• In 1993 the UN General Assembly passed the Declaration on the Elimination of Violence Against Women, UN Resolution 48/104 (444).
• At both the 1994 International Conference on Population and Development (ICPD) in Cairo and the 1995 Fourth World Conference on Women in Beijing, women's organizations from around the world advocated ending gender violence as a high priority (479). The Cairo Programme of Action recognized that gender violence is an obstacle to women's reproductive and sexual health and rights, and the Beijing Declaration and Platform for Action devoted an entire section to the issue of violence against women.
• In March 1994 the Commission on Human Rights appointed the first Special Rapporteur on Violence Against Women and empowered her to investigate abuses of women's human rights (479).
• In 1994 the Organization of American States (OAS) negotiated the Inter-American Convention to Prevent, Punish and Eradicate Violence Against Women. As of 1998, 27 Latin American countries had ratified the convention (82).
• In May 1996 the 49th World Health Assembly adopted a resolution (WHA49.25) declaring violence a public health priority (479). WHO is sponsoring, together with the Center for Health and Gender Equity (CHANGE) and the London School of Hygiene and Tropical Medicine, a multicountry study on women's health and domestic violence.
• In September 1998 the Inter-American Development Bank (IDB) brought together 400 experts from 37 countries to discuss the causes and costs of domestic violence, and policies and programs to address it. The IDB currently funds research and demonstration projects on violence against women in six Latin American countries.
• In 1998 UNIFEM launched regional campaigns in Africa, Asia/Pacific, and Latin America designed to draw attention to the issue of violence against women globally (502). UNIFEM also manages The Trust Fund in Support of Actions to Eliminate Violence Against Women, an initiative that has disbursed US$3.3 million to 71 projects around the world since 1996 (503).
• In 1999 the United Nations Population Fund declared violence against women “a public health priority” (445).

A Framework for Understanding Partner Violence

The model can best be visualized as four concentric circles. The innermost circle represents the biological and personal history that each individual brings to his or her behavior in relationships. The second circle represents the immediate context in which abuse takes place—frequently the family or other intimate or acquaintance relationship. The third circle represents the institutions and social structures, both formal and informal, in which relationships are embedded—neighborhood, workplace, social networks, and peer groups. The fourth, outermost circle is the economic and social environment, including cultural norms.

A wide range of studies agrees on several factors at each of these levels that increase the likelihood that a man will abuse his partner:

• At the individual level these include being abused as a child or witnessing marital violence in the home (218, 310), having an absent or rejecting father (118), and frequent use of alcohol (30, 263, 291, 310, 339, 352).
• At the level of the family and relationship, cross-cultural studies have cited male control of wealth and decision-making within the family (275, 339) and marital conflict as strong predictors of abuse (215, 219).
• At the community level women's isolation and lack of social support, together with male peer groups that condone and legitimize men's violence, predict higher rates of violence (159, 255, 339).
• At the societal level studies around the world have found that violence against women is most common where gender roles are rigidly defined and enforced (210) and where the concept of masculinity is linked to toughness, male honor, or dominance (95, 393). Other cultural norms associated with abuse include tolerance of physical punishment of women and children, acceptance of violence as a means to settle interpersonal disputes, and the perception that men have “ownership” of women (210, 275, 310, 340).

Culture: A Double-Edged Sword

In recent years, for example, dowry has become an expected part of the marriage transaction in some countries, with future husbands demanding ever-increasing dowry both before and after marriage. Dowry demands can escalate into harassment, threats, and abuse; in extreme cases the woman is killed or driven to suicide, freeing the husband to pursue another marriage and dowry (237, 368, 407).

Elsewhere, husbands are expected to pay “bridewealth” to compensate the bride's family for the loss of labor in her natal home. In parts of Africa and Asia this exchange has likewise become commercialized, with inflated bridewealth leaving many men with the impression that they have “purchased” a wife. In a recent survey in the Eastern Cape Province of South Africa, 82% of women said it is culturally accepted that, if a man pays lobola (bridewealth) for his wife, it means that he owns her. Some 72% of women themselves agreed with this statement (235).

