Abstract
This systematic review sought to describe the prevalence of intimate partner violence (IPV) victimization among immigrants in the United States (U.S.) and the prevalence of IPV perpetration among immigrants in the U.S. PsycInfo, PubMed, Global Health and Scopus databases were searched for peer-reviewed literature that quantitatively examined IPV in relation to immigration. Twenty-four articles were included in the final review. Past-year IPV victimization rates among immigrants ranged from 3.8% to 46.9% and lifetime IPV victimization rates ranged from 13.9% to 93%; past-year IPV perpetration rates ranged from 3.0% to 24.8% and the one lifetime IPV perpetration rate was 12.8%. Estimates varied widely by country of origin, type of violence measured, and measure used to quantify IPV. Reliance on small convenience samples is problematic when trying to determine the true prevalence of IPV among immigrants. Epidemiological research is needed to improve the accuracy and representativeness of findings.
Introduction
Intimate partner violence (IPV) is a critical public health problem that transcends borders and is associated with a myriad of health consequences (Adhia et al., 2019; Bacchus et al., 2018). The Centers for Disease Control and Prevention (CDC) defines IPV as any physical, sexual, or psychological violence perpetrated by a current or former intimate partner and estimates that one in four women and 1 in 10 men in the United States (U.S.) experience intimate partner violence victimization (IPV-V) in their lifetime (CDC, 2021; Smith et al., 2018). Globally, one in three women experience lifetime IPV-V (Violence against Women Prevalence Estimates, 2018, 2021), and 25.4% to 80% of men report lifetime intimate partner violence perpetration (IPV-P) (Fulu et al., 2013). Global estimates for IPV-V of men and IPV-P by women are not readily available (Scott-Storey et al., 2022).
Known risk factors for IPV-V and IPV-P include low income, low education, being part of a minority group, experiencing stress, alcohol use, lack of social support, and living in communities where violence and crime are common (Spencer et al., 2022; Stith et al., 2004). These risk factors are prevalent among immigrant populations in the U.S. (Sabri et al., 2018) and create a bidirectional relationship between IPV and immigration. The term immigrant, is used in this study to encompass asylum seekers (a legally defined term referring to individuals who were forcibly displaced or who have fled their country of origin because of a threat to their lives or liberties and are seeking international protection), refugees (a legally defined term to refer to asylum seekers whose request for asylum has been granted) and migrants (a general term used to refer to individuals who leave their country of origin by choice typically to rejoin family or seek economic opportunities). While there may be unique risk factors for each of these groups, many of the risk factors of IPV are present among asylum seekers, refugees, and migrants (C. Kim & Schmuhl, 2020; Li et al., 2020; Moynihan et al., 2008; Sanz-Barbero et al., 2019). Acculturation theory and culture theory identify acculturative stress, lack of social capital, social isolation stress, language barriers, low social support, poor financial means, low education status, cultural barriers, and fear of deportation as specific risk factors for IPV among immigrants (C. Kim & Schmuhl, 2020; Sabri et al., 2018). The immigration and the resettlement process can create stressful environments exacerbating tension between intimate partners and resulting in increased IPV (Li et al., 2020). Isolation can be the result of physical and geographic separation and/or social seclusion from supportive networks created by a perpetrator, language barriers, and financial limitations, and can restrict a victim’s ability to seek help to address on going IPV and to prevent future violence (Li et al., 2020; Marrs Fuchsel & Brummett, 2020; Midlarsky et al., 2006; Moynihan et al., 2008). Additionally, many immigrants may not know their legal rights in the country where they resettle, leading to fear of deportation (Ingram, 2007; Li et al., 2020; Marrs Fuchsel & Brummett, 2020; Moynihan et al., 2008). One study also found that immigration status (either by forcible displacement or by choice) was a specific risk factor for emotional IPV-V as victims reported perpetrators threatening them with deportation (Rai & Choi, 2021). Furthermore, immigrants women from strongly patriarchal societies are at increased risk of IPV-V in part due to men holding traditional gender norm beliefs that may sanction the use of violence within relationships or position women as possessions of men (Sikweyiya et al., 2020; Tonsing & Tonsing, 2019). Known rates of IPV from patriarchal societies are likely underestimates as women may be less likely to report instances of IPV-V, and if they do report, their social groups may dismiss the event as acceptable (Li et al., 2020; Midlarsky et al., 2006; Moynihan et al., 2008; Sabri et al., 2018).
