Love in the Time of War: Identifying Neighborhood-level Predictors of Intimate Partner Violence from a Longitudinal Study in Refugee-hosting Communities

Abstract

A high prevalence of intimate partner violence (IPV) has been documented among women living in conflict-affected and refugee-hosting areas, but why this occurs is not well understood. Conflict and displacement deteriorate communities’ social cohesion and community connectedness; these neighborhood social environments may influence individual IPV outcomes. We explored neighborhood-level social disorganization and cohesion as predictors of recent IPV in refugee-hosting communities in northern Ecuador by conducting multi-level logistic regression on a longitudinal sample of 1,312 women. Neighborhood social disorganization was marginally positively associated with emotional IPV (AOR: 1.17, 95% CI: .99, 1.38) and physical and/or sexual IPV (AOR: 1.20, 95% CI: .96, 1.51). This was partially mediated by neighborhood-level civic engagement in the case of emotional IPV. At the household level, perceived discrimination and experience of psychosocial stressors were risk factors for both types of IPV, whereas social support was protective. To our knowledge, this is one of the first studies to examine how neighborhood social factors influence IPV outcomes in refugee-hosting communities or in South America. As the world grapples with the largest number of displaced people in history, this research can inform prevention and response programming and reinforces the critical importance of promoting acceptance of refugees and immigrants and positively engaging all community members in civic life in refugee-hosting settings.

Keywords

intimate partner violence social cohesion South America migration displacement

Background

Colombia experienced over 50 years of violent internal conflict between the government military, left-wing guerilla groups, and right-wing paramilitary groups, which began in the late 1940s and formally ended with a peace deal ratified in 2016. During the conflict, exposure to violence and stressors was ubiquitous, and violence often spilled across the Ecuadorian border (Conway, 2013; ABColombia, 2013). The conflict displaced over 7.6 million Colombians internally and abroad; an estimated 120,000 to 450,000 Colombian refugees resettled in Ecuador (Carvajal, 2017; UN High Commissioner for Refugees (UNHCR), 2016).

Conflict and miss migration have significant consequences for communities’ social fabric and safety nets—eroding social ties and networks, generating distrust, and creating social divisions (Horn, 2010; Hynes et al., 2016; International Organization for Migration Migration Health Department, n.d.). Many refugees and migrants lose their social status, support network, and communities, and face severe psychosocial stressors, such as poverty, food, and housing insecurity, violence, and discrimination (Horn, 2010; Shultz et al., 2014; UNHCR, 2016). Qualitative research found that in addition to facing substantial violence in Colombia, Colombians faced violence from police, sexual harassment at work, and low quality or limited access to health care in Ecuador (Shedlin et al., 2014). Over the course of the conflict, Ecuadorian public opinion regarding refugees soured and the percent of applications approved for asylum status declined dramatically—from 80 to 90% granted at the start of the conflict to less than 10% in later years (González et al., 2011; U.S. Department of State Bureau of Democracy Human Rights and Labor, 2013). Journalistic research reports that refugees also feared continued violence from other Colombians in Ecuador, and therefore severely restricted their own social interactions with other Colombians (Leutert, 2012).

Studies have found a high prevalence of intimate partner violence (IPV), or emotional, physical and sexual abuse by a current or former partner (Ellsberg & Heise, 2005), among women living in communities impacted by conflict and migration, but why this occurs is not well understood (Colorado-Yohar et al., 2012; Colorado-Yohar et al., 2016; Wirtz et al., 2014). Social disorganization theory provides a strong theoretical framework to explore how community-level social factors influence IPV in conflict-affected areas. Social disorder—traditionally defined as concentrated disadvantage (e.g., poverty) and residential instability—can weaken social bonds between neighbors. Social disorganization then limits collective community action and deteriorates the community’s informal social mechanisms to control violent behavior (Shaw & McKay, 1969). Social cohesion—or the level of trust and reliance between community members—acts as a mediator between neighborhood disorder and the likelihood of collaboration to address neighborhood challenges (Benson et al., 2003; Beyer et al., 2015). Social cohesion can foster the exchange of resources and information, build social capital, and increase collective efficacy (i.e., the belief a group can work together to achieve shared goals). This social process can serve to improve health outcomes; alternatively, neighborhoods and their corresponding social dynamics can foster negative outcomes, such as health inequalities, delinquency, or crime (Bernard et al., 2007; Sampson et al., 1999; Thulin et al., 2020).

In the context of IPV, scholars theorize that when living in a community with weak social bonds and low social cohesion, neighbors may not know when IPV is occurring or may not feel an obligation to intervene (Benson et al., 2003; Beyer et al., 2015; Edwards et al., 2014). Living in a violent environment could also normalize violence in the household. In communities with high levels of distrust in other neighbors or in institutions, women may be less likely to reach out for help or access police or social services (Edwards et al., 2014). Conversely, social resources and cohesion may enhance a community’s ability to prevent IPV (Browning, 2002; Dekeseredy et al., 2003; Edwards et al., 2014; Kirst et al., 2015; O’Campo et al., 1995; Stueve & O’Donnell, 2008; Thulin et al., 2020; VanderEnde et al., 2012). Alternatively, the disruption caused by conflict, migration, or other large-scale disasters, could lead to more violence through other channels, including the increase in individual and household stressors and familial conflict, displacement of anger by partners also exposed to violence, and barriers to leaving an abusive relationship such as precarious or liminal legal status, lack of resources, social isolation and discrimination (Bermudez et al., 2019; Menjívar & Salcido, 2002).

Research, mostly conducted in the United States, has found that community disorganization, weak community sanctions, and social norms supportive of violence increase the risk of IPV (Benson et al., 2003; Beyer et al., 2015; Heise & Kotsadam, 2015; O’Campo et al., 1995; Stueve & O’Donnell, 2008). Although social disorganization theory specifically identifies social cohesion as a mediator between social disorganization and delinquency outcomes, there is a dearth of studies examining social process indicators, such as social cohesion (VanderEnde et al., 2012). Emerging data suggest that social support can serve to reduce the risk of IPV (Browning, 2002; Dekeseredy et al., 2003; Kirst et al., 2015). Importantly, interventions have been successful at increasing a community’s solidarity, reciprocity, and social cohesion, suggesting that this is a modifiable protective factor and may be appropriate for IPV prevention purposes (Mousa, 2020; Pronyk et al., 2008; Valli et al., 2018).

This study examines the relationship between neighborhood social disorganization and social cohesion factors, and individual IPV outcomes among women living in refugee-hosting communities in northern Ecuador. We further identify if neighborhood social cohesion partially mediates the relationship between neighborhood social disorder and individual-level IPV, after accounting for individual, partner, and household characteristics.

