Strengthening Health System Responses to Gender-based Violence in Eastern Europe and Central Asia

A resource package

4.1. The definition of a referral system

Women who have experienced GBV have multiple and complex needs. This includes medical care, safe accommodation, psychosocial counselling, police protection and/or legal advice. Therefore, an effective response to GBV requires a comprehensive set of services. Since it is virtually impossible for a single organization to provide all services in the required quality and specialization, a multi-sectoral response that coordinates the services by all relevant service providers helps to ensure the availability of comprehensive support for survivors of GBV. Referrals are an important step in case management as part of multi-sectoral work, besides case review, monitoring and after-case/follow-up support. An important prerequisite for the design and implementation of effective referrals is the existence of an institutionalized referral mechanism.

Referrals in general describe the processes of how a woman gets in touch with an individual professional or institution about her case and how professionals and institutions communicate and work together to provide her with comprehensive support. Partners in a referral network usually include different government departments, women’s organizations, community organizations, medical institutions and others. There is also the possibility of self-referral, where a woman approaches an agency herself, which is low in case of restricted mobility. As a principle of good clinical practice, referrals should happen with the consent of the woman concerned. However, in some cases, it may be justified that referrals by a family member or an agency occur without the woman’s consent, in cases where her life is at risk, like high risk of suicide, threat of being killed or child marriage.

A referral system can be defined as a comprehensive institutional framework that connects various entities with well-defined and delineated (albeit in some cases overlapping) mandates, responsibilities and powers into a network of cooperation, with the overall aim of ensuring the protection and assistance of survivors, to aid in their full recovery and empowerment, the prevention of GBV and the prosecution of perpetrators (the so-called 3 p’s). Referral mechanisms work on the basis of efficient lines of communication and establish clearly outlined referral pathways and procedures, with clear and simple sequential steps (UNFPA 2010).

Referral systems should involve governmental, non-governmental and, as appropriate, relevant international organizations (see section 4.2 for an overview of relevant services and providing organizations). Their work, at the levels of both the multi-sectoral structure and individual agencies, should be normatively based on international human rights principles and standards (see sections 2.1 and 2.5). In order to ensure that cooperation among stakeholders is grounded in sustainable structures, rather than relying on the contributions of committed individuals, the operation of referral mechanisms should be grounded in legislation or standardized protocols that define the roles and responsibilities of all organizations involved (see also section 2.3.2).

Referral mechanisms may operate at different levels - national, regional and/or municipal/community. Local and/or regional referral mechanisms should be grounded in national laws and/or policies. See box 23 for an example from Kyrgyzstan, illustrating a step-by-step-approach for setting up a referral mechanism for cases of domestic violence at the local level.

Box 23: Establishing a referral mechanism to prevent and respond to domestic violence at the local level, Kyrgyzstan

Since 2010, UNFPA has been partnering with NGOs, the government of Kyrgyzstan to set up a system of multi-sectoral cooperation for the prevention of and response to GBV at the municipal level. At the outset, the following steps were undertaken:

  •  A Coordinating Council was set up by the mayors’ office in capital city Bishkek; it approved local level work plans for the prevention of and response to domestic violence.
  • Models for inter-agency cooperation were developed and tested in specified pilot districts/sub-districts in Bishkek. The pilot was coordinated by the municipal administration in the respective pilot district/sub-district.
  • Sector-specific action plans and departmental instructions for health care and the municipality administration (including police, health, education and local self-governance including social services) on how to work with survivors of domestic violence were adopted.
  • All involved professionals were trained on understanding GBV, relevant local and international laws, and how to effectively respond to GBV. Training of health care providers targeted both, doctors and nurses.
  • Standard forms for reporting and tracking domestic violence for use by health care institutions and municipal administration were adopted.
  • Quarterly coordination meetings of all sectors involved have been organized and facilitated by the deputy mayor and UNFPA. These meetings served to register identified cases of women and children survivors of domestic violence; further, service providers discussed and agreed on further support measures.
  • A number of public awareness raising activities were held. For example, health facilities organized information stands to inform patients and health care professionals on available services for survivors and contact information of organizations providing support. Education departments organized a series of activities in schools among students and children, those who have experienced domestic violence were provided with psychosocial support.
  • Local self-governance bodies worked on identification and prevention of domestic violence in their respective communities.

