Abstract
Domestic violence (DV) experienced by immigrant women is a public health concern. In collaboration with a community agency, researchers undertook a retrospective review of 1,763 client files from 2006–2014. The three aims were to document the incidence of DV, service needs associated with DV, and identification of risk factors associated with DV in the extracted file data. About 41% reported DV and required multiple services. Separated and divorced women, and women on visitor/temporary visas showed the highest risk. The results underscore the value of research partnerships with community-based service agencies in increasing our understanding of DV among immigrant women.
Globally, violence against women is a serious public health problem. Such violence can occur in multiple forms (e.g., targeted sexual violence in the context of civil unrest, in the workplace, dating violence); however, there has been more public awareness and governmental agency attention in Canada to domestic violence (DV) experienced by women. The definition of DV used in this article was guided by the expanded definition of DV developed initially by the World Health Organization (WHO) for its 2005 multicountry study, where DV was defined as violence by an intimate partner experienced by women. Included in this definition were acts of physical, sexual, and emotional abuse by a current or former intimate male partner, whether cohabiting or not. The WHO has since updated its definition to reflect that others in the household could be involved in acts of violence in addition to the intimate partner, as is the case in such countries as India where multigenerational households are common (Fernández, 2006; Kalokhe et al., 2015).
There has also been increased awareness that men can experience DV (Coker et al., 2002; Corbally, 2015; Emery et al., 2010; Wallace et al., 2019) and that both partners can engage in abusive behaviors (couple violence; Budd et al., 2017). Equivocal data from multiple studies regarding gender have been reported, with some studies showing that DV may be as prevalent in men as in women (Archer, 2000; Chan, 2011), whereas others report a gender differential, depending on the type of violence (Brownridge et al., 2008; Romans et al., 2007). However, there is general consensus that women experience higher rates of sexual violence and that DV results in more serious physical, mental, and financial consequences for women compared with men (Brownridge et al., 2008; Coker et al., 2002; Garcia-Moreno et al., 2003; McDonald, 1999). DV experienced by women was chosen for this article over the other common term intimate partner violence (IPV), as we believe it acknowledges the broader impacts of experiencing or witnessing physical, sexual, and/or emotional violence and abuse on the health of women and their children.
Epidemiological data on the enormity of DV as a public health issue and its impact on individuals and society vary across regions, nations, cultures, and populations. Large population surveys have made major contributions to increased awareness. For example, the WHO Multi-country Study on Women’s Health and Domestic Violence Against Women (Garcia-Moreno et al., 2003) reported lifetime prevalence rates of DV for 24,000 women in 10 different countries, with rates ranging from 15–71% with markedly higher rates among rural versus urban women. A follow-up WHO report (2019) found that 35% of women worldwide had experienced physical or sexual violence and that most of these women did not disclose nor seek any type of service related to their experiences. Various national prevalence studies also report variability, depending on when the study was completed, the regions involved, and the inclusion/exclusion of specific subpopulations of women (Hetling et al., 2006; see Raj & Silverman, 2002, for a review). For example, American national data from 2002 indicated that 28.9% of women had experienced DV in their lifetime (Coker et al., 2002), whereas data collected between 2010 and 2012 revealed 37.5% of American women had experienced DV at some time in their lives (Smith et al., 2017). Garcia-Moreno et al. (2003) reported a lifetime prevalence rate of 60% in a large sample of women in Bangladesh. Variability in the definition of DV used in studies, the study design and data collection tools used, the study time frame, and the strategies used to recruit potential participants also prevent pooling of data across studies.
