Skip to main content
Download PDF
Download ePub

Drivers of men’s use of intimate partner violence in conflict-affected settings: learnings from the Democratic Republic of Congo

Conflict and Health volume 18, Article number: 9 (2024) Cite this article

Abstract

Background

Intimate partner violence against women (IPVAW) is prevalent in conflict-affected settings. Yet, there is limited knowledge about the risk factors that influence men’s use of IPVAW in conflict-affected settings. This paper adopts a transdisciplinary perspective to understand how experiences hypothesized to increase men’s use of IPVAW relate to each other and to men’s use of IPVAW. The findings may help researchers and interventionists to better select and target interventions for IPVAW in conflict-affected settings.

Methods

We used baseline data from the Tushinde Ujeuri project in the Democratic Republic of Congo. Men with at least partial data for the variables of interest were included in the analysis (n = 2080). We estimated a structural equation model that explored how five constructs – interpersonal violence, mental health, socioeconomic adversity, gender inequitable attitudes, and conflict violence – influenced men’s self-reported past-year use of physical and/or sexual IPVAW.

Results

The model had acceptable fit (χ2 = 1576.574, p = 0.000; RMSEA = 0.041; CLI = 0.882; SRMR = 0.055). There was a statistically significant path from interpersonal violence to IPVAW (β = 0.875; OR = 2.40). Interpersonal violence also was linked to gender inequitable attitudes (β = 0.364), which were linked to increased use of IPVAW (β = 0.180; OR = 1.20). Moreover, interpersonal violence was linked to trauma symptoms (β = 0.331), which were linked to increased use of IPVAW (β = 0.238; OR = 1.27). Use of IPVAW decreased as conflict exposures increased (β=-0.036; OR = 0.96), and there was no path from socioeconomic adversity to IPVAW.

Conclusions

Our findings suggest interpersonal violence exposures, trauma symptoms, and gender inequitable attitudes are all risk factors for the use of IPVAW in a conflict-affected setting. While continuing to focus on gender inequitable attitudes and norms, interventionists should also consider addressing men’s experiences of victimization and mental wellbeing. Doing so can help to improve trauma symptoms and may hold promise to reduce IPVAW in conflict-affected settings.

Background

Intimate partner violence (IPV) against women (IPVAW) is a critical public health issue in conflict-affected settings. Defined as behavior within an intimate relationship that threatens or causes physical, psychological, or sexual harm to women [1], IPVAW perpetrated by men is the most common form of violence against women in these settings [2]. Emerging evidence suggests exposure to conflict violence and living in conflict-affected communities may increase women’s risk for experiencing IPVAW [3,4,5,6,7], and the confluence of multiple adversities and insufficient protective factors may compound the effect of IPVAW on women’s physical and mental health [8, 9].

Understanding risk factors for men’s use of IPVAW in conflict-affected settings is critical for understanding why this violence occurs and for informing interventions to address it. The mechanisms driving men’s use of IPVAW in conflict-affected settings, however, are not well understood, potentially reflecting the relative paucity of research on men’s use of violence compared to women’s experience of violence [2]. Perhaps consequently, few IPVAW interventions engaging men have proven efficacious. Among ten studies engaging men to reduce IPVAW in conflict and post-conflict settings [10,11,12,13,14,15,16,17,18,19], three have shown a statistically significant reduction in any type of IPVAW when compared to a control group: community-based dialogue groups [19], gender dialogue groups added to group savings [10], and trauma-informed psychotherapy [17].

Current evidence has identified multiple risk factor targets for prevention of IPVAW in conflict-affected settings [20]. Some risk factors are common to non-conflict settings, such as age [21,22,23,24], polygamous marriage [24], income [24], childhood abuse [24], witnessing intra-parental violence [22, 23], permissive attitudes toward IPV [22], and psychological symptoms and substance misuse [23,24,25,26]. There also are risk factors more specific to conflict settings including exposure to torture [27], history of migration or displacement [24], and exposure to conflict violence [3, 4, 6, 28, 29].

Few studies model relationships among the different experiences and characteristics that impact men’s use of IPVAW in conflict-affected settings [27]. Understanding relationships among experiences and characteristics related to men’s use of IPVAW can improve interventions by helping to identify if an experience or characteristic is more proximal or distal, exerts an independent effect or functions through a mediator, and what other things will be affected by changing the experience or characteristics, thereby identifying potential intervention mechanisms, synergies, or untoward effects. Among studies that consider relationships among experiences and characteristics, none explicitly adopts a transdisciplinary perspective, investigating the multiple ways that exposures affect IPVAW in conflict-affected settings (i.e., they typically focus on experiences and characteristics from a single domain while IPVAW requires more complex modeling). In this paper, we present a model that examined how multiple constructs (i.e., experiences represented by one or more variables), drawn from different domains, interrelate to increase men’s use of IPVAW in a conflict-affected area of the Democratic Republic of Congo.

Different health and social science disciplines (e.g., public health, clinical psychology, econometrics) have developed hypotheses that focus on different domains (e.g., attitudes, behaviors, and norms; experiences of traumatic stress; experiences of poverty) to explain men’s use of violence. We focused on five constructs from different domains that may impact IPVAW in conflict-affected settings: men’s experiences of interpersonal violence, mental health, socioeconomic adversity, gender inequitable attitudes, and conflict violence. Briefly, interpersonal violence, such as child abuse, can prime intergenerational cycles of violence [30], exacerbate socioeconomic adversity [31], contribute to mental health symptoms [32], and heighten gender inequitable attitudes (e.g., where witnessing interparental IPV teaches gender inequitable attitudes like the belief that there are times when a woman deserves to be beaten) [33]. Conflict-related experiences also can produce mental health symptoms that compromise emotional regulation, lead to attribution of negative partner intent, or create additional stress in relationships, increasing risk for IPVAW [34, 35]. Armed conflict can exacerbate socioeconomic adversity [36], and economic stressors can overwhelm available resources, thus generating a crisis in families leading to IPVAW [37]. Gender inequitable attitudes, behaviors, and norms [38] can result in unequal power that increases risk for the use of IPVAW [39, 40]. Exposure to conflict violence can normalize violence [41], such that aggression may be seen as an effective way to resolve conflict [42].