Methods

This study draws on publicly available longitudinal data from a study conducted in 2011 in northern Ecuador (International Food Policy Research Institute [IFPRI], 2016a, 2016b). The data originated from a cluster randomized controlled trial of a World Food Program Cash, Voucher, and Food Transfer intervention conducted by IFPRI; additional information on the intervention can be found in previous publications (Hidrobo et al., 2014; Hidrobo et al., 2016). The current study does not use the data to evaluate the intervention but takes advantage of existing longitudinal data to build our empirical knowledge base about predictors of IPV.

Study Setting

The data were collected in two provinces in northern Ecuador near the Colombian border—Carchi and Sucumbíos. Carchi is in the northern highlands and Sucumbíos is in the Amazonian lowlands, representing differing geographic, cultural, and socio-economic contexts. Within the two provinces, seven urban and peri-urban areas were included, which are characterized by a high concentration of refugees and a poverty index that exceeds 50%. Neighborhoods, as defined by local government, were then selected for inclusion based on having a high percentage of Colombian and low-income households.

Sample Description

The sample includes 80 neighborhoods; Maas and Hox (2005) have found that estimates derived from multi-level models with Level 2 sample size of 50 or higher provide reliable estimates. Neighborhoods were divided into 148 clusters. Households were mapped in each cluster and a one-page questionnaire with basic demographic and socio-economic questions was conducted to determine inclusion; 20–27 low-income households were then randomly selected to be surveyed. Colombian and Colombian-Ecuadorian households were oversampled. At Time 1, 2,357 households agreed to be surveyed; 2,122 households participated again approximately seven–eight months later. In total, 1,312 women were partnered at both time points and were either the head of the household or their spouse, thus qualifying to be asked the IPV module.

Data Collection and Ethics

Paper surveys were administered and verbal consent was collected. Field staff were trained in accordance with the WHO ethical guidelines for conducting research on IPV. Training topics included ethical recruitment, consent, and data collection procedures; ensuring participant safety, privacy, and confidentiality; and safe data practices. Specific procedures were in place to ensure the safety of the participant during data collection. For example, IPV questions were only asked if no other family members were present, including her partner. In addition, all women, regardless of if the women had disclosed IPV or not, were offered anonymized referral information with contact information for local services. IRB approval was obtained from the IFPRI and in-country data collection partner, Centro de Estudios de Población y Desarrollo Social. The data are publicly available and the ethics review board at the University of North Carolina at Chapel Hill approved secondary data analysis.

Measures

Measures are described in detail in Table 1. In our analysis, we take advantage of both Time 1 and Time 2 data. Independent variables (e.g., social disorganization, household psychosocial stressors) were measured at Time 1, while our dependent (IPV) variables were measured at Time 2. The dependent variables assess emotional and physical and/or sexual IPV in the last six months (yes/no) using the WHO Violence Against Women Instrument (Ellsberg & Heise, 2005).

Table 1.

Description of Key Measures.

Variable	Description
Intimate partner violence (dependent variable, Time 2)
Intimate partner violence in the last six months (Time 2)	Thirteen measures of physical, sexual, or emotional intimate partner violence that occurred in the last six months. This was be dichotomized as yes/no. For example, “When two people are married or live together, they usually have good and bad moments. Has your spouse/partner: accused you of infidelity; tried to strangle you; utilized force to have sex when you did not want to.”
Household Variables (independent variables, Time 1)
Economic stability
Household socio-economic status	Following Demographic Health Survey guidance, a score of household wealth was constructed from a series of questions regarding the household asset data such as physical infrastructure of the house (household construction material, material of the flooring, number of rooms in the house, electricity, type of electricity, cooking type, type of bathroom, and type of water consumed), and ownership of land, tools, animals and 11 consumer items (cell phones, TVs, etc.)
Food security	Food security was measured by using the food consumption score, a continuous measure (ranging from 0 to 112) calculated using the frequency of food groups consumed by the household during the seven days before the survey, and weighted to capture relative nutritional importance.
Female/male labor participation	A participant was considered as participating in the labor force if they indicated that had (a) farmed their own land; (b) worked in an agricultural position for pay; (c) worked in a non-agricultural position for pay; or (d) worked on non-agricultural activities, on their own, like a small business, in the last six months.
Psychosocial stressors and supports
Psychosocial stressors in the last 6 months	Number of adverse situations (out of 13) in the last 6 months and 6–12 months ago that has affected the household; as this was only measured at Time 2, the 6–12-month measure serves as a proxy for Time 1 measures. Indicators included: robbed of cash; robbed of food; robbed of other goods; death of male adult household members; death of female adult household members; permanent debilitating injuries or illness of household member; physical attack; drought or flood; destruction of property or land due to natural disaster; destruction of property due to violence; alcoholism or drug addiction of family member; job loss or failed business; prison for household member.
Discrimination in the last six months	Coded as yes if participant responded positively to anyone or ten questions about experiencing discrimination based on different characteristics. Each question began with: “Has anyone in the house felt discriminated against because of their [e.g., nationality]?” Indicators included: race; gender; economic status; occupation; political affiliation; disability; nationality; religion; appearance; other.
Civic engagement	Coded as yes if participant indicated household participation in (a) agricultural or business associations, (b) religious or spiritual groups, (c) community or neighborhood associations, (d) political groups, and/or (e) other groups (e.g., NGOs, education or cultural).
Social support	Participants were asked: If you needed $100 in the case of an emergency, how many people (outside the household) could you ask for help?
Migration history	Household moved to the neighborhood in the last 20 years (yes/no)
Male/female nationality	Whether the individual is Ecuador or Colombian. Due to the small number of participants reporting Ecuadorian-Colombian nationality (n = 39), those participants were included with Ecuadorians; the small number of participants reporting “other” as their nationality (n = 11) were included with Colombians
Neighborhood variables^a (independent variables, Time 1)
Social disorganization	Neighborhood SES (neighborhood average of household SES scores)
	Percent of female-headed households in the neighborhood
	Percent of owner-occupied housing in the neighborhood
	Residential instability (percent of households moving to the neighborhood in the last 20 years)
	Neighborhood average of reported household psychosocial stressors
Social cohesion (refer to Supplementary Table A)	Trust in individuals (aggregated household factor score)Trust in institutions and community connectedness (aggregated household factor score)Percent of reporting recent civic engagementPercent of households reporting experiences of discriminationNeighborhood social support (neighborhood average of household-level indicator of social support)

Note. ^aNeighborhood variables constructed from entire sample (n = 2,357) and weighted to adjust for oversampling of Colombian households.

