Patterns of Gender-Based Violence in Conflict-Affected Ukraine: A Descriptive Analysis of Internally Displaced and Local Women Receiving Psychosocial Services

Abstract

Since 2014, a protracted armed conflict has afflicted eastern Ukraine, resulting in the displacement of over 1.4 million residents. The resulting humanitarian crisis has placed women, particularly displaced women, at greater risk of gender-based violence (GBV). In Ukraine, reports of GBV were higher following the start of the conflict (22.4% in 2014 vs. 18.3% in 2007), with displaced women suffering from GBV nearly three times more than non-displaced residents (15.2% vs. 5.3%). Many GBV incidents in Ukraine have been reported along the “contact line,” the border separating government from non-government-controlled areas. This study compares types of GBV experienced by displaced and local (non-displaced) women receiving psychosocial support in order to identify the gaps in services during a time of conflict. Data was collected by mental healthcare providers from 11,826 women (25.5% displaced; 74.5% local) aged 15 to 69 receiving psychosocial services in five conflict-affected regions from February 2016 to June 2017. Group differences were assessed using Pearson’s chi-squared or Fisher’s exact tests for categorical variables and Wilcoxon rank-sum tests for continuous variables. Overall, almost half of the women experienced intimate partner violence and psychological abuse. Compared to residents, displaced women were more likely to report non-domestic GBV incidents involving sexual and economic violence. Almost 8% of violent incidents against displaced women occurred at checkpoints or at reception centers for internally displaced persons (IDP) and 20% were perpetrated by armed men. Consistent with the literature, this study suggests that displaced women are more vulnerable to attacks by persons outside the home and by armed groups. Our findings underscore the need to expand violence prevention programs to address the unique vulnerabilities of displaced women before, during, and after displacement. Programs should be tailored to prevent violence within and outside the home. Increased prevention efforts are needed in areas with high concentrations of armed men, along the contact line, and at IDP reception centers to protect displaced women. This is particularly urgent in the context of increased GBV due to COVID-19.

Background

Since 2014, a protracted armed conflict has afflicted eastern Ukraine, claiming the lives of over 3300 civilians, injuring over 7000 others, and forcing the displacement of more than 1.4 million persons (Ministry of Social Policy (MoSP) & United Nations High Commissioner for Refugees (UNHCR), 2018; United Nations Office for the Coordination of Humanitarian Affairs (UNOCHA), 2020). Currently, 3.4 million persons (60% of whom are women and children), living principally along the contact line that separates government- from non-government–controlled territories, are experiencing extreme economic insecurity and are in need of humanitarian assistance (UNOCHA, 2020). A survey of internally displaced persons (IDPs) in Ukraine found that a majority of respondents described their household economic situation as bad or very bad (59%), and only 22% held regular employment (Roberts et al., 2017).

Women affected by conflict, and particularly women displaced by conflict, are at increased risk for gender-based violence (GBV) due to economic strain, changes in familial and community support structures, and normalization of violence in the home (Ager et al., 2018). GBV, defined as “any act that is perpetrated against a person’s will and based on gender norms and unequal power relationships” can be “physical, emotional, psychological, or sexual in nature,” as well as economic (UNHCR, 2001). Globally, 35% of women have experienced violence by an intimate partner or sexual violence by any perpetrator in their lifetimes (Abrahams et al., 2014; Garcia-Moreno et al., 2006), with dire mental, physical, and reproductive health consequences (Asgary et al., 2013). Findings from systematic reviews indicate that the prevalence of physical violence against women in humanitarian settings ranges from 3% to 52% (Rubenstein et al., 2020; Stark et al., 2017). Meta-analytic findings estimate a 21% prevalence of sexual violence among female refugees and IDPs (Vu et al., 2014).

Gender-Based Violence in Eastern Ukraine Before and After the Start of the Conflict

While the precise magnitude of GBV has yet to be adequately assessed in Ukraine, the patterns and intensity of GBV appear to have shifted since the onset of the conflict. Since February 2014, calls to a domestic violence hotline and filed police reports increased dramatically in Ukraine compared to before the beginning of the conflict (Benigni, 2016; Pozarni & Rohwerder, 2016). A 2014 national survey conducted shortly after the start of the conflict found that 19% of 15–49 year old women had experienced violence since the age of 15 compared to 17% in 2007 (Martsenyuk et al., 2014). And whereas the share of women experiencing domestic violence remained unchanged, the overall increase was largely due to violence committed by perpetrators outside the home (3% in 2007 vs. 5% in 2014) (Martsenyuk et al., 2014).

