Lancet series - The health-systems response to violence against women

The health-systems response to violence against women

Dr Claudia García-Moreno, MD

,

Prof Kelsey Hegarty, PhD

,

Ana Flavia Lucas d'Oliveira, PhD

,

Jane Koziol-McLain, PhD

,

Manuela Colombini, PhD

,

Prof Gene Feder, MD

Published Online: 20 November 2014

DOI: http://dx.doi.org/10.1016/S0140-6736(14)61837-7

Article Info

• Summary
• Full Text
• Tables and Figures
• References
• Supplementary Material

Summary

Health systems have a crucial role in a multisector response to violence against women. Some countries have guidelines or protocols articulating this role and health-care workers are trained in some settings, but generally system development and implementation have been slow to progress. Substantial system and behavioural barriers exist, especially in low-income and middle-income countries. Violence against women was identified as a health priority in 2013 guidelines published by WHO and the 67th World Health Assembly resolution on strengthening the role of the health system in addressing violence, particularly against women and girls. In this Series paper, we review the evidence for clinical interventions and discuss components of a comprehensive health-system approach that helps health-care providers to identify and support women subjected to intimate partner or sexual violence. Five country case studies show the diversity of contexts and pathways for development of a health system response to violence against women. Although additional research is needed, strengthening of health systems can enable providers to address violence against women, including protocols, capacity building, effective coordination between agencies, and referral networks.

This is the second in a Series of five papers about violence against women and girls

Introduction

Violence against women is a global public health and clinical problem of epidemic proportions.¹ It is also a gross violation of women's human rights. Violence affects the health and wellbeing of women and their children, with vast social and economic costs.^{2, 3, 4} Its adverse physical, mental, and sexual and reproductive health outcomes^{5, 6} lead women who are abused to make extensive use of health-care resources.^{4, 7} Health-care providers frequently, and often unknowingly, encounter women affected by violence.

The health-care system can provide women with a safe environment where they can confidentially disclose experiences of violence and receive a supportive response. Furthermore, women subjected to intimate partner violence identify health-care providers as the professionals that they trust with disclosure of abuse.⁸ However, the crucial part that health-care providers and services can play to address violence against women is often not recognised or implemented. Health systems need to strengthen the role of providers as part of a multisectoral response to violence against women.⁹

This Series paper is based on evidence on the health-care response to violence against women, experience of the implementation of services to address violence against women in diverse countries, and consultations with those involved in the planning or delivery of services in resource-poor settings. We describe the challenges involved in engagement of the health sector and make recommendations to integrate effective care for women experiencing violence.

Rationale for a health-care response

As noted in the 2013 WHO report, Global and regional estimates of violence against women,⁶ one in three women worldwide who have ever had a partner report physical or sexual violence, or both, by an intimate partner. This violence contributes to the burden of women's ill health in many ways.^{5, 6} Women with a history of intimate partner violence are more likely to seek health care than are non-abused women.^{4, 10, 11} For example, Bonomi and colleagues⁴ showed that women who were physically abused used more mental health, emergency department, hospital outpatient, primary care, pharmacy, and specialty services.

Key messages

• •
  The health-care system has a key part to play in a multisectoral response to violence against women; that role, however, remains unfulfilled in many settings.
• •
  Violence against women needs to have higher priority in health policies, budget allocations, and in training and capacity building of health-care providers.
• •
  Although evidence of effective interventions in health-care services remains scarce, especially for resource-poor settings, there is a global consensus that health-care professionals should know how to identify patients experiencing intimate partner violence and provide first-line supportive care that includes empathetic listening, ongoing psychosocial support, and referral to other services, as well as comprehensive post-rape care for sexual assault victims.
• •
  The health system needs to ensure the enabling conditions for providers to address violence against women, including good coordination and referral networks, protocols, and capacity building.
• •
  No model of delivery of health-care response to violence against women is applicable to all settings, and countries should develop services that take into account resources and the availability of specialised violence-support services.
• •
  Violence against women should be integrated into medical, nursing, public health, and other relevant curricula, and in-service training should ensure that health-care providers know how to respond appropriately and effectively; this training needs to be sustained and supported by ongoing supervision and mentorship.
• •
  Health policy makers should show leadership and raise awareness of the health burden of violence against women and girls and the importance of prevention among health-care providers, managers, and the general public.
• •
  More research is needed to be able to quantify the health burden associated with different forms of violence, and to assess and scale up interventions to prevent, and respond to, violence against women.