An ongoing issue related to DV and a limitation of many previous large-scale epidemiology studies have been the exclusion or underrepresentation of minority subpopulations, such as immigrant women. This is especially problematic in countries like Canada, a multicultural nation with federally mandated high rates of immigration (Immigration, Refugees and Citizenship Canada, 2018 Annual Report to Parliament on Immigration). In 2011, more than 6.8 million individuals living in Canada were born outside of Canada (Salami et al., 2019). With the reported total population of Canada in 2011 of 33.48 million, immigrants represented more than 20% of the Canadian population. In 2018, 45,758 Permanent Residents were admitted into Canada. The top three source countries (representing 42% of the total) were India, the Philippines, and China (2019 Annual report from Immigration, Refugees and Citizenship, Canada). In the province from which our data were drawn, similarities to the federal data were observed in terms of the percentage of the population born outside of Canada (20.8%) and the top source countries.
Canadian prevalence data related to DV collected via the 1999 version of the Canada General Social Survey indicated that the last 5-year rates of DV increased among women for all types of violence except psychological violence (Romans et al., 2007). A review of 11 Canadian studies reported annual prevalence rates of DV as varying from 0.4–23% (Clark & Du Mont, 2003). Neither of these studies reported data specifically related to immigrant women. Other studies, many of which involved smaller sample sizes, were designed to examine whether immigrant women are more likely to experience DV compared with Canadian-born women. Cohen and Maclean (2003) reported that rates of DV were lower among immigrant women in Canada, but the risk increased with length of stay (Hyman et al., 2006a; Hyman et al., 2006b). This contrasts with other studies (e.g., Brownridge & Halli, 2002) that found a higher prevalence of DV among immigrant women, especially among those from developing countries. This latter finding is consistent with an American study that found higher rates of DV among Latino, South Asian, and Korean immigrants (Raj & Silverman, 2002). Other Canadian studies used pregnancy-related surveys (e.g., Canadian Maternity Experience Survey) or samples of pregnant women to estimate prevalence rate differences in DV between immigrant and Canadian-born women. Again, results have been inconsistent. Some data suggested that pregnant immigrant women were less likely to experience DV (Daoud et al., 2012), whereas other data suggested that long-term immigrants (Miszkurka et al., 2012) or recent immigrants had the highest prevalence rates of DV (Stewart et al., 2012). However, some researchers have cautioned against simplistic comparisons between immigrant and nonimmigrant women in relation to DV, raising concerns that these equivocal data can reinforce negative stereotypes regarding immigrant communities and the risk of oversimplification by inferring that immigrant women are a homogeneous group (McDonald, 1999; Okeke-Ihejirika et al., 2018; Sharma, 2001).
Reluctance by immigrant women to disclose DV experiences is often cited as contributing to equivocal epidemiological findings. Their reluctance may be partially explained by the existence of patriarchal social structures with traditional gender roles where family harmony and unity are considered a female responsibility (Wachholz et al., 2000); individual educational and socioeconomic status and neighborhood characteristics (Daoud et al., 2012; Stewart et al., 2012); societal tolerance to DV and support for silencing of women (Erez, 2000; Hyman et al., 2011; Shirwadkar, 2004); and women’s lack of awareness about their rights (McDonald, 1999; Oxman-Martinez et al., 2005; Thurston et al., 2013). In addition, immigrant women may be reluctant to seek services from government institutions (e.g., legal and justice systems, health care, children’s services, shelters for abused women), depending on their immigration status (Barrett & St Pierre, 2011; Oxman-Martinez et al., 2005; Shirwadkar, 2004), due to concerns about the impact on their immigration status (Alaggia et al., 2009; Salami et al., 2019). In addition, government-funded health, law enforcement, and social agency policies emphasize themes of “rescue & prosecute,” which may not suit immigrant women, their families, or their communities (Okeke-Ihejirika et al., 2018; Wachholz et al., 2000).
Immigrant women who do share their experiences of DV often seek informal services from family or friends, religious or community leaders, or community-based nongovernmental organizations (NGOs; Barrett & St Pierre, 2011). While NGOs provide a broad range of services and support, they often face unpredictable funding levels and typically focus their resources on the provision of programs and services. Although NGOs are viewed by some women as safer than government-funded services, data from NGOs and other community-based services are typically unavailable to researchers and government policymakers.