Using the full sample at Time 1 (n = 2,357), exploratory and confirmatory factor analyses were conducted in MPLUS to identify underlying latent constructs of social disorganization and social cohesion. Both EFA and CFA analyses accounted for categorical response distributions and the clustered structure of the dataset. We identified one underlying factor of social disorganization, which included neighborhood averages of wealth, female-headed households, household ownership, and residential instability (i.e., concentration of households that had moved to the neighborhood in the last 20 years). Social cohesion included four domains: trust in individuals; trust in institutions and community connectedness; experiences of discrimination; and civic engagement. Indicators of model fit were high. Social cohesion CFA results can be found in Supplemental Table A; Supplemental Table B documents domain correlations.

Indicators of migration history, psychosocial stressors, experiences of recent discrimination, household wealth, food security, social support, and civic engagement were included at the household level. Key household-level independent variables were centered within the neighborhood. Neighborhood-level independent variables were grand-mean centered (i.e., centered around the entire sample’s mean). Centering allows us to parse out within-neighborhood variance and between-neighborhood variance, therefore allowing us to understand the compositional and contextual effects, respectively, of these variables on IPV outcomes (Enders & Tofighi, 2007). We also weighted neighborhood-level aggregates to account for the over-representation of Colombian households in our sample. Neighborhood scores (e.g., social disorganization) were standardized to reflect a change in odds ratio per one standard deviation change in predictor, and neighborhood averages (e.g., percent of renters) were scaled to reflect change in odds ratio per 10% change in predictor.

Analysis

We conducted an attrition analysis to examine if women who were partnered at Time 1 and then lost to follow-up at Time 2, were different from our study sample. Overall, 12.8% of women that were partnered at Time 1 were lost to follow-up or refused to participate at Time 2. Supplementary Table C presents associations between attrition and key independent variables. We found some evidence of differential attrition: households with Colombian male partners and households that had moved to the neighborhood in the last 20 years were more likely to attrit.

We then assessed the pattern of missing data: 3.6% and 2.1% of women were missing data for IPV at Time 1 and Time 2, respectively. We conducted joint multi-level multiple imputations, accounting for categorical variables and for the hierarchical nature of the dataset. We then conducted multi-level logistic regression to examine the variance in IPV outcomes attributable to individual, partner, household, and neighborhood-level characteristics. Maximum likelihood parameter estimates were generated with standard errors robust to non-normality and non-independence of observations. Random intercepts were included in the models to account for unexplained neighborhood clustering related to IPV. We conducted a 2-2-1 cross-level mediation analysis to see if social cohesion mediated the relationship between social disorganization and IPV. All models controlled for female age, education, and ethnicity; province; intervention status (pooled treatment); and IPV at Time 1.

Results

Sample characteristics are listed in Table 2. At Time 2, 29.4% and 15.6% of women reported emotional and physical or sexual IPV in the last six months, respectively. Social disorganization was marginally, positively associated with emotional IPV (AOR: 1.17, 95% CI: .99, 1.38) and trended in the same direction for physical and/or sexual IPV (AOR: 1.20, 95% CI: .96, 1.51; Tables 3 and 4, Models 1a and 2a). In other words, a woman living in a neighborhood with a social disorganization score one standard deviation higher than the average neighborhood, had around 17% and 20% higher odds of experiencing emotional and physical and sexual IPV, respectively, although this did not reach statistical significance at p < .05. Social disorganization and social cohesion were negatively associated (AOR: .76, .60, .96), but social cohesion did not act as a mediator of the relationship between social disorganization and IPV outcomes (Tables 3 and 4, Models 1a and 2a, Figure 1a). Some individual domains of social disorganization and social cohesion were associated with IPV (Tables 3 and 4, Models 1b and 2b). Residential instability was marginally, positively associated with emotional (AOR: 1.10, 95% CI: .98, 1.23) and physical and/or sexual IPV (AOR: 1.13, 95% CI: .98, 1.30). Neighborhood civic engagement was negatively associated with emotional IPV (AOR: .84, 95% CI .74, .96) and physical and/or sexual IPV (AOR: .85, .72, 1.00), such that a woman living in a neighborhood with 10% more households that participate in a community group, compared to the average neighborhood, had around 15% lower odds of experiencing IPV. However, neighborhood “trust in institutions and community connectedness” was positively associated with emotional IPV (AOR: 1.25, 95% CI: 1.07, 1.47).

Figure 1.

(a) Path model of neighborhood social disorganization, neighborhood social cohesion, and individual recent emotional and physical and/or sexual IPV. (b) Path model of neighborhood social disorganization, neighborhood civic engagement, and individual recent emotional IPV.

Table 2.

Sample Characteristics (at Time 1) of Poor Ecuadorian and Colombian Refugee Households (n = 1,312).

	n	%
Emotional IPV (Time 2)^a	378	29.42
Physical and/or sexual IPV (Time 2)^a	201	15.64
Female demographics
Female age (mean years)		35.91
Female highest education level
Primary or less	827	62.99
Secondary or higher	486	37.01
Female-headed households	28	2.13
Female nationality
Ecuadorian	836	63.67
Colombian	477	36.33
Household characteristics
Moved to neighborhood in last 20 years	561	42.73
Female participation in labor force	427	32.52
Male participation in labor force	1,252	95.43
Experienced discrimination in the last six months	499	38.00
Race	67	5.10
Gender	39	2.97
Economic status	314	23.91
Occupation	161	12.26
Political affiliation	52	3.96
Disability	109	8.30
Nationality	180	13.71
Religion	84	6.40
Appearance	53	4.04
Other	26	1.98
Psychosocial stressors experienced in the last six months (mean)		0.30
Any in the last six months	303	23.08
Robbed of cash	27	2.06
Robbed of food	11	.84
Robbed of other goods	48	3.66
Death of male adult household members	2	.15
Death of female adult household members	0	.00
Permanent debilitating injuries or illness of household member	144	10.97
Physical attack	6	.46
Drought or flood	49	3.73
Destruction of property or land due to natural disaster	22	1.68
Destruction of property due to violence	5	.38
Alcoholism or drug addiction of family member	32	2.44
Job loss or failed business	38	2.89
Prison for household member	6	.46
Participated in community activities in last six months	661	50.34
Number of people can ask to borrow $100 (mean)		1.90
Neighborhood (n = 80) ^b
Percent of female-headed households		1.86
Percent of renters		58.35
Residential instability		42.83

Note. Refer to Table 1 for indicator definitions.

^an = 1,285, missings not imputed. bNeighborhood variables weighted to extrapolate to the full eligible sample of the cash, voucher and food transfer program.

Table 3.

Multi-level Logistic Regression Predicting Effect of Social Disorganization (at Time 1) on Emotional Intimate Partner Violence (at Time 2) Among a Sample of Poor Ecuadorian and Colombian Refugee Women (n = 1,312).