Evidence suggests that displaced women are prone to heightened risk and may also experience different patterns of GBV risk compared to local women. A study among conflict-affected women found that the proportion of displaced women experiencing violence (15%) was three times higher than that of local (non-displaced) women (5%) (Ukrainian Center for Social Reforms (UCSR), 2018). Whereas a majority (66%) of local GBV survivors knew their perpetrators personally, displaced women were proportionally more likely to experience violence at the hands of strangers and of men involved in armed conflict (i.e., active and demobilized soldiers). Entry–exit checkpoints along the contact line posed an area of heightened risk for GBV against displaced women (UCSR, 2018).

Addressing Gaps in Gender-Based Violence Response Care and Support Services

Initial research following onset of the conflict identified an unmet need for support services for GBV survivors, including mental healthcare, psychosocial support, and shelters, near Ukraine’s conflict zones (UCSR, 2018; United Nations Human Rights Monitoring Mission in Ukraine, 2017). Limited access to support services was the main reason cited by Ukrainian GBV survivors for not seeking assistance and remaining in violent situations (Organization for Security and Co-operation in Europe (OSCE), 2019). Restrictive hours of operation and rigid eligibility criteria (i.e., only for women younger than 35 years old) (Immigration and Refugee Board of Canada, 2015) also limited access. These barriers may disproportionately affect displaced women. A national study found that almost three-quarters of IDPs who experienced mental health distress had not received mental healthcare; further, almost half of those who received mental healthcare, did so from pharmacists, rather than from mental health professionals (Roberts et al., 2017).

In order to meet the psychosocial support service needs of conflict-affected women in Ukraine, a coalition of partners including the United Nations Population Fund (UNFPA), HealthRight International, the Ukrainian Foundation for Public Health (UFPH), and the Ukrainian Ministry of Social Policy (MoSP) established mental health and psychosocial mobile teams in November 2015. These mobile teams were stationed in the five most conflict-affected oblasts (i.e., administrative regions) (Donetsk, Luhansk, Kharkiv, Dnipropetrovsk, and Zaporizka), where nearly three-quarters (72%) of IDPs were registered at the time (MoSP & UNHCR, 2018). Mobile teams publicize the program through public advertisements, as well as community-outreach campaigns targeting local health clinics, police departments, and social services. These local organizations then refer women to the mobile teams, who provide direct counseling and psychosocial services, case documentation, and referrals to other wrap-around services. Women can also self-refer to services.

Study Aims

Rigorous GBV research in humanitarian settings has been sparse in Eastern Europe and Central Asia. Further, in conflict-afflicted Ukraine, there is limited evidence regarding the characteristics and unique vulnerabilities of women who experience GBV. Early surveys indicate that displaced women may experience a disproportionate burden of violence. However, more nuanced data regarding GBV incidence and types of risk is needed for more informed intervention design and policy-making. In addition, little is known about help seeking and reporting behaviors among women survivors of GBV who accessed care in this setting.

This study seeks to identify the drivers and the patterns of GBV in Ukraine in the context of conflict and displacement. Using cross-sectional data collected from GBV survivors in care based upon the United Nations Gender-Based Violence Information Management System (GBV-IMS) intake form (UNFPA et al., 2011), the study defines and compares patterns of GBV by residency status. The key factors examined were: 1) the sociodemographic characteristics associated with experiencing violence in conflict areas; 2) the characteristics of perpetrators (i.e., age, relationship to survivor, and occupation) and incidents (i.e., periodicity, location, and type) of violence; and 3) the patterns of reporting and referrals, by residency status (displaced and local women).

Based upon the current literature, we anticipated that the patterns of violence experienced by displaced women in the context of protracted armed conflict would be different from those of local women. Furthermore, we hypothesized that among GBV survivors: 1) proportionally more violent acts against displaced women would be non-domestic and associated with combat operations (i.e., demobilized and active governmental and non-governmental soldiers); 2) displaced women would be more likely to experience sexual violence than local women; and 3) patterns of reporting and referrals would differ depending on a woman’s residency status. The study findings are intended to inform the design of culturally appropriate violence prevention and GBV survivor support programs.

Materials and Methods

Data Collection

All study data were collected using a single intake and assessment form administered by the trained psychologists and social workers who staff the mobile teams at the initial visit. These mobile team care providers received annual training on completion of the intake form. The intake process took between 20 to 90 minutes to complete. To reduce the risk of re-traumatization, providers elicited information while providing psychosocial support. The intake and assessment form was based on a tool developed by the UN GBV-IMS Global Team (UNFPA et al., 2011) intended to harmonize GBV data collected during service delivery in humanitarian settings. A group of humanitarian agencies and non-governmental organizations under the leadership of UNFPA, and including HealthRight International and UFPH, translated, adapted, and validated the tool for the Ukrainian context (e.g., the education, employment, and residence status categories were tailored) over a 2-month period. The adaptation process followed the recommendations of the GBV-IMS Rollout Guidelines (UNFPA, n.d.) and the UN Women’s Framework for emergency response and preparedness (UN Women, 2013), and entailed piloting the tool with several mobile teams and incorporating the feedback from the field. As secondary analysis of de-identified data, the current study does not constitute human subjects research.