In summary, despite increased public awareness, service agencies and policymakers are severely hampered in their understanding of DV in various immigrant populations, in specific strategies that can support disclosure and the development of a full range of services. A research partnership developed between a researcher and an NGO provided a unique opportunity to contribute to our knowledge of DV in a population of immigrant women living in a large urban center in Canada.
Changing Together: A Center for Immigrant Women
Changing Together is a nonprofit charitable organization that was established in May 1984 as an urban resource and support center for immigrant women, both citizens and newcomers to Canada. A voluntary Board of Directors and an Executive Director provide leadership. Most staff are immigrant women themselves and provide multilevel English language education, pre-employment information, help with gaining employment, basic computer training, and help with immigration processes and counseling. Specific counseling services provided by the Social Worker (SW) may include help with food and/or financial insecurity, housing, immigration problems, parenting, legal concerns, and DV issues.
Documentation for each woman who books an appointment with the SW or “walked in” and requested service includes a paper-based intake form, plus the SW’s written summary notes on the presenting problem, contextual issues, and the suggestions, advice, or supports provided. These cumulative files have been kept in filing cabinet drawers and boxes since 2006 with no formalized method to capture the data electronically or to track the boxes as the NGO relocated several times over the years. Lack of sufficient computer knowledge and workload limited the NGO’s ability to develop an electronic intake form, which would have allowed the agency to characterize the population they serve, to evaluate their programs, and to gather detailed information for funding applications.
A long-term research partnership between the SW (N.Z.) working at Changing Together and the lead researcher (K.H., an experienced mental health nurse) provided a unique opportunity to identify both research-related and organizational objectives for a research project to advance our understanding of the extent and impacts of DV experienced by immigrant women seeking services from Changing Together. This long-term research partnership was based on reciprocity. For example, N.Z. and K.H. would meet to discuss counseling cases and potential mental health needs of the women and their families. A financial contribution from the project budget was also made to acknowledge that some of the SW time was taken up with this project and that there was no capacity in the NGO budget to support research activity. The ability to design a research study to examine the data reported here (N = 1,763) was a unique opportunity that would not have occurred if not for this professional relationship/research partnership.
The aims of the current study included both research- and organization-related objectives and outputs. Our research questions were the following: (a) What was the incidence of DV experienced by immigrant women who sought social work services at Changing Together between 2006 and 2014? (b) what types of service needs are associated with women who have experienced DV? and (c) what individual and environmental factors are associated with risk of DV in this population? Organizational outputs aimed at improving the agency’s ability to characterize their population and the services provided were also developed.
This article adds to the current literature in that it reports on the firsthand experiences from a community agency dedicated to providing community and social services to a culturally diverse population of immigrant women.
Method
This project involved a retrospective document review (using the method described by Vassar & Holzmann, 2013) of immigrant women’s interactions with the SW at Changing Together over a 9-year period (2006–2014). Research questions to be addressed by the file audit, audit time frame, and standardized definitions of variables were developed in conjunction with Changing Together staff. Research funding was obtained from the Women and Children’s Health Research Institute (WCHRI) in Alberta, Canada. Ethics approval was secured from the University Research Ethics Board, and all research team members signed confidentiality agreements with the agency. Distinct from a research study, where the study design lays out a consistent method of data collection, the data extracted in this study do not reflect standardized questions or self-reports, but rather initial assessments or counseling sessions between an immigrant woman and the community agency counselor and then recorded on the intake form postsession. The intake form included basic demographic information (name, address, contact information, age, marital status, employment outside the home, immigration status, current sources of support, education, and family composition) and contextual factors (country of origin, years in Canada, referring agency if relevant, police involvement). Check boxes on the form included types of services requested or suggested and a direct question regarding past or current experiences of DV. A standardized data extraction form along with definitions of all quantitative variables was developed for this project and used for data extraction. All SW notes were entered into a separate database for future qualitative analysis.