This paper undertook structural equation modeling to understand how these five different constructs may contribute to IPVAW, and how they may relate to each other. To date, nearly all studies have used multivariable logistic regression, which does not account for complex relationships, such as the presence of mediating variables. By considering multiple domains that are the focus of different disciplines (i.e., adopting a transdisciplinary perspective) and using this advanced statistical approach, we can better understand the relationship of these constructs to IPVAW in a conflict-affected setting, helping researchers and interventionists to better select and target interventions for IPVAW in conflict-affected settings.

Methods

Design and setting

DRC is a large central African country roughly equivalent in size to western Europe. Much of the country has experienced longstanding armed conflict [43] including periods of intense conflict during the first (1996–1997) and second Congo Wars (1998–2003). Although the intensity of armed conflict has vacillated since the official end of the armed conflict, militias have continued to proliferate [44]. Gender-based violence also is prevalent in DRC. 57% of ever-married women of reproductive age (15–49 years old) reported lifetime IPVAW, as defined by physical, sexual, and/or emotional violence from their husband/partner. Of these women, 77% reported IPVAW within the past year [45]. This is compared to less than one quarter of women who reported non-partner sexual violence [29, 46].

This secondary analysis used baseline data from the Tushinde Ujeuri project funded by the United States Agency for International Development (USAID, DRG Learning, Evaluation, and Research I award). Additional information about the study is available elsewhere [47]. In brief, the Tushinde Ujeuri project aimed to help communities prevent and respond to gender-based violence in eastern DRC. At baseline, villages were selected purposively based on eligibility for project implementation. To select households for baseline interviews, field teams used a random walk methodology to select approximately 20 households for each of 224 villages. One woman or man from each household was selected at random from a list of usual members of the household. Male interviewers administered the survey to men, and female interviewers administered the survey to women. In total, baseline data was collected from 2108 men, 2114 women, and 1 nonbinary person. This study focused on the male sample, using 2080 men who had at least partial data on variables of interest for this analysis. The sample was not restricted by relationship status because it was not adequately measured or assessed (33% missingness). Moreover, 22% of men who identified as “single, never married” reported using IPVAW, suggesting restriction to currently partnered/married men would miss an important demographic, even if restriction were made possible through imputation.

Measures

Several individual-level questions relevant to IPVAW were included in the survey. Among these were: (1) demographics, (2) exposure to interpersonal violence such as witnessing interparental IPV, (3) mental health symptoms, (4) socioeconomic experiences, (5) gender inequitable attitudes and social norms, (6) conflict exposures such as experiences of human insecurity, and (7) use of IPVAW. We selected variables related to the five constructs outlined in the introduction: interpersonal violence, mental health, socioeconomic adversity, gender inequitable attitudes, and conflict violence.

Constructs hypothesized to contribute to IPVAW

We operationalized each construct in one of two ways: (1) by using a previously developed scale or index or (2) by examining all measured variables related to the construct and reducing to a core set. Scales or indices were available to organize measured variables for mental health symptoms and gender inequitable attitudes. For three constructs (interpersonal violence, socioeconomic adversity, and conflict violence), we used exploratory factor analysis (EFA) to identify variables that could represent the construct (see below for additional description of EFA).

Interpersonal violence

For interpersonal violence, four variables were available: familial physical child abuse (never, sometimes, often), witnessing parental IPV (no/yes), lifetime sexual violence victimization (no/yes), and adult physical violence victimization (no/yes).

Mental health symptoms

Mental health symptoms, hereafter referred to as trauma symptoms, were measured by the Harvard Trauma Questionnaire. The scale was selected, adapted, and tested in-country using an approach described elsewhere [47]. The Harvard Trauma Questionnaire included 16 items measuring posttraumatic stress disorder (PTSD) symptoms (Cronbach’s ⍺ = 0.88). For each item, respondents were asked to describe how much the symptom had bothered them in the last month: not at all (0), a little (1), moderately (2), or a lot (3). The average of the 16 items was used (0–3).

Socioeconomic adversity

For socioeconomic adversity, we investigated five variables: poverty (210,000+, 150,000-209,999, 60,000-149,999, 30,000–59,999, 10,000–29,999, 1-9999, 0 Congolese franc), standard of living (count of 11 household assets such as a radio or electricity; treated as a continuous variable), unemployment (no/yes), perceived household living conditions compared to neighboring households in the village (much worse, somewhat worse, about the same, somewhat better, much better), and perceived village living conditions compared to neighboring villages (much worse, somewhat worse, about the same, somewhat better, much better).

Gender inequitable attitudes

Gender inequitable attitudes were measured through an index of 14 statements to which respondents indicated their degree of agreement. For example, respondents were presented with the statement, “a woman should obey her husband in all things,” and asked to respond whether they strongly disagree (-2), somewhat disagree (-1), neither agree nor disagree (0), somewhat agree (1), or strongly agree (2). The average of these 14 items was used (-2–2).