Measures

The intake and assessment form includes sociodemographic information, details about the incident and the perpetrator, and past contact with and referrals to relevant services (e.g., police, medical, and social), as described below.

Residency status:

Women were asked about their residency status at the time of the incident, coded as 0 = non-displaced local resident (local); 1 = IDP (displaced). Ukrainian law defines an internally displaced person as “a citizen of Ukraine, a foreigner or a stateless person who is in the territory of Ukraine legally and has the right to reside permanently in Ukraine, and who was forced to leave his place of residence due to armed conflict, temporary occupation, widespread violence, human rights violations or emergencies of natural or man-made nature” (On Ensuring the Rights and Freedoms of Internally Displaced Persons, 2014). Officially registered IDPs are entitled to humanitarian assistance, housing, and financial support. However, not all IDPs are registered. In our study, residency status category was based on self-reported residency status and was not verified by the mobile team service providers.

Sociodemographic variables were coded thus: 1) Age at incident (continuous); 2) current marital status (0 = single or widowed; 1 = married or cohabitating; 2 = divorced or separated); 3) number of children under 18 living with the client (0 = no children; 1 = 1 child; 2 = 2+ children); and 4) current occupation (6 nominal categories, where unpaid labor was defined as non-remunerated domestic labor, such as child and elder care). Age was maintained as a continuous variable to minimize information loss (Osborne, 2017) and the median and interquartile range (IQR) were reported given a skewed distribution.

The incident description included two variables coded as: recurrence (0 = assaulted once; 1 = assaulted multiple times or violence was ongoing); place of incident (0 = home; 1 = outdoors; 2 = checkpoint or IDP reception center; 3 = all other indoor locations not included in the previous categories, including institutions such as healthcare clinics or social services buildings).

GBV type:

Determination of GBV type was made by mobile team members using the GBV-IMS classification tool (UNFPA et al., 2011). The form instructs providers to select only one GBV type per case based on a series of questions asked in a specific order, as follows: 1) rape (if any type of penetration occurred); 2) sexual assault (if there was unwanted sexual contact); 3) physical assault (if there was physical battery); 4) forced marriage; 5) economic violence (in cases of denial of resources, opportunities, or services); 6) psychological or emotional abuse (if the incident involved insults, name-calling, and humiliation); and 7) no GBV (if none of the above). If, for example, a woman reported experiencing unwanted sexual contact, the provider would classify the case as “sexual assault” and continue to the following section. Supplementary Appendix A presents the instructions given to providers to determine the GBV type and Supplementary Appendix B describes in detail each GBV category. We created a four-category variable, as follows: sexual violence (includes rape and sexual assault); physical violence; economic violence; and psychological abuse. Even though there are important distinctions between survivors of rape and sexual assault, these categories were combined because of the statistically low numbers (n = 56 rape and n = 219 of sexual assault incidents). Forced marriage (n = 8) and non-GBV (n = 537) were deemed outside of the scope of the current analysis and not reported.

Perpetrator characteristics included: 1) number of perpetrators (0 = 1; 1 = 2+; 2 = unknown); sex (0 = male; 1 = female; 2 = both); 2) age of perpetrator in five age groups: aged under 18; 18−25; 26−40; over 40; unknown); 3) relationship to perpetrator (0 = intimate partner; 1 = family member; 2 = non-domestic, defined as perpetrators with no intimate or familial relationship to the survivor; 3 = unknown); 4) perpetrator’s occupation (8 nominal categories, including governmental and non-governmental soldier); and 5) demobilized soldier (0 = no; 1 = yes; 2 = unknown). For age of perpetrator, the categories of 41−60 and 61+ were combined because of low numbers.

Reporting and referrals variables included: 1) Whether the incident was reported, assessed with the question “Has the client reported this incident anywhere else?” (0 = no, 1 = yes); if yes, the provider followed up with a question about where it was reported (4 nominal categories: 0 = psychosocial services; 1 = police; 2 = social protection services; 3 = other institutions, such as health services). Multiple options were possible; 2) Clients were asked if they had been referred to the services. If self-referred, clients were asked how they learned of the mobile teams; 3) Clients were also asked if they had or were currently receiving psychosocial support from providers other than the mobile teams; 4) The GBV-IMS form also asked providers to record what services they had referred the clients to during the visit.

Analytic Sample

We conducted a secondary analysis of intake data collected from 11,826 GBV survivors who sought services from the mobile teams in five oblasts between February 2016 and June 2017. During this period the mobile teams served 17,095 clients; of them 3299 did not consent to sharing their data for research purposes and 1970 did not meet the eligibility criteria defined for the current analysis. The sample includes all women aged 15–69 years old whose self-reported residency status at the time of the incident was either displaced or local resident. Supplementary Appendix C presents the flow diagram.