In keeping with the university’s Research Ethics Board, two databases were created: one for the agency, containing names, home addresses, and contact information, which would be given to the agency at the end of the project, and a research database with all identifying variables removed to ensure anonymity. Data entry was completed by a single person (A.M.), data cleaning was completed by A.M. and K.H., and data analysis and interpretation by all authors. Any data interpretive concerns were resolved in discussion with the principal investigator (K.H.).
Data Analysis
Data extracted included demographic (age, marital status, employment outside the home, immigration status, DV experiences, current sources of support, education, and family composition) and contextual factors (country of origin, years in Canada, referring agency if relevant, police involvement and past or current use of a shelter if reported DV). Although data entry spanned the years between 2006 and 2014, some files from 2008 and 2009 were lost during an agency relocation.
Data analyses included descriptive statistics and frequency data for all standardized variables from the intake form and the SW’s written notes. Incidence rates were calculated based on percent of the total sample (N = 1,763) who report DV. Univariate analysis for each of the standardized variables was used to compare women without and with DV, and risk ratio analysis was used to build best-fit models that predict risk for experiencing DV.
Results
There were 2,147 files available, of which 1,763 files contained sufficient data to be included in the analysis. However, some variables were not recorded in more than 50% of the 1,763 files and thus were not included in the overall demographics nor in the data analysis. These included age, education, current sources of support, and arrival date in Canada.
The demographic data are presented in Table 1. The marital status of all the women (N = 1,763) were grouped into six categories, as the relational dynamics between couples can create different risk factors associated with violence. Most women were married or in common-law relationships, had children living in the home, and were not employed outside their own home. The country-of-origin data did not necessarily reflect the country of birth as some women reported the country from which they came to Canada was not always their birth country. These data were categorized into regional groups based on the WHO Regions for our analysis. Immigration status was reported using six categories with the most reported status being permanent resident.
Demographic Details.
Table 2 reflects the services offered by the SW (as listed on the intake form) and the number of women who utilized each of the available services. More than 40% of women either came to the agency in relation to DV or disclosed having experienced DV when asked (as per routine agency policy). Each woman could request or be recommended to access more than one of the services, such that the total for all services exceeds 100%.
Changing Together Services and Utilization.
Table 3 presents the calculated significant risk ratios. Compared with single women and while holding all other variables constant, women who were separated from their spouses had 2.08 times the risk of reporting DV, divorced women had 1.71 times the risk, and married women or those living common-law had 1.33 times the risk. Compared to women with Canadian citizenship and holding all other variables constant, women whose reported immigration status was Visitor/Temporary or Permanent Resident had an increased risk of reporting DV (1.65 and 1.37 times higher, respectively).
Significant Effects on Risk Ratio.
Note. DV = domestic violence; CI = confidence interval.
Significant.
As stated previously, women could seek out any number of services, and as such, each service (housing, senior, immigration, finance, food, health, education, employment, parenting, legal, or other) was treated as a separate variable, with a separate effect. The only services that were not significantly associated with a change in the risk of DV were those related to being a senior and “other services” that did not fit into the other categories. Seeking housing, finance, parenting, or legal services each similarly increased the risk of reporting DV when compared with those that did not seek that specific service. The calculated risk ratios that reflected increased risk varied between 1.65 and 1.39, with seeking legal or parental services having the highest risk ratio in this cohort of women. Area of origin and women with children were the two variables that did not show any significant risk of DV within the risk ratio modeling.
Seeking other service categories (education, food, immigration, health, or employment services) significantly decreased the risk of reporting DV. The calculated risk ratios that reflected decreased risk were similar in relative impact to those that increased risk (range = −0.71 to −0.40). The current contexts of the women’s lives may have significantly contributed to these specific findings about decreased risk. Seeking information about how to improve one’s English, how to apply for a change in immigration status, and how to gain outside employment suggests a sense of settling into the community and looking to the family’s stability and future prosperity. It is less likely that women experiencing DV have these same priorities in terms of information needs. In addition, the women had not left their marital home; thus, food insecurity may not have been a priority.