Conversely, experiencing IPV-V in one’s country of origin is a risk factor for immigration when IPV victims immigrate in order to flee abuse (C. Kim & Schmuhl, 2020; Li et al., 2020; Moynihan et al., 2008). Other risk factors for immigration include low income, financial stress, poor job prospects, and living in environments experiencing high levels of crime, violence, and political unrest (Efendic, 2016; Parkins, 2010). Each of these risk factors is also a risk factor for IPV-V (C. Kim & Schmuhl, 2020; Sabri et al., 2018). History of IPV-V is also associated with future risk of IPV-V (Black et al., 2011). Since IPV-V is common in areas of conflict and instability, immigrants who are fleeing such areas may have higher rates of previous IPV-V experience, putting them at increased risk for future IPV-V (Manjoo & McRaith, 2011; Stark & Ager, 2011).
Prior research has examined the prevalence of IPV-V and IPV-P among specific groups of immigrants (Gonzalez-Guarda et al., 2013; Marrs Fuchsel & Brummett, 2020; Raj & Silverman, 2003). For example, one study found that the past-year prevalence estimate among Asian-Indian immigrant women was 46.9% (Munisamy, 2010). However, there is no review that synthesizes research on prevalence rates of IPV-V and IPV-P among all immigrants in the U.S. Given that IPV-V may be one of the motivations for immigrating, and that once one immigrates, additional IPV-V risk factors arise (e.g., language barriers), we conducted this review to assess if and how immigration status varied as a risk factor for all immigrant populations within the U.S. These estimates may help to better describe the relationship between IPV and immigration, and the demographics of individuals affected by IPV.
Understanding the relationship between IPV and immigration in the U.S. could have implications for tailoring preventative interventions and designing immigrant-specific referral resources. Understanding the prevalence of IPV among immigrants is crucial to appropriately prioritizing resources for those most at risk. Quantifying the nature and scope of this problem allows us to identify gaps and patterns that may inform future practice, policy, and research. Therefore, this systematic review aims to synthesize existing data to provide two estimates: the prevalence of IPV-V and the prevalence of IPV-P reported among immigrants in the U.S. In addition, we present prevalence rates by region, report on sampling methods, IPV assessment tools, and immigrant country of origin.
Methods
Selection Criteria
The systematic review protocol is available in Appendix 1. The review follows Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting guidelines (Figure 1). Studies were eligible for inclusion if they reported results of a primary research study and contained quantitative data on the prevalence of IPV-V or IPV-P among a sample of first-generation immigrants in the U.S. Studies were excluded if they: (1) did not include a prevalence estimate, (2) were not primary research or secondary data analyses (e.g., case reviews or systematic reviews, gray literature, editorials), (3) were not written in or translated into English, (4) reported on immigrant populations settled outside of U.S., or (5) did not contain information about IPV (i.e., reported on violence not perpetrated by an intimate partner). When screening articles reporting on the prevalence of IPV-V or IPV-P among immigrants, we included articles that contained a sample or sub-sample of only immigrants. No restrictions were placed on year of data collection or publication, participant country of origin, or participant age.
PRISMA Diagram.
Search Strategy and Data Abstraction
PsycInfo, PubMed, Global Health, and Scopus databases were searched in September 2021 using a combination of Medical Subject Headings (MeSH) and text words. Search terms were grouped into four main themes: IPV, immigrants and immigration, survivors and victims, and perpetrators. The four groups of search terms were combined, and articles were further filtered to include only articles available in English. The exact search terms and strategy can be found in Appendix 2. Returned articles were downloaded from each database, duplicates were removed, and articles were uploaded into Covidence, a systematic review automation manager. All titles and abstracts were independently reviewed by at least two of the first three authors (AM, JC, or LS). Each reviewer was blinded to the other reviewer’s decisions until screening was complete and discrepancies were flagged by the automation manager. All three reviewers met to resolve discrepancies and agreement was reached regarding which articles met the criteria for a full text review. Two reviewers independently assessed the full text of each potentially relevant article and all reviewers met again to resolve discrepancies. Reviewers also hand searched the reference sections of each included paper for additional studies to include before data abstraction occurred. The one hand-identified study underwent a full-text review by two reviewers for eligibility. Reviewers designed a data abstraction matrix and independently abstracted data from the same article to assess internal consistency. Discrepancies were addressed before data abstraction continued. Data from each article with a clearly presented prevalence rate was abstracted by a single reviewer. In articles where the prevalence rates were not immediately apparent, all three reviewers met to collectively abstract data from these articles. The data abstracted included: article title, author names, publication year, prevalence question addressed, years of data collection, sample size, relevant sample demographics (age, gender, number of years in the U.S.), participant country of origin, tool used to assess IPV, and types of IPV measured.