Statistical Analysis

This analysis compares displaced (n=3020, 25.5%) and local women (n=8806, 74.5%) who received care from the mobile teams in terms of sociodemographic characteristics, incident and perpetrator characteristics, and access to care patterns. Differences between the groups were assessed using Pearson’s chi-squared or Fisher’s exact tests for categorical variables and two-sided Wilcoxon rank-sum tests for non-normally distributed continuous variables. All analyses were conducted in Stata 15.1 (Stata Corporation LP, College Station, TX).

Results

Sociodemographic Characteristics

Selected sociodemographic characteristics of the sample are detailed by residency status at time of the incident in Table 1. The median age was 36 (IQR=25–45), with no significant differences between the groups. A majority of the women in our sample were married (56.3%). However, displaced women were more likely than local women to be single or widowed (29.0% vs. 24.4%, p<0.001) and had on average fewer children (26.3% of IDPs had two or more children vs. 32.5% of local women, p<0.001). More than one in five (21.6%) women who experienced violence were unemployed, with no differences between the groups. Overall, slightly less than one-third (30.7%) of the women engaged in unpaid labor such as elder and childcare, with significantly higher proportions among local women. Proportionally more displaced women had a professional occupation (24.6% vs. 20.0%, p<0.001).

Table 1.

Sociodemographic characteristics of 15–69-year-old women receiving psychosocial services for GBV in eastern Ukraine by residency status, N = 11,826.

	Total	Local	IDP	p-valuea
	n (%)	n (%)	n (%)
	11,826 (100.0)	8806 (74.5)	3020 (25.5)
Age at incident [median (IQR)]	36 (29–45)	36 (29–45)	35 (29–45)	0.200
Marital status
Single or widowed	3023 (25.6)	2146 (24.4)	877 (29.0)	<0.001
Married or cohabitating	6658 (56.3)	5151 (58.5)	1507 (49.9)	<0.001
Divorced or separated	2145 (18.1)	1509 (17.1)	636 (21.1)	<0.001
Number of children under 18
No children	4588 (39.0)	3385 (38.7)	1203 (40.1)	0.180
1 child	3531 (30.0)	2522 (28.8)	1009 (33.6)	<0.001
2+ children	3636 (31.0)	2845 (32.5)	791 (26.3)	<0.001
Employment
Professional	2495 (21.2)	1757 (20.0)	738 (24.6)	<0.001
General labor	2328 (19.8)	1825 (20.8)	503 (16.8)	<0.001
Unpaid labor	3624 (30.7)	2760 (31.4)	864 (28.8)	0.007
Unemployed	2547 (21.6)	1867 (21.2)	680 (22.7)	0.107
Student	262 (2.2)	200 (2.3)	62 (2.1)	0.498
Retired	528 (4.5)	374 (4.3)	154 (5.0)	0.046

Notes: IDP=internally displaced person; IQR=interquartile range.

^ap-value based on Pearson’s chi-square, Fisher’s exact tests or two-sided Wilcoxon rank-sum tests.

Type of Incident and Perpetrator

In the overall sample, most incidents were documented as a single occurrence of violence (60.4%), which happened at home (79.1%) by a singular (80.4%) perpetrator. More than three-quarters (78.3%) of women reported that a man was the perpetrator. In nearly half of the cases, the perpetrator was an intimate partner (49.5%); and in roughly one in five (21.8%) a family member. Psychological abuse (48.4%) was reported by almost half of the women (See Table 2).

Table 2.

Description of GBV incident and perpetrators among 15–69-year-old women receiving psychosocial services in eastern Ukraine by residency status, N = 11,826.

