is important to note that any single act of physical violence qualifies as intimate partner violence if it happens between intimate partners, but an individual violent act is often a precursor to repeated violent episodes, hence the definition as a pattern of behaviors.

Obviously this definition can be applied without regard to gender or marital status, but considerable debate occurs because past work has failed to agree on such basic matters. As noted in the introduction to the Centers for Disease Control and Prevention’s (CDC) recent publication, Intimate Partner Violence Surveillance: Uniform Definitions and Recommended Data Elements, "researchers have used terms related to violence against women in different ways and have used different terms to describe the same acts."^5p1 Indeed the CDC’s document represents the first step in attempting to develop a consistent foundation for use by researchers, policy makers, and health care providers.
Back to Top

Violence directed at women. According to Heise and colleagues⁶ surveys from around the world indicate that 10% to 50% of women report that they have been physically abused by a male partner, meaning the problem is not unique to the United States. In this country, commentaries about women who are battered by partners each year typically assert that the annual number of cases ranges from about 2 million to 8 million. However, the most reliable national surveys indicate that the numbers are lower, although the impact of definitional issues and methodological differences among surveys cannot be determined.

The National Family Violence Surveys (NFVS) are perhaps the best known source of national data on family violence. Part of the ongoing efforts of the Family Research Laboratory at the University of New Hampshire, the surveys collected data on spousal and parent-child violence and psychological abuse in two national household surveys in 1975 and 1985. Using data from the NFVS, Straus and Gelles⁹ reported that some 1.8 million women were been beaten by their partners annually.

More recent national data were collected in the National Violence Against Women Survey (NVAWS) during 1995-1996, which examined both intimate partner and non-intimate partner violence. It collected information on physical assaults and rape but excluded other sexual assaults, murders, and robberies. In addition, the NVAWS employed an altered version of the Conflict Tactics Scale that Straus had used in the NFVS to measure interpersonal conflict, including domestic physical assaults, within the home. Based on the resulting NVAWS data, 1.9 million women are physically assaulted each year, and another 300,000 are raped. An intimate partner (current or former husband, cohabiting partner, or date) victimizes most of these women, with women suffering 1.3 million physical assaults and 200,000 rapes at the hands of an intimate partner annually.¹⁰ The NVAWS remains the only government survey focused specifically on violence, and was a joint effort of the Departments of Justice and Health and Human Services.

Other national data come from the most recent National Crime Victimization Survey (NCVS), which provides data on all types of crime experienced by the public through a national sample survey. It includes measures of violent assaults by intimate partners, including sexual assault and rape; robbery; aggravated assault; and simple assault, which itself accounts for about two thirds of the total victimizations. These data are supplemented with information on homicides from the Uniform Crime Reports. Data from the latest NCVS indicate that about 876,000 women were victimized by intimate partners in 1998; this figure represents a rate of about 7.7 per 1000 women aged 12 years or older.¹¹ One advantage of this survey is that it is ongoing, allowing one to make comparisons over time, and the latest figure is about 21% lower than the 1993 statistic of 1.1 million (9.8 per 1000). Violent victimization of women by their intimate partners declined throughout this time period except for a slight increase from 1997 to 1998 (848,000 to 876,000).

Homicide is an extreme form of interpersonal violence, and 1320 women were murdered by their intimate partners in 1998.¹¹ Women are far more likely than men to be murdered by an intimate partner: in 1998 one third of all murdered women were killed by an intimate partner while only 4% of murdered men died at the hands of their intimate partners. Slightly over half of all victims of intimate partner homicide are killed by their spouses.¹¹

The NVAWS and NCVS are generally regarded as the best sources of current data on violence between intimates, although like all sample surveys they are subject to limitations. Clearly, the two surveys describe the current situation somewhat differently. Some discrepancies between the NVAWS and NCVS may stem from methodological differences,^{10 p4} including the fact that the NCVS is presented as a study about crime, a characterization that could easily lead respondents to discount incidents of intimate partner violence. (See below for issues related to perpetrators and controversy over these surveys.)
Back to Top

Violence directed at men. Both qualitative and quantitative differences exist in comparing men and women as victims of intimate partner violence. Information on such violence against men is available from the same sources as for women, and to avoid complications due to methodological differences in surveys, those data are the focus of this section.

The NFVS reports that rates of intimate partner assault in married and cohabiting couples are about the same for men and women.⁹ Intimate partner assaults include any act of aggression as measured by the Conflict Tactics Scale. Given the same rates, the numbers of cases would be similar as well. While this finding has held up across numerous surveys, considerable controversy surrounds it as will be discussed below.

Both the NVAWS and the NCVS also provide data on intimate partner violence for men. In the NVAWS, parallel data were collected for 8000 men and 8000 women specifically to facilitate comparisons. The results indicate that about half again as many men as women had been physically assaulted in the previous 12 months; 3.15 million men had been physically assaulted and 90,000 raped.¹⁰ However, for men, such incidents were less likely to be the result of intimate partner violence. According to respondents to the NVAWS, only about one quarter of physically assaulted men (835,000) were by attacked by an intimate partner in that period. (The number of reported intimate partner rapes of males was too small to make reliable estimates.)

The NCVS similarly shows that men were less likely to be victims of intimate partner violence. The 1998 survey found about 157,000 cases of intimate partner violence directed against men.¹¹ Like women, about two thirds of the cases of intimate partner violence were simple assaults. For men, most of the remaining cases were aggravated assaults whereas women were at substantial risk of rape and robbery by their intimate partners. The overall rate of intimate partner violence for men was about 146 cases per 100,000, which was about one fifth the rate for women (767 per 100,000). The figure for men has remained rather stable throughout the period 1993 to 1998, while for women the rate has declined about 21%.

Murder by an intimate partner is a rare event for both men and women, with 0.5 cases per 100,000 men and 0.8 cases per 100,000 women. At the same time, intimate partner homicide made up only 4% of the murders of men but fully one third of the murders of women. This pattern holds across all types of violence as reported in the NCVS: women are more likely to be victimized by a nonstranger—a friend, family member, or intimate partner—than men, who are more likely to experience violence at the hands of a stranger.^{11p10
Back to Top}

Injuries and Reports to Authorities. A considerable literature on violence between intimates has been based on studies of emergency department visits or police reports. In general, individuals with more severe problems are more likely to present for treatment, and men are less likely than women to make police reports.¹² As such, while useful for some purposes, their reliance on convenience samples and self-selection makes these studies unsuitable for determining the scope of the problem.

Fortunately, both the NVAWS and the NCVS have information on these topics. In the NVAWS, respondents who had been assaulted were asked if they had suffered any injuries during the assault. Among those raped, nearly 1 in 3 women (31.5%) had been injured (other than the rape) compared to about 1 in 6 (16.1%) men. The difference was not quite as great for physical assaults, with corresponding figures being 39% for women and approximately 25% for men. However, these figures include all rapes and assaults, not only those perpetrated by intimates. When only assaults by intimates are considered, women are more than twice as likely as men to report being injured: 41% to 19%.¹⁰ Men’s greater average strength probably accounts for some of this difference.

Data from the NCVS present a similar picture, with half of female victims of intimate partner violence reporting injuries compared to one third of male victims. This is remarkable because the survey reports that about two thirds of both male and female victims of intimate partner violence were physically attacked. Thus, women are much more likely to report injuries.

The most common injuries for both men and women who are victims of intimate partner violence are relatively minor; e.g., bruises and scratches. However, more serious injuries are not unknown, and the NCVS reports similar rates of serious injuries (knife wounds, internal injuries, broken bones, loss of consciousness) for both men and women (4% and 5% respectively).^{11 p6} In addition, as noted above, homicides are not unknown, although women are more likely than men to die in intimate partner violence. As for other injury patterns, Stark and colleagues¹³ two decades ago noted that victims of domestic violence were 13 times more likely to sustain an injury to the breast, chest or abdomen than were accident victims. Long-term, chronic conditions may also be associated with abuse by intimate partners, and the mental health effects of such violence may be particularly devastating.¹⁴

Medical care for physical injuries is, however, not the norm. A recent report from the NVAWS indicates that about one third (31%) "of the women injured during their most recent intimate partner rape received some type of medical care," and for those injured in their most recent physical assault by an intimate partner, the figures for women and men are about one quarter (28.1%) and one fifth (21.5%) respectively.^15p45 Clearly, the majority of victims of intimate assaults do not seek medical assistance.