Discussion
The reported DV incidence rate in this file review was 41%. This figure is in the range of some reports from national surveys and studies (Du Mont et al., 2012; Kalokhe et al., 2017; WHO, 2019). However, our figure of 41% is higher than reports from other countries, including Malawi (Palamuleni, 2019), Nigeria (Orpin et al., 2020), Finland (Kivelä et al., 2019), and Thailand (Chuemchit et al., 2018), and lower than that observed in Iran (Afkhamzadeh et al., 2019). It is likely that diversity in the populations involved the definition of DV, and the data collection methods used contribute to the broad range of reported rates of DV across countries.
Most of the studies related to the prevalence of DV completed in various countries include women within their countries of origin, rather than specific to immigrant women, so the data are not directly comparable with our findings. In addition, immigrant women from different cultures than those represented in our study may experience DV within a different context. A recent American study highlights such potential differences. Gonzalez-Guarda and colleagues (2011) aimed to describe the experiences of substance abuse, DV, and risky sexual behavior in Hispanic women in the United States. They postulated that for this population of immigrant women, these three types of experiences are interdependent. In unpublished work (by K.H.) from the same community agency as the findings presented in this article, a diagnostic interview based on the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association, 1994) was a component of the research design (N = 80). The incidence of substance use was <1%. As well, the dominant belief and value system precluded sexual activity outside of marriage. Interestingly, despite the likely differences in the relationship between substance abuse, DV, and risky sexual behaviors between this group and Hispanic female immigrants, the themes identified in the American study (uprooted in another world, the breeding grounds of abuse and breaking the silence) are similar to the unpublished work referred to here. Perhaps cultural contexts are important to consider in defining DV, in terms of decreasing barriers for disclosure, in identifying women who experience DV and in the services required to support women who chose to leave their partner; however, some of the psychosocial, emotional, and financial impacts may be similar across cultures.
Despite our finding that four of 10 women had experienced DV, it is likely that it underrepresents the extent of DV in the female immigrant population (Oxman-Martinez et al., 2005; Salami et al., 2019; Thurston et al., 2013; Wachholz et al., 2000). In addition, our findings do not suggest that immigrant women experience higher rates of DV than do Canadian-born women. What our findings do underscore is that DV is a relatively common experience for immigrant women and that having access to a dedicated immigrant women’s community-based social service agency may support disclosure and help-seeking to address the impacts on women and their children.
Immigration status was important to note as there are potential consequences should DV be reported to local law enforcement. For example, the current law in Canada states that if a DV-related incident is reported to the police, the police will take the responsibility of laying charges. This is significant as a criminal charge can influence immigration status or immigration applications for both partners. Immigrant women who have succeeded in securing permanent residency or Canadian citizenship are less likely to experience DV (Bhuyan, 2008; Hrick, 2012; Okeke-Ihejirika, 2017; Souto et al., 2019). It may be that while moving toward a more secure immigrant status, women learn more about their rights and about DV. For example, the Canadian Citizenship Study Guide includes this statement: In Canada, men and women are equal under the law. Canada’s openness and generosity do not extend to barbaric cultural practices that tolerate spousal abuse, “honour killings,” female genital mutilation, forced marriage or other gender-based violence. Those guilty of these crimes are severely punished under Canada’s criminal laws. (Citizenship and Immigration Canada, 2012)
Other potential contributors to the decreased risk for immigrant women who have gained citizenship include time spent in Canada, increased awareness of governmental and nongovernmental services, and increased access to services not available to noncitizens. Although time in Canada has been shown to increase or decrease risk of DV (Hyman et al., 2011; Stewart et al., 2012), these studies did not consider the role of becoming a citizen as a modifying factor in the time spent in Canada.