Women also experience other forms of violence, including rape and other sexual violence at the hands of acquaintances, friends, and strangers; physical and sexual violence from relatives; trafficking; female genital mutilation; early and forced marriage; and murders in the name of so-called honour.¹²

All of these forms of violence can bring women into contact with the health-care system, which must be prepared to respond. This Series paper focuses on intimate partner and sexual violence because they are the most common worldwide and have most evidence for effective interventions.

Data sources

In addition to the literature search (search strategy and selection criteria panel),^{13, 14, 15, 16, 17} this paper is based on consensus in meetings of experts for the development and implementation of the WHO clinical and policy guidelines,^{18, 19} and included lessons learned from different countries in building a health system response. Five case studies (appendix) show different challenges, policies, and processes, although not based on formal assessment (table).

Table 

Summary of country case studies

What can health systems do?

The main role of health-care systems for women, and their children, facing the health effects of violence is to provide supportive care. This supportive care can contribute to prevention of violence recurrence and mitigation of the consequences, address associated problems, such as substance misuse and depression, and provide immediate and ongoing care. The health system also has a part to play in primary prevention (ie, prevention of violence occurring before it starts), through documenting violence against women, emphasising its health burden, and advocating coordinated action with other sectors (figure 1).

Figure 1

The role of the health system to address violence against women

View Large Image | View Hi-Res Image | Download PowerPoint Slide

Implementation of health-care policies and training programmes for providers to address violence against women face individual and system barriers.^{20, 21, 22, 23} Evidence suggests that information dissemination or training in isolation do not create consistent, sustainable change,^{23, 24, 25} and that a comprehensive systems approach is needed.^{23, 24, 25, 26, 27, 28, 29}

Figure 2³⁰ provides an overview of the necessary elements at the level of the providers and services, and of the health system more broadly, organised by core components (or building blocks): service delivery, health workforce, health information, infrastructure and access to essential medicines, financing, and leadership and governance.³¹

Figure 2

Elements of the health system and health-care response necessary to address violence against women
Adapted from Colombini and colleagues,³⁰ by permission of BioMed Central. SA=sexual assault.

View Large Image | View Hi-Res Image | Download PowerPoint Slide

Many countries have begun to address violence against women in health care with varying success, as shown by the case studies (appendix). The case studies also show that progress in the integration of violence against women into health systems is slow and incremental. In many countries, social and cultural barriers need to be overcome (eg, Lebanon [appendix]), and in most countries, health system barriers such as high staff turnover and limited resources must be addressed (eg, India and South Africa [appendix]). Traditional biomedical approaches are inadequate and inappropriate to address violence against women,³² so changes will be needed (eg, India [appendix] and Spain [panel]).^{33, 34, 35, 36, 37, 38, 39}

PanelSexual and domestic violence against women in the Spanish health-care system

Country context

The health-care system's commitment to address violence against women has been a central element in Spain's multisectoral response to sexual and domestic violence. A strong legal and normative framework are provided by the Organic Act 1/2004 of 28 December on Integrated Protection Measures against Gender-Based Violence³³ that was passed unanimously by the Spanish Parliament, the creation of the State Observatory for Violence against Women to monitor the magnitude of the problem and progress, and the establishment of a national gender-based violence awareness and prevention plan (2006).³⁴ This plan was developed by a multisectoral group involving government, civil society organisations, and other experts. The plan covers primary, secondary, and tertiary prevention, and includes objectives for the judicial system, security forces, health services, social services, information systems, education system, and the media. Cross-cutting areas include research, training and funding measures, mobilisation of actors, coordination, follow-up, and evaluation. By law, all regional governments must include service provision for gender-based violence in the Regional Health Service.³⁵

Health system context

Spain has taken a systematic and standardised approach to the implementation of a health-care system response to violence against women. The incorporation of care for violence against women in the 2006 National Health System portfolio was a key driver. Ministry and National Health System leaders created a commission against gender-based violence to provide technical support, coordinate actions, and assess healthcare performance across the National Health System regions (autonomous communities).³⁵ This commission operates through the Observatory on Women's Health and includes several working groups:

• •
  Epidemiologic Surveillance Group, to reach consensus on indicators and standardised records design.
• •
  Healthcare Aid Protocols Group, to develop a common protocol.
• •
  Ethical and Legal Aspects Group, to address confidentiality and safety.
• •
  Healthcare Professionals' Training Group, to develop educational objectives, and training content, materials, and quality criteria.
• •
  Performance Evaluation Group, to develop information systems, implementation of protocols, processes, training, and coordination and continuity of care in addition to accreditation and dissemination of good practices.