The NCVS also reports that most victims of intimate partner violence do not seek medical treatment for their injuries. Three in five victims, both male and female, did not seek treatment. Among those who do seek treatment, most are treated at home or at the scene. Although the gender differences were not statistically significant, very few male victims reported visiting a physician’s office or hospital, while 1 in 20 women visited a physician and 1 in 7 visited a hospital or emergency department.^11p6

Reports to the police are, of course, another mechanism for seeking help for intimate partner violence, and again, differences between male and female victims are relatively small. According to the NCVS, between 1993 and 1998, about half of victims made police reports (53% of women, 46% of men). The figure for women has increased over that time period, from 48% (in 1993) to 59% (in 1998) but remained essentially unchanged for men.¹¹ Data from the NVAWS show a lower frequency of reports to the police than does the NCVS: only about one quarter of women (26.7%) and one seventh of men (13.5%).^15p49

Among those who did not report to the police, an interesting gender difference emerges. Nearly 1 in 5 (19%) of the women who did not report expressed a fear of reprisal from the perpetrator as the reason for not reporting, while virtually no men stated such a concern. The most common reason for not making a report was the belief that intimate partner violence is a "private or personal matter."^{11 p7} This was true for about half (52%) the men and one third (32%) of the women. Protecting the perpetrator was another common reason for not reporting the violence (11% males, 3% females). (These figures are from the NCVS and are based on respondent reports of their actions; data based on actual police reports would show a preponderance of reports by women.) 
 Back to Top

Perpetrators. There is considerable debate over the relative rates of male-on-female versus female-on-male intimate partner violence. Notwithstanding the findings of the NVAWS and the NCVS, a number of earlier surveys suggest nearly equal participation rates for men and women in intimate partner violence. Best known is the NFVS, which reports that female-on-male rates of aggression are as high or higher than male-on-female rates,⁹ and Straus¹⁶ notes that his review of some 30 studies using non-self-selected samples yielded the same conclusion. According to Straus^16p211 the "near equality in assault rates cannot be attributed to a gender bias in reporting."

A common criticism of these studies is that they are generally lacking information on the context of the violence. That is, information on self-defense as the purpose of the violence or on injuries is absent.¹⁶ Common claims are that women are using violence for self-defense and that the violence they utilize tends to be less severe, including behavior such as slapping, but given the lack of context, these assertions are impossible to assess.

The NFVS does include data on some of these issues. Straus¹⁶ reports an analysis, using only female respondents, that suggests a minimum of 25% to 30% of intimate partner violence is initiated—that is, would not generally be construed as self-defense—by women. Moreover, the women involved in violent relationships report that they struck the first blow in slightly more than half (53%) of the cases while men struck first in 42% of the cases.^16,17 (Who struck first was not determined in the remaining incidents.) Straus¹⁶ cites a number of other supportive studies, and while all are now more than 10 years old, there is no reason to believe that more recent data would provide a contrary view. At the same time, as a number of researchers have pointed out, the fact that one struck the first blow does not in and of itself preclude self-defense.^16,18 A situation perceived as a serious threat might well warrant defensive action, including physical assault, in which case striking the first blow might be justified.

An alternative path of inquiry deals with patterns of injury. Stets and Straus¹⁷ report that assaults by women caused an injury only 0.4% of the time, while assaults by men caused injury in 3% of cases. In other words, intimate partner assaults by males were more than seven times as likely to cause injury as were such assaults by women. Obviously, some part of this difference can be traced to the greater average physical strength of men, but additionally, evidence suggests that men are somewhat more likely to commit severe assaults and are more likely to engage in repeated attacks.¹⁶ As Tjaden and Thoennes^10p10 point out:

[B]ecause women are primarily raped and physically assaulted by intimate partners, the injury and medical utilization estimates … pertain mostly to rapes and physical assaults perpetrated by current and former husbands, cohabiting partners, dates, and boyfriends. Because men are primarily raped and physically assaulted by male strangers and acquaintances, the injury and medical utilization estimates for men pertain to violence perpetrated by men other than intimate partners.

Regardless of injury rates or issues of self-defense, differences between the NFVS on the one hand and other surveys, particularly the NVAWS, which used a variation of the Conflict Tactics Scale, have created an uproar. Some groups decry the NFVS findings, alleging bias and methodological faults, while others use the findings to rail about other perceived slights. Likewise, the NVAWS has its proponents and opponents, with each side seeking advantage over the other.

In part, the differences can be traced to the definitional issues raised earlier, but in addition, slight changes in methodologies may yield large changes in results. Mathematically, this is particularly true for events with low prevalence. Thus, for example, when the NCVS was revised to improve reporting of intimate partner violence, the rate of domestic assaults quadrupled.¹⁹ In fact, the redesigned survey captured more incidents of violence across all categories than did the old questionnaire.²⁰ The changes had been expected, and could be attributed in part to changing the "demand characteristics" of the survey; that is, the specific instructions or cues given to respondents about what sort of information is requested. According to Straus,¹⁹ most respondents do not consider domestic battery as a crime unless it is chronic and ongoing or there is serious injury. Moreover, men tend to downplay assaults by women. These conditions would tend to increase the ratio of male to female assaults (i.e., assaults by males on females).

As for the NVAWS, methodological issues are yet to be resolved. Straus argues that the instructions to respondents cued them to think more about "criminal" acts of intimate partner violence than about the "routine" differences that arise in interpersonal, including intimate, relationships.^19,21 Additionally, the NVAWS altered its use of the Conflict Tactics Scale by asking only about victimizations and not about acts committed by both the respondent and his or her partner. Tjaden and Thoennes¹⁵ dispute Straus’s first point and focus on the ways the NFVS and NVAWS frame their screening questions. The particulars are not relevant for this report, and it is sufficient to note that a number of factors may explain the discrepant findings between the NFVS and the more recent NVAWS and the NCVS. In fact, the CDC and the National Institute of Justice continue to examine methodological differences among these surveys in order to assess their effects on the study findings.^15,22

Conclusion. There is little question that the rate of intimate partner violence has declined over the last 10 to 15 years, particularly for women victims and probably for men victims, although not as much for the latter as the former. However, while the difference in rates between genders is unclear, the following points are supported by most evidence:

• Women are more likely to be injured in cases of intimate partner violence; they are 6 to 7 times more likely to report being injured in these cases. Women are much more likely to require medical attention for such injuries.
• Men are likely to inflict a higher number of assaults on their intimate partners, although the difference is probably less than 2 assaults per year (8 assaults by males, 6 by females). Nonetheless, this higher frequency combined with the greater likelihood of injuries stemming from male assaults will obviously adversely affect men’s female partners.
• Men are far more likely to be the assailants in cases of intimate partner violence involving former partners.
• Men are undoubtedly victims of intimate partner violence. Current data suggest that at least 15% of cases of intimate partner violence have male victims, most of whom have female partners.
  Back to Top

Violence Between Same-Sex Partners

Compared to the voluminous literature on intimate partner violence involving heterosexual couples, the literature on same-sex partner violence is miniscule. In part this is due to the sensitive nature of such inquiries, but aside from the fact that disclosures about same-sex relationships are sensitive, not all people who engage in same-sex activities define themselves as homosexual. This means the classification of homosexual relationships is subject to error, which in turn further complicates study of the issue. Despite these difficulties, some studies have emerged, and it is generally believed that intimate partner violence is at least as prevalent among homosexual couples—male or female—as among heterosexual couples.^23,24

Likewise, instances of intimate partner violence in same-sex couples share a number of other similarities with cases in heterosexual couples. The violence frequently escalates over time and the victims will oftentimes remain in the relationship despite ongoing abuse.^23-26 More importantly, the reasons underlying the abuse parallel the reasons for the abuse in other relationships: one partner is trying to exert power and control over the other. Contrary to commonly accepted myths, violence in same-sex relationships is not part of any desired sexual activity.