Both marriage traditions undermine the ability of women to escape abusive relationships. For example, parents on the Indian subcontinent are reluctant to allow their daughters to return home for fear of having to pay a second dowry, whereas in bridewealth cultures, women's parents must repay the man if their daughter leaves the marriage. As an abused woman in India observed, “One often feels like running away from it all. But where does one go? The only place is your parents' house, but they will always try to send you back” (451).

Cultural attitudes toward female chastity and male honor also serve to justify violence against women and to exacerbate its consequences. In parts of Latin America and the Near East, a man's honor is often linked to the sexual “purity” of the women in his family. If a woman is “defiled” sexually—either through rape or by engaging voluntarily in sex outside of marriage—she disgraces the family honor.

For example, in some Arab societies the only way to “cleanse” the family honor is to kill the “offending” woman or girl.A study of female homicide in Alexandria, Egypt, found that 47% of all women killed were murdered by a relative after they had been raped (190). At a recent conference in Jordan, experts from six Arab countries estimated that at least several hundred Arab women die each year as a result of honor killings (231).

Culture is neither static nor monolithic, however. Women's rights activists argue that communities must dismantle those aspects of culture that oppress women while preserving what is good. In the words of Ghanaian lawyer Rosemary Ofibea Ofei-Afboagye, “A culture that teaches male mastery and domination over women must be altered” (332).

Women at the forefront of the women's human rights movement point out that appeals to culture are often anexcuse to justify practices oppressive to women. Sudanese physician Nahid Toubia asks, “Why is it only when women want to bring about change for their own benefit that culture and custom become sacred and unchangeable?” (211)

Although culture can aggravate women's vulnerability, it can also serve as a creative resource for intervention. Many traditional cultures have mechanisms—such as public shaming or community healing—that can be mobilized as resources to confront abuse. Activists from Canada's Yukon Territory, for example, have developed Circle Sentencing, an updated version of the traditional sanctioning and healing practices of the Canadian aboriginal peoples. Within the “circle,” crime victims, offenders, justice and social service personnel, as well as community residents, listen to the victim's story and deliberate about how best to “restore justice” to the victim and the community. Sentencing often includes reparation, community service, jail time, treatment requirements, and community healing rituals (22, 289).

Activists in India and Bangladesh likewise have adapted the salishe—a traditional system of local justice—to address domestic violence. For example, when a woman is beaten, the West Bengali NGO Shramajibee Mahila Samity sends a female organizer to the village to consult with the individuals and families involved. The organizer then facilitates a salishe, attempting to steer the discussion in a pro-woman direction. Collectively, the community arrives at a proposed solution, which is formalized in writing and monitored by a local committee (102).

Reproductive Health Programs in the Lead

South Africa: Addressing violence as part of “life skills” workshops. The Planned Parenthood Association of South Africa (PPASA), together with AVSC International's Men as Partners Program, has developed a program that integrates participatory activities on gender, sexual power, and intimate relationships into PPASA's “life skills” workshops. The program began after a survey of 2,000 South African men found that 58% believed that the concept of rape did not apply to a husband forcing his wife to have sex, 48% thought the way a woman dressed caused her to be raped, and 22% approved of a man hitting his partner (compared with 5% who approved of a woman hitting her partner) (371).

Latin America: Integrating violence issues into other reproductive health care. The IPPF Western Hemisphere Region is currently working with affiliates in the Dominican Republic, Peru, and Venezuela to integrate attention to gender-based violence into other sexual and reproductive health programming. For example, in Venezuela PLAFAM has trained service providers, redesigned patient routing forms, and created new case registration forms (12).

Peru: Women listening to women's voices. ReproSalud, an innovative reproductive health program of the Peruvian women's organization Manuela Ramos, helps rural women organize to address reproductive health issues that they identify as most important. Of the 51 communities that had held diagnosticos as of March 1998, 12 communities had identified domestic violence as one of their three most important problems (262).

The Philippines: Organizing against violence. The Davao City Coordinating Council on Violence Against Women has carried out activities to reduce violence at all levels of society. These activities range from puppet shows that encourage community dialogue about gender-based violence to city-wide training for police, health workers, and government officials (70). In 1997 the Davao City Council passed the Women's Development Code, a landmark ordinance that promotes and protects the rights of women and includes extensive provisions on gender-based violence, including comprehensive counseling, medical and legal support for victims, and women's desks in all Davao City police departments (109).