	Total	Local	IDP	p-valuea
	n (%)	n (%)	n (%)
	11,826 (100.0)	8806 (74.5)	3020 (25.5)
Recurrence of incident
Single incident	6903 (60.4)	5036 (58.9)	1867 (65.0)	<0.001
Periodic/recurrent	4519 (39.6)	3514 (41.1)	1005 (35.0)
Place of incident
Home	9358 (79.1)	7485 (85.0)	1873 (62.0)	<0.001
Outdoors	1189 (10.1)	703 (8.0)	486 (16.1)
Checkpoint or IDP center	303 (2.6)	65 (0.7)	238 (7.9)
Institutions	976 (8.2)	553 (6.3)	423 (14.0)
Type of GBV
Rape or sexual assault	276 (2.4)	182 (2.1)	94 (3.1)	<0.001
Physical assault	3840 (32.5)	3124 (35.5)	716 (23.8)
Economic	1975 (16.7)	1271 (14.4)	704 (23.4)
Psychological abuse	5719 (48.4)	4222 (48.0)	1497 (49.7)
Number of perpetrators
1	9508 (80.4)	7641 (86.8)	1867 (61.8)	<0.001
2+	1437 (12.2)	851 (9.7)	586 (19.4)
Unknown	881 (7.4)	314 (3.5)	567 (18.8)
Sex of perpetrator
Male	9258 (78.3)	7202 (81.8)	2056 (68.1)	<0.001
Female	1335 (11.3)	954 (10.8)	381 (12.6)
Both	1233 (10.4)	650 (7.4)	583 (19.3)
Age of perpetrator
Under 18	164 (1.4)	117 (1.3)	47 (1.6)	<0.001
18–25	766 (6.5)	611 (6.9)	155 (5.1)
26–40	5362 (45.3)	4288 (48.7)	1074 (35.6)
Over 40	3899 (33.0)	3114 (35.4)	785 (26.0)
Unknown	1635 (13.8)	676 (7.7)	959 (31.7)
Relationship to perpetrator
Intimate partner	5858 (49.5)	4913 (55.8)	945 (31.3)	<0.001
Family member	2572 (21.8)	2169 (24.6)	403 (13.3)
No relation (non-domestic)	3090 (26.1)	1617 (18.4)	1473 (48.8)
Unknown	306 (2.6)	107 (1.2)	199 (6.6)
Perpetrator’s occupation
Professional	2152 (18.2)	1494 (17.0)	658 (21.8)	<0.001
General labor	3626 (30.7)	3032 (34.4)	594 (19.7)
Unemployed	2681 (22.7)	2288 (26.0)	393 (13.0)
Ukrainian army soldier	456 (3.9)	226 (2.6)	230 (7.6)
Police	55 (0.5)	43 (0.5)	12 (0.4)
Non gov. Soldier	617 (5.2)	242 (2.7)	375 (12.4)
Other	1080 (9.1)	839 (9.5)	241 (8.0)
Unknown	1159 (9.8)	642 (7.3)	517 (17.1)
Demobilized soldier
Yes	679 (5.7)	608 (6.9)	71 (2.3)	<0.001
No	9659 (81.7)	7377 (83.8)	2282 (75.6)
Unknown	1488 (12.6)	821 (9.3)	667 (22.1)

Notes: IDP=internally displaced person.

^ap-value based on Pearson’s chi-squared or Fisher’s exact tests.

There were significant differences in the patterns of violence by residency status (See Table 2). Displaced women were less likely than local women to report multiple or ongoing situations of violence (35.0% vs. 41.1%, p<0.001). Displaced women were more likely than local women to report that the incident occurred outside the home. An alarming proportion of violence against displaced women occurred at IDP reception centers or at entry/exit checkpoints (7.9%). Compared to local women, proportionally more displaced women reported an incident of rape or sexual abuse (3.1% vs. 2.1%, p<0.001) or economic abuse (23.4% vs. 14.4%, p<0.001).

About half of displaced women reported non-domestic violence (48.8%), defined as violence perpetrated by persons outside the survivor’s domestic sphere, such as friends, acquaintances and strangers; fewer reported domestic violence, including by intimate partners (31.3%) and family members (13.3%). Proportionally more displaced than local women reported violence by members of the military (7.6% vs. 2.6%, p<0.001) and by non-governmental soldiers (12.4% vs. 2.7%, p<0.001).

Reporting and Referrals

As shown in Table 3, we examined the history of GBV reports and the patterns of referrals made to and by the mobile teams. Only 38.7% of women had reported the GBV incident prior to obtaining services from the mobile teams, with no differences by residency status. There were marked differences between the groups among survivors who had previously reported the GBV incident. Displaced women were more likely than local women to have reported the incident to psychosocial counselors and to social protection services (46.4% vs. 30.1% and 25.5% vs. 5.8%, respectively, both p<0.001); but were less likely to have filed a police report (25.4% vs. 55.7%, p<0.001) (Table 3).

Table 3.

Reporting and referrals among 15–69-year-old women receiving psychosocial services for GBV in eastern Ukraine by residency status, N = 11,826.

