Developing an Inclusive Conceptual Model for Preventing Violence in the Home in Humanitarian Settings: Qualitative Findings From Myanmar and the Democratic Republic of Congo

Abstract

This qualitative study sought to explore the shared risk factors and social norms that may underpin the co-occurrence of intimate partner violence, child maltreatment, and violence against other marginalized household members in the home. Data are drawn from participants who completed either in-depth interviews (N = 51 men, N = 52 women participants) or focus groups (N = 22 men, N = 23 women participants) and were living in two distinct humanitarian settings: North Kivu, Democratic Republic of the Congo, and Northern Shan State and Southern Kachin State, Myanmar. Within this overarching objective, attention to these shared drivers for violence in the home, which may arise from people having multiple and interacting social identities, such as disability status, gender, and age, was explored through an inclusion lens and inductive coding approaches. Findings point to risk factors of violence in the home at all levels of the ecological model, which are underpinned at the macro level by gender inequality, armed conflict, and political instability, among other factors. With the community and family levels, gender inequality manifested in norms related to the acceptability of violence, family reputation, and aged and gendered power hierarchies within the home. Shared risk factors of violence at these levels also included displacement/migration-related stressors, inconsistent income, and lack of community resources to support families, especially for those with disabilities. At the most time-proximal level, interactions between role (non)fulfillment, inability of families to meet their basic needs, alcohol and substance abuse, and abusive expressions of anger were found to catalyze instances of violence in the home. The conceptual model also illuminates potential levers and inclusive entry points that prevent violence in the home for diverse women, children, persons with disabilities, and older persons. Key attention to addressing gender inequality and acceptance of violence must be a cornerstone of sustainable programming, alongside complementary approaches that address other shared risk factors.

Keywords

intimate partner violence child abuse disabilities qualitative Democratic Republic of Congo Myanmar ecological

Introduction

Globally, one in three women experience some form of intimate partner violence (IPV) in their lifetime (World Health Organization, 2005) and nearly one out of every two children experience abuse. Importantly, a growing body of literature recognizes that these forms of violence against women and children, particularly physical and emotional child maltreatment and neglect, often share risk factors, co-occur within the same home, and have intergenerational linkages (Guedes et al., 2016; Guedes & Mikton, 2013; Namy et al., 2017).

Guided by both an ecological and feminist perspective (Heise, 1998), shared risk factors for these forms of violence in the home may be rooted in norms related to patriarchy and gender inequality, and may be exacerbated by men’s use of alcohol use or economic strain at the household level that interplay across societal, community, familial, and individual levels. Furthermore, not only do these types of violence share risk factors, each manifestation of violence may directly increase the likelihood of perpetuating the intergenerational cycle of violence within homes. For instance, witnessing parental IPV as a child or other forms of childhood trauma has consistently been linked to increased risk of IPV perpetration among men, and victimization among women as adults (Fulu et al., 2017; Roberts et al., 2010). Within this body of evidence, attention to the unique needs of adolescent girls is also warranted as they face dual burdens related to IPV and abuse from caregivers (Stark et al., 2017).

Within humanitarian settings, a recently published systematic review documented several common factors that cut across levels of the ecological model that could exacerbate violence perpetration and victimization in the home including conflict exposure, economic status, and limited social support (Rubenstein et al., 2017). These risk factors, some existing before a crisis, also have the potential to accumulate and converge during an armed conflict or a natural disaster, as was seen in a qualitative analysis from post–Hurricane Haiti (Bermudez et al., 2019). Recent qualitative research from displaced settings in Colombia also demonstrates the importance of considering relocation as a risk of violence, and how it may relate with other community and family factors, such as availability of community services or use of alcohol and other substances (Mootz et al., 2019).

Concurrent to the increased attention to the co-occurrence of IPV and child maltreatment, within both stable and humanitarian settings, a growing body of public health research is recognizing the importance of understanding how multiple marginalized identities create compounding and cumulative layers of oppression and discrimination. This draws from intersectionality theory (Crenshaw, 1989), which posits that multiple identities of an individual cannot be artificially separated and these identities “are not simply the sum of their parts” (Bauer, 2014). Examining the co-occurrence of violence in the home through this lens may deepen our understanding of shared risk factors related to these identities. For instance, women with disabilities have an amplified risk of IPV in comparison with women without disabilities (van der Heijden et al., 2016; van der Heijden & Dunkle, 2017). Similarly, children with disabilities also face heightened risk of violence (Jones et al., 2012), with magnified risk for female children with disabilities compared with children without disabilities (van der Heijden & Dunkle, 2017). Yet, little investigation has applied this analytic approach to understanding how shared risks of different forms of violence in the home may interplay across multiple levels of the ecological model in humanitarian settings and has not yet been fully utilized to highlight opportunities for inclusive violence prevention programming.

Therefore, to further examine and understand risk factors of violence in the home, this qualitative study seeks to explore the shared risk factors, including social norms, that may underpin the co-occurrence of IPV, child maltreatment, and violence against other marginalized household members in the home in two distinct humanitarian settings: North Kivu, Democratic Republic of the Congo (DRC), and Northern Shan State and Southern Kachin State, Myanmar. Within this overarching objective, attention to these shared risk factors for experiencing violence in the home, which may arise from people having multiple and interacting marginalized identities related to disability status, gender and age, was explored as compounding risks of experiencing violence at the hands of others in the home. These shared risk factors for violence in the home were then situated within an ecological conceptual framework that highlights drivers of violence in the home in humanitarian settings and underscores inclusive entry points and programming opportunities.

Method

Setting

DRC is a protracted humanitarian setting that has experienced armed conflict and instability for more than two decades. Data for the present study were collected in two sites in North Kivu, Eastern DRC, which hosts a population that has experienced multiple instances of displacement and is among the most conflict-affected areas of the country and home to numerous armed groups. IPV and child abuse are also prevalent at epidemic levels. Secondary analyses of the 2007 DRC Demographic and Health Survey demonstrate an estimated 68.2% women aged 15 to 49 years who were married or cohabitating reported experiencing physical, emotional, and/or sexual IPV in their lifetime (Tlapek, 2015). Furthermore, a study conducted in North Kivu found approximately 45% of ever-partnered men reported perpetrating physical IPV in their lifetime (Fleming et al., 2015). Child abuse is also pervasive as nearly 60% of adult men and women had experienced physical violence as a child and approximately 80% had experienced psychosocial or economic violence at home as a child in North Kivu (Slegh et al., 2014). In addition, among adolescent girls in South Kivu, DRC, aged 13 to 14 years, approximately one third reported physical violence in the past year and nearly 35% reported emotional abuse (Stark et al., 2017).