Among the more recent work is that of Tjaden and colleagues²⁷ who, using data from a national telephone survey, compared the lifetime prevalence of violence in same-sex and opposite-sex cohabiting couples. They report that respondents with same-sex intimate partners were more likely than respondents in opposite-sex relationships to have experienced violence of all types, including physical assaults by their intimate partners. Their study found a higher prevalence of intimate partner violence among same-sex cohabiting males but found less such violence in same-sex cohabiting female couples than in opposite-sex cohabiting couples.

Data from a program in New York City suggest that the patterns outlined above for intimate partner violence apply for the most part to violence within same-sex relationships.²⁸ For example, about half (49%) of the clients of the Anti-Violence Project (AVP) reported that their current partners were the perpetrators of intimate partner violence while former partners were the perpetrators in one quarter to one third of the incidents, a pattern not dissimilar to the one described above. Similarly, just over half of the AVP’s clients were not injured in the most recent occurrence of intimate partner violence, although one quarter (27%) received minor injuries and 10% suffered serious injuries. About three fifths of the injuries were untreated.

An analysis of national data by Community United Against Violence²³ on lesbian, gay, bisexual, and transgender domestic violence reports that most of its participating programs report stable rates of case reporting, although most programs are working at capacity, meaning the stable rates are likely an artifact. Among those cases, however, this analysis finds roughly equal numbers of male and female cases.

Conclusion. Intimate partner violence is at least as common in same-sex relationships as in heterosexual relationships, although the prevalence may vary by gender of the couples, a question that awaits better data. Regardless, the health care needs of gay males and lesbians in abusive relationships are no different than for heterosexuals in such relationships: identification and appropriate documentation and referral of victimized patients is critical.⁸ At the same time, given the sensitive nature of intimate partner violence and the stigmatization of same-sex behaviors, clinicians will need to be appropriately sensitive in their interaction with these patients. Some of these issues are addressed in CSA Report 8 (I-94), "Health Care Needs of Gay Men and Lesbians in the United States," which was published in JAMA.^{29
Back to Top}

Elder Abuse

Like most other forms of interpersonal family violence, the magnitude of the problem of elder abuse is often downplayed. In fact, only over the last decade has elder abuse been generally acknowledged as an important public health concern. Although it remains poorly documented, Lachs and colleagues³¹ have estimated that about 2 million elderly are abused in this country each year. The best, most often cited prevalence data suggest that 32 of every 1000 adults aged 65 years and over experience at least one incident of maltreatment after reaching age 65, and the same data yield an estimate of 26 new cases per 1000 persons aged 65 or more each year.³¹ These data are now over a decade old, but no more recent survey is available.

However, according to the National Center on Elder Abuse (NCEA), reports of elder abuse have climbed steadily since the mid-1980s, more than doubling to a total of 293,000 cases in 1996,³² suggesting that the incidence of elder abuse may be increasing. Approximately two thirds (64.2%) of reports are substantiated.³³

The NCEA data indicate that neglect is the most common form of elder abuse, accounting for more than half (55%) of reported cases, while physical abuse accounts for about 1 in every 7 (14.6%) case reports.³² About 1 in 8 (12.3%) reports concerns exploitation, and 1 in 12 (7.7%) deals with emotional abuse. About two thirds of victims are female, and the median age is nearly 80 years.³³

Approximately equal numbers of males and females perpetrate cases of elder abuse and maltreatment.³² Unfortunately, there are no data on gender differences, if any, for the perpetrators of different types of elder abuse. However, adult children are most often abusers of the elderly, being the abusers in more than one third (36.7%) of cases. Spouses are named as the abusers in about 13% of cases, while other family members commit the abuse in about 10% of cases.³² Nearly 4 elderly abuse cases in 10 are carried out by non-family members, including friends, neighbors and paid caregivers.

Physicians encounter the effects of this elder abuse on a regular basis, and health care providers—not necessarily physicians—make nearly a quarter of the reports of elder mistreatment.³³ In most states physicians are among those who are mandated reporters, although laws on elder abuse reporting vary considerably more than the laws dealing with child abuse reports. For example, although all states have mandatory reporting laws, seven states do not require reporting of suspected abuse, and only 30 states have penalties for failure to report.³⁴ According to Conlin³⁵ physicians report only 1 in every 13 cases they identify. The AMA’s guidelines on elder abuse call for routine questions about abuse and neglect even among patients with cognitive impairment in order to improve the identification of cases and implement appropriate treatment and referral. Cognitive impairment should not necessarily preclude inquiries, since diminished cognitive capacity does not necessarily negate the elderly person’s ability to describe mistreatment.⁷

Like cases of intimate partner violence, elderly patients who are being abused may be unwilling to disclose this information. While some of this may be due to embarrassment, fear of retaliation by the abuser would be common as well, just as in cases of intimate partner violence. In addition, definitions of abuse are likely to vary, based on birth cohorts, cultural background, and ethnic beliefs, meaning that victims may not define their experiences as abuse.³⁴

The critical need to identify elder abuse and maltreatment is nowhere better supported than by the work of Lachs and colleagues³⁶ who examined the mortality associated with elder mistreatment in a cohort of community-dwelling adults. Those elders who had experienced mistreatment were more likely to die than those who were self-neglectful or those who had not been seen for any type of abuse or neglect. This finding held even after controlling for other health-related characteristics such as chronic disease, marital status, and cognitive status.

Conclusion. The incidence and prevalence of elder abuse are not well known, but there is some evidence to suggest that the numbers are increasing. Like other forms of intimate and partner violence, elder abuse appears to affect women more often than men, although some portion of this difference may be due to women’s longer life spans; nearly three fifths (59%) of the population aged 65 years and older is female. Physicians are no doubt regularly seeing victims of elder abuse among their patients, but may not be identifying all cases, so routine screening continues to be recommended.
Back to Top

Immigrant and Ethnic Groups

A recent UNICEF report on violence against women and girls points out that "violence against women is present in every country, cutting across boundaries of culture, class, education, income, ethnicity and age."^37p2 Under these circumstances, one could reasonably expect to find instances of intimate partner violence and other abuse within immigrant and ethnic population groups in the United States, not because of any individual’s status as an immigrant or ethnic minority but simply because violent and abusive behaviors are so pervasive.

Unfortunately, however, data on violence in immigrant and ethnic groups is sparse. Reports on only a handful of immigrant or ethnic groups have appeared in the published literature. Furthermore, ethnic groups are much more heterogeneous than homogeneous, labels notwithstanding. Consider, for example, that "Asians" would include Chinese, Japanese, Vietnamese, Indians, Pakistanis, and Turks, among others, or that "Native Americans" includes people from more than 500 federally recognized tribes, each with a unique history and cultural heritage.

Clearly, many of the issues surrounding cases of intimate partner violence—power and control on the one hand, appropriate referral and treatment on the other—arise for individuals who are members of particular immigrant or ethnic groups as well. However, the issues are complicated by cultural definitions of appropriate behavior, including factors that encourage or discourage helpseeking behavior, a topic that is well beyond the bounds of this report. Where there are particular issues of domestic violence in immigrant and ethnic populations, involvement of the local community is critical to support the victims.³⁷

For the United States as a whole, the most recent data come from the NVAWS. That study reports that American Indian/Alaska Native women were "significantly more likely to disclose rape and physical assault victimization than women of other racial/ethnic backgrounds, while Asian/Pacific Islander women were significantly less likely" to make such reports.^{10 p5} It should be noted that the numbers of American Indian/Alaska Native women in the sample were too small to make reliable estimates, however. Worse, these figures ignore the heterogeneity among the various American Indian tribes, lumping all these women into a single category. Data on ethnic groups and intimate partner violence rather than all physical assaults are not found in the NVAWS reports.