	Total	Local	IDP	p-valuea
	n (%)	n (%)	n (%)
	11,826 (100.0)	8806 (74.5)	3020 (25.5)
Reported incident	4580 (38.7)	3411 (38.7)	1169 (38.7)	0.980
If reported, where?
Psychosocial services	1570 (34.3)	1028 (30.1)	542 (46.4)	<0.001
Police	2197 (48.0)	1900 (55.7)	297 (25.4)	<0.001
Social protection services	496 (10.8)	198 (5.8)	298 (25.5)	<0.001
Other institutions	861 (18.8)	573 (16.8)	288 (24.6)	<0.001
Client referred by
Self	5252 (44.4)	3796 (43.1)	1456 (48.2)	<0.001
Psychosocial services	2597 (22.0)	2035 (23.1)	562 (18.6)	<0.001
Police	1360 (11.5)	1173 (13.3)	187 (6.2)	<0.001
Other institutions	2617 (22.1)	1802 (20.5)	815 (27.0)	<0.001
If self-referred, learned about service from
Public advertisement	2583 (49.2)	1943 (51.2)	640 (44.0)	<0.001
Mass Media	659 (12.6)	496 (13.1)	163 (11.2)	0.067
Internet or social media	685 (13.1)	367 (9.6)	318 (21.8)	<0.001
Referral card	753 (14.3)	565 (14.9)	188 (12.9)	0.068
Other	572 (10.9)	425 (11.2)	147 (10.1)	0.252
Receiving psychological support	2000 (16.9)	1501 (17.1)	499 (16.5)	0.509
Client referred to b
Domestic violence shelter	149 (1.3)	127 (1.4)	22 (0.7)	0.002
Medical care	777 (6.6)	549 (6.2)	228 (7.6)	0.012
Psychosocial support	3625 (30.7)	2755 (31.3)	870 (28.8)	0.011
Legal support	1551 (13.1)	1129 (12.8)	422 (14.0)	0.105
Police	727 (6.2)	605 (6.9)	122 (4.0)	<0.001
Social protection services	1487 (12.6)	1104 (12.5)	383 (12.7)	0.835

Notes: IDP=internally displaced person.

^ap-value based on Pearson’s chi-squared or Fisher’s exact tests.

^bnon-exclusive choices; multiple referrals possible.

Overall, a majority of the sample sought services independently (i.e., self-referral). However, a slightly larger proportion of displaced women self-referred (48.2% vs. 43.1%, p<0.001). Among the self-referrals, roughly half learned about the mobile teams through public advertisements (49.2%), while proportionally more displaced women learned about the services through social media and the Internet (21.8% vs. 9.6%, p<0.001). Displaced women were slightly more likely to be referred to medical services (7.6% vs. 6.2%, p=0.012) compared to local women (Table 3).

Discussion

GBV is a grave human rights violation that affects an estimated one million women annually in Ukraine (Barrett et al., 2012). Social disruption and frail economic conditions in humanitarian settings further aggravate women’s vulnerability to violence, particularly for displaced women (Stark & Ager, 2011; Stark et al., 2017). This analysis supports our primary hypothesis that the experience of violence differs by survivors’ residency status. Specifically, we found differences in terms of relationship to the perpetrator, type of violence experienced and access to care between local and displaced women.

Relationship to Perpetrator and Residency Status

In our sample, nearly half of the women reported violence at the hands of an intimate partner, although local women were more likely than displaced women to report violence by intimate partners and family members. This was broadly similar to the findings in the 2019 OSCE survey that found that in Ukraine intimate partners were the most common perpetrators of violence against women aged 15 and older, and that local women were less likely to experience non-domestic violence than displaced women (5.3% vs. 15.2%).

Notably, 20.0% of displaced women in our sample experienced violence at the hands of armed men compared to 5.3% of local women. We also found that checkpoints between government-controlled and non-government–controlled areas and IDP reception centers posed a particular risk for displaced women in our study. Our findings are in alignment with other surveys in which displaced women reported military checkpoints as dangerous locations in Ukraine (UCSR, 2018; Organization for Security and Co-operation in Europe OSCE, 2019).

Displaced Women are Disproportionately Affected by Sexual Violence

Population-based surveys in Ukraine found that 53% of women had experienced psychological violence, 23% physical violence and 7% sexual violence since turning 15 (OSCE, 2019). Consistent with this literature, psychological violence was the most common type of GBV experienced by women in our sample, followed by physical and sexual violence. Whereas sexual violence was the least common type of reported violence, 38% more displaced women reported experiencing sexual violence than local women. It is worth noting the differences from the UNFPA survey, which found higher prevalence of non-domestic sexual violence among local than displaced women (7% vs. 4%) (UCSR, 2018). However, the authors warn of likely undereporting of sexual violence. Overall, the UNFPA survey found that prevalence of GBV from any perpetrator was three times higher among displaced women compared to local women, highlighting the vulnerability of displaced women to GBV.

These findings underscore the need to extend violence prevention efforts to protect displaced women from violence outside of the home. In the context of the conflict in eastern Ukraine, strengthening protection efforts along the contact line and travel routes, for example, by improving infrastructure and lighting in the area, could serve to protect displaced women traveling alone.