Progress

Health-care protocols

A Common Protocol for the Healthcare Response to Gender-Based Violence was published in 2007,³⁶ establishing standardised performance guidelines for the National Health System. Each region adapts the Common Protocol to its own context and offers information on local resources. The Common Protocol was updated in 2012³⁷ to include recommendations for the treatment of children exposed to domestic violence and for other people at risk such as disabled women, immigrants, pregnant women, and mentally disabled people.

Health professional training

A training of trainers strategy has been implemented with resource materials and quality control criteria.^{38, 39} The National Healthcare School and Women's Institute provide support for the training of trainers, including core and advanced training. Many health-care professionals have undergone training, with priority given to primary care providers, but also hospital professionals, emergency care service providers, midwives, and mental health professionals. The training duration and content differ according to their roles. Regions have developed their own training plans, integrated in programmes of continuous education, and delivered at workplaces through regional teams of trainers. Funding is provided by the regional health services and Ministry of Health.

Knowledge sharing

Good practices are identified, collected, and disseminated to share across regions.

Challenges

• •
  There is a need to sustain and reinforce basic training, awareness, and competence to manage victims, and training should be extended to hospital professionals; support and supervision by experts are also needed after the training.
• •
  Intersectoral coordination to establish clear referral pathways to relevant services should be further improved, especially for women who have been sexually assaulted.
• •
  Continued work is needed to improve information systems and data management, including incorporation of gender-based violence in electronic clinical reporting and protection of confidentiality.
• •
  Research and assessment are needed to show the outcomes of interventions for women and their children, including changes in women's status, their health and wellbeing, and use of resources.

Lessons learned

• •
  Brief workshops and clinical case sessions are highly valued.
• •
  Raising of awareness and training of professionals increases detection and improves the health-care response.
• •
  Inter-institutional and intersectoral coordination improves case management.
• •
  Women and victims' associations should be taken into account when processes to help victims are developed.
• •
  Institutional leadership helps with the implementation of measures.
• •
  Structural conditions should be improved, since excessive caseloads and too little time are obstacles for identification and care in some settings, in addition to the need for financing of training and support resources.
• •
  Interventions should include mechanisms for communication of the evidence generated and the best practices to health-care professionals.

What can health providers do?

Overview

The appropriate response by health-care providers will vary depending on the women's level of recognition or acknowledgment of the violence, the type of violence, and the entry point or level of care where the survivor is identified. Actions by health-care providers include identification, initial supportive response to disclosure or identification, and provision of clinical care, follow-up, referral, and support for women experiencing intimate partner violence, in addition to comprehensive post-rape care and support for victims of sexual assault.

Different women will have different needs, and the same woman will have different needs over time. She might present with an injury to the accident and emergency department, with depression or functional symptoms in primary health care, with an unwanted pregnancy or for a termination of pregnancy in sexual and reproductive health-care services, or with various physical problems to an outpatient department in a secondary or tertiary hospital. In addition to provision of clinical care for the condition presented, identification of violence as the underlying problem is important.

Identification of intimate partner violence

Identification of women and girls who are, or have been, subjected to violence is a prerequisite for appropriate treatment and care, and referral to specialised services where these exist.

Identification in health-care settings could be increased if all women were asked about intimate partner violence; however this is only effective (and safe) if followed by an appropriate response. Disclosure is low relative to best estimates of prevalence of partner violence¹⁷ and some studies have reported that, despite training for universal screening, most providers ask selectively.^{23, 40} WHO does not recommend universal screening,¹³ rather it recommends that health-care providers should be trained in how to respond and be aware of the mental and physical health indicators associated with violence, and ask about violence when they are present.¹³ Insufficient evidence exists for a universal screening policy, with three randomised clinical trials that directly tested screening programmes reporting no evidence of reductions in violence or improvement in health outcomes.¹⁷ Moreover, in settings or countries where prevalence of present violence is high and referral options are scarce, universal enquiry might bring little benefit to women and overwhelm health-care providers. A systematic review⁴¹ of studies in high-income countries reported that most women (whether or not they have experienced intimate partner violence) find routine questions about abuse acceptable. However, a systematic review⁴¹ of health-care professionals noted that they are less willing to undertake screening or routine enquiry than women are to be questioned.

