The purpose of this research was to understand the types, perpetrators, and contexts of GBV experienced by female refugees in multiple settings that should be included in a screening tool to confidentially identify survivors and barriers to effectively respond to GBV in humanitarian settings. Among urban and camp-based refugee populations, findings reveal multiple types of violence, settings and contexts in which violence occurs, and a range of perpetrators. Reports, provided by survivors and service providers alike, included a range of physical, sexual, and psychological violence and concur with other findings of GBV reported by female refugee and IDPs in other studies [2–4, 8, 31, 32]. While other studies have provided evidence of abduction during conflict by armed combatants , also reported here, these findings are among the few to document the experience of abduction and threats of abduction within the host country or across borders from the host country humanitarian setting into the country of origin. These findings highlight the need for ongoing protection and attention to risks, even when refugees are considered to be settled in a secure setting.
Research efforts specific to GBV among refugees have commonly focused on either the experience of GBV in conflict [21, 23, 32–35] or mental health outcomes [35–38]. Here, we investigate GBV that occurs across the continuum of the refugee experience and whether GBV was reported to obtain justice and/or to access GBV services. Evidence is suggestive that even after escaping rape and physical violence perpetrated in conflict settings, women may continue to be subjected to GBV, such as domestic violence and sexual violence perpetrated by intimate partners, neighbors or other trusted individuals in the camp or urban host setting. Similar observations have quantified temporal transitions among East Timor refugees; for example in the post-conflict period, there was a 75% reduction in violence perpetrated by individuals outside of the family while some levels of IPV remained consistent or increased . Our qualitative research suggests that the change in gender and family roles that occur with displacement, such as the husband’s loss of productivity and financial and community status, may escalate violence in an abusive relationship or contribute to the acceptance of domestic violence observed in refugee camp settings. Experiences of violence may vary according to the length of time a participant has resided in the camp or undergone displacement, and response efforts should be able to identify and adapt to these variations. As a result, services need to be comprehensive, for example able to respond to the multiple reproductive and mental health needs of women who experience gang rape as well as capable of providing protection and timely health services to women who disclose violence by their husbands in the camp setting.
Within humanitarian settings, GBV response is predominantly governed by several sets of guidelines, including but not limited to the Inter Agency Standing Committee’s Guidelines on Guidelines for Gender-based Violence Interventions. For GBV, key steps include coordination and mapping of services and making reproductive health supplies available,  providing post-exposure prophylaxis (PEP) in areas with HIV prevalence greater than 1%,  and ensuring availability of counseling services. Training on GBV and human rights are included in guidelines for response. Further, humanitarian workers are trained and required to adhere to policies to prevent gender-based violence, including sexual exploitation and abuse. Monitoring and evaluation provides the feedback loop to assess trends and gaps in prevention and response . During these times of increased attention on GBV due to consistently high levels of GBV in humanitarian settings, low access to services by those who experience GBV,  and faced by funding restrictions and fiscal challenges, implementers may find greater success in the use of evidenced-based interventions.
A review of studies on refugee camp density and size shows that camp design may help to mitigate structural vulnerability to GBV , such as reported here. Other proven GBV prevention strategies have been implemented in non-humanitarian settings. These include income enhancement and gender training, [44, 45] male-targeted, community-based training sessions to address GBV and HIV risk behaviors,  training sessions with participatory learning and communications skills, [44, 45] and community mobilization (trial underway) . Though still few in number, research to inform evidence-based response efforts has been implemented in humanitarian settings, including studies of psychological support interventions for rape survivors,  community-led mobile clinic and psychological support services for male and female survivors of GBV,  and support of women and families through economic empowerment by village-led microfinance (trial underway) . The dearth of data on evidence-based prevention and response for GBV in humanitarian settings sets a research agenda to develop a base with which to inform programming to provide the most effective and efficient responses in times of limited resources.
We found that significant barriers exist to GBV reporting and service utilization in the study population. Under-reporting of GBV experiences and low uptake of available services by survivors challenge GBV prevention and response programs implemented in humanitarian settings. To actively and confidentially overcome the barriers to disclosing GBV and referral, we used these findings to develop a screening tool that is multi-dimensional and captures broad domains and constructs of GBV, which we are now testing for validity and feasibility of use. Six questions covering psychological violence and threats of violence, physical violence, sexual violence, forced pregnancy, and forced marriage were included in the screening tool. Additional questions to identify locations and perpetrators are included and can be used to inform referrals for health, protection, and psychosocial services. For example, understanding the location or perpetrator of a sexual violence event can quickly inform service providers as to whether the survivor is still at risk and in need of additional protection. Such a screening tool may be integrated into existing GBV referral and surveillance mechanisms. Given the broad definition of GBV,  this proposed screening tool would not identify all forms of GBV but is designed to capture those widely reported and likely needing response from the health, psychosocial service providers, and/or protection officers. Other screening tools may be better positioned to identify other forms of GBV, for example, screening of female children may include questions to identify and respond to risk or recent experience of early marriage and female genital mutilation/cutting.
Like existing screening tools for intimate partner violence (IPV) or domestic violence, the proposed GBV screening tool for refugees may be implemented within the health settings , or may be implemented where confidential interviews may be conducted by trained service providers with refugees and where appropriate referrals are accessible to survivors. As with other screening tools, careful consideration should be given the setting where it will be implemented; for example, the proposed screening tool should only be used where skilled providers and GBV services are available, and where confidentiality and protection can be assured. This could include, for example, camp registration, during child–parent tracing interviews, and nutritional programs. Additional benefits to GBV screening include linkages to other data collection and documentation that is needed for program planning . Finally, routine screening of women can serve a secondary purpose of changing the social norms that currently sustain GBV and increasing awareness of rights and services for those who have experienced GBV.
The findings should be viewed in light of several limitations.
Men and children are known to experience substantial forms of violence [35, 52–54]. These populations may have different vulnerabilities, experiences, and interpretations of violence; thus, separate research studies for these groups are warranted. The research presented here focuses on female populations; however; the authors are currently collaborating to develop a screening tool for refugee and displaced men and boys.
Research was conducted among female refugees residing in urban and camp-based settings in Ethiopia. Refugees are often referred from camps to the urban setting to address serious health or protection needs. To overcome this potential bias, we used purposive sampling methods to recruit and interview participants with a range of GBV experiences. Because we used qualitative methods, selected GBV survivors for interviews, and did not include the general refugee population for interviews, we did not establish estimates of the prevalence of GBV among the refugee population studied. Refugee camps on the Somali border were selected for research activities and the findings derived from those settings may be limited to the Somali refugee experience. Sites were selected on the basis of several logistical, security, and ethical considerations (e.g. ensuring that established and quality referral services are available when survivors are identified). To ensure a range of experiences were captured, attention was given to include refugees from other countries (Sudan, Eritrea, Burundi, and DRC) in the interviews conducted in Addis Ababa.
An additional limitation is associated with the recruitment and eligibility of survivor participants. Survivors were recruited from the population of female refugees (15 years and older) who had reported their case of GBV and were receiving/had received GBV services from our implementing partners. Thus, these individuals had successfully overcome many of the barriers described above. Experiences and barriers reported in this manuscript may, therefore, be different among those who were not receiving services or had not disclosed GBV to service providers. This recruitment method and eligibility criteria, however, was identified in collaboration with the implementing partners working in the camps as ethically appropriate as talking about the GBV can be distressing, therefore, the survivors invited and consented to participate were prepared to discuss GBV and had ongoing access to services.