Results
The study selection process is presented in the PRISMA diagram (Figure 1). The initial database search returned 1,760 unique articles and of these, 716 were removed as duplicates, and 869 were excluded based on title and abstract. One additional article was identified through citation searching. A total of 176 articles were marked for full-text review. Of these, 18 articles could not be retrieved due to lack of institutional access, 33 were excluded because they were not primary research, 29 were excluded because they did not contain a relevant prevalence estimate, 36 were comprised of the wrong population (e.g., the entire sample was immigrant victims of IPV, thus no prevalence estimate could be calculated), 19 had no full-text available (due in part to lack of institutional access), 16 were not set in the U.S., and 1 was not available in English. The final sample contained 24 articles all with unique sample groups. All included studies assessed IPV or immigration status at a single point in time. Studies used differing timeframes for assessing IPV including past-year and lifetime prevalence. Only one study reported on whether the IPV (in this case, IPV-P) occurred prior to or after immigration.
Studies were grouped based on which of the two research questions they answered (Tables 1 and 2). Of the 24 studies, 20 provided data on the prevalence of IPV-V among immigrants in the U.S. and 6 provided data on the prevalence of IPV-P among immigrants in the U.S. Two studies included estimates of both the prevalence of IPV-V and IPV-P among immigrants and are thus included in both groups.
Prevalence of IPV-V among Immigrants.
Note. IPV, intimate partner violence; IPV-V, intimate partner violence victimization
Prevalence of IPV Perpetration among Immigrants.
Note. IPV, intimate partner violence; IPV-P, intimate partner violence perpetration.
Although IPV can occur in all types of intimate partnerships, the studies included in this manuscript did not specifically comment on the sex, gender identities, or sexual orientations of the participants or their partners. Therefore, in this manuscript, we refer to participants mainly as “men” or “women” to remain consistent with the studies’ original language.
Prevalence of IPV Victimization among Immigrants
The 20 studies that reported the prevalence of IPV-V estimated rates ranging from 3.8% (Kimber et al., 2015) to 46.9% (Munisamy, 2010) for past-year IPV-V and 14.3% (Shibusawa & Yick, 2007) to 93% (Barkho et al., 2011) for lifetime IPV-V. Half of the 20 studies (50%), measured IPV-V using the Conflict Tactics Scale (or an adaptation), 12 (60%) measured more than one form of IPV-V (e.g., physical violence and sexual violence), 18 (90%) used more than one item to assess IPV-V, 13 (65%) offered study materials in a language other than English, and 14 (70%) used convenience sampling. There were 14 (70%) studies that had a sample comprised entirely of women. Of the five studies including men and women, only one provided estimates of IPV-V separated by gender. Among studies that measured only one kind of IPV-V (e.g., only physical violence), past-year and lifetime prevalence rates ranged from 5.6% (Shibusawa & Yick, 2007) to 32.9% (Finno-Velasquez & Ogbonnaya, 2017) and 13.9% (Shibusawa & Yick, 2007) to 39% (Finno-Velasquez & Ogbonnaya, 2017) respectively.