Self-Referrals as a Key Driver of Access to Care

In humanitarian settings, women affected by conflict have been found to rely on digital channels, such as informal online support groups, to learn about and access psychosocial support services in response to GBV (Phiri & Junior, 2018; Vuningoma et al., 2020; Walker et al., 2015). In Zambia, programs have effectively used social media to build awareness about services for GBV survivors (Phiri & Junior, 2018). Consistent with the literature, a large proportion of women in our study self-referred to the GBV mobile teams, rather than accessing the services through traditional referral channels, such as the police or social services. Notably, compared to local women, greater numbers of displaced women sought mobile team services in response to social media and the Internet. Anecdotally, mobile team staff indicated that many women also found out about the mobile teams through word-of-mouth. Targeted outreach approaches that involve virtual support groups that are accessible via mobile devices may be a promising strategy to increase program awareness in this setting (Vuningoma et al., 2020; Walker et al., 2015).

Differences in Incident Reporting to the Police

Studies in conflict-affected Ukraine found that a majority of survivors were unwilling to report GBV incidents to the police, particularly among internally displaced women (UCSR, 2018). Reasons cited included mistrust of law enforcement bodies and fear of further violence (UCSR, 2018). Whereas we did not collect information on why the incidents were not reported prior to obtaining services from the mobile teams, we found that displaced women were less than half as likely than local women to have filed a police report. Furthermore, displaced women were less likely than local women to obtain referrals to the police. This may confirm displaced women’s unwillingness to pursue further legal actions related to the GBV incidents as well as general distrust of law enforcement.

Other sources indicate that high concentrations of demobilized soldiers are associated with GBV risk (Lucas et al., 2017; Tabarovsky, 2016). However, a lack of information about perpetrators precluded us from making conclusions about an association between military involvement of perpetrators and violence. It is important to note that the situation in eastern Ukraine is in constant flux with ongoing changes in the make-up of military groups.

Strengths and Limitations

There are several study limitations that are important to consider. First, as a cross-sectional analysis, no causal inferences can be made. Second, since data only includes persons who sought care for GBV, we cannot make conclusions with regards to prevalence of violence. Furthermore, there may be systemic differences between women who did and did not access GBV services, so findings may not be generalizable to the broader population. For example, among GBV survivors in Ukraine, younger women seek services for GBV more often than older women (41% of those aged 15–29 vs. 26% those aged 40–49) (Martsenyuk et al., 2014). Therefore, this analysis is not representative of all women experiencing violence.

Multiple mobile team providers collected data at different locations, over extended time periods. Therefore, data collection errors and variability may have been introduced. However, all mobile team personnel were trained annually on the use and proper completion of the GBV-IMS intake form, partially controlling for these errors. Further, all data is based on self-report and prone to biases. Given documented stigma and normalization of IPV in Ukraine, women experiencing IPV may not have been willing to report their partners as the perpetrator (Centers for Disease Control and Prevention & ORC Macro, 2003). Studies in complex emergency settings have found stigma among GBV survivors, normalization of domestic violence during times of conflict, unwillingness to report men living in the home for fear of forced military recruitment, and reluctance to involve law enforcement as major reporting barriers, especially among displaced women survivors of violence (Ager et al., 2018; Stark & Ager, 2011). For example, in a post-conflict survey, 40% of Ukrainian women reported being single compared to 17% of women in 2007 (UCSR, 2018). In this context, non-domestic violence as well as missing perpetrator data may be misreported instances of IPV. This study has a limited geographic scope (five areas in Ukraine affected by conflict). Provision of services by the mobile teams did not discriminate based on sex, gender, age, abilities, race/ethnicity, or nationality. Anybody seeking services from the mobile teams received the needed support. For the purposes of the analysis, only persons who identify as women between the ages of 15 and 69 were included. No questions related to religion, sexual orientation, or race/ethnicity were collected. An important limitation of the GBV-IMS form is that it only allows providers to input one type of GBV per respondent (UNFPA et al., 2011). As a result, all types of violence with the exception of rape and sexual assault are underreported, in particular psychological violence. To the best of our knowledge, the extent of underreporting resulting from the use of this form has not been evaluated. While this manner of classification facilitates standardization and consistency of data capture across multiple service providers, sites, and countries, thus improving comparability, it comes at the cost of not capturing the multiple forms of GBV that women may have experienced (UNFPA et al., 2011).

In spite of these limitations, a strength of this study is the large sample size, as well as the use of a validated assessment form, developed to document service provision in humanitarian emergencies and adapted to this population. In addition, the findings contribute to the literature in an understudied context, with recommendations that can be utilized to improve policies and programs to prevent and respond to GBV. Another strength is that it is one of the first analysis of reports to utilize data from all the oblasts surrounding the ongoing conflict.