Conflict volence

For conflict violence, we investigated eight variables: adult physical violence victimization (no/yes), lifetime sexual violence victimization (no/yes), witnessing conflict violence (count of six events such as witnessing harassment and torture; treated as a continuous variable), experiencing conflict violence (count of 10 events such as experiencing looting or theft of assets/supplies; treated as a continuous variable), perceived safety of the village (very unsafe, unsafe, neither safe nor unsafe, safe, very safe), human insecurity index (average of seven questions for which respondents selected never, sometimes, often), current or lifetime displacement as a result of armed conflict (no/yes), and time to reach the nearest drivable road (minutes). We included the last variable as a proxy for community violence exposure; past findings suggested relationships between road density and exposure to conflict violence in Africa [51].

Outcome

The primary outcome was past-year use of physical and/or sexual IPVAW. We selected this timeframe to reduce temporal bias. Men were asked: “In your relationship with your (last) wife/girlfriend/partner do (did) the following happen never, sometimes, or frequently: You hit, slapped, kicked, or did anything else to hurt your current or previous wife or girlfriend physically? You forced your current or previous wife or girlfriend to have sex with you or perform another sexual act when she did not want to?” After each question, men who answered sometimes or frequently were asked if they had done any of these things in the past 12 months. Men who answered yes to either physical or sexual IPVAW use within the past 12 months were coded as using past-year IPVAW.

Statistical analyses

Exploratory factor analysis and descriptive statistics

Where there was not an available scale or index to represent a construct (i.e., interpersonal violence, socioeconomic adversity, and conflict violence), we performed EFA using the WLSMV estimator, with adjustment for clustering at the village level. As applied in this study, EFA is a statistical technique that facilitates iterative examination of a group of variables and enables removal of variables until there is a cluster of intercorrelated variables that load on a single (latent) factor. We privileged a single factor loading so that each group of variables would represent a single construct. We used the resulting group of variables (i.e., measurement model) to represent the construct in the structural equation model (SEM). EFA was chosen over confirmatory factor analysis because there was not significant theory and empirical findings to test hypothesized groups of variables; rather, we aimed to uncover relationships present in the data.

We then calculated descriptive statistics for variables representing interpersonal violence, trauma symptoms, socioeconomic adversity, gender inequitable attitudes, conflict violence, and the outcome.

Structural equation model

We used a SEM approach to explore how interpersonal violence, trauma symptoms, socioeconomic adversity, gender inequitable attitudes, and conflict violence might influence IPVAW. SEM is a statistical technique in which one or more groups of variables representing a latent construct (i.e., measurement model) predict one or more outcomes (i.e., structural model). The model commonly is drawn as a series of ovals (latent variables) and rectangles (observed variables) connected by directional arrows. Each arrow specifies a hypothesized causal relationship, which is one assumption of the statistical model [48]. To define these casual relationships, we drew on the empirical literature and expert knowledge of the field.

Primary hypotheses

We hypothesized that interpersonal violence, trauma symptoms, socioeconomic adversity, gender inequitable attitudes, and conflict violence would directly affect IPVAW, as described in the introduction.

Secondary hypotheses

We also hypothesized relationships among constructs. These are detailed in Fig. 1, which presents the SEM without the measurement models. We hypothesized:

Fig. 1
figure 1

Hypothesized relationships among five constructs and men’s use of IPVAW

Full size image
  1. 1.

    interpersonal violence exposure would lead to trauma symptoms and gender inequitable attitudes, the latter because experiences such as child abuse and witnessing interparental IPV can normalize familial violence;

  2. 2.

    socioeconomic adversity would exacerbate gender inequitable attitudes: for example, when unemployment threatens a man’s fulfillment of traditional gender roles leading him to reclaim his masculine identity through increased justification and use of IPVAW. By contrast, socioeconomic adversity cannot cause trauma symptoms (as defined by the Diagnostic and Statistical Manual 5);

  3. 3.

    trauma symptoms would function in the opposite direction: exacerbating socioeconomic adversities where symptoms result in functional impairment;

  4. 4.

    conflict violence would increase trauma symptoms and gender inequitable attitudes, the latter when conflict violence causes multiple types of violence including IPVAW to be normalized and justified;

  5. 5.

    reciprocal relationships would exist between (a) interpersonal violence and socioeconomic adversity and (b) conflict violence and socioeconomic adversity, and displacement would partially mediate the relationship between conflict violence and socioeconomic adversity.

After estimating the model, we examined fit statistics (χ2, RMSEA, CLI, and SRMR) and modification indices. We centered theory and past empirical findings when selecting model adjustments. The final model balanced model fit and parsimony.

Sensitivity and additional analyses

We conducted two sets of sensitivity analyses. One controlled for age. The other estimated the main SEM on a restricted sample that excluded anyone who reported (a) adult physical violence and experience of being physically beaten by an armed group or (b) lifetime sexual violence and experience of sexual abuse by an armed group. This sensitivity analysis was undertaken because we were unable to fully distinguish between interpersonal violence and conflict violence exposures for men who reported both exposures (i.e., men who reported adult physical violence may have endorsed this item due to their experience of physical violence within conflict), leading to possible misclassification bias. We also conducted an analysis using depression/anxiety symptoms in place of trauma symptoms. We have chosen to present these findings in an online supplement to streamline the analysis. We privilege the analysis with trauma symptoms because these symptoms have a stronger theoretical and empirical basis for a causal effect.

Descriptive statistics were calculated in Stata 14. All EFA and SEM analyses were conducted in MPlus 8. Full information maximum likelihood estimation was used in EFA and SEM models to use all available data. Statistical significance was evaluated at the conventional p < 0.05 level. The Tushinde Ujeuri baseline and endline surveys were approved by the NORC Institutional Review Board.