Disclosure of violence is more likely if women are asked in a compassionate and non-judgmental manner, in private, and in an environment where the person feels safe and confidentiality can be protected.^{8, 11} Clinicians can be trained on when and how to ask, and how to provide a first-line response consisting of empathetic listening, validation of the patient's experience, and support, consistent with what women have been reported to want.^{8, 13}

However, intimate partner violence is a very stigmatised problem and women and girls often have realistic fears for their safety if they disclose the violence, so specific conditions must be met. These conditions include that women can be asked safely, that the abusive partner is not present, that providers are regularly trained in how to ask and respond, and that protocols, standard operating procedures, and a referral system are in place.¹³

Initial response to intimate partner violence

So far, research has not addressed the effectiveness of the initial response to disclosure or identification. However, a meta-analysis of qualitative studies suggests that women want health-care providers to provide first-line support: attentive listening, sensitive non-judgmental enquiry about their needs, validation of women's disclosure without pressure, enhancement of safety for the woman and her children, and provision of support and help to access resources (eg, India [appendix]).⁸ WHO guidelines recommend that all health-care providers be trained in women-centred first-line support, to respect a woman's right to decide on her own pathway to safety.¹³ This approach is consistent with so-called psychological first aid, a first response to individuals undergoing traumatic events.^{42, 43} A supportive response from a well trained provider can act as a turning point on the pathway to safety and healing.⁴⁴

Ongoing response to intimate partner violence

Women need different responses at various points in the course of violence and relationships.^{44, 45} A prerequisite for a woman to accept help is her awareness or recognition that what she is experiencing is abuse (figure 3).^{44, 45, 46} Health-care providers can help women to name what is happening to them as abuse through inquiry and validation of their experiences. They can help to empower women to make even small changes that might improve their self-efficacy.⁴⁶ Furthermore, health-care providers can provide ongoing support and potentially empower women to take action to safely improve their lives (figure 3).

Figure 3

Women's (non-linear) trajectory to safety: health professional's response to women's readiness for action
Women's pathway to safety is not linear and health professionals need to respond at different time points to where a woman is currently at, in terms of her readiness to take action.

View Large Image | View Hi-Res Image | Download PowerPoint Slide

Beyond first-line support, other health-care interventions are supported by evidence, such as advocacy by health-care providers with additional training or by specialist partner violence advocates,^{15, 16, 47, 48} safety planning,^{47, 49} motivational interviewing,⁵⁰ and cognitive behaviour techniques and other trauma-informed mental health interventions.^{16, 51, 52, 53}

Advocacy interventions aim to help abused women directly by providing them with information and support to help them to access community resources. These interventions usually link survivors with legal, police, housing and financial services, and many also include psychological or psychoeducational support. Trials of advocacy or support interventions for women facing intimate partner violence in high-income countries report some reduction in violence and possible improvement in mental health outcomes.^{47, 48} The health-care provider might continue to offer ongoing support, but the patient also benefits from the expertise of a domestic violence advocate or support worker.^{23, 40} Little evidence⁴⁹ exists for safety planning that is delivered face to face by health practitioners or by telephone counsellors. Various counselling approaches, such as motivational interviewing and empowerment counselling strategies, provide support and can help women to discuss safety and reduce depressive symptoms.^{46, 47, 50}

Referral for intimate partner violence

Linking of health-care providers with specialist support or advocacy services increases the likelihood of the providers asking about, and identifying, patients with a history of violence.^{23, 54} Furthermore, trauma-informed cognitive behavioural therapy has been shown to work for women who have post-traumatic stress disorder and who are no longer experiencing violence.¹³ Evidence suggests that children who have been exposed to intimate partner abuse are likely to benefit from referral for psychotherapeutic interventions,⁵⁵ but more research is needed to develop effective interventions for these children.