Myanmar is the site of ongoing conflict since its independence in 1948, and violent conflict between the Myanmar military and ethnic armed groups has displaced more than 100,000 people in Kachin and Northern Shan states. Political instability is further complicated by fighting between numerous ethnic armed groups with shifting alliances. Internal displacement camps included in this study, located in Northern Shan State and Southern Kachin State, house internally displaced persons of Kachin and Ta’ang ethnic origin on land owned by the church, cultural organizations, and militias. Internally displaced persons (IDPs) are caught between the conflicting sides with armed groups frequently moving through their site including forcibly recruiting men to serve, while the military establishes bases and security check points close to the camps. The militarization of the area results in an insecure and unsafe environment for IDPs, which in turn limits their freedom of movement and ability to access assistance and support. In Southern Kachin State, the IDPs are further restricted to services being unable to access services in Kachin State due to road blockages and insecurity, IDPs instead have to travel into Northern Shan State to receive services. Women and girls also face an elevated risk of violence inside the home; according to the Myanmar Demographic and Health Survey (2015–2016), 21% of married women have ever experienced spousal violence, 37% of married women who have experienced spousal violence report suffering physical injuries, and 37% have never told anyone about the violence (Ministry of Health Sports and ICF, 2017).

Study Design and Target Population

The present analyses uses qualitative data from a formative research effort to guide the development of an International Rescue Committee (IRC) program, Safe at Home, that seeks to prevent the co-occurrence of IPV and child maltreatment in the home by addressing shared risk factors. Respondents were purposively sampled in both sites by IRC to ensure inclusion of adult men and women of older ages and with disabilities. Participants were not selected from a larger sampling frame, but were identified via local staff members and coordination with community-based organizations in DRC or camp management in Myanmar. Overall, 73 adult men and 75 adult women participated in six focus group discussions in the DRC and in-depth interviews in both countries (see Table 1). In DRC, focus group participants were distinct from in-depth interview participants. In response to fears that gathering in groups may make men vulnerable to targeting for forced recruitment into armed groups in the Myanmar study sites, researchers did not conduct focus groups in Myanmar. Children and adolescents were not included within the study sample due to budgetary and logistical limitations; further exploration of this limitation is included in the “Discussion” section.

Table 1.

Demographic Characteristics of the Study Sample.

Characteristics	DRC		Myanmar
	Men	Women	Men	Women
	M (SD)	M (SD)	M (SD)	M (SD)
Demographic
Age	43.0 (16.6)	34.6 (14.2)	32.3 (14.7)	33.9 (15.5)
Years of education	6.2 (4.6)	3.8 (4.6)	6.1 (3.6)	6.8 (3.8)
Number of children	5.4 (4.2)	4.6 (3.0)	1.8 (2.1)	2.4 (2.6)
	% (N)	% (N)	% (N)	% (N)
Married	89.5% (51)	71.2% (42)	62.5% (10)	56.3% (9)
Single	7.0% (4)	11.9% (7)	31.3% (5)	25.0% (4)
Widowed	3.5% (2)	10.2% (6)	6.25% (1)	18.8% (3)
Have a disability (self-identified)	19 (33.3)	10 (16.9)	4 (25.0)	1 (6.3)
Activity
Focus group	22 (38.6)	23 (39.0)	–	–
Individual interview	35 (61.4)	36 (61.0)	16 (100.0)	16 (100.0)
Total	57	59	16	16

Note. DRC = Democratic Republic of the Congo.

Study Tool

In-depth interviews used an open-ended vignette approach to elicit information on shared risk factors and social norms related to violence in the home. The vignette began by verbally describing a hypothetical family, consisting of a husband and wife, their three children, and the husband’s sister, an older woman with a physical disability. The youngest of the three children in the vignette was 5 years old and also had a physical disability. Probes were developed to uncover whether participants’ responses would vary based on whether the child was a girl or boy child. Although the UN Convention on the Rights of Persons with Disabilities also includes mental, intellectual, and sensory impairments (UN General Assembly, 2007), physical disability was selected for the vignette to align with inclusion of participants for the study, and because phrases referring to physical disabilities may be more readily understood and consistently used by both research teams and participants than other forms of disability. Furthermore, because the study sought to include people with disabilities, a physical disability might be associated with less stigma in the two contexts than other forms of disability, which, in turn, limit risk of harm to persons with disabilities when disclosing disability status during recruitment and interviews. Participants were asked to hypothesize sequences of events that might occur within the household, as well as community and macro-level factors related to behaviors and relationships between family members. Focus groups were conducted to explore community perceptions of potential referent groups and influencers and other potential shared risk factors. For the purposes of this article, family and household are used interchangeably. Of note, sexual violence within the family was not explored within this study as it was hypothesized to have distinct risk factors.

Ethics

All participants completed oral informed consent in their local language before engaging in research activities. Because levels of education and literacy were low among participants overall, oral consent was more appropriate than written consent. Interviewers sought consent to audio record interviews, and all but one participant agreed to audio recording. For the participant who did not consent to audio recording, extensive notes were taken during the interview and transcribed the day of the interview. Interviews were completed by an interviewer (and translator in Myanmar) in private rooms managed by organizations, out of earshot of other persons in the community. In both contexts, interviewers debriefed participants on available services and asked whether participants wanted a referral after interview completion. The study protocol was approved by IRC Institutional Review Board, Comité National d’Ethique de la Santé (CNES)–Direction Provinciale du Sud-Kivu in DRC, and an advisory group of community leaders in Myanmar. No respondents were paid for their participation.

Analysis

All audio-recorded data were transcribed and translated from the local language to English. In DRC, transcripts were also coded from the local language to French to English. One interview in Myanmar was not audio recorded at the request of the participant, and notes were taken and transcribed in English. During data collection, the second and fourth authors led discussions on emerging themes and points of clarification, with feedback from the first author. The first, second, and third authors reviewed completed transcripts to familiarize themselves with the data; independently coded the same transcript using broad codes developed a priori to reflect previous research on intersections of violence against women and children and related social norms and risk factors (Bacchus et al., 2017; Guedes et al., 2016; Heise, 2011), and illuminate components of social norms as described by Cislaghi and Heise (2018); jointly discussed discrepancies in code applications and perceptions of code meanings; and developed consensus on creation or elimination of codes, code meanings, and code applications. The first three authors met regularly during coding, and as new themes emerged from the data, adapted the codebook based on joint consensus. During analysis, authors sought to distinguish participants’ perceptions of more proximal factors and norms related to violence from more distal factors and norms through participants’ own descriptions of how different factors might relate to each other, and by noting participants’ more immediate reactions to scenarios and questions as more time proximal. Code frequency and descriptions from participants informed most salient factors and relationships between factors. After review of transcripts and codes, the first three authors agreed that the ecological model served as an appropriate framework for the data, and jointly constructed a visual model (Bronfenbrenner, 1979). Once developed, the ecological model and accompanying quotes were presented to research and programmatic teams based in the DRC and Myanmar for feedback, contextualization, and validation. All analyses were conducted in Dedoose.