Results

Table 1 describes the 2080 male study participants. Overall, the sample was highly trauma exposed. Most men reported physical childhood abuse (83.8%) and witnessing interparental IPV (51.0%). Approximately one-third reported adult physical violence (33.1%). Lifetime sexual violence was least commonly reported (8.4%). On average, men witnessed or experienced 3.8 conflict-related events, and more than half (53.8%) of the sample reported experiencing displacement in their lifetimes.

Table 1 Descriptive statistics for variables included in the structural equation model
Full size table

Regarding socioeconomic disadvantage, although there was a broad range of income, 11.5% of men reported no household income, and 67.1% of men reported unemployment. Perception of relative household and community disadvantage was clustered in the middle, with most men having responded somewhat worse (household: 29.6%, community: 21.8%), about the same (household: 20.6%, community: 31.5%), or somewhat better (household: 38.8%, community: 37.8%) than neighboring households or communities.

On average, there was slight agreement with gender inequitable attitudes (0.12; scale range − 2–2). Men reported none to little trauma symptoms on average (0.69; scale range 0–3). 16% (16.2%) of men reported past-year use of physical and/or sexual IPV.

Exploratory factor analysis

We performed EFA to select variables to represent three constructs: interpersonal violence, socioeconomic adversity, and conflict violence. In the EFA for interpersonal violence exposure, all variables were retained in a one-factor model (χ2 = 5.111, p = 0.078; RMSEA = 0.027; CLI = 0.991; SRMR = 0.039). For socioeconomic adversity, the final solution was a one-factor model with all variables except standard of living (χ2 = 7.705, p = 0.021; RMSEA = 0.037; CLI = 0.996; SRMR = 0.027). We found a two-factor solution with reasonable fit for conflict violence exposure. This solution, however, had many cross-loadings and few/low loadings on the second factor, suggesting the absence of two factors measured by distinct variables. We, therefore, decided to use a measured variable that was a combination of witnessing and experiencing conflict violence (0–16 events; treated as a continuous variable). We also included lifetime displacement as a separate variable.

Structural equation model

The SEM had acceptable fit (χ2 = 1576.574, p = 0.000; RMSEA = 0.041; CLI = 0.882; SRMR = 0.055) (Fig. 2). Regarding primary hypotheses: Interpersonal violence was linked directly to IPVAW (β = 0.875; OR = 2.40). Interpersonal violence also was linked to trauma symptoms (β = 0.331), which were linked to use of IPVAW (β = 0.238; OR = 1.27). There was no path, however, from socioeconomic adversity to use of IPVAW. There was a path from interpersonal violence to gender inequitable attitudes (β = 0.364) to increased use of IPVAW (β = 0.180; OR = 1.20). Whereas there was a direct relationship between conflict exposure and use of IPVAW, use of IPVAW decreased as conflict exposures increased (β=-0.036; OR = 0.96).

Fig. 2
figure 2

Structural equation model of men’s use of IPVAW (n = 2080)

Full size image

Sensitivity analyses

In the sensitivity analyses controlling for age (Supplement, Fig. 1), patterns of results did not change. In the sensitivity analyses considering potential overlap in physical or sexual violence experiences and conflict violence exposures (n = 1888) (Supplement, Fig. 2), the primary relationships for the five constructs and IPVAW did not change.

Discussion

Our study is among the first to empirically assess relationships among constructs, each of which speaks to a different hypothesis of drivers for the use of IPVAW in a conflict-affected setting. If the model’s assumptions are correct, including the direction of causality, men’s experiences of interpersonal violence lead to increased use of IPVAW. Moreover, men’s experiences of interpersonal violence lead to trauma symptoms and gender inequitable attitudes, which lead to IPVAW use. By contrast, men’s experiences of socioeconomic adversity do not contribute directly or indirectly to the use of IPVAW, and men’s experiences of conflict violence are inversely related to the use of IPVAW.

Interpersonal violence

The link between select types of interpersonal violence exposure (e.g., child abuse) and men’s use of IPVAW is well established in the current literature [3, 4, 29, 30, 49, 50], although less well documented in literature focused on conflict-affected settings. Scholars have explained this connection in varied ways, including through social learning theories that emphasize how direct experiences of violence contribute to the internalization of complex scripts condoning violence [51]. Through exposure to violence, violence can be normalized [41], and aggression may be seen as an effective way to resolve conflict [42]. When these internalized scripts include patriarchal attitudes justifying IPVAW (e.g., when a boy witnesses his father use IPVAW), the relationship between interpersonal violence and men’s use of IPVAW could be mediated through gender inequitable attitudes, consistent with this SEM.

Interventions focused on the relationship between men’s experiences of interpersonal violence and men’s use of IPVAW in conflict-affected settings are relatively rare. One recent exception is Safe at Home, a discussion group intervention in DRC, that explicitly recognized shared structural drivers for IPV and harsh parenting [19], thereby addressing multiple interpersonal violence exposures (e.g., child abuse and witnessing IPVAW) that could contribute to IPVAW use in the present and subsequent generations.

Trauma symptoms

Our finding of a significant relationship between trauma symptoms and the use of IPVAW adds to a large body of research that demonstrates a relationship between PTSD symptoms or diagnosis and IPV [35, 49, 50]. Most data, however, are limited to high-income countries [49]. Among studies in conflict and post-conflict settings, our findings agree with studies from Liberia, where posttraumatic stress symptoms were associated with the use of IPV after controlling for exposure to conflict events [15], and DRC, where treatment for posttraumatic stress reduced violent behavior [27]. They diverge from a study from Uganda, in which posttraumatic stress symptom severity was not associated with IPVAW [16]. It is possible that multiple contextual factors shape the relationship between posttraumatic stress symptoms and IPVAW. For example, in northern Uganda, women with reexperiencing symptoms had less IPV risk when their husbands also had posttraumatic stress symptoms due to reduced stigmatization of women by their male partners [16]; thus, male posttraumatic stress symptoms were protective in some cases.