The limited research that is available has generally been restricted to small convenience samples in particular locales. For example, Rodriguez and colleagues³⁸ interviewed groups of Latina and Asian women in San Francisco about domestic violence and health care. They concluded that the women sought "open and supportive patient-provider relationships, encouragement by providers to discuss abuse issues, and continuity of care." ^(38p310) Both groups of women were reluctant to initiate such discussions, preferring that the issues be raised by the physician. These preferred behavior patterns are the same as those voiced by abused women in general⁸ and are an indication of commonalities in caring for victims of abuse. While this is a reasonable conclusion, the further generalization of these results to other groups, including other Latina or Hispanic women, is at best uncertain.

A recent review calls for emergency physicians to "be prepared to screen for, diagnose, and treat women in the international community who are victims of domestic violence."^{39 p617} The same article notes that supportive mechanisms found in one’s homeland only rarely accompany immigrants to this country, meaning that former supports are lost and access to care in this country may be limited. Further complicating the picture for recent immigrants are laws that restrict access to welfare benefits⁴⁰ and laws that limit the legal status of non-citizen spouses for a period after coming to this country.^39,40 Such legal limitations present special challenges to victims and constrain their ability to remove themselves from the abuse. When these factors combine with the more common situations involving financial dependence, limited education, fear of retribution and the like, the role of the health care system in providing safety is further complicated.

Conclusion. Immigrants and members of ethnic groups include many individuals who likely are victims of intimate partner violence, and women immigrants may be at particular risk.³⁹ There is certainly no reason to believe that such groups are experiencing less intimate partner violence than are members of the majority public. Often these victims will be without traditional support systems, and they may be from backgrounds in which certain behaviors are not defined as abusive. Physicians who encounter immigrant patients or patients from some ethnic groups need to be especially vigilant and must be ready to refer such patients to community-based support groups if available. In the absence of data, cultural sensitivity and familiarity with issues of intimate partner violence are key. Improving care for such groups will require rigorous research with representative samples of diverse populations that allow the examination of both intra- and inter-group diversity as well as the assessment of group-specific risk factors along with the same groups’ strengths and protective mechanisms.
Back to Top

Child Maltreatment

That there are significant gaps in the knowledge base of family and intimate partner violence might be surprising until one realizes that even the study of child maltreatment is less than 40 years old. Only after Kempe and associates⁴¹ published their now classic study of battered children in 1962 did the medical community fully recognize and appreciate the significant health effects of child maltreatment.

Child maltreatment continues to be a significant health problem in this country, with both short- and long-term consequences and implications for the health of both victims and succeeding generations through a cycle of violence.⁴³ While statutory definitions vary from state to state, child maltreatment includes physical, sexual, and emotional abuse, as well as neglect. The most recent report from the National Child Abuse and Neglect Data System (NCANDS) estimates that there were 903,000 cases of child maltreatment nationwide in 1998, or 12.9 cases per 1000 children.⁴³ This figure is down from a high of 15.3 cases per 1000 children in 1993 and down as well over the 1996 and 1997 rates of 15 and 13.9 per 1000 respectively.

The most common type of child maltreatment is neglect, which is an act or failure to act that results in serious harm or an imminent risk of harm. Physical neglect is the most common type of neglect, but instances of educational, emotional, and medical neglect are also known. In 1998, just over half (53.5%) of substantiated cases involved one of the first three types of neglect—usually physical neglect—and about another 6% were cases of medical neglect. Thus 3 cases in 5 have to do with neglect. For the remaining 40% of cases, one quarter (22.7%) are cases of physical abuse, about 1 in 8 (11.5%) cases has to do with sexual abuse, and the remaining cases deal with emotional abuse.⁴³

While these proportions would probably remain relatively constant, the true prevalence of child maltreatment is probably greater than the 903,000 cited by the Department of Health and Human Services. In 1998, 2.8 million reports of possible child maltreatment were received by child protective agencies, but more than half (53.1%) were not substantiated upon investigation. The number of suspected cases not reported is indeterminate, but it is likely that cases perceived as less severe are less likely to be reported to the authorities.

Some national surveys have also collected data on the incidence and prevalence of child maltreatment. In fact, the NFVS asked respondents about their behaviors toward their own children and found 2% of parents admitted to behaviors that would likely be physically abusive; this yields an estimate of 1.5 million abused children annually.⁴² Similar self-report data are not available for most other types of abuse, particularly neglect. In 1986 and 1993, information was collected from samples of professionals who deal with child maltreatment in the National Incidence Studies. The later data indicated that nearly 900,000 children had suffered neglect in 1993, and nearly another 400,000 had experienced physical abuse. Figures for all types of abuse had increased dramatically over the 1986 data.⁴² However, these data are somewhat limited by who gets reported to the child protection system and what data states actually collect. There is a need for a more comprehensive surveillance system to know whether child maltreatment is increasing or decreasing; without doubt, the figures are incomplete.

Even child maltreatment deaths are not known with certainty. The number of child deaths caused by abuse or neglect has steadily climbed over the last decade, reaching an estimated total of 1100 children in 1998 (1.6 per 100,000 children).⁴³ More than one third (37.9%) of these deaths occurred in children under 1 year of age, and three quarters (77.5%) were under age 5 years. However, like other forms of maltreatment, even child fatalities may be underreported. A 1995 report from the US Advisory Board on Child Abuse and Neglect estimated that known and unknown (i.e., to child protective agencies) deaths due to child maltreatment numbered at least 2000 annually, a figure about twice the current official total.⁴⁴ Determining that a child’s death is due to maltreatment can be difficult, particularly for coroners without medical training but also for physicians whose forensic training tends to be with adult deaths. Not infrequently, parents’ explanations are accepted at face value simply because they are "nice people."^{44 p52-53} Consequently, most states have developed child fatality review teams, made up of key professionals from many fields, to examine every case of child death to determine the interplay between medical, criminal and child protective concerns, thereby better identifying child maltreatment deaths and preventing future cases.^44,45 The value of such teams has not yet been determined.

Perpetrators. Less is known about perpetrators since not all states report data on perpetrators. However, data from 16 states show interesting patterns when examining the different types of child maltreatment. On the one hand, looking at all cases of child maltreatment, women are more likely to be perpetrators than are men. In fact, three fifths (60.4%) of perpetrators of child maltreatment are women, and they are generally slightly younger than male perpetrators, with median ages of 31 years versus 34 years respectively.⁴³ On the other hand, compared to children who were victims of neglect, children who were victims of physical and sexual abuse were more likely to be mistreated by a male parent acting alone. Thus while the typical abuser may be female, what most people think of as more serious abuse tends to be carried out by male offenders. Distressingly, 7 cases in 8 (87.1%) involve maltreatment by one or both parents of the child.⁴³

Child maltreatment and intimate partner violence. Over the past decade there has been growing interest in the effects of children witnessing violence, particularly violence within the home in which the (step-) parents may engage in intimate partner violence. Available data indicate that children under age 12 years reside in about 43% of the homes where intimate partner violence occurs.¹¹ While direct measures of children witnessing violence are not generally available, it must certainly be the case that many of these young children witness their parents engaging in physical battle. NFVS data suggest that between 1.5 and 3.3 million children witness intimate partner assaults each year.⁹ It is now well known that intimate partner violence and child maltreatment can occur concurrently, and it is likely that men who are abusive of their intimate partners are likely to abuse the children of those intimate partners as well.⁴²

A careful review of the effects of domestic violence on children was recently published.⁴⁶ In short, the preponderance of evidence shows that children who experience such violence are encumbered in a number of ways. An overview by Fantuzzo and Mohr^47p27 reports that children "exposed to domestic violence tended to be more aggressive and to exhibit behavior problems … ranging from temper tantrums to fights." Other problems include mental health difficulties such as depression, anxiety, and suicide as well as "difficulty in schoolwork, and significantly lower scores on measures of verbal, motor, and cognitive skills."