Results

Table 1 present study sample characteristics, by various demographics across men and women and by country. The average age of women in DRC was 34.6 years (range = 18–70 years) and the average age of men was 43.0 years (range = 18–83 years). In Myanmar, the average age was 33.9 years (range = 18–58 years) and 32.3 years (range = 18–69 years) for women and men, respectively. The majority of participants were married with children. More men self-identified as having some form of a disability than women.

Based on the interviews and focus groups, we developed a conceptual framework using the ecological model to map shared risk factors, including social norms, for experiencing or perpetrating violence in the home (Figure 1). The findings illuminated a web of social norms and risk factors underpinning violence in the home at the macro level, community, family, and direct (or time-proximal) levels. To facilitate understanding of the social norms and other risk factors identified at each level, social norms are illustrated separately from other shared drivers. The ecological levels denote that all levels interact with each other and that shared social norm–related risk factors also interact with other types of risk factors at each level. Select quotes from the qualitative data are provided to illustrate the role of risk factors, including social norms at each level of the ecological model, as well as intersections between them. All quotes are from individual interviews unless stated otherwise.

Figure 1.

Ecological model of shared risk factors driving violence in the home in humanitarian context.

Macro Level

The key shared risk factor at the macro level presented itself as gender inequality, which materialized and drove manifestations of norms related to power and control over others at the community and family levels and exacerbated other shared risk factors for violence in the home. Throughout interviews, participants described gendered social roles within and outside of the home; although participants often described the operation of these norms at lower levels of the model, the identified norms were a reflection of traditional gender norms that reinforced the power of men over women and children in society. For example, participants almost exclusively identified men when asked about leaders and influential people whom they had respect for. A 20-year-old woman in the DRC explained that women should adhere to the desires of their husbands because

[a] man is always a man … Even if you are tired you shouldn’t put yourself higher than your husband, even if you find him acting wrong … He is the head of the family, so he should not allow disorders in his house even if he doesn’t bring anything [any financial resources] at home.

Further quotes on manifestations of gender inequality at more local levels are in the below sections on community and family norms and factors.

Other shared risk factors for violence in the home, rooted in patriarchal culture of power and control over others, were related primarily to armed conflict, the resultant instability, variable legal statuses, and belonging to a religion condoning gender inequitable attitudes and beliefs. For instance, in DRC, a 65-year-old woman spoke about her families’ experience of armed conflict:

… when we ran away to [Village 1] escaping the war, [man’s name] robbed me of them [my assets] and I was alone with my grandsons. And when we arrived in [Village 2], we spent three days without a shelter and without food. It is after that they opened houses and schools to us so we can sleep. And after five days they gave us food. He spent two months and they took us to [Village 1] and we spent two weeks without food. And we came back here in [Village 3] after some time and we found no house.

A 25-year-old man from Myanmar stated,

Before the armed conflict happened, we could go anywhere and find a job. Then, we could earn money and we felt comfortable. We didn’t face difficulties. After the armed conflict, we can’t go anywhere and find a job. We have faced more difficulties and poverty. We just stay at home mostly.

The pervasiveness of armed conflict and instability in the DRC created a climate of new and ever-evolving stressors, which emerged in the research as risk factors at the community and family levels for violence in the home. Examples of corruption and discrimination that can be situated at the macro level of the ecological model were perceived to exacerbate poverty and instability at the community level. For instance, participants in the DRC described the role of bribes to officials in settling disputes over land ownership, and described economic exclusion from nongovernmental organization (NGO) services; these challenges reinforced an environment of scarcity and lack of trust in existing systems. Violation of this norm in a context of economic scarcity facilitated mistreatment of persons in the home who could not contribute economically (discussed further below). In Myanmar, institutionalized pressure to convert to the dominant religion to find stable employment, obtain promotions, and access free schooling translated into local-level discrimination in hiring and withholding income or underpaying manual laborers.

Inequitable gender norms also magnified macro-level risk factors such as conflict and displacement. A 25-year-old woman in Myanmar related conscription into the military or another armed group to men’s reflections of their own power within their communities or homes and stated that men in armed groups were subject to the type of control that they might use against others within their own homes.

When they [men] are in the army they have to live with the discipline. They can’t do whatever they want. They would then reflect on what they have done while they were at home. He would think to himself that he could do whatever he wants to his followers, wife, and children while he was at home. He would think about how difficult it is to stay under the control of other people. He understands how hard it is to be under the control of others compared to staying at home.

Inequitable gender norms also interacted with other macro-level risk factors of violence in the home, which is illustrated by the vertical bidirectional arrow at each level in the model (in Figure 1). For example, subscription to inequitable religious beliefs exacerbated inequitable gender norms, as one 35-year-old man from DRC stated,

Show[ing] that he is the chief of the family that means the head of the house because the Bible says that a man should be the one in charge of the family and he should take good care of his house.

However, in both contexts, although some patriarchal religious beliefs were used to justify instances of violence or the position of the man in the home, religion was also referenced by both men and women as a moral deterrent to use of severe physical violence against both women and children. For instance, one 26-year-old woman from Myanmar stated, “The monks often explain [to] us that we shouldn’t be harsh and cruel. We should have mercy and compassion to each other.” Religious values were also the backdrop for ethical principles that denounced physical violence and neglect toward both children and adults with disabilities. Another 51-year-old woman from DRC stated, “No they wouldn’t let him [neglect a child with disabilities] because the boy has not asked to be born like that, God has decided that [the child] is born like that not because he has done something wrong.”

Community Level

At the community level, the most salient shared norms that emerged from the data included (a) acceptability of violence as harsh discipline, (b) family reputation in the community, and (c) men seen as economic providers of the home and head of household in the community, whereas woman were seen largely as noneconomic contributors to the home.

In regard to the acceptability of using violence as a mechanism of harsh discipline, whereas men may use violence against women and children and parents may use violence against children, participants largely asserted that others in their communities did not condone severe physical violence against either women or children. However, participants in both contexts indicated that acceptance of violence operated on a gradient: Although there was a range of violence behaviors toward women and children that were considered acceptable, extreme physical violence that would cause injury and/or was accompanied by screaming or crying that could be overheard by neighbors would prompt community action to protect both women and children. Closer proximity to neighbors in more densely populated environments such as camps for internally displaced persons increased detectability of violence in the home; participants frequently linked this detectability to intervention from community members.