Consensus evidence suggests that health services need to work closely with specialist services, including the police, to enhance safety for women and children.²⁷

Clinical care for sexual assault

Comprehensive post-rape care includes a set of clinical interventions to prevent pregnancy and possible infection with HIV or other sexually transmitted infections for those who seek care after an assault. This care is time sensitive so should be available in all secondary and tertiary care facilities and from primary health-care providers.^{13, 56, 57} Collection of forensic evidence when relevant, trauma-informed mental health care, and access to safe abortion are important services for survivors of sexual violence. Long-term follow-up for mental health problems might also be needed for some women.⁵⁸

Many survivors of sexual violence, however, face challenges to access essential medicines and post-rape services at health facilities.⁵⁹ These challenges can be because of scarcity of resources at health-care centres, fear of stigma, or further episodes of violence from the perpetrator. In rural areas, the distance to health-care centres and absence of adequately skilled staff are also barriers.⁶⁰ The stigmatising and discriminatory attitudes and practices of health providers themselves can be another barrier. Some countries have developed protocols and guidelines to improve access to post-rape care services,^{61, 62} but scarcity of training and equipment, poor coordination of services, and associated so-called out-of-pocket costs can make access to these services a challenge.⁶³ Access to a trained provider, coordination between services, including the police, and awareness about the importance of women seeking care immediately after the incident will increase access to, and use of, care.⁶⁴ What can health systems do in primary prevention?The health system can raise awareness about the need to address violence against women by reporting and publicising data for the prevalence, health burden, and costs of violence, and contribute to efforts to counter the acceptability of such violence. However, evidence to guide health-care organisations in primary prevention activities is scarce.⁶⁵ Although most primary prevention involves actions outside of the health sector (as explained by Lori Michau and colleagues in this Series⁶⁶), the health system can contribute to prevention of child maltreatment, for example through home visits and parenting programmes,⁶⁷ and actions to reduce, and provide treatment for, alcohol and substance use problems.⁶⁸ Antenatal classes for fathers to improve relationships and prevent violence are being used in Hong Kong, but need further assessment.⁶⁹

Health sector interventions with children (and their mothers) who witness domestic violence or who are abused^{70, 71} can potentially contribute to primary prevention because of the association between exposure to domestic violence and an increased risk of perpetration or experience of partner violence in adolescence and adulthood.^{72, 73}

How can services best be delivered?

Overview

Different models exist for the delivery of health care to women experiencing violence. Whatever model is used, a functional health system is needed for providers to deliver an effective and safe health-care response. All elements of the health system should adequately address violence against women (figure 2).

Leadership, political will, and governance

Violence against women is absent from many national health policies or budgets, and neither is health care always included in national plans to address violence against women. In some countries, no data exist and the issue is still not recognised. In others, the issue is not seen as a priority because of restricted health budgets and competing priorities. Recognition of the problem is an important first step, which can lead to the establishment of mechanisms to address violence against women, such as interdepartmental task forces or other coordinating bodies, or development of a national health policy and budgetary allocation. Ministries of women's affairs or gender and the women's movement have played an important part to engage the health sector in some settings (eg, Brazil [appendix]). In countries where they exist, there is a crucial role for national organisations that accredit health-care facilities or produce guidance on the commissioning of health services. A visible health-care response will not only encourage disclosure of violence against women to clinicians, but can convey a message to society as a whole that this violence is unacceptable.

Workplace prevention strategies that affect the climate, processes, and policies in a system or organisation⁷⁴ could be implemented in health workplaces, such as respectful relationship training, bystander education,⁶⁵ and displaying of posters that convey the unacceptability of violence against women.⁷⁵ The health-care system should prevent violence against women in the health workplace by putting policies (eg, on sexual harassment) in place, and training health-care workers on, and promotion of, respectful relationships in the workplace and with patients. Health-care organisations are large employers, especially of women (who can experience violence themselves), so personnel policies should also take this into account (eg, domestic violence leave). Further strategies that need testing include appointment of health centre champions, who will assist with improvements to the workplace climate, and peer support to address violence against women.^{23, 40}

Coordination

Women who experience violence can also have safety, social support, economic security, housing, and legal protection needs, so a multisectoral response is necessary. Irrespective of the point of entry, coordination within the health-care system and between the health system and other sectors is fundamental to provision of a holistic, seamless service.⁶⁵ In practice, however, many differences in language, goals, and institutional cultures need to be overcome.⁷⁶ Some countries, such as Philippines, Malaysia, and Malawi have developed specific guidelines to support a multisectoral response.^{77, 78, 79}

Examples also come from high-income countries where the health system has taken a lead role in a multisectoral response to domestic violence. Bacchus and colleagues³⁵ reviewed intervention models based on health care in seven European countries, and drew out key lessons for successful implementation, including committed leadership and organic growth from the bottom up, regular training of health-care professionals with feedback mechanisms, mandatory or motivated training attendance, creation of a pool of trainers for sustainability, and development of clear referral pathways between health care and the specialist domestic violence sector, to ensure input from survivors and document the process.