Aftereffects of violence. Perhaps the most disturbing aspect of children witnessing intimate partner violence or being the targets of violence themselves is the increased likelihood of their perpetrating such violence on their peers and their own partners and children later in life, the so called cycle of violence. Herrenkohl and colleagues⁴⁸ report that abuse victims are more likely than unvictimized children to become participants in violent, criminal behavior as teens and adults. Widom⁴⁹ found that physically abused children were more likely be arrested later for a violent crime, but victims of neglect were only slightly less likely to be arrested for such events. Given the greater incidence of maltreatment cases due to neglect compared to all other types of violence, a very large fraction of violent crime could be due in part to the experience of neglect early in life.

Similarly disturbing are the findings of the recent Adverse Childhood Experiences (ACES) Study. Felitti and colleagues⁵⁰ report that the greater the household dysfunction or the more often a child was victimized, the greater the likelihood of exhibiting health risk factors such as substance abuse, smoking, depression, and suicide attempts. Moreover, those with greater exposure to maltreatment were also more likely to experience adult disorders including heart disease, cancer, fractures, and liver disease. Obviously, the connections between childhood maltreatment and later sequelae are complicated, but exposure is clearly an independent risk factor for a wide variety of health and health-related problems.

Conclusion. Reported cases of child maltreatment appear to be declining, but the true incidence and prevalence of child maltreatment are unknown. Regardless, nearly a million children in this country are maltreated annually by their parents and guardians, and both mothers and fathers are guilty of this maltreatment. Of particular concern is the fact that child maltreatment and intimate partner violence frequently co-occur in the same families, and worse, maltreated children are at greater risk for a number of adverse events. These include a greater likelihood of engaging in criminal behavior, school difficulties, unhealthy lifestyles, and a propensity to victimize their own partners and children in later life.
Back to Top

Alcohol and Family or Intimate Partner Violence

At least two points are clear from the research literature and data examining the role that alcohol plays in intimate partner violence:

• The way alcohol interacts with family and intimate partner violence is complex and varies across types and severity of violence, populations and circumstances, and
• Alcohol is present in a substantial proportion although not a majority of intimate partner violence incidents. Nonetheless, it must play a major role in any discussion of the prevention or remediation of the effects of intimate partner violence.

Although there is a vast literature on the topic of alcohol and violence, a consistent theme is that a great deal more information is required before the exact nature and mechanisms of their relationship can be established. Many studies of alcohol and violence do not ascertain whether the alcohol use preceded or followed an incident, or whether the report of alcohol’s presence in an incident was only an indication of chronic alcohol problems surrounding an incident but not necessarily resulting in intoxication during the incident. Nevertheless, there is clearly a great deal of epidemiological, experimental, and observational data that alcohol is in many cases not a neutral but rather a positive (at times causal) factor in the initiation and exacerbation of violence. The frequency with which violence and alcohol are found together in a wide range of circumstances indicates that it is more than a correlational relationship.
Back to Top

Data from victim and perpetrator reports. A 1998 report from the US Department of Justice, "Violence by Intimates,"⁵¹ found that nationally, of 960,000 incidents involving violence by intimates over age 12 years (based on the average number of cases reported each year from 1992 to 1996), there was evidence of offender drinking in one quarter of all intimate assaults (24.9%), aggravated assaults (28%), and simple assaults (25.8%), and in one eighth (12.7%) of intimidation cases. In these cases about 85% of victims were women. Among state prisoners who had been drinking before they committed a crime against an intimate, half (49%) had spent 6-12 or more hours drinking, and one third (33.9%) of convicted jail inmates had consumed 10 or more drinks. About half of prison and jail inmates were drinking beer or liquor when they committed a violent crime against an intimate companion.

An analysis of national crime data for 1996 found that two thirds of victims who suffered violence by an intimate (current or former spouse, boyfriend or girlfriend) reported that alcohol was a factor.⁵¹ Among spouse victims, 3 out of 4 incidents were reported to involve an offender who had been drinking (versus 31% of offenders who were strangers and where the victim could determine if alcohol was present). Where the offender was a former spouse, boyfriend or girlfriend (1.1 million incidents), 34% of the offenders had used alcohol, another 7% used drugs and/or alcohol, and 5% used only other drugs. When the offender was a parent or child or other relative (481,800 incidents), the offender had used alcohol in 19%, drugs in 7%, and drugs and/or alcohol in 6+% of the cases. When the offender was a casual or well-known acquaintance (3.6 million incidents) the offender used alcohol in 16%, drugs in 5%, and drugs and/or alcohol in 5+% of the cases.
Back to Top

Alcohol and violence in the home. The Federal Bureau of Investigation’s 1995 National Incident-Based Reporting System (NIBRS) concluded that about half of incidents described by investigating officers as alcohol-related were between offenders and victims who were intimates.⁵² Seven out of ten alcohol-involved violent incidents occurred in a residence, and 2 in 10 involved the use of a weapon other than hands, fists, or feet. About half of convicted murderers in state prisons reported that alcohol was a factor in about half of the murders they committed (36% when the victim was a family member, 27% when the victim was an intimate, 24% when the victim was an acquaintance). Among the murdered victims, the average level of blood alcohol ranged from 0.24% to 0.30%. Nationally, of the 5.3 million convicted offenders under the jurisdiction of state, federal and local corrections agencies, about 36% (2 million) were estimated to have been drinking at the time of the offense.

A recent study and literature review of alcohol and violent death in the home found that the risks of violent death (homicide and suicide) increased when several indicators of chronic alcohol abuse were present.⁵³ The authors estimated that the use of alcohol is associated with an approximately two-fold increased risk of homicide within all age groups and that nondrinkers living in homes with alcohol users were at increased risk of homicide. Looking at other literature, they found that homicide and suicide victims between the ages of 25 and 64 years were more likely to be moderate and heavy drinkers than those who died from other causes and that this association was greater for men than for women.
Back to Top

Alcohol and child maltreatment. The literature concerning child abuse, neglect, and injury similarly indicates increased risk to children when problem drinking, alcoholism or other drug abuse is present in the home. The dynamics, however, vary according to the type of harm perpetrated. For example, maternal substance abuse can lead to intended harm or to supervisory neglect that allows the child to be injured in some other way. One national study estimated that children of problem drinking mothers had over twice the risk of serious injury compared to children of non-drinking mothers.⁵⁴ When mothers with problem drinking patterns were married to moderate or heavy drinking men, the relative risk of serious injury for their children was 2.7 compared with children of nondrinkers.

In general, the risks for child sexual abuse appear to be greatest when the perpetrator is a substance abuser who is not the father of the child (usually girls) but who either lives in the home or is a close relative or acquaintance of the mother. Risk appears to be higher if one or more adult primary caregivers are chemically dependent and were themselves physically abused or neglected when they were children. One extensive review of the literature in this area concluded that "children from alcoholic and heroin-dependent families may generally be at greater risk than the general population for all forms of abuse and neglect [although] the level and frequency of intoxication play important roles."^{55 p40}

A consequence of both child maltreatment and of parental alcohol abuse is that children are more likely to engage in similar behaviors when adults. A history of childhood abuse is frequently reported by male, and especially female, post-traumatic stress disorder patients who are drug abusers. Women rape victims are more likely to abuse drugs than are women who are not victims.⁵⁶ A number of studies and clinical reports have concluded that childhood victimization, especially sexual abuse, is significantly correlated with adult alcoholism in women.^55,57
Back to Top

How alcohol affects intimate partner violence. There are numerous explanations for the impact of alcohol on intimate partner violence, but none account for all individuals or types of violence. One set of theories posits that alcohol, through its drug effects, encourages aggression or violence by disrupting brain function; e.g., by weakening inhibitions and thus increasing aggressiveness.^58-60 Consequently, alcohol use impairs communications so that the user is less likely to pay adequate attention to what someone else is saying, is more likely to misinterpret information, and overall experiences impaired information processing. This may lead to more or intensified conflict and aggression and placing oneself in riskier (e.g., potentially volatile) situations. If the drinking patterns of the two individuals in a conflict-laden relationship are different, this may also exacerbate conflicts. However, there is also substantial information that the alcohol variables and intimate partner violence are also associated by common background causes.