	Model 1a				Model 1b				Model 1c
	OR	95% CI			OR	95% CI			OR	95% CI
Intercept	3.69	1.06	12.91		3.36	.94	11.93		3.83	1.09	13.49
Economic indicators
Female participation in labor force	1.17	.91	1.51		1.20	.93	1.56		1.19	.93	1.53
Male participation in labor force	1.31	.62	2.77		1.31	.61	2.80		1.31	.61	2.78
Household wealth	1.16	1.01	1.33		1.16	1.01	1.32		1.16	1.02	1.33
Food Consumption Score	1.01	1.00	1.01		1.01	1.00	1.01		1.01	1.00	1.01
Psychosocial stressors and supports
Number of psychosocial stressors in the last six months	1.38	1.13	1.69	**	1.37	1.12	1.68	**	1.37	1.13	1.68	**
Experienced discrimination in the last six months	1.49	1.12	2.00	**	1.49	1.12	2.00	**	1.49	1.11	1.99	**
Participated in community activities in last six months	1.00	.77	1.31		1.00	.77	1.30		1.01	.77	1.32
Number of people can ask to borrow $100	.95	.91	1.00	*	.95	.91	1.00	*	.95	.91	1.00	*
Migration-related indicators
Female Nationality (Colombian = 1; Ecuadorian = 0)	.96	.64	1.44		.95	.64	1.42		.98	.65	1.47
Male Nationality (Colombian = 1; Ecuadorian = 0)	.92	.59	1.43		.93	.60	1.44		.96	.62	1.49
Moved to neighborhood in last 20 years	.68	.47	1.00		.69	.47	1.01		.67	.46	.97	*
Neighborhood^a
Social disorganization^b	1.17	.99	1.38				1.08	.91	1.28
Community adversity^c	.76	.32	1.77	.91	.39	2.13
Community wealth^b					.98	.83	1.15
Percent of female-headed households^c					.98	.84	1.16
Percent of renters^b					.99	.90	1.10
Residential ⁱnstability^b					1.10	.98	1.23
Social cohesion^c	1.07	.91	1.26
Social support	.98	.85	1.12		.95	.83	1.09
Trust in individuals^c					1.04	.86	1.28
Trust in institutions^c					1.25	1.07	1.47	*
Experiences of discrimination^b					1.02	.88	1.17
Civic engagement^b					.84	.74	.96	*	.35	.13	.93	*
Mediation
Social disorganization on social cohesion	.76	.60	.96	*
Social disorganization on neighborhood civic engagement									.94	.91	.98
Indirect effect	.98	.94	1.03						1.06	1.00	1.14
Model fit
AIC (lower is better)	1603.85				1364.14				1303.88
BIC (lower is better)	1748.85				1524.68				1438.52

Note. *p < .05, **p < .01.

Model notes: Missing data imputed; all model use multilevel modeling to account for hierarchal structure; maximum likelihood parameter estimates with standard errors robust to non-normality and non-independence of observations are presented; all models controlled for emotional and physical/sexual IPV at Time 1, female age, female education, province, and treatment group.

Refer to Table 1 for indicator definitions.

^aNeighborhood variables weighted to extrapolate to the full eligible sample of the cash and voucher program; all neighborhood variables grand-mean centered.

^bStandardized to reflect a change in odds ratio per one standard deviation change in predictor.

^cScaled to reflect change in odds ratio per 10% change in predictor.

Table 4.

Multi-level Logistic Regression Predicting Effect of Social Disorganization (at Time 1) on Physical and/or Sexual IPV Intimate Partner Violence (at Time 2) Among a Sample of Poor Ecuadorian and Colombian Refugee Women (n = 1,312).

	Model 2a			Model 2b					Model 2c
	OR	95% CI			OR	95% CI			OR	95% CI
Intercept	1.51	.28	8.20		1.27	.24	6.80		1.77	.33	9.51
Economic indicators
Female participation in labor force	1.01	.69	1.49		1.01	.69	1.47		1.02	.70	1.50
Male participation in labor force	.57	.23	1.41		.55	.22	1.34		.57	.23	1.44
Household wealth	1.04	.86	1.25		1.04	.86	1.25		1.03	.86	1.24
Food Consumption Score	1.00	.99	1.01		1.00	.99	1.01		1.00	.99	1.01
Psychosocial stressors and supports
Number of psychosocial stressors in the last six months	1.63	1.25	2.12	**	1.63	1.25	2.12	**	1.64	1.26	2.15	**
Experienced discrimination in the last six months	1.58	1.13	2.23	**	1.60	1.14	2.25	**	1.58	1.12	2.23	**
Participated in community activities in last six months	.88	.62	1.24		.87	.62	1.23		.87	.62	1.23
Number of people can ask to borrow $100	.95	.89	1.01		.95	.89	1.01		.95	.90	1.01
Migration-related indicators
Female Nationality (Colombian = 1; Ecuadorian = 0)	.77	.47	1.26		.75	.46	1.22		.76	.47	1.22
Male Nationality (Colombian = 1; Ecuadorian = 0)	1.15	.67	1.98		1.10	.64	1.88		1.18	.68	2.04
Moved to neighborhood in last 20 years	.74	.48	1.14		.77	.50	1.18		.74	.48	1.15
Neighborhood^a
Social disorganization^b	1.20	.96	1.51						1.17	.95	1.44
Community adversity^c	1.98	.72	5.45		1.81	.57	5.75
Community wealth^b					1.02	.83	1.25
Percent of female-headed households^c					1.04	.85	1.27
Percent of renters^c					.99	.87	1.12
Residential instability^c					1.13	.98	1.30
Social cohesion^b	.95	.75	1.21
Social support	.92	.77	1.10		.91	.77	1.07
Trust in individuals^b					1.14	.86	1.51
Trust in institutions^b					.99	.83	1.20
Experiences of discrimination^c					1.01	.86	1.19
Civic engagement^c					.85	.72	1.00		.55	.14	2.20
Mediation
Social disorganization on social cohesion	.76	.60	.96	*
Social disorganization on neighborhood civic engagement									.94	.91	.98	**
Indirect effect	1.01	.95	1.09						1.04	.95	1.13
Model fit
AIC (lower is better)	1226.87				994.59				931.87
BIC (lower is better)	1371.87				1155.1				1066.51

Note. *p < .05, **p < .01.

Refer to Table 1 for indicator definitions.

^aNeighborhood variables weighted to extrapolate to the full eligible sample of the cash and voucher program; all neighborhood variables grand-mean centered.

^bStandardized to reflect a change in odds ratio per one standard deviation change in predictor.

^cScaled to reflect change in odds ratio per 10% change in predictor.