Another of the important findings of this study was that immigrant women who had experienced DV accessed or were identified as needing more services from Changing Together than women who had not had such experiences. Although this finding has been well recognized (Cohn et al., 2002; Du Mont et al., 2017; MacLeod & Shin, 1990; Novac, 2006; Salami et al., 2019), our data suggest a number of strategies for agencies and service providers to increase disclosure and better support women who experience DV. Policies that support inclusion of routine questions about DV should be implemented in all immigrant-serving agencies and services. However, such policies require carefully considered education for the staff, addressing staff comfort in discussing issues related to DV, and awareness of the broad range of services across the spectrum of available types of services and how to access or refer women. Screening without available support if the answer is “yes” has been shown to be ineffective in improving outcomes (MacMillan et al., 2009). For some women, very traditional patriarchal social contexts with clearly ascribed gender roles preclude them from being comfortable with disclosing DV in an immigrant-serving agency that employs both men and women. This perceived barrier must be addressed within the agency’s policy discussions. If universal screening regarding DV is not feasible, our findings suggest that identifying women who are accessing or seeking information about multiple services may help service providers ask specifically about DV in this high-risk group.
Other findings to note is that women who sought health services had a decreased risk of having experienced DV. One interpretation is that women who experience DV do not perceive health services as a supportive service. This would be in keeping with multiple other studies (Durbin et al., 2015; Kirmayer et al., 2007; Playfair et al., 2017; Weerasinghe, 2012) that reported lower rates of accessing health services, especially mental health services, and inequities in the quality of the services provided. A recent meta-analysis of interventions to improve immigrant health did not include any reference to DV, despite multiple other studies that suggest that it is an important issue for immigrant women’s health (Diaz et al., 2017).
Limitations
This data set includes records of information that were documented after the appointment with the SW had concluded. There were times when this was just not possible due to back-to-back appointments, urgent phone calls related to previously seen women or from other agencies, requirement to immediately access a space in a women’s shelter, and/or working with the police and Legal Aid. For the period reviewed, there was only one person doing the counseling (N.Z.), so time was also an issue that affected the level of detail in the files. The loss of an unknown number of files during an office relocation is also a limitation.
Some of the more common demographics were missing in many files (age, education, available social support, and arrival date in Canada). Previous studies have shown that these demographic variables are associated with risk for women to experience DV (Barrett & St Pierre, 2011; Brownridge et al., 2008; Oxman-Martinez et al., 2005; Thurston et al., 2013). We were unable to include these in our analyses, due to the percentage of missing data related to these variables. We hope that the new electronic database and user-friendly computer-based intake form will allow for more consistent reporting of these variables. However, this collaborative project has provided a glimpse into an important source of information about the issues related to DV for immigrant women. This project also highlights research partnerships with NGOs as an underacknowledged source of community-based research that can provide benefits to both the agency and a field of study.
Conclusion
This article provides some insight into the DV experiences of immigrant women seeking services from an NGO in Western Canada. Almost half of the immigrant women who sought services from the counselor were experiencing DV. The women who disclosed DV required multiple types of services compared with women who did not disclose DV. In terms of risk factor categories for experiencing DV, marital and immigration status were identified as significant risks, with separated and divorced women and women on a visitor or temporary visa showing the highest risk. The findings in this article are likely not broadly generalizable as they represent a subpopulation of immigrant women who sought services. It has been widely reported that women who experience DV across cultures and countries are reluctant to seek services. However, women’s stories of DV share commonalities across countries and cultures. Societal influences that perpetuate being silenced and feeling unheard, laws and policies that disadvantage women, and the many fears regarding leaving the relationship, language barriers, cultural norms, increased likelihood of financial dependence, and precarious immigration status all add to the burden of DV for immigrant women.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Women and Children’s Health Research Institute, University of Alberta, Edmonton, Alberta, Canada.