Policy and Program Implications

Findings from our study highlight the importance of residency status in understanding GBV risk and designing effective prevention and treatment programs and policies. A systematic review of effective interventions to prevent and respond to violence in conflict settings identified four strategies that may be applicable to Ukraine: increasing legal consequences for violent offenders; strengthening community engagement in violence prevention programs (e.g., by promoting bystander intervention and reporting); strengthening the availability of private and confidential resources for survivors; and building awareness of these resources (Spangaro et al., 2015). Ukraine has made significant advances in addressing GBV. Services offered by the mobile teams have partially served to address the gap in availability of confidential resources and the awareness campaigns organized by the teams have served to make these services known. The recently adopted domestic violence law has improved legal recourse for women who experience IPV (On Preventing and Counteracting Domestic Violence, 2018). However, other policy changes are needed to increase the legal consequences for perpetrators of violence both inside and outside the home, including in the context of sex work (Becker et al., 2019). Parallel efforts are needed to change cultural values normalizing domestic violence (Centers for Disease Control and Prevention, & ORC Macro, 2003). More investments to reach GBV survivors and prevent GBV are urgently needed in light of COVID-19, which has further deepened the impact of conflict on GBV (Inter-agency Standing Committee on Gender-Based Violence, 2020). Compared to before the pandemic, calls to the national domestic violence hotline increased by 26% after COVID-19 quarantine measures were implemented (UNFPA, 2020). At the same time, the mobile teams saw a surge of women seeking psychosocial support (UNFPA, 2020). We documented important differences in the types of GBV experienced by displaced versus local women, underscoring the need for services tailored to the prevention of GBV by residency status. Further research is needed to better understand these risk profiles so as to inform effective prevention programs.

Conclusion

Displaced women in Ukraine experienced the most violence in the conflict zones, which is also where women face the highest barriers to accessing social services, transportation, paid work, and basic goods (O’Sullivan, 2019; UNOCHA, 2019). Through a secondary analysis of UN GBV-IMS data collected during the second and third years of the conflict, we observed distinct differences in GBV survivors according to residency status. Of note is that displaced women experienced a greater share of sexual violence and of violence from armed men. Study results underscore the need to tailor and expand violence prevention programs to the unique vulnerabilities of displaced women prior, during, and after displacement.

Our findings echo UNFPA’s recommendations to implement novel outreach strategies to reach GBV survivors in conflict settings, as normal channels breakdown (UCSR, 2018). Programs differentially targeted to prevent violence within and outside the home may be more effective. In particular, increased prevention efforts in areas with high concentrations of armed men, along the contact line, and at IDP reception centers are recommended to reduce incidence of violence against displaced women.

With the spread of COVID-19, remote counseling for GBV via telemedicine is necessary, concomitantly with additional space to accommodate women safely at shelters. Expanding service providers’ capacity to address GBV and increasing coordination across different sectors (i.e., police, health sector, social protection, and psychosocial services) may increase timely identification, referrals, and access to appropriate GBV survivor services. Given the importance of self-referral, additional efforts to leverage social networks are needed to reach women seeking survivor care services, such as social media and integrated awareness raising campaigns.

Author biographies

Ariadna Capasso, MFA, is a PhD Candidate at New York University’s School of Global Public Health with over 15 years of experience implementing gender-based violence prevention programs. Prior to joining NYU, Ariadna was senior technical advisor at Management Sciences for Health, where she provided strategic leadership and managed a wide range of sexual and reproductive health projects in Latin America.

Halyna Skipalska, MA, is HealthRight Country Director for Ukraine and the Executive Director of the Ukrainian Foundation for Public Health. She has 13 years of experience working in non-governmental organizations in Ukraine, with a focus on prevention of HIV/AIDS and provision of care for vulnerable women and girls.

Urmi Chakrabarti, MA, MSW (Columbia University), is a clinical social worker/therapist with over 12 years of experience in the behavioral health/wellness field, serving in different capacities such as a therapist, activist, educator, workforce development, and public health professional. She brings her lived experience, multicultural, and multidisciplinary training to the field.

Sally Guttmacher, PhD, MPH, is a Professor Emeritus of Public Health at NYU and a Senior Technical Advisor for HealthRight International. Her domestic and international research includes policy and prevention of infectious diseases; poverty and public health; rights of migrants; reproductive health rights; women’s health and evaluation. She is co-author of, “Community Based Health Interventions.”

Peter Navario, PhD, MPH, is Executive Director at HealthRight International and Clinical Assistant Professor at NYU. He was previously technical advisor at UNAIDS and a Fellow for Global Health at the Council on Foreign Relations. He holds a PhD in Health Economics from the University of Cape Town, and an MPH in Global Health from Yale University.

Theresa P. Castillo, EdD, MA, CHES, has been working globally in the fields of gender, social justice, and health equity for over 20 years. Her expertise includes cultural rights, adolescent health, violence prevention, sexual & reproductive health. She is adjunct faculty at Columbia University, NYU, and Bard College. Dr Castillo holds a master’s and doctoral degree from Columbia University.

Data Availability

The data that support the findings of this study are available on request from the corresponding author.