We tested mediation for the individual domains that were associated with IPV outcomes. We found a marginally statistically significant indirect effect of social disorganization on emotional IPV through neighborhood civic engagement (AOR: 1.06, 95% CI: 1.00, 1.14). The results suggest that higher social disorganization is associated with lower civic engagement (AOR: .94, 95% CI: .91, .98) and higher civic engagement is associated with lower odds of emotional IPV (AOR: .35, 95% CI: .13, .93; Table 3, Models 1c, Figure 1b). We did not find the same relationship for physical and/or sexual IPV (Table 4, Models 2c). We also did not see indications of mediation for residential instability and social cohesion domains or “trust in institutions and community connectedness” and social disorganization domains (results not shown).

Within the household, psychosocial stressors and supports were associated with IPV outcomes (Tables 3 and 4, Models 1c and 2c). Women living in households reporting perceived discrimination at Time 1 were significantly more likely to report emotional IPV (AOR: 1.49, 95% CI: 1.11, 1.99) and physical and/or sexual IPV (AOR: 1.58, 95% CI: 1.12, 2.23) at Time 2, compared to households that did not report any discrimination. Women living in households reporting higher number of psychosocial stressors had significantly higher odds of emotional (AOR: 1.37, 95% CI 1.13, 1.68) and physical and/or sexual IPV (AOR: 1.64, 95% CI: 1.26, 2.15). Households reporting a higher number of people (outside of the household) in which they could borrow $100 in an emergency had significantly lower odds of reporting emotional IPV (AOR: .95, 95% CI: .91, 1.00) and marginally lower odds of reporting physical and/or sexual IPV (AOR: .95, 95% CI: .90, 1.01). Participation in community events, and male and female nationality were not associated with IPV outcomes. Household-level mobility was negatively associated with emotional IPV (AOR: .67, 95% CI: .46, .97).

Discussion

While theory and research illustrate the significant impact conflict has on community structures and social processes, the influence of neighborhood factors on IPV in refugee-hosting communities remains largely unexplored. We found that social disorganization was marginally associated with IPV, as has been seen in other settings (Pinchevsky & Wright, 2012; VanderEnde et al., 2012), and neighborhood civic engagement mediated the relationship between social disorganization and IPV. We found stronger associations with more proximal household-level indicators of recent discrimination, psychosocial stressors, and social support.

Our study results differ slightly from previous social disorganization and IPV research. Pinchevsky and Wright’s (2012) systematic review of studies examining neighborhood factors and IPV outcomes found that indicators of concentrated disadvantage were consistently associated with IPV outcomes, but indicators of immigrant concentration and residential instability were not. VanderEnde et al.’s 2012 review of literature on community-level correlates of IPV, found the same pattern among social disorganization literature. We found that residential instability, but not community wealth, percent of renters or percent of female-headed households, was related to physical and/or sexual IPV. The majority of social disorganization research has been conducted in the United States; it may be that different indicators are more relevant in international or conflict-affected areas. Differences could also be due to examining emotional IPV, while most other research has focused on physical IPV. While emotional and physical and/or sexual IPV are highly correlated, and often overlap, they are distinct constructs. As such, we would expect to see some differences in what predicts each outcome. This could be due to differences in the social acceptability of emotional abuse compared to physical or sexual abuse. Emotional abuse is also more difficult for health care providers, friends, or family to identify and intervene upon (Yoshihama et al., 2009). Finally, isolation, a common component of emotional IPV, reduces access to support, and increases vulnerability to more abuse.

We hypothesized that social disorganization would influence IPV outcomes through a social cohesion mediator. We saw a marginally significant indirect effect of social disorganization on IPV through neighborhood-level civic engagement, in the expected direction. Previous research has found that collective efficacy, a distinct but related domain of social cohesion, has similarly been identified as a protective factor for IPV in some, but not all, settings (Browning, 2002; Emery et al., 2011; Leddy et al., 2019). A neighborhood’s level of civic engagement may indicate that the community is more comfortable and practiced working together on shared concerns, and therefore may be more adept at addressing issues such as IPV. Neighborhood collective action has had profound effects in Latin America and around the world (e.g., Auyero & Sobering, 2017; Rose, 1992). Our findings provide empirical support for the critical benefits that neighborhood civic engagement can have in countering concentrated disadvantage with direct positive impacts on women’s safety within their homes.

However, our combined social cohesion measure did not act as a mediator between social disorganization and IPV. Further, in contrast to our hypothesis, we found that neighborhood-level trust in institutions was positively associated experiencing emotional IPV. It is possible that women who have previously experienced IPV could have successfully sought help from local institutions and therefore have higher trust in them. It is also possible that the relationship could reflect an endogenous placement of programs and services—local officials or community organizations could be responding to high rates of IPV prevalence in neighborhoods by purposefully placing and strengthening institutions there. Alternatively, these results could be a result of measurement. Our social cohesion measure was not previously validated. There is a wide range of social cohesion definitions and indicators and ours did not encapsulate all theorized domains; for example, collective efficacy and other social norms variables that may act as mediators between disorganization and IPV, such as norms supportive of violence, were not directly measured (Bruhn, 2009; Leites et al., 2017; National Research Council, 2014).

Results indicate that proximal stressors are related to short-term IPV outcomes, and these relationships are more robust as compared to measures of social disorganization. A large number of participants reported experiencing some form of recent discrimination (33.7%) or adversity (21.0%); both of which were positively related to IPV. A high percentage of households reported experiencing permanent debilitating injuries or illness in the past six months. This could reflect the violent and precarious environment in which these families are living. The border region has been described as a “lawless area,” with a “porous” border facilitating underground economies, violence, and the Colombian conflict spilling across the Ecuadorian border at times (Conway, 2013). External stressors on the household and partnership can increase tension in the family, as could displacement of anger by partners also exposed to violence (Bermudez et al., 2019; Wirtz et al., 2014). Wirtz’s qualitative work with internally displaced Colombian women found that women attributed IPV to their partner’s exposure to political violence, their partners’ lost employment, and economic insecurity as a cause of IPV (Wirtz et al., 2014). Our results are also consistent with other studies examining discrimination and IPV (Colorado-Yohar et al., 2012; Colorado-Yohar et al., 2016).

In line with previous research and our hypothesis, we found that household-level social support was protective against IPV (Browning, 2002; Dekeseredy et al., 2003; Kirst et al., 2015). Social support could reflect the household’s social capital within their immediate network, which would encourage intervening in abusive situations. It may also be that social support can help mitigate stress and conflict that otherwise would lead to IPV. Our results, as well as other research that has found that social support mitigates the consequences of IPV in terms of mental health and perceived quality of life (Beeble et al., 2009), suggest exploring interventions to increase women’s support networks and group-based programming to both prevent and respond to IPV (Brody et al., 2017; Gram et al., 2019).