For instance, one 25-year-old man from Myanmar stated,

If the parents beat the children severely, [community members] will tell parents not to beat children severely … not to overdo. They will tell them like … if it is required to beat the children, you should just beat them, not more than enough.

Perceptions that less severe violence was an appropriate response to disagreements with women were at times revealed when participants blamed women for violence used against them. A 20-year-old woman provided an example of deeply engrained patriarchal beliefs, stating that a woman would be expected to apologize to her husband for inciting his violent behavior,

If people found out that a man had beaten his wife, they] would come and apologize on behalf of [the man’s] wife and show her what she did wrong, and then she could ask for forgiveness to her husband and to God.

Although there was a range of violence against women and children that was acceptable, participants overwhelmingly condemned any physical violence and neglect of persons with disabilities. An 18-year-old man in DRC stated,

People could say [to the husband], “you deprive food to your sister although you are from the same womb?” And they will start speaking ill of him and it is bad. Because, if you deprive food to your sister, it is someone else who pities her who will give her food. Is that fair?

However, some noted that adult women with disabilities in the home were at risk of verbal abuse from another adult woman, such as a sister-in-law.

Protecting a family’s reputation was also a norm that drove violence against women and children at the community level. In both contexts, the actions of children could be attributed to their parents and could cause damage to the family’s reputation if the behavior was not in line with social expectations for children. In these cases, the use of corrective behaviors by caregivers, including violent discipline, was permissible and even condoned. In the DRC, participants were explicit that a daughter’s behavior before and after marriage is a reflection of her family of origin; this norm was connected to role fulfillment (a risk factor at the time-proximal level) of daughters as well as mothers. For example, one of a mother’s roles was to guide daughters on their gendered responsibilities as future wives and mothers. Failure to do so could lead the daughter’s future husband and his family to disparage the family name, which would bring shame to the daughter’s family. A 58-year-old man in DRC stated,

You know, we must take care of our daughters. They are ours; we are their parents. One has to punish her because tomorrow she will go to another family. You tell her: “you should not do what you are doing now when you are in your in-law family. If I do not punish you, you will be stupid in your in-law family and they will say it is because of an irresponsible parent.” That is why you must punish that girl by organizing a meeting about her. That is the punishment to subject to a girl.

In Myanmar, children’s disobedient behavior could reflect poorly on parents’ parenting skills, and could result in other community members’ refusal to allow their children to interact with that family. Family and community unity were often described as linked to each other, as participants felt that a lack of family unity could disrupt harmony in the community.

Further complicating family reputation was that it could at times serve as a protection factor against violence. For instance, within a home, neglect of persons with disabilities, whether an adult or child, would also damage a family’s reputation by bringing shame to men if other community members stepped in to contribute resources men were expected to provide to their families.

In addition, throughout both settings, norms regarding men being seen as the head of household and economic provider and women in a subservient role providing nonfinancial contributions to the home were prominent. When either party were seen as not fulfilling their role, instances of violence could occur, as explained in the “Time-Proximal Factors” section.

As illustrated in the conceptual model (Figure 1), these norms intersected with other risk factors for perpetration and victimization of violence in the home at the community level, as follows: (a) changes in stable employment or income opportunities for households, (b) overall political instability and displacement in communities, (c) moving to a more urban or population-dense environment either based in a community or camp, (d) availability of addictive substances (specific to Myanmar), and (e) availability of community resources to support families, especially for people with disabilities.

First, changes in stable employment related to armed conflict were challenges noted in both settings, which exacerbated stress within the family and subsequently could raise the risk of perpetration of violence at the time-proximal level. For example, norms supporting men’s role as economic providers, stemming from gender inequality influenced men’s reactions to employment instability—as a 60-year-old man in the DRC described, even if men knew that they were not generating much income from their farms, the perception that men should be economically productive would drive men to tend to their fields. Another 29-year-old man in DRC noted the potential stress of being a provider:

Since he is a farmer he could not have a good situation because he … has to wait for a whole year to start enjoying whatever he had worked for, being a man is knowing how the family would survive and giving them full support economically and psychologically

In addition, in DRC, displacement and loss of jobs seemed to have created a context of scarcity in which women or persons with disabilities who were not generating income could be mistreated.

Second, particularly in Myanmar, rural–urban migration due to displacement also resulted in economic shifts: Previous self-sustenance deriving from land ownership and farming was not feasible in a more crowded camp environment, and displaced men sought employment as day laborers, which created strain on families. This is illustrated by the following quote from a 19-year-old woman in Myanmar:

In the camp, we don’t have the jobs regularly because we can go to earn money only if the other people need labor. However, when we live in the village, we can work on farm regularly every year. Then, we can cut down the fire wood and take home. When we lived in the village, most people didn’t need to spend much money … [now] we have to buy everything because we don’t have enough places to grow vegetables. So, when we lived in the village, we didn’t need to buy curry because we had them grown in our own farm, whatever we need.

Third, this context of unstable employment and increased access to addictive substances such as heroin and methamphetamines, specifically in Myanmar, resulted in greater levels of drug use among men, which participants in turn associated with further economic instability and men’s use of violence in the home.

Finally, the fourth risk factor, lack of community resources to support families, could potentially exacerbate violence in the home if such services are not in place to prevent or mitigate violence or other risk factors. For instance, participants in Myanmar often cited closer proximity to schools and health facilities as factors that enabled them to provide better care for children and persons with disabilities, respectively, potentially reducing stress and abuse/neglect within the home. Trainings and services (nonspecified in transcripts) from NGOs serving the community were also listed as resources with potential to prevent violence in the home. Yet, meaningful inclusion of persons with disabilities to be able to access services was noted as a challenge, and participants often described familial support as the only means available to care for family members with disabilities. This lack of resources to support persons with disabilities exacerbated perceptions of persons with disabilities as an economic burden on the family.

Family/Relational-Level Factors

At the familial level, acceptance of three norms emerged: (a) power hierarchies related to age and gender in the home, (b) violence as an appropriate corrective response to women and children’s noncompliance with socially prescribed roles, and (c) that persons with disabilities or older persons have limited abilities.