In 2012, this model cooperation framework developed at the capital city level was then translated into action at the sub-national level, through a pilot in the cities of Osh and Jalal-Abad:

  •  Participating organizations formalized their cooperation based on a Memorandum of Understanding (MoU) which also includes implementation of GBV Standard Operating Procedures at the national level. The MoU was initiated under the IASC GBV sub-cluster during the interethnic conflict that occurred in June 2010, in order to set up a national network to respond to and prevent GBV in the aftermath of the conflict. The MoU was signed by the Ministries for Labour, Employment and Migration; Social Protection; Internal Affairs; Justice; and Health, as well as UNICEF, UNFPA and UN Women.
  • In both cities, the mayor’s offices set up Multi-sectoral Coordinating Councils; they approved local level work plans for the prevention of and response to domestic violence. Members include the deputy mayors, representatives of women crisis centres, women’s NGOs, NGOs providing legal support, health departments, local self-governance bodies, and municipal law enforcement bodies.
  • Municipal workers and service providers underwent training on the implementation of the MoU and the SOPs, based on the 2003 Law on Social and Legal Protection against Domestic Violence.

These efforts have resulted in identification and referrals of survivors of GBV, as follows:

  • In 2012, 500 cases of domestic violence were registered at the Emergency Medical Care Center (Osh), which included 329 women and 171 men.
  • 117 complaints of instances of physical domestic violence were registered in 2012 at the trauma unit of the City Hospital (Osh) and received medical support.
  • From August 2012 to March 2013, three women survivors of domestic violence sought help at the emergency room of Family Medical Center №1 (Osh). Two of them were referred to the Police Department of Osh.

While some progress has been made, more work still needs to be done to ensure that GBV survivors have access to comprehensive support. It is necessary to continue strengthening the capacity of local providers of health, psychosocial, legal and protection services, with a view to setting up a sustained and well-functioning referral pathway. For the health sector, emphasis should be put on sensitizing doctors to address GBV as a public health issue, rather than a family/private matter, in order to improve the identification of survivors. Further, improvements in the infrastructure of health facilities are urgently needed. Currently, many health facilities lack separate rooms for psychosocial consultations provided to patients who experienced GBV. This prevents doctors from asking confidential questions, and survivors from disclosing GBV due to shame and fear. Furthermore, sector-specific instructions on the response to GBV should be institutionalized through capacity building of ministerial workers on gender equality and GBV.

Lessons learned:

  • Political will on the part of central-level authorities that ensure accountability of local governments are important prerequisites to ensure that laws, policies and action plans documents are actually implemented and that appropriate financial and human resources are in place at the local level.
  • Political and financial support from the municipal administration not only helped to secure funds needed for the implementation of action plans but also serve to create a sense of ownership among those responsible for implementation.
  • Coordinating bodies need to meet regularly, in order to ensure ongoing implementation and monitoring of action plans and to enable them to address emerging issues in a timely manner. Ensuring broad-based membership of governmental, non-governmental and international organizations in coordination structures is desirable.

The added value of setting up a multi-sectoral cooperation mechanism and getting to know each other’s counter parts in the partner organizations is illustrated in the following statement from a district police inspector in Osh: “After we started working together, I realized that we are all doing the same job. Not only us, but others also work on resolving domestic violence. Now that we know each other, it’s easier. You can’t just send someone off to another agency without that.”