While we found that neighborhood residential stability was marginally positively associated with physical and/or sexual IPV, we did not see a relationship between nationality, a proxy for international migration, and IPV outcomes. We did see that household mobility was negatively associated with emotional IPV. It is possible that this estimation is biased as we found differential attrition based on household mobility. This analysis was not able to take into account the reason for moving to the neighborhood, years since relocating, or adversities experienced during migration. However, these likely influence IPV and merit further research.

Our research has some limitations. Mediation analyses ideally use three-time points to assure the temporal ordering of all variables. We only used two data points, and can only assess associations between social disorganization and social cohesion. While we cannot establish a temporal ordering of those two variables, the hypothesized model is in-line with social disorganization theory. Furthermore, our neighborhoods were defined by administrative units which may not reflect an accurate geographic boundary of women’s communities and the sample only includes relatively poor neighborhoods, not the full urban area—thus, we may be limited by the variation in some of the social disorganization indicators. Finally, one variable—has anyone in the household been physically attacked in the past six months—did not specify that this was perpetrated by a non-partner; it could be there is some overlap with IPV. As only ten households reported this, we do not expect this to meaningfully bias our results.

Implications, Future Research, and Conclusions

While the conflict in Colombia officially ended in 2016, the region is facing a new humanitarian crisis. Venezuela’s social, economic, and political instability resulted in the migration or displacement of 4.5 million Venezuelans to date, with the majority residing in the region (UN Refugee Agency, 2020). This mass migration creates significant stressors on host communities and the UN International Organization for Migration (IOM) reports extreme vulnerability among many displaced Venezuelans, including food insecurity, lack of access to education and medical care, poverty, exploitation, and fear of violence.

International agencies focus extensively on promoting social cohesion to support the mental and physical health of refugees and migrants, ensure social stability in host settings, and generate public support for social protection programs (UN Refugee Agency, 2020). Synergistic intervention approaches could combine economic and social supports—such as cash transfers, employment supports, efforts to increase access to health care and education—with community-level campaigns—such as social marketing, volunteerism, and community mobilization activities (de Berry & Roberts, 2018; Hidrobo et al., 2016; Leddy et al., 2019; Pronyk et al., 2008). These activities have been implemented in conflict-affected areas previously and could incorporate a gender lens to reduce IPV. Future research can continue expanding our understanding of social disorganization and cohesion in conflict-affected areas, as well as, other theoretically driven neighborhood-level factors the influence IPV among migrants and displaced people. We hope these findings will highlight the need to consider both social and economic factors at multiple levels in an effort to mitigate and prevent IPV for vulnerable populations.

Supplemental Material

Supplemental Material - Love in the Time of War: Identifying Neighborhood-level Predictors of Intimate Partner Violence from a Longitudinal Study in Refugee-hosting Communities

Supplemental Material for Love in the Time of War: Identifying Neighborhood-level Predictors of Intimate Partner Violence from a Longitudinal Study in Refugee-hosting Communities by Sarah Treves-Kagan, Amber Peterman, Nisha C. Gottfredson, Andrés Villaveces, Kathryn E. Moracco, and Suzanne Maman, in Journal of Interpersonal Violence

Footnotes

Acknowledgments

First and foremost, we thank the study participants for their time and for sharing their experiences with the study team. We thank the World Food Programme in Quito and Rome for their collaboration in undertaking the original study, and researchers at the International Food Policy Research Institute, and CEPAR for study implementation and making data available for analysis. Support for the original study including data collection was received from the Government of Spain, via the World Food Programme. Support for STK was provided by Royster Society of Fellows and the Injury and Violence Prevention Fellowship awarded by the Injury Prevention Research Center at UNC Chapel Hill. Support for NCG was provided by NIH K01DA035153.

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) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Sarah Treves-Kagan

Nisha C. Gottfredson

Author Biographies

Sarah Treves-Kagan, PhD, MPH, is a social scientist with a primary interest in advancing health equity and identifying structural determinants of violence among vulnerable populations to inform primary prevention programming and policies. She has conducted research, advocacy, service-delivery, and government programming on adverse childhood experiences, sexual assault, gender-based violence, and youth homicide in domestic and international settings.

Amber Peterman, PhD, is a research associate professor in public policy at UNC Chapel Hill and Co-lead of the intimate partner violence and cash transfer collaborative. Her research focuses on gender-based violence, economic empowerment, and social protection. She has lived and worked in over a dozen countries, with a focus on sub-Saharan Africa, and has published over 50 peer-reviewed articles and book chapters.

Nisha C. Gottfredson, PhD, is an assistant professor in the Department of Health Behavior at UNC Gillings School of Global Public Health in Chapel Hill. Dr. Gottfredson’s research focuses on the development and application of quantitative methods for the study of development and change in health and behavioral processes.

Andrés Villaveces, MD, PhD, MPH, is a physician and epidemiologist with over 25 years of experience researching injuries and violence prevention globally. He is currently a Senior Scientist at the Division of Violence Prevention, NCIPC, CDC, leading the Violence Against Children Surveys in Latin America. He has extensive experience working on violence and injury prevention including issues related to alcohol, firearms, and the built environment.

Kathryn E. Moracco, PhD, MPH, is an associate professor in the Department of Health Behavior at UNC Chapel Hill., and the associate director of the UNC Injury Prevention Research Center. Her current research includes federally funded evaluations of a community-based sexual violence prevention program, domestic violence protective orders (DVPOs), and a primary care-based IPV screening and brief intervention, as well as studies of e-filing of DVPOs, and firearm confiscation and surrender policies in criminal and civil domestic violence cases.

Suzanne Maman, PhD, MPH, is a professor of Health Behavior at the University of North Carolina Chapel Hill. Trained as a social scientist in public health, she has 20 years of research experience related to HIV/AIDS and gender-based violence in sub-Saharan Africa. Dr. Maman’s work focuses on two aspects of the association between HIV and violence: (a) violence as a barrier to implementing HIV risk reduction strategies and (b) violence as an outcome of HIV testing and HIV status disclosure to sexual partners.

References

ABColombia. (2013). Colombia: Women, conflict-related sexual violence and the peace process . ABColombia.

Auyero

, & Sobering

(2017, December). Violence, the state, and the poor: A view from the South. Sociological Forum , 32(1), 1018–1031.

Beeble

M. L.

, Bybee

, Sullivan

C. M.

, & Adams

A. E.

(2009). Main, mediating, and moderating effects of social support on the well-being of survivors of intimate partner violence across 2 years. Journal of Consulting and Clinical Psychology , 77(4), 718–729. https://doi.org/10.1037/a0016140

Benson

M. L.

, Fox

G. L.

, DeMaris

, & Wyk

J. van

. (2003). Neighborhood disadvantage, individual economic distress and violence against women in intimate relationships. Journal of Quantitative Criminology , 19(3), 207–235.