Participants routinely explained violence in the home in the context of hierarchies within the household that give men more power over women, and adults over children. These hierarchies conferred expectations that family members with less power (women and children) be “obedient” to those with more power (adult men); this resulted in a ladder of power and privilege in the home that placed men at the top, followed by their wives, and lastly, their children. Related to this, risks to family reputation that also emerged at the community level materialized as additional controls of behavior toward people lower on this ladder of power, particularly for girl children. Speaking of the relationship between a mother and her children, one 59-year-old woman from Myanmar stated,

She will beat all of the children who are misbehaving. Because they don’t listen to her. If she asks them to stop playing and they didn’t listen, if they didn’t come home on time, or if they didn’t study, she will beat them.

Women were at heightened risk of violence in the middle rung of this power ladder as men may use violence to ensure conformity to social expectations of the women’s gendered roles within the home as well as the fact that women were also held responsible for the disobedience or lack of conformity for children in their care. One 23-year-old woman in DRC stated, “He thinks, as he is the head of the family, he can come, starts insulting her and beating her because he knows he is responsible for everything.” As such, men are sanctioned to use multiple forms of violence against women and children when they perceive them to be disobedient: Children could be beaten by their fathers or mothers, and women could be beaten by husbands.

Having a disability complicated this power hierarchy, as interviews in both Myanmar and DRC revealed perceptions that persons with disabilities had limited perceived ability to both care for themselves and care for others. For children, such norms seemed to be protective against violence through two distinct mechanisms: perceptions that children with disabilities were incapable of causing mischief or disorder within the home and thus unable to disobey parents, and exemptions of requirements to complete chores due to perceptions of limited autonomy, which in turn meant these children were less likely to face maltreatment in response to lack of role fulfillment in completing chores. For example, a 45-year-old woman in Myanmar noted,

[The punishment] is different because we pity them very much … If I beat the eldest child about 4 and 5 times and I would beat the disable son once because I get angry. [You beat the disabled child less] because his is not like his brother. He is the poorest child among of them. We beat him because we are not satisfied their action and we have no choice.

In contrast, the perceived limited ability of adult women with physical disabilities (represented by the elderly sister in the vignettes) could exacerbate their risk of violence and neglect. In fact, inability to perform tasks related to age and potential mobility issues could reverse a power hierarchy between a wife and her in-laws. For example, a physical disability could make a sister-in-law dependent on her brother’s wife for care, which created risk that the wife could be abusive toward her sister-in-law. Illuminating this complicated relationship with older persons in the home is a quote from a 19-year-old woman in Myanmar: “She [the wife] will tell her husband ‘your elder sister can’t help us anything even little thing.’ She will feel that [her husband’s] elder sister is a burden.” In both settings, adults who were not working were considered burdensome to the family. This put persons with disabilities at particular risk of neglect due to perceptions of their lower position in the home if they were unable to economically contribute to the family.

These manifestations of inequitable power hierarchies within the home continued to interact with other shared risk factors for perpetration and victimization of violence in the home: (a) inconsistent income and contributions, (b) household structure and stress, (c) unhealthy problem-solving patterns, and (d) exposure to armed conflict related to the overarching humanitarian settings.

First, instability and economic insecurity at the community and macro levels of the ecological model led to inconsistent income and greater reliance on men and women alike to generate active income at the level of the family. The daily stress of inconsistent income, which for men often meant strenuous physical labor for little pay, was identified as a cause of an irritable demeanor that contributed to their use of violence in the home, as a 20-year-old woman from DRC states,

[A man will beat his wife and chase his children away because] he leaves the field tired and gets home when there is nothing to eat although he is hungry. He can get angry. It is bad life conditions. But if he has had a good job or a good activity which will not require much efforts, it cannot be that way.

Loss of land ownership and related ability to till one’s own land were cited as additional pressures that led to household stress, generating pressure for adults to find wage employment. In Myanmar, perceptions that all adults living in the home should substantially contribute to family functioning and security were often equated to income generation or other forms of tangible production that increased familial self-sufficiency; failure to do so could put one at risk of violence.

In Myanmar, this sometimes meant that parents of working age crossed the northern border to China to find employment, leaving their children in the care of grandparents or siblings who remained in the camps. Such arrangements could change the structure of the household and exacerbate shorter term stress in the hopes of gaining remittances later on. For example, older persons with more limited capacities to engage in physically burdensome work were caring for children who also were too young to work. Participants in DRC also noted having many dependents, such as children, as a cause of stress in the family that could then increase the likelihood of violence. In addition, this dynamic relationship between household structure and income interacted with a household having a person with a disability or an older person who may consume household resources, but be unable to provide economically to the well-being of the home: A 35-year-old man in DRC shared, in a context of economic scarcity, a person with disabilities’ perceived economic undervaluing could prompt neglect from her family:

What we mostly observe in such families, it’s about neglect, he [the husband] would neglect his sister and wouldn’t give her the consideration she deserves because he knows that she wouldn’t do anything for him and even for herself, he gets no benefits from her so he would neglect her.

Third, participants also cited unhealthy problem-solving patterns as factors that exacerbated violence, particularly related to not knowing useful strategies and thus resorting to yelling or using physical violence. Finally, exposure to armed conflict and traumatic experiences often negatively affected relationships within the family unit. This was particularly prominent in DRC when issues of sexual violence by nonpartners during the conflict was discussed. One 18-year-old female stated, “Sometimes a man can tell his wife to go back home without even thinking, but on my side, I think it’s hard to take back a woman who got raped.”

Time-Proximal Factors

Time-proximal factors in the ecological model illustrate risks of instances of violence occurring within the home. As described in the section on family-level risk factors, the generalized acceptance of violence as a corrective action, compounded by gender and age power hierarchies often led to men’s use of violence in the home when women or children were perceived as not fulfilling those prescribed roles. Expectations were fairly consistent; for example, participants indicated that cooking was a woman’s responsibility in more than 80% of individual interviews, and more than 60% of respondents referenced use of violence as a disciplinary behavior against women. Similarly, more than 70% of respondents discussed violent disciplinary strategies for children. Men were expected to be economic providers, women were expected to manage cleanliness and cooking in the home, and children were expected to complete chores to support their mother’s role in the home. With respect to a husband’s use of violence toward his wife, a 58-year-old woman from Myanmar stated men would use violence against a woman who had not fulfilled such duties:

Most of the men are impatient, so if they get hungry, he might beat his wife if the food is not ready. The wife should have the food ready because if the husband is working to earn money, it is her responsibility to cook the food. Also, her husband will be angry when he is hungry, so she should cook as soon as possible.

Regarding children, a 62-year-old man from DRC noted, “If children commit mistakes, the reasonable whip or punishment is to deprive them food. Twice or three times, they will change. If they do not change, you let them as they are.”

Expected roles intersected in divergent ways for children and women with disabilities. Although both children and adults or older persons’ disabilities were considered exempt from familial expectations, which protected them from physical violence, adult women with disabilities were at greater risk of other forms of violence such as emotional abuse or neglect because their limited autonomy at the familial level meant that they were violating expectations of their gendered role in the home.