Involvement of women's organisations and the community can raise awareness about violence and services available, and promote more respectful and equitable attitudes towards women and against violence. Where women's organisations exist, they are often a valuable resource for health systems.

Human resources and capacity building

Many low-income and middle-income countries struggle with scarcity of sufficiently qualified health-care providers, high staff turnover, and overstretched clinicians. This resource shortage is a barrier to designated staff taking on additional roles and implementation of services with specialist gender-based violence providers.⁷⁹

Training of health-care providers is central for any strategy to address violence against women in the health-care system. All staff working in health-care services need training to ensure an appropriate and safe initial response to women experiencing violence, and to provide acute care for sexual assault patients, although different responsibilities need different levels of training. Some evidence from high-income countries suggests that well trained providers can address this issue adequately and improve outcomes.^{46, 80}

Capacity building needs to include clinical knowledge and skills to respond to intimate partner violence and sexual violence, in addition to attitudes and values related to gender equality and violence against women.¹³ This process cannot, however, be confined to a single training event, because brief educational interventions improve knowledge but do not change behaviour.⁸¹ Ongoing support and reinforcement are needed to develop and maintain the competencies of the staff and be part of their continuing professional development education.

For example, Feder and colleagues⁵⁴ reported on a combined role where a domestic violence advocate provided care to survivors of abuse, but was also central to training and provided continuing support to primary care practices. The case study in India (appendix) describes a non-government organisation providing specialised services within a secondary hospital and training health professionals in the same hospital and other hospitals. This model has been replicated now in several other public hospitals in Mumbai and elsewhere in India.

The epidemiology of, and health-care response to, violence against women need to be integrated into the undergraduate and post-graduate curricula of nurses, doctors, midwives, and public health practitioners.¹³

Health-care delivery

Care for women subjected to violence can be delivered in health centres and clinics, district and regional hospitals, or multi-agency or hospital-based one-stop crisis centres. Colombini and colleagues⁸² have classified these approaches as provider integrated (where one provider delivers all services), facility integrated (where all services are available in one facility), or systems-level integrated (a coherent referral system between facilities). So far, there has been little assessment of these different models or approaches. The WHO clinical and policy guidelines on the health-system response to violence against women summarise the advantages and disadvantages of different models.¹⁴ No one model works in all contexts and the choice will depend on the availability of human resources, funding, and referral services. WHO recommends that, as much as possible, care for women experiencing intimate partner violence and sexual assault should be integrated into primary health-care services.¹⁴

The one-stop crisis centre model is, however, increasingly promoted in low-income and middle-income countries, despite not being well assessed or appropriate for all settings. This model is implemented in varying ways. In Malaysia, a hospital-based model has been used and is perhaps most effective in urban areas, but several challenges to implementation have been identified, such as budgetary and staffing constraints.³⁰ The appendix describes Dilaasa, a one-stop centre in Mumbai, India, based on a partnership between a non-government organisation and a public hospital.

Some countries have developed guidelines and standard operating procedures for providers and health-care systems, specifying the steps to follow in cases of sexual assault or domestic violence. Experience with implementation of these types of guidance or protocols suggests that they can help providers who might not feel comfortable addressing these issues, and provide a framework for actions to be taken, but this has not been formally assessed.⁸³

Protocols and guidelines can support providers by letting them know what actions to take. They should include clear guidance on documentation of violence against women (since this evidence is necessary to pursue legal action), maintaining confidentiality, enhancing safety of the survivor, and sharing of information without consent only when absolutely necessary, consistent with the country's legal framework.

Health-care infrastructure

At a minimum, a private and confidential space for consultation and a safe place for keeping records must be available (more detail in appendix). The necessary drugs (eg, emergency contraception for post-rape care) and other supplies and equipment also need to be available.

Financing

The existence of a specific budget allocation for violence against women services and for training and support of front-line clinicians underpins an effective response and is essential (eg, India [appendix]). This allocation creates capacity within health services, and represents a commitment from policy makers and managers of health-care services to address this important issue.

The existence of an explicit health budget line for the response to violence against women makes the service visible and provides a mechanism to monitor costs over time (more details in appendix). The very act of budgeting for system development and service delivery signals that violence against women services are a normal part of health service delivery and promotes a sustainable funding stream.

To support a health-care response to violence against women, costs will be incurred, such as those associated with possession of the appropriate equipment, supplies, and infrastructure, training of health workers, and provision of care, including specialist care. The existence of dedicated staff (including nurses and counsellors) who are paid by the health facilities in which the services for violence against women are integrated is a crucial step for the long-term sustainability of any interventions, and to increase staff motivation.