Social and cultural factors provide another set of explanations. Studies based on them conclude that alcohol consumption promotes aggression because people expect it to. In this case, alcohol is used as a justification for violence or to bolster aggressive behavior while avoiding accountability or providing the offender with an exemption from conduct expected when sober.^58.61 These changed expectations of violence (i.e., the belief that alcohol use leads to aggressive behavior) become a self-fulfilling prophecy. For example, a combination of beliefs that men can batter women and that alcohol both allows them to and excuses their behaviors is an example of this thinking. Increasing consumption to produce a higher level of intoxication can help prepare a perpetrator to commit a socially unacceptable or violent act, and at the same time it diminishes the perpetrator’s immediate self-control. Some studies indicate that alcohol abuse may be a marker for antisocial personality or a result of other mental illness and not the cause of the violence per se.⁵³

A growing body of research demonstrates that there is a relationship between greater availability of alcohol (e.g., through numerous outlets concentrated in a given geographic area) leading to higher consumption and a consequent wide range of crimes and forms of violence. Intimate partner violence-related research has especially shown this relationship in college campuses, their surrounding communities, and in inner city neighborhoods. The linkages of alcohol to injury, crime, and violence have been attributed to a combination of factors: increased heavy consumption and intoxication, easily accessible and low cost drinks, drinking behavior patterns, local normative expectations regarding alcohol and violence, and the use of alcohol outlets as a locus for other forms of crime and disruption. For example, on college campuses interpersonal conflicts (e.g., fights, rape, harassment, arguments) are higher on campuses with larger proportions of binge drinkers than on abstinent or low alcohol consumption campuses.⁶² Studies of Alaskan villages that, being relatively geographically isolated, have cut off supplies of alcohol, and of a Norwegian strike that cut off alcohol supplies, show decreases in domestic disruptions, violent crime (especially homicide), and injury leading to death.^63-65 Some economic studies have shown that higher prices for alcohol lead to a reduction in consumption and a consequent reduction in intimate partner violence and other crimes.⁶⁵

Conclusion. There is a growing consensus in support of a complex view incorporating all of the above explanations. These result in a perspective that while the data show a clear, strong correlation between alcohol use and substantial portions of many forms of domestic abuse, these incidents are also greatly influenced by situation, individual and cultural factors, biological factors, and even by population groups (e.g., subgroup norms, expectations and drinking behaviors) of perpetrators.⁶⁶ Alcohol may, therefore, increase the likelihood of violence or increase its severity, but is only involved in a substantial minority of intimate partner violence incidents. According to Leonard,⁵⁹ "The empirical evidence suggests that, like most factors in domestic violence, it is neither a necessary or sufficient cause, but rather a contributing cause."

Recognizing this complex relationship, perhaps the best model for looking at the whole issue is a public health one in which alcohol (the agent) interacts with a host (the individual) within an environment consisting of a complex of factors. From that perspective,

Recent research demonstrates that alcohol does play a causal role within a complex set of factors and circumstances. These include the drinking contexts, societal and cultural attitudes, alcohol’s pharmacological effects, and factors peculiar to the personality of those involved in the violence. … the greatest progress in preventing alcohol-related violence will come from multiple, interrelated strategies that take complex relationships into account.⁶⁷

Based on all the above, the treatment and prevention of intimate partner violence must always recognize the strong possibility that alcohol was involved, must be ascertained and, if a repetition of the behavior is to be avoided, must be treated and otherwise intervened with. Public health messages need to indicate both that violence against others is not acceptable and that the use of alcohol will not excuse it.
Back to Top

CONCLUSION

The Council on Scientific Affairs believes that existing AMA policy accurately reflects the situation in US society in which physicians, particularly those providing primary care to patients, are likely to regularly encounter patients who are victims of family violence. At the same time, existing policy was developed somewhat piecemeal, and the CSA believes that policy consolidation is appropriate.

RECOMMENDATIONS (Adopted AMA Policy)

As recommended by the Council on Scientific Affairs, the following was adopted by the AMA House of Delegates as a consolidation of AMA policy in December 2000:

Policy H-515.965 Family and Intimate Partner Violence

1. The American Medical Association believes that all forms of family and intimate partner violence are major public health issues and urges the profession, both individually and collectively, to work with other interested parties to prevent such violence and to address the needs of victims. Physicians have a major role in lessening the prevalence, scope and severity of child maltreatment, intimate partner violence, and elder abuse, all of which fall under the rubric of family violence. To support physicians in practice, the AMA will continue to campaign against family violence and remains open to working with all interested parties to address violence in US society. The AMA’s efforts will be guided, in part, by its Advisory Council on Family Violence.
2. The AMA believes that all physicians should be trained in issues of family and intimate partner violence through undergraduate and graduate medical education as well as continuing professional development. The AMA, working with state, county and specialty medical societies as well as academic medical centers and other appropriate groups such as the Association of American Medical Colleges, should develop and disseminate model curricula on violence for incorporation into undergraduate and graduate medical education, and all parties should work for the rapid distribution and adoption of such curricula when developed. These curricula should include coverage of the diagnosis, treatment, and reporting of child maltreatment, intimate partner violence, and elder abuse and provide training on interviewing techniques, risk assessment, safety planning, and procedures for linking with resources to assist victims. The AMA supports the inclusion of questions on family violence issues on licensure and certification tests.
3. The prevalence of family violence is sufficiently high and its ongoing character is such that physicians, particularly physicians providing primary care, will encounter victims on a regular basis. Persons in clinical settings are more likely to have experienced intimate partner and family violence than non-clinical populations. Thus, to improve clinical services as well as the public health, the AMA encourages physicians to: (a) Routinely inquire about the family violence histories of their patients as this knowledge is essential for effective diagnosis and care; (b) Upon identifying patients currently experiencing abuse or threats from intimates, assess and discuss safety issues with the patient before he or she leaves the office, working with the patient to develop a safety or exit plan for use in an emergency situation and making appropriate referrals to address intervention and safety needs as a matter of course; (c) After diagnosing a violence-related problem, refer patients to appropriate medical or health care professionals and/or community-based trauma-specific resources as soon as possible; (d) Have written lists of resources available for victims of violence, providing information on such matters as emergency shelter, medical assistance, mental health services, protective services and legal aid; (e) Screen patients for psychiatric sequelae of violence and make appropriate referrals for these conditions upon identifying a history of family or other interpersonal violence; (f) Become aware of local resources and referral sources that have expertise in dealing with trauma from victimization; (g) Be alert to men presenting with injuries suffered as a result of intimate violence because these men may require intervention as either victims or abusers themselves; (h) Give due validation to the experience of victimization and of observed symptomatology as possible sequelae; (I) Record a patient’s victimization history, observed traumata potentially linked to the victimization, and referrals made; (j) Become involved in appropriate local programs designed to prevent violence and its effects at the community level;
4. Within the larger community, the AMA (a) Urges hospitals, community mental health agencies, and other helping professions to develop appropriate interventions for all victims of intimate violence. Such interventions might include individual and group counseling efforts, support groups, and shelters. (b) Believes it is critically important that programs be available for victims and perpetrators of intimate violence. (c) Believes that state and county medical societies should convene or join state and local health departments, criminal justice and social service agencies, and local school boards to collaborate in the development and support of violence control and prevention activities.
5. With respect to issues of reporting, the AMA strongly supports mandatory reporting of suspected or actual child maltreatment and urges state societies to support legislation mandating physician reporting of elderly abuse in states where such legislation does not currently exist. At the same time, the AMA opposes the adoption of mandatory reporting laws for physicians treating competent, non-elderly adult victims of intimate partner violence if the required reports identify victims. Such laws violate basic tenets of medical ethics. If and where mandatory reporting statutes dealing with competent adults are adopted, the AMA believes the laws must incorporate provisions that (a) do not require the inclusion of victims' identities; (b) allow competent adult victims to opt out of the reporting system if identifiers are required; (c) provide that reports be made to public health agencies for surveillance purposes only; (d) contain a sunset mechanism; and (e) evaluate the efficacy of those laws. State societies are encouraged to ensure that all mandatory reporting laws contain adequate protections for the reporting physician and to educate physicians on the particulars of the laws in their states.
6. Substance abuse and family violence are clearly connected. For this reason, the AMA believes that: (a) Given the association between alcohol and family violence, physicians should be alert for the presence of one behavior given a diagnosis of the other. Thus, a physician with patients with alcohol problems should screen for family violence, while physicians with patients presenting with problems of physical or sexual abuse should screen for alcohol use. (b) Physicians should avoid the assumption that if they treat the problem of alcohol or substance use and abuse they also will be treating and possibly preventing family violence. (c) Physicians should be alert to the association, especially among female patients, between current alcohol or drug problems and a history of physical, emotional, or sexual abuse. The association is strong enough to warrant complete screening for past or present physical, emotional, or sexual abuse among patients who present with alcohol or drug problems. (d) Physicians should be informed about the possible pharmacological link between amphetamine use and human violent behavior. The suggestive evidence about barbiturates and amphetamines and violence should be followed up with more research on the possible causal connection between these drugs and violent behavior. (e) The notion that alcohol and controlled drugs cause violent behavior is pervasive among physicians and other health care providers. Training programs for physicians should be developed that are based on empirical data and sound theoretical formulations about the relationships among alcohol, drug use, and violence.