Bermudez

L. G.,

Stark

L.,

Bennouna

C.,

Jensen

C.,

Potts

A.,

Kaloga

I. F.

, Tilus

R.,

Buteau

J. E.,

Marsh

M.,

Hoover

, & Williams

M. L.

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

Bernard

P.,

Charafeddine

R.,

Frohlich

K. L.,

Daniel

M.,

Kestens

Y.,

& Potvin

(2007). Health inequalities and place: A theoretical conception of neighbourhood. Social Science and Medicine , 65(9), 1839–1852.

Beyer

K.,

Wallis

A. B.

, & Hamberger

L. K.

(2015). Neighborhood environment and intimate partner violence: A systematic review. Trauma Violence Abuse , 16(1), 16–47. https://doi.org/10.1177/1524838013515758

Brody

, Hoop

T. D.,

Vojtkova

M.,

Warnock

, Dunbar

, Murthy

, & Dworkin

S. L.

(2017). Can self-help group programs improve women’s empowerment? A systematic review. Journal of Development Effectiveness , 9(1), 15–40.

Browning

C. R.

(2002). The span of collective efficacy: Extending social disorganization theory to partner violence. Journal of Marriage and Family , 64(4), 833–850. https://doi.org/10.1111/j.1741-3737.2002.00833.x

10.

Bruhn

(Eds.). (2009). The concept of social cohesion. In The group effect (pp. 31–48). Springer.

11.

Carvajal

(2017). As Colombia emerges from decades of war, migration challenges mount . Migration Policy Institute. https://www.migrationpolicy.org/article/colombia-emerges-decades-war-migration-challenges-mount

12.

Colorado-Yohar

S. M.,

Agudelo-Suárez

A. A.

Huerta

J. M.

, & Torres-Cantero

A. M.

(2016). Intimate partner violence and its associated factors in a sample of Colombian immigrant population in Spain. Journal of Immigrant and Minority Health , 18(4), 904–912.

13.

Colorado-Yohar

S.,

Tormo

M. J.,

Salmerón

D.,

Dios

S.,

Ballesta

, & Navarro

(2012). Violence reported by the immigrant population is high as compared with the native population in Southeast Spain. Journal of Interpersonal Violence , 27(16), 3322–3340. https://doi.org/10.1177/0886260512441260

14.

Conway

J. A.

(2013). Lago Agrio (Nueva Loja), Ecuador: A Strategic Black Spot? (Master’s thesis, United States Army War College, Carlisle, PA, USA). https://apps.dtic.mil/sti/pdfs/ADA589128.pdf

15.

de Berry

J. P.

, & Roberts

(2018). Social cohesion and forced displacement: A desk review to inform programming and project design . World Bank Group. http://documents.worldbank.org/curated/en/125521531981681035/Social-cohesion-and-forced-displacement-a-desk-review-to-inform-programming-and-project-design

16.

Dekeseredy

W. S.,

Alvi

, & Tomaszewski

E. A.

(2003). Perceived collective efficacy and women’s victimization in public housing. Criminal Justice , 3(1), 5–27. https://doi.org/10.1177/1466802503003001453

17.

Edwards

K. M.,

Mattingly

M. J.

, Dixon

K. J.

, & Banyard

V. L.

(2014). Community matters: Intimate partner violence among rural young adults. American Journal of Community Psychology , 53(1–2), 198–207.

18.

Ellsberg

, & Heise

(2005). Researching violence against women: A practical guide for researchers and activists . World Health Organization and PATH.

19.

Emery

C. R.,

Jolley

J. M.

, & Wu

(2011). Desistance from intimate partner violence: The role of legal cynicism, collective efficacy, and social disorganization in Chicago neighborhoods. American Journal of Community Psychology , 48(3–4), 373–383. https://doi.org/10.1007/s10464-010-9362-5

20.

Enders

C. K.

, & Tofighi

(2007). Centering predictor variables in cross-sectional multilevel models: A new look at an old issue. Psychological Methods , 12(2), 121–138.

21.

González

G. G.,

Schiavon

J. A.

, & Maldonado

(2011). The Americas and the World 2010–2011: Public opinion and foreign policy in Brazil, Colombia, Ecuador, Mexico, and Peru . https://www.semanticscholar.org/paper/The-Americas-and-the-world-2010-2011%3A-public-and-in-Schiavon-Crow/5106b724ee71f13e1588d5ec04d208ea63bd3a6a

22.

Gram

L.,

Fitchett

, Ashraf

A.,

Daruwalla

, & Osrin

(2019). Promoting women’s and children’s health through community groups in low-income and middle-income countries: A mixed-methods systematic review of mechanisms, enablers and barriers. BMJ Global Health , 4(6). https://dx-doi-org.web.bisu.edu.cn/10.1136/bmjgh-2019-001972

23.

Heise

, & Kotsadam

(2015). Cross-national and multilevel correlates of partner violence: An analysis of data from population-based surveys. Lancet Global Health , 3(6), e332–340. https://doi.org/10.1016/s2214-109x(15)00013-3

24.

Hidrobo

M.,

Hoddinott

J.,

Peterman

, Margolies

, & Moreira

(2014). Cash, Food, or Vouchers? Evidence from a randomized experiment in northern Ecuador. Journal of Development Economics , 107. https://doi.org/10.1016/j.jdeveco.2013.11.009

25.

Hidrobo

M.,

Peterman

, & Heise

(2016). The effect of cash, vouchers and food transfers on intimate partner violence: Evidence from a randomized experiment in northern Ecuador. American Economic Journal: Applied Economics , 8(3), 284–303.

26.

Horn

(2010). Responses to intimate partner violence in Kakuma refugee camp: Refugee interactions with agency systems. Social Science and Medicine , 70(1), 160–168. https://doi.org/https://doi.org/10.1016/j.socscimed.2009.09.036

27.

Hynes

M. E.,

Sterk

C. E.,

Hennink

M.,

Patel

S.,

DePadilla

, & Yount

K. M.

(2016). Exploring gender norms, agency and intimate partner violence among displaced Colombian women: A qualitative assessment. Global Public Health , 11(1–2), 17–33.

28.

International Food Policy Research Institute. (2016a). Impact evaluation of cash, food vouchers, and food transfers among Colombian refugees and poor Ecuadorians in Carchi and Sucumbios: Baseline survey. Harvard Dataverse. https://dx-doi-org.web.bisu.edu.cn/10.7910/DVN/YW4WIT

29.

International Food Policy Research Institute. (2016b). Impact evaluation of cash, food vouchers, and food transfers among Colombian refugees and poor Ecuadorians in Carchi and Sucumbios: Endline survey. Harvard Dataverse. https://dx-doi-org.web.bisu.edu.cn/10.7910/DVN/AXGCHT

30.