Monitoring and assessment

Monitoring and assessment are important to strengthen a health system's response to violence against women. They provide local information for training of health practitioners (eg, feeding back referral data), to monitor progress, help with funding, and, ultimately, contribute to knowledge of what works. Progress can be monitored in terms of budget allocation (which suggests the level of commitment), staff training, proportion of health centres that can provide first-line support, and post-rape care, among others.

Challenges and lessons learned from country implementation

Few countries have developed a comprehensive health-care policy integrated into a multisectoral societal response to violence against women, although some are moving in that direction. For example, Spain's 2004 gender-based violence law led to the development of standard health-care protocols, training of providers, and indicators to monitor progress at the national level for regional adaptation and implementation (panel). In other countries, such as Brazil, India, and South Africa (appendix), sexual violence has been the entry point, in part because post-rape care includes explicit clinical interventions. All three countries have faced challenges as they seek to expand their services to include intimate partner or domestic violence.

The biomedical model that predominates in most health-care settings does not help with the disclosure of domestic violence by women or enable an appropriate response from providers. Violence is often seen as solely a social or criminal-justice problem, and not as a clinical or public health issue.⁸⁴ Linked to this is the failure to understand inequalities, in particular those faced by women, as social determinants of health, and how the health system itself can reproduce (or help to change) some of these inequalities.^{32, 85} Health providers, both male and female, might share the predominant sociocultural norms that sanction male dominance over women and the acceptability of violence—attitudes that reinforce violence against women.^{86, 87, 88} Additionally, although many policy responses to domestic violence acknowledge gender inequality as a root cause of intimate partner violence and sexual assault against women, other forms of discrimination faced by women and girls are often invisible. The overrepresentation of indigenous women and non-white women (in dominant white societies) in violence statistics in many countries is an expression of the intersection of several types of discrimination—eg, by gender, class, caste, race, and (dis)ability—that needs to be addressed in health policies.⁸⁹

Disrespect and abuse of women, especially in reproductive health services or when they are transgressors of social norms, is documented.⁹⁰ Health-care providers should model non-abusive behaviours in their interactions with patients (and other staff and colleagues) by, for example, listening respectfully, validating the patient's experience, and not imposing treatments or solutions. To respond to violence against women, the health-care system must deal with the violence that is perpetrated within health care.

The scarcity of resources available to the health sector worldwide, and especially in poor countries, is a major challenge. However, effective responses to violence against women can occur with available resources through the development of partnerships (eg, India [appendix]), while advocacy continues for additional funding consistent with the magnitude of the health effects of violence against women.

Discussion

Violence against women is a global health problem that needs an integrated health-system response. The evidence base for effective interventions, however, is small and comes largely from a handful of high-income countries. In high-income countries, intimate partner violence and sexual assault services developed separately, and have struggled to integrate. In low-income and middle-income countries, where resources are more scarce, the primary care provider will be confronted with both forms of violence, with a large proportion of sexual violence perpetrated by partners. Clinicians should therefore be equipped to deal with both issues. In some countries, sexual violence might be especially difficult to disclose (eg, Lebanon [appendix]), while in others it seems to be an easier entry point to health services than intimate partner or domestic violence (eg, India, Brazil, and South Africa [appendix]) because of a medicolegal mandate, such as in India, or because it fits a biomedical model.

Women who have experienced violence can access services through different entry points, and one model does not fit all settings or countries.^{13, 82} The services used most frequently by women, such as antenatal care, family planning, gynaecological, and post-abortion services,⁹¹ and children's services offer obvious entry points, as does family medicine where this exists. Emergency services are likely to see women with injuries or who have been raped. HIV counselling and testing services and mental health or psychiatric settings also need to know how to respond.^{92, 93} One limitation of current intervention models, for both women and men, is their typically vertical nature. Intimate partner violence, child abuse, and services to treat alcohol and drug misuse problems are usually delivered in professional silos, despite often involving the same individuals and families.^{93, 94} Medical service models often promote a simple health-care response with inadequate attention to multimorbidity.⁹⁵ Only a few trials that reported an intervention effect for intimate partner violence also integrated interventions for comorbidities.⁹⁶

A second, related, limitation is that present models of health care often do not adequately take into account the context—family and social—in which individuals are located.⁸⁸ These limitations can be overcome through engagement with the community, challenging of gender and other discriminations, and through a patient-centred⁹⁷ approach based on each woman's needs. Changing the wider social context, especially where violence against women is widely accepted, is a crucial element (eg, in Lebanon [appendix]).