The AMA reaffirms Policy H-515.983, Physicians and Family Violence, which outlines ethical considerations for physicians in dealing with instances of family violence.
Back to Top

CSAPH home page
1999-2000 reports
Reports by topic

---

References

1. Coker AL, Smith PH, McKeown RE, King MJ. Frequency and correlates of intimate partner violence by type: physical, sexual, and psychological battering. Am J Public Health. 2000;90:553-559.
2. Bensley L, Macdonald S, Eenwyk JV, Wynkoop Simmons K, Ruggles D. Prevalence of intimate partner violence and injuries—Washington, 1998. MMWR. 2000;49:589-592.
3. National Research Council. Understanding Violence Against Women. Washington, DC: National Academy Press; 1996.
4. Chalk R, King PA, eds. Violence in Families: Assessing Prevention and Treatment Programs. Washington, DC: National Academy Press; 1998.
5. Saltzman LE, Fanslow LJ, McMahon PM, Shelley GA. Intimate Partner Violence Surveillance: Uniform Definitions and Recommended Data Elements. Atlanta: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention; 1999.
6. Heise L, Ellsberg M, Gottemoeller M. Ending Violence Against Women. Population Reports, Series L, No 11. Baltimore: Johns Hopkins University School of Public Health, Population Information Program; 1999.
7. American Medical Association. Diagnostic and Treatment Guidelines on Elder Abuse and Neglect. Chicago: American Medical Association; 1992.
8. American Medical Association. Diagnostic and Treatment Guidelines on Domestic Violence. Chicago: American Medical Association; 1992.
9. Straus MA, Gelles RJ. How violent are American families? Estimates from the National Family Violence Resurvey and other studies. In: Straus MS, Gelles RJ, eds. Physical Violence in American Families: Risk Factors and Adaptations to Violence in 8,145 Families. New Brunswick, NJ: Transaction; 1990:95-112.
10. Tjaden P, Thoennes N. Prevalence, Incidence, and Consequences of Violence Against Women: Findings from the National Violence Against Women Survey. Washington, DC: US Department of Justice, November 1998. NCJ 172837.
11. Rennison CM, Welchans S. Intimate Partner Violence. Bureau of Justice Statistics Special Report. Washington, DC: US Department of Justice. May 2000. NCJ 178247.
12. Kaufman Kantor G, Straus MA. Response of victims and the police to assaults on wives. In: Straus MA, Gelles RJ, eds. Physical Violence in American Families: Risk Factors and Adaptations to Violence in 8,145 Families. New Brunswick, NJ: Transaction; 1990:473-486.
13. Stark E, Flitcraft A, Frazier W. Medicine and patriarchal violence: the social construction of a "private" event. Int J Health Serv. 1979;9:461-493.
14. Council on Scientific Affairs. Mental Health Effects of Family Violence. Presented at the 1993 Annual Meeting of the House of Delegates.
15. Tjaden P, Thoennes N. Extent, Nature, and Consequences of Intimate Partner Violence: Findings From the National Violence Against Women Survey. Washington, DC: National Institute of Justice; 2000. NCJ 181867.
16. Straus MA. Physical assaults by women partners: a major social problem. In: Walsh MR, ed. Women, Men and Gender: Ongoing Debates. New Haven: Yale University Press; 1997:210-221.
17. Stets JE, Straus MA. Gender differences in reporting marital violence and its medical and psychological consequences. In: Straus MA, Gelles RJ, eds. Physical Violence in American Families: Risk Factors and Adaptations to Violence in 8,145 Families. New Brunswick, NJ: Transaction; 1990:151-167.
18. Browne A. When Battered Women Kill. New York: Free Press; 1987.
19. Straus MA. The controversy over domestic violence by women: a methodological, theoretical, and sociology of science analysis. In: Arriaga XB, Oskamp S, eds. Violence in Intimate Relationships. Thousand Oaks, CA: Sage; 1999:17-44.
20. Bachman R, Saltzman LE. Violence Against Women: Estimates from the Redesigned Survey, Special Report. Washington, DC: US Department of Justice, Bureau of Justice Statistics; 1995. NCJ 154348.
21. Straus MA. Characteristics of the National Violence Against Women Study that might explain the low assault rate of both sexes and the even lower rate for assaults by women. Unpublished manuscript. Family Research Laboratory, University of New Hampshire; 1999.
22. Bachman R. A Comparison of Annual Incidence Rates and Contextual characteristics of Intimate-Partner Violence Against Women from the National Crime Victimization Survey (NCVS) and the National Violence Against Women Survey (NVAWS). Violence Against Women. 2000;6:839-867.
23. Community United Against Violence. Part 2, national section. In: Lesbian, Gay, Bisexual and Transgender Domestic Violence in 1998. New York: National Coalition of Anti-Violence Programs; 1999:23-41.
24. Brand PA, Kidd AH. Frequency of physical aggression in heterosexual and female homosexual dyads. Psychol Rep. 1986;59:1307-1313.
25. Morrow J. Identifying and treating battered lesbians. San Francisco Medicine. 1994;67(4):17, 20-21.
26. Letellier P. Identifying and treating battered gay men in a medical setting. San Francisco Medicine. 1994;67(4):16, 18-19.
27. Tjaden P, Thoennes N, Allison CJ. Comparing violence over the life span in samples of same-sex and opposite-sex cohabitants. Violence Vict. 1999;14:413-425.
28. Dolan-Soto D. Part 1, New York City Section. In: Lesbian, Gay, Bisexual and Transgender Domestic Violence in 1998. New York: National Coalition of Anti-Violence Programs; 1999:7-20.
29. Council on Scientific Affairs. Health care needs of gay men and lesbians in the United States. JAMA. 1996;275:1354-1359.
30. Lachs MS, Berkman L, Fulmer T, Horwitz R. A prospective community-based pilot study of risk factors for the investigation of elder mistreatment. J Am Geriatr Soc. 1994;42:169-173.
31. Pillemer K. Finkelhor D. The prevalence of elder abuse: a random sample survey. Gerontologist. 1988;28:51-57.
32. Tatara T, Kuzmeskus LM, Duckhorn E. Trends in elder abuse in domestic settings. Elder Abuse Information Series, No 2. Washington, DC: National Center on Elder Abuse; 1997.
33. Tatara T, Kuzmeskus LM, Duckhorn E. Reporting of elder abuse in domestic settings. Elder Abuse Information Series, No 3. Washington, DC: National Center on Elder Abuse; 1997.
34. Clarke ME, Pierson W. Management of elder abuse in the emergency department. Emerg Med Clin North Am. 1999;17:631-644.
35. Conlin M. Silent suffering: a case study of elder abuse and neglect. J Am Geriatr Soc. 1995;43:1303-1308.
36. Lachs MS, Williams CS, O’Brien S, Pillemer KA, Charlson ME. The mortality of elderly mistreatment. JAMA. 1998;280:428-432.
37. UNICEF (United Nations Children’s Fund). Domestic violence against women and girls. Innocenti Digest No. 6 (Preliminary Edition). Florence: Innocenti Research Center. May 2000.
38. Rodriguez MA, Bauer HM, Flores-Ortiz Y, Szkupinski-Quiroga S. Factors affecting patient-physician communication for abused Latina and Asian immigrant women. J Fam Pract. 1998;47:309-311.
39. Clarke ME, Pierson W. Domestic violence against women in the international community. Emerg Med Clin North Am. 1999;17:617-630.
40. Ivey SL, Kramer EJ. Immigrant women and the emergency department: the juncture with welfare and immigration reform. J Am Womens Assoc. 1998;53(2):94-95,107.
41. Kempe CH, Silverman FN, Steele BF, Droegemueller W, Silver HK. The battered-child syndrome. JAMA. 1962;181:1-17.
42. English DJ. The extent and consequences of child maltreatment. Future Child. 1998;8(1):39-53.
43. US Department of Health and Human Services. Child Maltreatment 1998: Reports from the States to the National Child Abuse and Neglect Data System. Washington, DC: US Government Printing Office; 2000.
44. US Advisory Board on Child Abuse and Neglect. A Nation’s Shame: Fatal Child Abuse and Neglect in the United States. Washington, DC: Department of Health and Human Services, Administration for Children and Families, US Advisory Board on Child Abuse and Neglect; 1995.
45. Interagency Council on Child Abuse and Neglect. Interagency Council on Child Abuse and Neglect Web Site. Available at: http://child.cornell.edu/ncfr/home.html. Accessed July 19, 2000.
46. Domestic violence and children. Future Child. 1999;9(3):entire issue.
47. Fantuzzo JW, Mohr WK. Prevalence and effects of child exposure to domestic violence. Future Child. 1999;9(3):21-32.
48. Herrenkohl RC, Egolf BP, Herrenkohl EC. Preschool antecedents of adolescent assaultive behavior: a longitudinal study. Am J Orthopsychiatry. 1997;67:422-32.
49. Widom CS. The Cycle of Violence. Washington, DC: National Institute of Justice; 1992. NCJ 136607.
50. Felitti VJ, Anda RF, Nordenberg D, Williamson DF, Spitz AM, Edwards. V, Koss MP, Marks JS Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: the Adverse Childhood Experiences (ACE) Study. Am J Prev Med. 1998;14:245-258.
51. Greenfeld LA, Rand MR, Craven D, et al. Violence by Intimates: Analysis of Data on Crimes by Current or Former Spouses, Boyfriends, and Girlfriends. Washington, DC: US Department of Justice, Bureau of Justice Statistics; 1998. NCJ 167237.
52. Greenfeld LA. Alcohol and Crime: An Analysis of National Data on the Prevalence of Alcohol Involvement in Crime. Washington, DC: US Department of Justice, Bureau of Justice Statistics; 1998. NCJ 168632.
53. Rivara FP, Mueller BA, Somes G, Mendoza CT, Rushforth NB, Kellerman AL. Alcohol and illicit drug abuse and the risk of violent death in the home. JAMA. 1997;278:569-575.
54. Bijur PE, Kurzon M, Overpeck MD, Scheidt PC. Parental alcohol use, problem drinking, and children’s injuries. JAMA. 1992;267:3166-3171.
55. Yoast RA, McIntyre K. Alcohol, Other Drug Abuse and Child Abuse and Neglect. Madison, WI: Board of Regents, University of Wisconsin System, Wisconsin Clearinghouse; 1991.
56. Swan N. Exploring the role of child abuse in later drug abuse. NIDA Notes. 1998;13;2:1,4-6.
57. Widom CS, Ireland T, Glynn PJ. Alcohol abuse in abused and neglected children followed-up: are they at increased risk? J Stud. Alcohol. 1995;56:207-217.
58. National Institute on Alcohol Abuse and Alcoholism. Alcohol, Violence, and Aggression. Alcohol Alert No. 38. Rockville, MD: US Department of Health and Human Services, PHS, NIH, NIAAA ; no date.
59. Leonard K. Domestic violence and alcohol – what is known and what do we need to know to encourage environmental interventions. In: Commissioned Papers, Alcohol Policy XII Conference: Alcohol & Crime, Research and Practice for Prevention. Washington, DC: National Crime Prevention Council; 2000:69-91.
60. Pihl RO, Peterson J. Alcohol and aggression: three potential mechanisms of the drug effect. In: Martin SE, ed. Alcohol and Interpersonal Violence. NIAAA Research Monograph No. 24. NIH Pub. No. 93-3496. Rockville, MD: NIAAA; 1993:149-159.
61. Lang AR. Alcohol-related violence: psychological perspectives. In: Martin SE, ed. Alcohol and Interpersonal Violence. NIAAA Research Monograph No. 24. NIH Pub. No. 93-3496. Rockville, MD: NIAAA; 1993:121-147.
62. Wechsler H, Lee JE, Kuo M, Lee H. College binge drinking in the 1990’s: a continuing problem. Results of the Harvard School of Public Health 1999 College Alcohol Survey. J Am Coll Health. 2000;48:199-210.
63. Berman M, Hull T, May P. Alcohol control and injury death in Alaska Native communities: wet, damp and dry under Alaska’s local option law. J Stud Alcohol. 2000;61:311-319.
64. Landen MG, Beller M, Funk E, Propst M, Middaugh J, Moolenaar RL. Alcohol-related injury death and alcohol availability in remote Alaska. JAMA. 1997;278:1755-1758.
65. Cook, PJ, Moore MJ. Economic perspectives on reducing alcohol-related violence. In: Martin SE, ed. Alcohol and Interpersonal Violence. NIAAA Research Monograph No. 24. NIH Pub. No. 93-3496. Rockville, MD: NIAAA; 1993:193-212.
66. Roizen J. Issues in the epidemiology of alcohol and violence. In: Martin SE, ed. Alcohol and Interpersonal Violence. NIAAA Research Monograph No. 24. NIH Pub. No. 93-3496. Rockville, MD: NIAAA; 1993:3-36.
67. Mosher JA, Jernigan DH. Making the link: a public health approach to preventing alcohol-related violence and crime. In: Commissioned Papers, Alcohol Policy XII Conference: Alcohol & Crime, Research and Practice for Prevention. Washington, DC: National Crime Prevention Council; 2000:1-21. Back to Top
    