International Organization for Migration Health Department. (n.d.). Mental health, psychosocial assistance, and cultural integration in emergency and displacement: IOM perspective . https://www.iom.int/jahia/webdav/shared/shared/mainsite/activities/health/Mental-Health-Psychosocial-Assistance.pdf

31.

Kirst

M.,

Lazgare

L. P.,

Zhang

Y. J.

, & O’Campo

(2015). The effects of social capital and neighborhood characteristics on intimate partner violence: A consideration of social resources and risks. American Journal of Community Psychology , 55(3–4), 314–325. https://doi.org/10.1007/s10464-015-9716-0

32.

Leddy

A. M.,

Lippman

S. A.,

Neilands

T. B.,

Twine

R.,

Ahern

J.,

Gomez-Olive

F. X.,

DeLong

S. M.,

MacPhail

C.,

Kahn

, & Pettifor

A. E.

(2019). Community collective efficacy is associated with reduced physical intimate partner violence (IPV) incidence in the rural province of Mpumalanga, South Africa: Findings from HPTN 068. Journal of Epidemiology and Community Health , 73(2), 176–181.

33.

Leites

M.,

Pereira

G.,

Rius

A.,

Salas

, & Vigorito

(2017). Protocol: The effect of cash transfers on social solidarity: A systematic review. Campbell Systematic Reviews , 13(1), 1–49.https://doi.org/10.1002/CL2.181

34.

Leutert

(2012). Dispatches: Ecuador’s invisible refugee population. Americas Quarterly . https://www.americasquarterly.org/fulltextarticle/dispatches-ecuadors-invisible-refugee-population/

35.

Maas

C. J.

, & Hox

J. J.

(2005). Sufficient sample sizes for multilevel modeling. Methodology , 1(3), 86–92.

36.

Menjívar

, & Salcido

(2002). Immigrant women and domestic violence: Common experiences in different countries. Gender & Society , 16(6), 898–920.

37.

Mousa

(2020). Building social cohesion between Christians and Muslims through soccer in post-ISIS Iraq. Science, 369(6505), 866–870.

38.

National Research Council. (2014). Civic engagement and social cohesion: Measuring dimensions of social capital to inform policy . National Academies Press.

39.

O’Campo

, Gielen

A. C.,

Faden

R. R.,

Xue

X.,

Kass

, & Wang

M.-C.

(1995). Violence by male partners against women during the childbearing year: A contextual analysis. American Journal of Public Health , 85(8), 1092–1097.

40.

Pinchevsky

G. M.

, & Wright

E. M.

(2012). The impact of neighborhoods on intimate partner violence and victimization. Trauma, Violence, & Abuse , 13(2), 112–132.

41.

Pronyk

P. M.,

Harpham

T.,

Busza

J.,

Phetla

G.,

Morison

L. A.

, Hargreaves

J. R.

, Kim

J. C.

, Watts

C. H.

, & Porter

J. D.

(2008). Can social capital be intentionally generated? A randomized trial from rural South Africa. Social Science and Medicine , 67(10), 1559–1570.

42.

Rose

(1992). Where women are leaders: The SEWA movement in India . Zed Books Ltd.

43.

Sampson

R. J.,

Morenoff

J. D.

, & Earls

(1999). Beyond social capital: Spatial dynamics of collective efficacy for children. American Sociological Review , 64(5), 633–660.

44.

Shaw

C. R.

, & McKay

H. D.

(1969). Juvenile delinquency and urban areas [Revised ed.]. The University of Chicago Press.

45.

Shedlin

M. G.,

Decena

C. U.,

Noboa

, & Betancourt

Ó.

(2014). Sending-country violence and receiving-country discrimination: Effects on the health of Colombian refugees in Ecuador. Journal of Immigrant and Minority Health , 16(1), 119–124. https://doi.org/10.1007/s10903-013-9777-9

46.

Shultz

J. M.,

Garfin

D. R.,

Espinel

Z.,

Araya

R.,

Oquendo

M. A.,

Wainberg

M. L.,

Chaskel

R.,

Gaviria

S. L.

, Ordóñez

A. E.,

Espinola

M.,

Wilson

F. E.,

García

N. M.,

Ceballos

A. M. G.,

Garcia-Barcena

Y.,

Verdeli

, & Neria

(2014). Internally displaced “victims of armed conflict” in Colombia: The trajectory and trauma signature of forced migration. Current Psychiatry Reports , 16(10), 475.

47.

Stueve

, & O’Donnell

(2008). Urban young women’s experiences of discrimination and community violence and intimate partner violence. Journal of Urban Health , 85(3), 386–401.

48.

Thulin

E. J.,

Heinze

J. E.,

Kusunoki

, Hsieh

H.-F.

, & Zimmerman

M. A.

(2020). Perceived neighborhood characteristics and experiences of intimate partner violence: A multilevel analysis. Journal of Interpersonal Violence . https://doi.org/10.1177/0886260520906183

49.

UN High Commissioner for Refugees (UNHCR). (2016). Colombia Situation update, December 2016 . http://www.refworld.org/docid/58627cd24.html

50.

UN Refugee Agency. (2020). Refugee and Migrant Response Plan 2020, for refugees and migrants from Venezuela (January–December 2020) . https://reliefweb.int/report/colombia/refugee-and-migrant-response-plan-2020-refugees-and-migrants-venezuela-january

51.

U.S. Department of State Bureau of Democracy Human Rights and Labor. (2013). Ecuador Human Rights Report 2013 . https://2009-2017.state.gov/documents/organization/220651.pdf

52.

Valli

E.,

Peterman

, & Hidrobo

(2018). Economic transfers and social cohesion in a refugee-hosting setting . UNICEF Office of Research. https://www.unicef-irc.org/publications/pdf/Valli%20et%20al%20for%20WEB.pdf

53.

VanderEnde

K. E.,

Yount

K. M.,

Dynes

M. M.

, & Sibley

L. M.

(2012). Community-level correlates of intimate partner violence against women globally: A systematic review. Social Science and Medicine , 75(7), 1143–1155.

54.

Wirtz

A. L.,

Pham

K.,

Glass

N.,

Loochkartt

S.,

Kidane

T.,

Cuspoca

D.,

Rubenstein

L. S.,

Singh

, & Vu

(2014). Gender-based violence in conflict and displacement: Qualitative findings from displaced women in Colombia. Conflict and Health , 8, 10. https://doi.org/10.1186/1752-1505-8-10

55.

Yoshihama

M.,

Horrocks

, & Kamano

(2009). The role of emotional abuse in intimate partner violence and health among women in Yokohama, Japan. American Journal of Public Health , 99(4), 647–653.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.11 MB