Ultimately, a societal response to violence against women needs engagement with perpetrators, including legal sanctions against sexual violence and against intimate partner violence, which is still tolerated in some societies. Evidence from high-income countries suggests that perpetrators of intimate partner violence, including femicide perpetrators, are frequently seen in health-care settings and that, therefore, an opportunity to intervene exists,^{98, 99, 100} especially in mental health, drug and addiction, general practice, and emergency services, in addition to health system employee assistance programmes,¹⁰¹ although no evidence exists yet for the effectiveness of these interventions.

A functional and well financed health system is necessary to both prevent violence against women and to respond to victims and survivors in a consistent, safe, and effective manner to enhance their health and wellbeing.

Conclusions

The health system has a key part to play in a multisectoral response to violence against women. Governments need to develop or strengthen multisectoral national plans of action to address violence against women that include health system actions, budgets, and staffing.

Violence against women needs to receive higher priority in health policies, budgets, and the training of health-care providers and public health officials. To overcome this largely hidden epidemic, health policy makers and programme planners should draw on the growing evidence of effective interventions in high-income countries and experience of programme implementation in low-income and middle-income countries, combined with new research in all settings.

A non-judgmental, compassionate, and equitable response to women experiencing violence, with an emphasis on their safety and wellbeing and that of their children, is needed, in addition to improvement of longer term outcomes. An effective health-system response needs to complement society-wide policies to prevent violence. These society-wide policies need to include adequate allocation of national budgets and senior level commitment. International funders should support the efforts of ministries of health and others to address violence against women.

All clinicians, including primary care, sexual and reproductive health (eg, family planning and post-abortion care), and mental health service providers should be trained pre-service and in-service to, at least, know when and how to ask about violence, what first-line care to provide, and how to refer for additional support. Although recognition of this goal might not be realistic in many settings, colocation in health services of champions or advocates for prevention of violence against women can enhance the care received by women and support health-care providers.

Services should be monitored to assess access, acceptability, and quality of care provided to female survivors of violence. These services should collect information in a safe and confidential way, but also use it to inform policies, monitor services, and improve their response.

Research is needed to identify what works, assess promising practices, and develop new strategies for prevention and responses to violence against women, with a particular focus on low-income and middle-income settings.

An effective health-care response to violence against women can contribute to achievement of the Millennium Development Goals, in particular those on gender equality and reduction of maternal and child mortality and HIV.¹⁰² The post-2015 agenda should include strategies to reduce, eliminate, and respond to violence against women. An inadequate response to violence against women from health-care services has economic and social costs.

The time has come for health systems to play their part in a multisectoral response to violence against women that is consistent with their countries' commitments to promotion of public health and human rights.

Search strategy and selection criteria

We based this Series paper on the systematic review linked to the WHO 2013 guidelines Responding to intimate partner violence and sexual violence against women¹³ and the systematic reviews¹⁴ informing the UK National Institute for Health and Care Excellence (NICE) Domestic violence and abuse 2014 guidelines, and other relevant systematic reviews.^{15, 16, 17} To update the evidence base on interventions for violence against women, we searched PubMed and Google Scholar for relevant trials and systematic reviews from May 1, 2012 (NICE reviews), or Dec 1, 2011 (WHO reviews) to June 30, 2014, with the keywords “intimate partner violence” or “domestic violence” or “gender violence” or “violence against women”, and “healthcare” without language restrictions. We have prioritised systematic reviews and trials in our citations.

This online publication has been corrected. The corrected version first appeared at thelancet.com on February 6, 2015

Contributors

CG-M led the writing of the manuscript with substantive inputs from KH and GF. All authors have reviewed and commented on drafts and all have read and agreed on the final manuscript. We thank the authors of the case studies (AFLd'O [Brazil], Padma Bhate-Deosthali [India], Jinan Usta, [Lebanon], Ruxana Jina [South Africa]; and M Carmen Fernández-Alonso [Spain]).

Declaration of interests

CG-M is a staff member of WHO. The author alone is responsible for the views expressed in this publication and they do not necessarily represent the decisions or policies of WHO. We declare that we have no competing interests.

Supplementary Material

Title	Description	Type	Size
Supplementary appendix		pdf	.2 MB