Tanzania: Organizing to protect refugee women. The International Rescue Committee (IRC) has launched a project on sexual abuse and gender-based violence among the Burundian refugee women housed in camps in the Kibondo district of Tanzania. The project has used participatory research and peer outreach workers to organize the camp communities to deal with gender-based violence. The project provides counseling, 24 hour a day medical services, and access to emergency contraception through four drop-in centers (324).

Liberia: Training traditional birth attendants. In 1993 Mother Patern College of Health Sciences in Monrovia, Liberia, joined with Women's Rights International, a US-based NGO, to address the aftermath of rape during Liberia's seven-year civil war. The project's Liberian staff developed a participatory program for traditional birth attendants. The program uses exercises such as “Kaymah's Trouble,” a story of a woman raped during the war, to help traditional birth attendants expand their roles as community leaders to address violence against women (474).

Nicaragua: Researching the reproductive health consequences of violence. Since its 1991 inception, the research col- laboration between Ume— University, Sweden, and the Faculty of Medicine in León, Nicaragua, has yielded some of the richest data available anywhere on the reproductive health consequences of violence against women. Working closely with the Nicaraguan Women's Network Against Violence, researchers integrated questions on violence into a series of studies exploring infant mortality, adolescent pregnancy, HIV risk, and low birth weight. As frequent references in this report indicate, these pioneering studies have produced a wealth of information (129).

1. Do more than train. While training health care providers is important, training alone is seldom enough to change providers' behavior toward victims of domestic violence (298, 435). Although training can improve providers' knowledge and practice in the short term, the impact of training generally erodes unless a variety of other measures also are taken that support and sustain new approaches (203, 298).

2. Adopt a systems approach. Achieving lasting change requires transforming the health system itself as well as changing the behavior of individual providers (40, 89). When managers, administrators, and the health care system itself encourage and reward new, caring behavior towards victims of abuse, providers will feel better able to recognize and address violence (61, 355, 398, 491).

Adopting a systems approach to addressing violence means developing policies and protocols and ensuring that they become expected practice throughout a health care system, from the top policy makers to the front-line providers. (For a description of systems approaches in reproductive health care, see Population Reports, Family Planning Programs: Improving Quality, J-47, November 1998).

3. Make procedural changes in client care. Often, making such procedural changes as adding prompts for providers on medical charts (e.g., stickers asking about abuse, or a stamp that prompts providers to screen) or including appropriate questions on intake forms and interview schedules can encourage attention to domestic violence (329).

For example, in one US study identification rates almost doubled after staff were given a one-hour presentation on domestic violence and a violence screening question was added to the emergency department patient record chart. Evaluation showed that the addition of the chart prompt, rather than the training, made the difference (335). In another US study identification of abused women in a primary care clinic rose from none, with discretionary inquiry, to 12% when a single question on abuse was added to the client health history form (160).

4. Confront underlying attitudes and beliefs. Most training programs for health care workers have focused on the clinical management of victims. This approach yields limited results, however, because providers themselves generally share the same biases, prejudices, and fears regarding abuse as the society at large. As programs have gained experience, it has become clear that providers must examine their own attitudes and beliefs about gender, power, abuse, and sexuality before they can develop new professional knowledge and skills about dealing with victims (252, 277).

In South Africa, for example, the Agisanang Domestic Abuse Prevention and Training Project (ADAPT) and its partner, the Health Systems Development Unit of the University of Witwatersrand, developed a gender training program to be incorporated into a four-week reproductive health curriculum for nurses. The program focused first on the nurses, not as health care professionals but rather as men and women themselves. It used role-playing, popular sayings, and wedding songs to help participants analyze common notions about violence and about the proper roles of men and women. Only then did the training turn to the nurse's responsibilities as health professionals. A post-training survey found that participants no longer believed that beating a woman was justified and that most accepted the concept of marital rape (252).

5. Redefine success. Health workers often feel reluctant to address cases of domestic violence because it is a problem that cannot easily be cured or even addressed. In response, some training projects have tried to help the provider reframe their role from “fixing” the problem and dispensing advice to providing support. Revising expectations in this way has helped providers overcome feelings of resentment and impotence in addressing domestic violence (374).