Mental health interventions have been underutilized as a tool to reduce the use of IPVAW in conflict-affected settings. In combination with the current literature, our findings suggest trauma-focused mental health interventions could be integrated and tested as part of IPVAW interventions in conflict-affected settings. For example, the Common Elements Treatment Approach (CETA), a transdiagnostic psychotherapeutic intervention, has been used successfully to reduce IPV in a low resource setting [51] and may represent one model that can be combined with current gender equity-focused interventions. Likewise, narrative exposure therapy was proven to reduce violent behavior among male former combatants with PTSD in DRC [27] but may have applicability to the general population.

Gender inequitable attitudes

Men’s gender inequitable attitudes have been consistently linked to IPVAW [52], and researchers have suggested this link may be amplified in conflict-affected settings [53]. One line of reasoning suggests that, as conflict-affected communities experience changes around social roles that challenge gender inequity (e.g., women working outside the home), men may use IPVAW to maintain power in their relationship, home, and larger community [54].

Dedicated attention to gender norms and attitudes in IPV interventions in conflict-affected settings is critical. Although interventions focusing on gender inequitable attitudes and related social norms have shown mixed effects in controlled studies, some studies have observed robust effects [19] or changes in factors that may precede IPV, including changes in attitudes toward violence 12. Our findings suggest targeting gender inequitable attitudes alongside mental health may be a useful strategy. Aligned with this recommendation, Living Peace in DRC used psychotherapeutic groups to address men’s trauma experiences and gender inequities [55]. Evaluation findings have been positive [18], although a quantitative impact assessment was not available at the time of writing. We emphasize the combination of gender inequitable attitudes and mental health here because interpersonal violence exposures were heterogeneous in terms of timing and setting, suggesting more research is needed on which exposures drive the relationship between interpersonal violence and IPVAW prior to programmatic recommendations.

Socioeconomic adversity and conflict violence

No association between men’s reported experience of socioeconomic adversity and IPVAW was found, and a statistically significant association was found between increased conflict violence and decreased IPVAW. The latter finding differs from at least one previous study that linked exposure to war-related events with the use of IPVAW by men [15]. It is possible, however, that trauma symptoms explain the association between exposure to conflict-related events and IPVAW. Rees and colleagues [18] found that mental disturbance (including symptoms of PTSD, depression/anxiety, and alcohol abuse) mediated the relationship between experiences of torture and IPVAW in Timor-Leste [18]. If the posited direction of effect is correct for these exposures, microfinance, cash transfers, and similar interventions targeting men may be less useful in this setting. Likewise, focusing on conflict violence exposures may be less fruitful for IPVAW than addressing interpersonal violence exposures: for example, when identifying higher-risk men for selective prevention interventions.

Strengths and limitations

Our study has several strengths and limitations. To our knowledge, this is the first study to simultaneously explore multiple different disciplinary perspectives on the use of IPVAW in a conflict-affected setting. In so doing, it extends a literature dominated by univariable and multivariable regression models exploring variables correlated with women’s experiences of IPVAW and generates actionable results relevant to public health and humanitarian practice.

Among limitations, our measure of the use of IPVAW captures men’s self-report. Men who report using IPVAW may differ systematically from men who use, but do not report, IPVAW. The direction of bias is unclear as the characteristics of men who use IPVAW but do not report its use are unknown in this setting. It is possible, however, that any such misclassification attenuates the magnitude of our findings. For example, if there is a relationship between gender inequitable attitudes and IPVAW, findings would be attenuated if men demonstrating this relationship were incorrected classified as not using IPVAW. Additionally, our research only included variables measured at the individual level, whereas IPVAW is affected by experiences across the social ecology [56]. Furthermore, no data on alcohol use were available. This is important as alcohol may, at least partly, mediate the relationship between trauma symptoms and use of IPVAW, suggesting an additional focus for intervention. Future studies should attend to the potentially complex interaction between alcohol use and trauma symptoms to guide intervention components. Given multiple exposures were specific to childhood (i.e., witnessing IPV between parents and child abuse), our findings underscore the possibility that addressing violence against children may be a key IPVAW prevention strategy in this setting. More research is needed, however, to disentangle the effects of childhood and adult or lifetime exposures on men’s use of IPVAW in conflict-affected settings.

SEM posits directional relationships, and we employed cross-sectional survey data. We therefore made assumptions about the direction of effect based on theory and empirical literature. It is possible, however, that some relationships could be bidirectional (suggesting endogeneity or simultaneity bias). For example, we modeled the impact of gender inequitable attitudes on the use of IPVAW; the use of IPVAW also may heighten and entrench gender inequitable attitudes. We were unable to model bidirectionality in all cases due to the statistical complexity of the resulting model. Although reverse causality also is possible, this is less likely given the variables used in the model have strong empirical or theoretical bases for influencing the use of IPVAW.

Conclusions

Continuing to address women’s mental health needs and promote women’s advancement is imperative. Simultaneously, men increasingly are being engaged in IPVAW interventions in conflict-affected settings with mixed efficacy. In this paper, we sought to understand why IPVAW occurs to inform interventions engaging men to address it in conflict-affected settings. By using EFA, we better represented the complex constructs that may influence men’s use of IPVAW. Through a SEM approach, we modeled complicated relationships among these constructs, simultaneously considering domains emphasized by different disciplines and reducing bias by correctly specifying relationships (as opposed to multivariable regression where analyses including all variables would control for mediators, for example).