When children and women did not fulfill their expected roles, risk of experiencing violence was exacerbated by other shared risk factors, which could serve as direct catalysts for violence: (a) instances when a family was unable to meet basic needs, (b) men’s alcohol and/or substance use, and (c) men’s or women’s abusive expressions of anger. These are further described below.

As previously described, macro-level political instability and displacement resulted in migration to more urban areas and a dearth of stable employment opportunities in these new environments, which could have generated economic pressures that exacerbated household stress, changed household structures, and meant that families had less consistent income or other means of self-sustenance. These economic challenges were described by more than 50% of individual interview participants. The amalgamation of these factors resulted in an inability of individuals in the family to meet basic needs, spanning from sufficient caloric intake to clothes and school fees. Tension arose between husbands and wives, and parents and children, on the best uses of available household resources, which could result in physical and emotional violence against women, children, and persons with disabilities.

Men’s abuse of alcohol and substance use were discussed by participants as a key factor related to violence perpetration in the home in all six focus groups, and in approximately 40% of individual interviews. For instance, a 28-year-old woman in DRC noted,

They beat their wives, if a man is drunk, when he gets home he starts asking for food, if you take time to give him whatever he wants he would slap you, he would start beating you and sometime he would get you out of the house, the mother and her children … He kicks you out of the house and you spend the night out when he is alone in the house; he won’t give you any money to buy food but you would see him bringing other women in the house (prostitutes).

In addition to alcohol use by men, participants in Myanmar described heroin use as a trigger of violence, as one 26-year-old woman stated, “If the father is a drug-addicted man, no one can [say] anything to them when they are hungry of heroin and drug. If someone tells them something, they will not listen and they will beat us.”

Abusive expressions of anger of the perpetrator, such as difficulties self-regulating feelings of anger, exhaustion, and frustration, were also cited as factors that would proximally lead to violence in response to perceptions of disobedience and role violation at the family level. Failure to demonstrate obedience by women, children, or other marginalized household members in the ladder of power described above would prompt anger. This was demonstrated by men toward women through use of violence, and by both parents toward children. For example, one 45-year-old woman in Myanmar stated, “He is hungry and tired. He might kick them. When we are hungry, when we are sleepy and busy, if the children misbehave, we would probably beat [them].”

A 56-year-old woman in Myanmar describes the interactions of men’s perceived views of a woman stepping outside of her expected role in the home alongside other proximal shared risk factors:

From what I have seen, if the man goes to work and the wife is wasting at home and she didn’t cook. And the husband has nothing to eat although he is hungry. They will fight, and he will beat his wife … He will continually yell and beat his wife. As he gets angry, he will throw their properties away and he will tell his wife that you are useless woman. After that he will expel his wife. He will say, “You can go wherever you want. You are just wasting time in my house … ” Because, he has to work and got tired. He will beat his wife if she doesn’t cook food because she has nothing to do at home. If I were him, I will also get angry because I am hungry. So, if the wife doesn’t cook, he can get really mad. So, he will beat [them].

Discussion

Overall, this study demonstrated significant shared risk factors and norms that underlie multiple forms of violence in the home, including IPV, child abuse and neglect, and abuse toward persons with disabilities or older persons in the home. As illustrated in Figure 1, mapping these risk factors onto the ecological model demonstrates the presence of higher level risk factors, such as gender inequality as a driving force at the macro and community levels as well as its influence at the family level and the time-proximal level.

Gender inequality and social norms present at each level interacted with and exacerbated other shared risk factors for violence in the home. Furthermore, most shared risk factors were neither necessary nor sufficient for IPV, child abuse, or abuse against other at-risk household members if gender inequality and social norms supporting these factors were not present. For instance, at the most time-proximal level, the shared risk factors such as inability to meet basic needs and alcohol and substance use, were directly linked to men’s perceived lack of women’s role fulfillment and parents’ of children, for which violence was used as a corrective measure for “undesirable” behavior.

The conceptual model also includes risk factors of violence related to the humanitarian context. Most notably, displacement and migration resulted in new stressors and vacillating relationships with community members and their relevant norms. The instability catalyzed by armed conflict and displacement also directly led to challenges in maintaining economic stability and consistent sources of income to support family members. Such humanitarian-related risk factors for IPV and child maltreatment have also been found in other settings (Annan & Brier, 2010; Bermudez et al., 2019; Mootz et al., 2019). Other salient norms in these conflict-affected settings also included family reputation, particularly as it relates to female sexuality. Similar results related to these norms have been found in South Sudan and Somalia (Perrin et al., 2019). Within humanitarian settings in particular, subscription to patriarchal norms, such as hegemonic masculinity or power imbalances between heterogeneous groups of people may also be reaffirmed or challenged by other drivers at the macro level such as armed conflict itself (Kirby & Henry, 2012).

Furthermore, the analyses revealed a clear power hierarchy in the home based on norms dictating familial roles and responsibilities and the use of violence against women and children; violence was condoned if such roles were not “fulfilled.” Similar to an analysis examining intersections of IPV and violence against children (VAC) in Uganda, Namy, et al. also highlight the role of patriarchal culture in violence in the home, which often manifested as men having power and control over others and adults having power and control over children (Namy et al., 2017). If this power hierarchy is thought of as a ladder of power, children were the lowest rung on the ladder, and had not only fewer responsibilities but also very little power or influence within the home. Within Myanmar and DRC, mothers were often framed as the “middle man” between fathers and children, and as such, women were in the uncomfortable position of being held responsible for children’s wrongdoings within the home and also being subject to violence for not fulfilling other gendered roles.

As highlighted by italicized font in the model, risk or protective factors for people with a disability indicated the need for further nuance when examining power dynamics within the home. For example, women or children with disabilities or being an older person complicated this relationship such that although norms at higher levels in the ecological model were perceived to be protective against the use of violence toward them in the home, dissonant attitudes or time-proximal factors such as economic stress within the family may still act as risk factors for experiencing violence. In these instances, role fulfillment related to gender superseded norms operating at higher levels of the model, such as religious beliefs that encouraged compassion for persons with disabilities. The relationship between disability, older age, and risk of IPV against women is particularly complex (Scolese et al., 2020), and given approximately 15% of the global population is estimated to be living with some form of a disability (World Health Organization, 2011), further research is warranted.