---

Resolution 410, I-99

Resolution 410, I-99, introduced by the District of Columbia Delegation, was referred. The original resolution read:

RESOLVED, That the American Medical Association (AMA) study and update the full spectrum of domestic abuse in light of recent research using evidence-based and peer-reviewed, reliable and valid methodologies for the purposes of obtaining an accurate picture of abuse among intimates, focusing especially on what have been previously hidden targets of violence between intimates, such as: ethnically diverse and underserved immigrant populations, adolescents, children, elders, same sex partners and men; and be it further

RESOLVED, That the AMA report back at I-00.

The recommendation from the Reference Committee was that the item be referred, but an amendment was offered on the floor of the House. The Reference Committee’s annotated report states the following as House action, with the text of the floor amendment underlined:

HOD ACTION: Resolution 410 referred with a report back to the House of Delegates, at I-2000 or in the alternative, a report on the status of AMA efforts to implement the aims of Resolution 410.

The Reference Committee in its report made the following comments about the testimony heard during its hearing:

Considerable testimony was heard on the need for the AMA to update its policies and publications on family violence. At the same time, testimony questioned the adequacy of available data to carry out this task and the resources to conduct the comprehensive study apparently called for in the resolution. While the sponsor’s position was that a "book-length" report on this subject was neither expected nor needed, your Reference Committee is uncertain of the breadth and depth of the research necessary to address the resolution. Thus, it may be more appropriate to review the recent literature, discuss the issue with key groups, and then prepare a report on this subject for the House with recommendations for further action. Back to Top

---

CSAPH home page
1999-2000 reports
Reports by topic

---