Whereas changing gender attitudes and norms and/or economic resources commonly have been foci for intervention [56], our findings suggest men’s interpersonal violence exposures, trauma symptoms, and gender inequitable attitudes are critical to their use of IPVAW in a conflict-affected setting. If the causal assumptions of our model are correct (e.g., the hypothesized direction between constructs is correct) and to the extent results are generalizable, interventions aiming to reduce IPVAW in conflict-affected settings should adopt new foci on men’s experiences of victimization and mental wellbeing. As experiences of interpersonal violence used in the model were heterogenous regarding timing and setting, however, we recommend additional research to disentangle the impact of violence victimization on IPVAW use, while considering the potential applicability of interventions addressing intergenerational cycles of violence and trauma symptoms. Importantly, these are among the few strategies engaging men that have had positive effects on IPVAW in conflict-affected settings [17, 19].

Data availability

The data that support the findings of this study now are publicly available.

Abbreviations

DRC:

Democratic Republic of Congo

EFA:

Exploratory factor analysis

IPV:

Intimate partner violence

IPVAW:

Intimate partner violence against women

PTSD:

Posttraumatic stress disorder

SEM:

Structural equation model

References

  1. World Health Organization. Intimate partner violence 2022 Available from: https://apps.who.int/violence-info/intimate-partner-violence/.

  2. Stark L, Ager A. A systematic review of prevalence studies of gender-based Violence in complex emergencies. Trauma Violence & Abuse. 2011;12(3):127–34.

    Article  Google Scholar 

  3. Kelly JTD, Colantuoni E, Robinson C, Decker MR. From the battlefield to the bedroom: a multilevel analysis of the links between political conflict and intimate partner Violence in Liberia. BMJ Global Health. 2018;3:1–11.

    Article  Google Scholar 

  4. Kelly J, Colantuoni E, Robinson C, Decker MR. From political to personal Violence: links between conflict and non-partner physical Violence in post-conflict Liberia. Glob Public Health. 2019;14(12):1639–52.

    Article  PubMed  Google Scholar 

  5. Gutierrez IA, Gallegos JV. The effect of civil conflict on Domestic Violence: the case of Peru. Santa Monica, CA: RAND Corporation; 2016.

    Book  Google Scholar 

  6. Ekhator-Mobayode U, Hanmer L, Rubiano E, Arango D. The effect of armed conflict on intimate partner Violence: evidence from the Boko Haram insurgency in Nigeria. World Dev. 2022;153:1–16.

    Google Scholar 

  7. Annan J, Brier M. The risk of return: intimate partner Violence in northern Uganda’s armed conflict. Soc Sci Med. 2010;70(1):152–9.

    Article  PubMed  Google Scholar 

  8. Falb KL, McCormick MC, Hemenway D, Anfinson K, Silverman JG. Suicide ideation and victimization among refugee women along the Thai-Burma border. J Trauma Stress. 2013;26(5):631–5.

    Article  PubMed  Google Scholar 

  9. Meinhart M, Seff I, Troy K, McNelly S, Vahedi L, Poulton C, et al. Identifying the impact of intimate partner Violence in humanitarian settings: using an ecological framework to review 15 years of evidence. Int J Environ Res Public Health. 2021;18(13):1–19.

    Article  Google Scholar 

  10. Gupta J, Falb KL, Lehmann H, Kpebo D, Xuan Z, Hossain M, et al. Gender norms and economic empowerment intervention to reduce intimate partner Violence against women in rural Côte d’Ivoire: a randomized controlled pilot study. BMC Int Health Hum Rights. 2013;13(1):1–12.

    Article  Google Scholar 

  11. Hossain M, Zimmerman C, Kiss L, Abramsky T, Kone D, Bakayoko-Topolska M, et al. Working with men to prevent intimate partner Violence in a conflict-affected setting: a pilot cluster randomized controlled trial in rural Côte d’Ivoire. BMC Public Health. 2014;14:1–13.

    Article  Google Scholar 

  12. Vaillant J, Koussoubé E, Roth D, Pierotti R, Hossain M, Falb KL. Engaging men to transform inequitable gender attitudes and prevent intimate partner Violence: a cluster randomised controlled trial in North and South Kivu, Democratic Republic of Congo. BMJ Global Health. 2020;5(5):1–14.

    Article  Google Scholar 

  13. Glass N, Perrin N, Marsh M, Clough A, Desgroppes A, Kaburu F, et al. Effectiveness of the Communities Care programme on change in social norms associated with gender-based Violence (GBV) with residents in intervention compared with control districts in Mogadishu, Somalia. BMJ Open. 2019;9(3):1–10.

    Article  Google Scholar 

  14. Green EP, Blattman C, Jamison J, Annan J. Women’s entrepreneurship and intimate partner Violence: a cluster randomized trial of microenterprise assistance and partner participation in post-conflict Uganda. Soc Sci Med. 2015;133:177–88.

    Article  PubMed  Google Scholar 

  15. Glass N, Perrin NA, Kohli A, Campbell J, Remy MM. 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. 2017;2(1):1–11.

    Article  Google Scholar 

  16. Gurman TA, Trappler RM, Acosta A, McCray PA, Cooper CM, Goodsmith L. By seeing with our own eyes, it can remain in our mind’: qualitative evaluation findings suggest the ability of participatory video to reduce gender-based Violence in conflict-affected settings. Health Educ Res. 2014;29(4):690–701.