Using an intersectional lens, it is apparent these potential associations varied by age and gender of the person in the vignettes and may have conferred even lower power based on different identities in the ladder of power within the home. For example, younger children were seen as having lower power in the home than older children and roles and expectations around fulfillment of such roles varied for girl and boy children. However, in a recent analyses examining experiences of violence against adolescent girls in South Kivu, DRC, and among refugee adolescent girls residing in Ethiopia, age was not a statistically significant predictor of reported experiences of violence (Stark et al., 2017). Magnified risk of abuse based on identities of age, gender, and disability for children should be further explored, including quantitatively to assess these relationships.

The conceptual model also illuminates potential ways in which practitioners might focus inclusive programmatic efforts that seek to prevent and reduce co-occurring violence in the home. For instance, a recent scoping review that examined programs measuring both IPV and child maltreatment outcomes found that nearly all interventions were delivered in the context of parenting programs, which included some emphasis on gender norms (Bacchus et al., 2017). The current model posits that a deeper engagement with power dynamics in the home as influenced by patriarchy, including a critical examination of the relationship between gender and power, within the home may maximize positive impacts on multiple outcomes. Alternatively, approaches to transform gender norms to prevent IPV may also benefit from the addition of other strategies including economic empowerment approaches, sustained reduction of alcohol use by men, and skill-building related to coping with stress and emotions and problem solving within families. Developing community-based approaches to reduce violence, including social norm strategies that collaborate with influential figures such as monks in Myanmar, may be a novel approach to prevent IPV. These approaches can also be further deepened through an inclusive programming lens.

Given economic stress highly interacted with inequitable norms and underscored many perceived antecedents to violence, economic-based interventions that promote shared decision making among couples about how resources are used may be particularly helpful in settings affected by armed conflict where both instability and poverty are often found. For example, within villages of South Kivu, DRC, household economic instability, among other factors, has also been found to be attributed to increased risk of IPV (Kohli et al., 2015). Upon participation in an animal husbandry program, clear downward trends in IPV were noted (Glass et al., 2017). However, care must be taken within interventions that may challenge men’s traditional roles as providers in the home to not exacerbate risk of violence if women are perceived to be gaining power (Vyas & Watts, 2009). In addition, these economic-based approaches may also be differentially beneficial for households that include older persons or persons with disabilities, whereas the proportion of income-producing adults may be lower as these economic interventions may have a magnified effect. Heterogeneity of impact should be explored, alongside potential reductions of child abuse, in addition to IPV.

Limitations of the study should be noted. First, although the study examined multiple forms of family violence from the perspectives of male and female adults, with and without disabilities, and at various ages, children and adolescents were not included in the data collection efforts; thus, their voices are absent in the analyses. Furthermore, there was limited representation of women with disabilities in Myanmar within the target population. In addition, although this study focused on developing an understanding of co-occurring and shared risk factors of violence, it does not include a longitudinal lens through which one could examine the perpetuation of violence over generations nor does it include distinct insights on sexual violence within families. In particular, the relationship between witnessing violence in the home, particularly for children observing instances of IPV, and behavior as adults is lacking. Further insight into this area would be helpful to identify appropriate intervention points and could address potential mechanisms such as negative role modeling or bystander trauma (Namy et al., 2017). Further research is also needed to unpack the experiences and intersectional risks marginalized household members may endure, including further analyses of those with disabilities, older persons, and persons with diverse sexual orientation, gender identities, and sex characteristics. We also want to note that the purpose of this study was to identify shared risk factors within the two contexts; as such, factors specific to only one context are underrepresented in our conceptual model. Given these limitations, this model should be viewed as a first step toward incorporating an inclusive and intersectional lens toward examining risks of violence, but should be expanded upon in later iterations.

In conclusion, the ecological model developed through this study offers new insight into the shared risk factors and norms of violence in the home, including the incorporation of an intersectionality-informed inclusion lens that purposively includes deeper analyses and sampling methodology based on age, disability status, and gender. Future research should include a robust disability lens alongside recognition and investigation of other diverse identities of household members, with attention to how various manifestations of different types and severity of disabilities, could potentially influence risk of abuse in the home (Gupta et al., 2018). Finally, the model demonstrates potential levers and entry points in which to influence co-occurring violence in the home for women, children, and other marginalized household members through inclusive programming approaches. Key attention to address root causes of violence, and in particular, working to address the norms related to gender, power, and violence that result from gender inequality alongside complementary approaches that address other shared risk factors may be most effective to improve the immediate well-being of families in humanitarian contexts.

Footnotes

Authors’ Note

The content of this article is the responsibility of the International Rescue Committee and does not necessarily reflect the views of United States Agency for International Development (USAID) or the U.S. Government.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was made possible by the generous support of the American people through the United States Agency for International Development (USAID).

Author Biographies

Kathryn Falb, ScD, is a social epidemiologist focusing on understanding what works to prevent violence against women and children and leads the International Rescue Committee’s violence research portfolio.

Khudejha Asghar, MPH, is a mixed methods researcher on violence against women and children, pursuing a doctoral degree at Johns Hopkins University’s Bloomberg School of Public Health.

Natalia Maria Pardo is a master’s candidate in development studies at the Graduate Institute, Geneva, and contributed to research analyses of the study.

Jean De Dieu Hategekimana has a bachelor’s degree in development management and is supervisor for the Women’s Protection and Empowerment Program with the International Rescue Committee.

Haja Kakay holds a master’s in international conflict resolution and works for the International Rescue Committee’s Women’s Protection and Empowerment Program in Myanmar.

Danielle Roth, MA in international development, supports International Rescue Committee’s work in family violence prevention and response.

Meghan O’Connor, MA, is a social worker by training and serves as senior technical advisor for the Violence Prevention and Response Unit at the International Rescue Committee.

ORCID iD

Kathryn Falb

References

Annan

Brier

(2010). The risk of return: Intimate partner violence in Northern Uganda’s armed conflict. Social Science & Medicine, 70, 152–159.

Bacchus

L. J.

Colombini

Contreras Urbina

Howarth

Gardner

Annan

, … Watts

(2017). Exploring opportunities for coordinated responses to intimate partner violence and child maltreatment in low and middle income countries: A scoping review. Psychology, Health, and Medicine, 22, 135–165.

Bauer

G. R.

(2014). Incorporating intersectionality theory into population health research methodology: Challenges and the potential to advance health equity. Social Science & Medicine, 110, 10–17.

Bermudez

L. G.

Stark

Bennouna

Jensen

Potts

Kaloga

I. F.

, … Williams

M. L.

(2019). Converging drivers of interpersonal violence: Findings from a qualitative study in post-hurricane Haiti. Child Abuse & Neglect, 89, 178–191.

Bronfenbrenner, U. (1979). The Ecology of Human Development: Experiments by Nature and Design. Harvard University Press.