Reframing the provider's role also helps promote women's self-determination. Counseling concerning abuse, like contraceptive counseling, should be nondirective and respect women's choices. As one advocate put it, “We are trying to work through the frustration of providers who don't understand that it takes time for a battered woman to take action. When we ask a woman to make a decision within 10 minutes, we are saying, `We know what's good for you.' This is no different from the batterer who makes all the decisions for her” (452).

6. Provide opportunities to model new behavior. Two major barriers to asking clients about abuse are providers' belief that violence is uncommon among their clients and providers' fear of how the clients will respond (151, 428). Opportunities to practice new behavior can help overcome both barriers. In working with medical students, for example, Pakistani physician Fariyal Fikree often issues a challenge: “Go out and ask your next five clinic patients a simple screening question for abuse. With this direct experience base, you will be in a better position to evaluate the utility of this practice.”

This exercise breaks down the student's resistance, replaces assumptions with experience, and stimulates their interest in learning more about family violence. Generally, students come back from the experience amazed at how many women disclosed abuse and how willing women were to discuss such matters (151).

7. Be strategic about where you start. Changing health systems is difficult. Thus the best practice is usually to start where success is most likely. Often this strategy means choosing to undertake pilot interventions first in settings where there is substantial internal and external support for change.

Internally, it is important to gain the commitment and support of top managers early. Efforts to integrate concern for sexuality into family planning programs have shown that institutional support is absolutely essential to program success (24, 398).

Externally, it is best to undertake pilot interventions where support and referral services for abuse victims already exist. This will not be possible in all instances, but, given that there are so few pilot initiatives yet in resource-poor settings, it makes sense to begin where there are community resources to draw upon.

8. Plan for staff turnover. In most health systems, particularly in developing countries, staff members routinely rotate in and out of clinics and other health centers. Thus policies on violence must be institutionalized, and training will be needed for new staff members on a continuing basis (329).

9. Follow up. Programs should provide continuing support to individuals and institutions attempting to reform their response to domestic violence. Projects that have attempted to spark change by using a “train the trainer model”—inviting providers to attend a centralized training and then expecting them to duplicate the training in their home setting—have generally found that such schemes do not work well without substantial continuity and support (61).

High-Priority First Steps Reproductive health professionals often feel that the issue of violence against women is too complex and too overwhelming to tackle. But fundamental change can—and often must—begin incrementally. A graduated response to violence could begin with the following steps: Priorities for Donors Research into vaginal microbicides. Changing the power balance between women and men in sexual relationships will take time—time that women at risk of HIV and other sexually transmitted infections today do not have. Thus a high-priority investment by donors must be research into vaginal microbicides—substances, similar to today's spermicides, that women could use to protect themselves from infection—if necessary, without the knowledge or cooperation of their sex partners. Scientists predict that a first-generation microbicide could be developed within 5 years given sufficient investment. Presently, research in this area is inadequate. Women's and AIDS groups have organized the Global Campaign for HIV/STI Prevention Alternatives for Women to demand more investment in microbicide development (495). Pilot projects. More must be learned about how to integrate concern for gender-based abuse into other reproductive health programs. Immediate support is needed for pilot projects with strong evaluation components to discover what works best in different settings, particularly where resources are few. Priorities for Program Planners Integration into ongoing training. The most effective way to improve training about abuse for reproductive health care providers is to integrate it into current training, especially when training addresses quality of care, counseling, and male involvement. At a minimum all training for providers can add sensitization exercises about gender, sexuality, and abuse. Make new norms a program objective. Measurable indicators of reproductive health program success can include, for example, changes in the percentage of women and men who agree that a married woman has a right to refuse sex. The DHS now include such questions. With new norms as a program objective, managers will focus attention on how best to encourage changes in public attitudes about women's autonomy and men's behavior. Priorities for Providers Dicuss with women clients how much they can control sexual encounters. This is a crucial consideration in choice of a family planning method. Providers can point to methods that a woman can use without her partner's knowledge or if she cannot anticipate sex. Also, providers can emphasize that sex—including sex within marriage—should be wanted by both parties, not forced by the man. Do not require spousal approval for contraceptive use. Many providers require a woman to have her husband's consent to obtain contraception, even when policies do not (499). Such requirements undermine women's autonomy and put them at risk of violence. In Ethiopia, when family planning clinics stopped requiring spousal consent, women's attendance soon rose 26% (491).