    Article  PubMed  Google Scholar 

  17. Koebach A, Carleial S, Elbert T, Schmitt S, Robjant K. Treating trauma and aggression with narrative exposure therapy in former child and adult soldiers: a randomized controlled trial in Eastern DR Congo. J Consult Clin Psychol. 2021;89:143–55.

    Article  PubMed  Google Scholar 

  18. Tankink M, Slegh H. Living peace in Democratic Republic of the Congo: an impact evaluation of an intervention with male partners of women survivors of conflict-related Rape and intimate partner Violence. Washington, DC: Promundo-US; 2017.

    Google Scholar 

  19. Falb KL, Khudejha A, Alexandra B, Simon B, Martin N, Danielle R, et al. Improving family functioning and reducing Violence in the home in North Kivu, Democratic Republic of Congo: a pilot cluster-randomised controlled trial of Safe at Home. BMJ Open. 2023;13(3):1–12.

    Article  Google Scholar 

  20. Swaine A, Spearing M, Murphy M, Contreras-Urbina M. Exploring the intersection of Violence against women and girls with post-conflict statebuilding and peacebuilding processes: a new analytical framework. J Peacebuilding Dev. 2019;14(1):3–21.

    Article  Google Scholar 

  21. Bhatta DN. Shadow of Domestic Violence and extramarital sex cohesive with spousal communication among males in Nepal. Reproductive Health. 2014;11(1):1–8.

    Article  Google Scholar 

  22. Fleming PJ, McCleary-Sills J, Morton M, Levtov R, Heilman B, Barker G. 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. 2015;10(3):1–18.

    Article  Google Scholar 

  23. Fulu E, Jewkes R, Roselli T, Garcia-Moreno C. Prevalence of and factors associated with male perpetration of intimate partner Violence: findings from the UN multi-country cross-sectional study on men and Violence in Asia and the Pacific. Lancet Global Health. 2013;1(4):e187–207.

    Article  PubMed  Google Scholar 

  24. Wirtz AL, Perrin NA, Desgroppes A, Phipps V, Abdi AA, Ross B, et al. Lifetime prevalence, correlates and health consequences of gender-based Violence victimisation and perpetration among men and women in Somalia. BMJ Glob Health. 2018;3(4):1–12.

    Article  Google Scholar 

  25. Saile R, Neuner F, Ertl V, Catani C. Prevalence and predictors of partner Violence against women in the aftermath of War: a survey among couples in northern Uganda. Soc Sci Med. 2013;86:17–25.

    Article  PubMed  Google Scholar 

  26. Vinck P, Pham PN. Association of exposure to intimate-partner physical Violence and potentially traumatic war-related events with mental health in Liberia. Soc Sci Med. 2013;77:41–9.

    Article  PubMed  Google Scholar 

  27. Rees S, Mohsin M, Tay AK, Steel Z, Tam N, da Costa Z, et al. Risk of perpetrating intimate partner Violence amongst men exposed to Torture in conflict-affected Timor-Leste. Global Mental Health. 2018;5:1–13.

    Google Scholar 

  28. Torrisi O. Young-age exposure to armed conflict and women’s experiences of intimate partner Violence. J Marriage Family. 2022;85:7–32.

    Article  Google Scholar 

  29. Kelly JTD, Holmes M, Gibbons N, Matabaro N, Voors M. Conflict, displacement and overlapping vulnerabilities: Understanding risk factors for gender-based violence among displaced women in eastern Democratic Republic of Congo. Washington, DC: World Bank; 2021. Policy research working paper (No. 9819).

  30. Li S, Zhao F, Yu G. A meta-analysis of childhood maltreatment and intimate partner Violence perpetration. Aggress Violent Beh. 2020;50:101362.

    Article  Google Scholar 

  31. Zielinski DS. Long-term socioeconomic impact of child abuse and neglect: Implications for policy. 2015.

  32. Cohen P, Brown J, Smailes E. Child Abuse and neglect and the development of mental disorders in the general population. Dev Psychopathol. 2001;13(4):981–99.

    Article  CAS  PubMed  Google Scholar 

  33. Widom CS, Wilson HW. Intergenerational transmission of violence. Violence and mental health: Its manifold faces. 2014:27–45.

  34. Murphy CM. Social information processing and the perpetration of intimate partner Violence: it is (and isn’t) what you think. Psychol Violence. 2013;3:212–7.

    Article  Google Scholar 

  35. Spencer C, Mallory AB, Cafferky BM, Kimmes JG, Beck AR, Stith SM. Mental health factors and intimate partner Violence perpetration and victimization: a meta-analysis. Psychol Violence. 2019;9:1–17.

    Article  Google Scholar 

  36. Kohli A, Perrin N, Mpanano RM, Banywesize L, Mirindi AB, Banywesize JH, et al. Family and community driven response to intimate partner Violence in post-conflict settings. Soc Sci Med. 2015;146:276–84.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Fox G, Benson M, DeMaris A, Wyk J. Economic distress and intimate Violence: testing family stress and resources theories. J Marriage Fam. 2002;64:793–807.

    Article  Google Scholar 

  38. Connell R. Gender and power: Society, the person and sexual politics. John Wiley & Sons; 2013.

  39. Willie TC, Kershaw TS. An ecological analysis of gender inequality and intimate partner Violence in the United States. Prev Med. 2019;118:257–63.

    Article  PubMed  Google Scholar 

  40. González L, Rodríguez-Planas N. Gender norms and intimate partner Violence. J Econ Behav Organ. 2020;178:223–48.

    Article  Google Scholar 

  41. Dubow EF. Contagion of collective Violence: contagion from ethnopolitical Violence to other forms of aggression and Violence. Contagion of Violence: workshop summary [Internet]. Washington, DC: The National Academies Press; 2013.