Cislaghi, B., & Heise, L. (2018). Four avenues of normative influence: A research agenda for health promotion in low and mid-income countries. Health Psychology, 37(6), 562–573.

Crenshaw

(1989). Demarginalizing the intersection of race and sex: A black feminist critique of antidiscrimination doctrine, feminist theory, and antiracist politics. University of Chicago Legal Forum, 1989(1), Article 8.

Fleming

P. J.

McCleary-Sills

Morton

Levtov

Heilman

Barker

(2015). Risk factors for men’s lifetime perpetration of physical violence against intimate partners: Results from the International Men and Gender Equality Survey (IMAGES) in eight countries. PLOS ONE, 10(3), Article e0118639.

Fulu

Miedema

Roselli

McCook

Chan

K. L.

Haardorfer

, … UN Multi-Country Cross-Sectional Study on Men and Violence Research Team. (2017). Pathways between childhood trauma, intimate partner violence, and harsh parenting: Findings from the UN Multi-Country Study on Men and Violence in Asia and the Pacific. Lancet Global Health, 5(5), e512–e522.

10.

Glass

Perrin

N. A.

Kohli

Campbell

Remy

M. M.

(2017). Randomised controlled trial of a livestock productive asset transfer programme to improve economic and health outcomes and reduce intimate partner violence in a postconflict setting. BMJ Global Health, 2(1), Article e000165.

11.

Guedes

Bott

Garcia-Moreno

Colombini

(2016). Bridging the gaps: A global evidence review of intersections of violence against women and violence against children. Global Health Action, 9, Article 31516.

12.

Guedes

Mikton

(2013). Examining the intersections between child maltreatment and intimate partner violence. Western Journal of Emergency Medicine, 14(4), 377–379.

13.

Gupta

Cardoso

L. F.

Ferguson

Shrestha

P. N.

Harris

, … Clark

C. J.

(2018). Disability status, intimate partner violence and perceived social support among married women in three districts of the Terai region of Nepal. BMJ Global Health, 3(5), Article e000934.

14.

Heise

(1998). Violence against women: An integrated, ecological framework. Violence Against Women, 4(3), 262–290.

15.

Heise

(2011). What works to prevent partner violence? An evidence overview. https://www.oecd.org/derec/unitedkingdom/49872444.pdf

16.

Jones

Bellis

M. A.

Wood

Hughes

McCoy

Eckley

, … Officer

(2012). Prevalence and risk of violence against children with disabilities: A systematic review and meta-analysis of observational studies. The Lancet, 380(9845), 899–907.

17.

Kirby

Henry

(2012). Rethinking masculinity and practices of violence in conflict settings. International Feminist Journal of Politics, 14(4), 445–449.

18.

Kohli

Perrin

Mpanano

R. M.

Banywesize

Mirindi

A. B.

Banywesize

J. H.

, … Glass

(2015). Family and community driven responses to intimate partner violence in post-conflict settings. Social Science & Medicine, 146, 276–284.

19.

Ministry of Health Sports and ICF. (2017). Myanmar Demographic and Health Survey 2015-16. https://dhsprogram.com/pubs/pdf/FR324/FR324.pdf

20.

Mootz

J. J.

Stark

Meyer

Asghar

Roa

A. H.

Potts

, … Bennouna

(2019). Examining intersections between violence against women and violence against children: Perspectives of adolescents and adults in displaced Colombian communities. Conflict and Health, 13, Article 25.

21.

Namy

, ‘ Carlson

C. O.

Hara

Nakuti

Bukuluki

Lwanyaaga

, … Michau

(2017). Towards a feminist understanding of intersecting violence against women and children in the family. Social Science & Medicine, 184, 40–48.

22.

Perrin

Marsh

Clough

Desgroppes

Yope Phanuel

Abdi

, … Glass

(2019). Social norms and beliefs about gender based violence scale: A measure for use with gender based violence prevention programs in low-resource and humanitarian settings. Conflict and Health, 13, Article 6.

23.

Roberts

A. L.

Gilman

S. E.

Fitzmaurice

Decker

M. R.

Koenen

K. C.

(2010). Witness of intimate partner violence in childhood and perpetration of intimate partner violence in adulthood. Epidemiology, 21(6), 809–818.

24.

Rubenstein

L. Z. N.

MacFarlane

Stark

(2017). Predictors of interpersonal violence in the household in humanitarian settings: A systematic review. Trauma Violence Abuse, 21(1), 31–44.

25.

Scolese

Asghar

Pla Cordero

Roth

Gupta

Falb

K. L.

(2020). Disability status violence against women in the home in North Kivu Democratic Republic of Congo. Global Public Health. Advance online publication. https://doi.org/10.1080/17441692.2020.1741661

26.

Slegh

Barker

Levtov

(2014). Gender relations, sexual and gender-based violence and the effects of conflict on women and men in North Kivu, Eastern Democratic Republic of Congo: Results from the International Men and Gender Equality Survey (IMAGES). Promundo-US and Sonke Gender Justice.

27.

Stark

Asghar

Bora

Baysa

A. A.

Falb

K. L.

(2017). Prevalence and associated risk factors of violence against conflict-affected female adolescents: A multi-country, cross-sectional study. Journal of Global Health, 7(1), Article 010416.

28.

Tlapek

S. M.

(2015). Women’s status and intimate partner violence in the Democratic Republic of Congo. Journal of Interpersonal Violence, 30(14), 2526–2540.

29.

UN General Assembly. (2007). Convention on the rights of persons with disabilities: Resolution/adopted by the General Assembly (A/RES/61/106). https://www.un.org/disabilities/documents/convention/convoptprot-e.pdf

30.

Van der Heijden

Abrahams

Harries

(2016). Additional layers of violence: The intersections of gender and disability in the violence experiences of women with physical disabilities in South Africa. Journal of Interpersonal Violence, 34(4), 826–847.

31.

Van der Heijden

Dunkle

(2017). What Works Evidence Review—Preventing violence against women and girls with disabilities in lower- and middle-income countries. https://www.whatworks.co.za/resources/evidence-reviews/item/349-what-works-evidence-review-preventing-violence-against-women-and-girls-with-disabilities-in-lower-and-middle-income-countries-lmics

32.

Vyas, S., & Watts, C. (2009). How does economic empowerment affect women’s risk of intimate partner violence? A systematic review of published evidence. Journal of International Development, 21(5), 577–602.

33.

World Health Organization. (2005). WHO multi-country study on women’s health and domestic violence against women: Summary report of initial results on prevalence, health outcomes, and women’s responses.

34.

World Health Organization. (2011). World report on disability.