Source: UNFPA 2013, information provided by UNFPA Country Office Kyrgyzstan, June 2014

Referral systems benefit both the patient who experienced GBV and the health care provider. Guiding a survivor through a referral system enables her to access comprehensive and specialized care and support, tailored to her individual needs. From the perspective of health care professionals, the establishment of clear and simple referral routes

  • can offer relief to their daily work load, as they can count on support provided by other referral partner agencies;
  • can increase the confidence of health care professionals to ask about violence, as shown in following quote from a general practitioner : “[I]t has really made a huge difference being able to [refer patients to a specialized domestic violence support service]. We are all much more willing to ask the question which might open a can of worms knowing that there’s help if we do – and our patients have benefitted enormously” (Johnson 2010); and
  • enables them to adequately act upon the identification of a survivor of GBV, keeping in mind that the existence of a system of referring survivors to further services constitutes a minimum requirement for health care professionals inquiring about GBV (see section 3.1.3); “[i]ndeed, what is the point of asking a woman whether she has been subjected to gender-based violence if no appropriate help is available?" (UNFPA 2010).

In practice, health care professionals, confronted with the reality of limited time and resources in busy clinical settings, might be inclined to perceive the referral process for patients who experienced GBV as complicated and time-consuming. In this situation, it may help to approach GBV like any other health symptom that they would enquire, diagnose and refer for specialist treatment, as they do in their regular clinical work.

Effective referrals require that health care professionals:

  • Are able to recognize and facilitate the disclosure of GBV (see section 3.1) and provide first-line support (see section 3.2.1).
  • Are able to assess the individual situation and needs of the patient, in particular the risk of further or escalating violence (for more information on risk assessment, see section 3.4), and tailor the type of referral accordingly. If the assessed risk is high, the survivor requires immediate crisis intervention, such as immediate medical or psychological support and/or access to a shelter. If the assessed risk is not high, referrals to other social, psychological or legal support might be appropriate.
  • Are knowledgeable about the existing referral system and services and support the patient in identifying the best options. To this end, health care professionals should have at hand the contact details from relevant service providing institutions, in particular shelters and police. A useful tool is a referral directory (see annexes 4.3.2 and Annex 7), which should also specify any protocols on acceptance to shelters. In some countries, shelters accept survivors only upon referral via specific state bodies only. Seemingly small details may matter a lot in actual referral. Health care professionals also should keep in mind that survivors might have limited opportunities for visiting several locations, due to lack of money, time and freedom to travel. Therefore, they should try to offer the most efficient route and give very clear directions, so as to minimize the number of contacts and enable the patient to receive as much care and support as possible on the first contact.
  • Are knowledgeable about national lawson GBV, including definitions of relevant criminal offences, about available protection measures and any reporting obligations on their part. For more information in mandatory reporting, refer to section 2.5.2.
  • Obtain the consent of the survivor before sharing informationabout her case with other agencies or service providers and follow the procedure that protects the woman’s confidentiality.

At the level of health facilities, effective referrals benefit from:

  • Ongoing capacitybuildingof health care staff to ensure effective referrals and service delivery. Multi-sectoral trainings can serve as a first step towards establishing a working partnership as well as to develop and maintain capacities to deliver services and provide effective referrals.
  • A coordinated monitoring mechanism, such as a joint database for monitoring the system of response and improving the quality of services provided to women. The analysis of the data will allow to track the use of different services by individual survivors, to identify women who need support over a long period of time and those who drop out and to obtain critical information to improve management and planning of services. Particular emphasis should be put on collecting feedback from survivors who use the services provided and on ensuring safety and confidentiality of personal data of the survivor. For more information on setting up a monitoring and evaluation system, please refer to chapter 5 and IPPF 2010.

Box 24: Capacity building improves the effectiveness of referrals – an example from UK (IRIS intervention)

As demonstrated in the case of the IRIS (Identification & Referral to Improve Safety) intervention, training of health care professionals in general practices combined with the provision of technical support to practice teams and the establishment of a simple referral pathwayto a specialist domestic violence support organization led to improvements in the clinical response to domestic violence. The intervention was implemented from 2007-2010 and targeted doctors, nurses and reception staff in 24 general practices in London and Bristol, UK. The evaluation compared the performance of this group of health care professionals (intervention practices) with 24 control practices who did not participate in the training and support programme. Following the intervention, the number of referrals recorded in intervention practices was 21 times larger than that recorded in control practices. Further, intervention practices recorded 3 times more identifications than control practices (Feder et al 2011).