    Google Scholar 

  42. Powers RA, Cochran JK, Maskaly J, Sellers CS. Social learning theory, gender, and intimate partner violent victimization: a structural equations approach. J Interpers Violence. 2017;35(17–18):3554–80.

    PubMed  Google Scholar 

  43. Kongolo M, Zamberia AM. State Fragility and Capacity Building in Sub-saharan Africa: the case of the Democratic Republic of Congo. In: Olowu D, Chanie P, editors. State Fragility and State Building in Africa: cases from Eastern and Southern Africa. Cham: Springer International Publishing; 2016. pp. 183–207.

    Chapter  Google Scholar 

  44. Raeymaekers T. Why history repeats itself in Eastern DR Congo. 2012.

  45. ICF. The DHS Program STATcompiler: Funded by USAID.; 2012 [January, 15, 2022]. Available from: http://www.statcompiler.com.

  46. Slegh H, Barker G, Levtov R. Gender relations, sexual and gender-based Violence and the effects of conflict on women and men in north Kivu. Eastern Democratic Republic of the Congo: results from the International Men and Gender Equality Survey (IMAGES): Promundo-US; 2014.

    Google Scholar 

  47. NORC at the University of Chicago. Countering gender based violence in Eastern Democratic Republic of Congo: Baseline report. 2019.

  48. Bollen KA, Pearl J. Eight myths about causality and structural equation models. Handbook of causal analysis for social research. Springer; 2013. pp. 301–28.

  49. Kimber M, Adham S, Gill S, McTavish J, MacMillan HL. The association between child exposure to intimate partner Violence (IPV) and perpetration of IPV in adulthood—A systematic review. Child Abuse Negl. 2018;76:273–86.

    Article  PubMed  Google Scholar 

  50. Roberts AL, Gilman SE, Fitzmaurice G, Decker MR, Koenen KC. Witness of intimate partner Violence in childhood and perpetration of intimate partner Violence in adulthood. Epidemiology. 2010;21(6):809–18.

    Article  PubMed  PubMed Central  Google Scholar 

  51. Straus MA. Social stress and marital Violence in a national sample of American families. Ann N Y Acad Sci. 1980;347(1):229–50.

    Article  CAS  PubMed  Google Scholar 

  52. Jewkes R. Intimate partner Violence: causes and prevention. The Lancet. 2002;359(9315):1423–9.

    Article  Google Scholar 

  53. Gibbs A, Dunkle K, Ramsoomar L, Willan S, Jama Shai N, Chatterji S, et al. New learnings on drivers of men’s physical and/or sexual Violence against their female partners, and women’s experiences of this, and the implications for prevention interventions. Global Health Action. 2020;13(1):1–12.

    Article  Google Scholar 

  54. Manji K. Articulating the role of social norms in sustaining intimate partner Violence in Mwanza. Tanzania: London School of Hygiene & Tropical Medicine; 2018.

    Google Scholar 

  55. Equimundo. Living Peace n.d. Available from: https://www.equimundo.org/programs/living-peace/.

  56. Heise LL. Violence against women: an integrated, ecological framework. Violence against Women. 1998;4(3):262–90.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

We are deeply grateful to all the research participants who shared their time, experiences, and expertise with the research team. We also would like to thank the diligent and dedicated team of data collectors who made this work possible. The NORC research team in collaboration with Maarten Voors and Morgan Holmes stewarded this work throughout the process. Finally, we particularly would like to acknowledge the role of a wonderful friend and colleague, Jean Paul Zibika, whose commitment to human rights remains an inspiration.

Funding

CB was supported by training grant T32MH103210 from the National Institute of Mental Health. The funders had no role in designing the study; collecting, analyzing, and interpreting data; or writing the manuscript.

Author information

Authors and Affiliations

  1. Department of Epidemiology, Columbia University Mailman School of Public Health, New York, NY, USA

    Christine Bourey, Rashelle J. Musci & Judith K. Bass

  2. Johns Hopkins University School of Nursing, Baltimore, MD, USA

    Nancy Glass

  3. Action Kivu, Bukavu, Democratic Republic of Congo

    Amani Matabaro

  4. Harvard Humanitarian Initiative, Harvard University, Boston, MA, USA

    Jocelyn T. D. Kelly

  5. Brigham and Women’s Hospital, Boston, MA, USA

    Jocelyn T. D. Kelly

Authors
  1. Christine Bourey
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Rashelle J. Musci
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Judith K. Bass
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Nancy Glass
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Amani Matabaro
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Jocelyn T. D. Kelly
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

CB conceptualized the study, analyzed the data, and wrote the first draft of the manuscript. RJM advised on methodology, and JKB advised on content. All authors contributed to substantive revisions of the manuscript and read and approved the final manuscript.

Corresponding author

Correspondence to Christine Bourey.

Ethics declarations

Ethics approval and consent to participate

The baseline and endline surveys were approved by NORC at the University of Chicago Institutional Review Board. All participants provided informed consent to participate in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Bourey, C., Musci, R.J., Bass, J.K. et al. Drivers of men’s use of intimate partner violence in conflict-affected settings: learnings from the Democratic Republic of Congo. Confl Health 18, 9 (2024). https://0-doi-org.brum.beds.ac.uk/10.1186/s13031-023-00562-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://0-doi-org.brum.beds.ac.uk/10.1186/s13031-023-00562-5

Keywords

Conflict and Health

ISSN: 1752-1505