Among the 14 studies that used convenience samples, the past-year estimates for any IPV-V ranged from 21.2% (Raj & Silverman, 2007) to 46.9% (Munisamy, 2010), while the 6 studies that used randomly selected samples reported past-year estimates from 3.8% (Kimber et al., 2015) to 32.9% (Finno-Velasquez & Ogbonnaya, 2017). Studies that used convenience samples reported estimates of lifetime IPV-V ranging from 0% (Cuneo et al., 2021) to 93% (Barkho et al., 2011), and studies that used random samples reported lifetime estimates ranging from 13.9% (Shibusawa & Yick, 2007) to 39% (Finno-Velasquez & Ogbonnaya, 2017). Most studies (n = 24, 92%) had fewer than 1,000 participants. The two studies with more than 1,000 participants were random samples (Ingram, 2007; Kimber et al., 2015). Of these, one reported a past-year prevalence rate of 8.7% (Kimber et al., 2015) and the other reported a lifetime prevalence rate of 43.5% (Ingram, 2007).
A convenience sample of 55 Iraqi immigrant women report the highest rate of lifetime IPV-V among all studies examined: 93% of participants reported experiencing controlling IPV-V behavior in their life (Barkho et al., 2011). A study conducted by Kimber et al. (2015) that used a random sample of 5,128 immigrants from unspecified countries measured past-year exposure to physical and sexual IPV-V among immigrants who had and had not experienced child maltreatment. Findings from this study reflect the lowest reported rates of past-year IPV-V among all studies examined (3.8%) (Kimber et al., 2015).
Studies with immigrant participants originally from China, Taiwan, Hong Kong, and Vietnam reported past-year estimates between 5.6% (Shibusawa & Yick, 2007) and 21% (Li et al., 2020) and lifetime estimates between 13.9% (Shibusawa & Yick, 2007) and 53.3% (Tran, 1997). Estimates from studies with participants from India, Nepal, Pakistan, Sri Lanka, and Bhutan report past-year estimates between 21.2% (Raj & Silverman, 2007) and 46.9% (Munisamy, 2010) and lifetime estimates between 48% (Rai & Choi, 2021) and 50% (Pallatino, 2018). Countries with participants from Central and South America reported a past-year prevalence rate of 17% (Sabina et al., 2021) and lifetime prevalence rates between 0% (Cuneo et al., 2021) and 60.0% (Hass et al., 2000).
Prevalence of IPV Perpetration among Immigrants
Seven studies reported on the estimated prevalence of IPV-P among immigrants (Table 2). Six of these (86%) assessed IPV-P using some form of the Conflict Tactics Scale, 4 (57%) offered study materials in a language other than English, and 4 (57%) were composed of participants selected using convenience sampling. Of the six studies that reported on the prevalence of IPV-P among immigrants, four reported on past-year IPV-P with estimates ranging from 3.0% (Kimber et al., 2015) to 24.8% (Gupta et al., 2010). One study of immigrant men reported on lifetime prevalence measured with a single item and found that 12.8% reported ever hitting or throwing things at their partner (Sorenson & Telles, 1991). Two studies used mixed-gender samples to examine IPV-P and reported past-year rates ranging from 3.3% to 18.9% and lifetime prevalence of 12.8% (Sorenson & Telles, 1991; Vaughn et al., 2015). Two studies included only men and reported past-year IPV-P rates from 9.5% to 24.8% (Gupta et al., 2009, 2010).
The lowest rate of past-year IPV-P is derived from a nationally representative sample that measured physical and sexual IPV-P among 5,128 immigrants from unspecified nations (Kimber et al., 2015). This study stratified the sample by those who had ever experienced child maltreatment, and those who had never experienced child maltreatment. They found that only 3.0% of respondents who had never experienced child maltreatment and 9.5% of respondents with a history of child maltreatment reported past-year IPV-P (Kimber et al., 2015). The highest estimate of past-year physical or sexual IPV-P (24.8%) is derived from a study by Gupta et al. (2010) in which the countries of origin were not specified for the 1,668 immigrant men in the sample. One prevalence rate of sexual coercion among a convenience sample of 98 immigrant women from India was higher at 30.6%, but the timeframe was not specified (Munisamy, 2010). One study examined IPV-P using convenience samples of immigrants from Central and South America, the Caribbean, and Puerto Rico, and found the prevalence of IPV-P to range from 9.5% (physical violence perpetration among men) to 17.9% (any violence perpetration among men) (Gupta et al., 2009).
Discussion
This systematic review found wide-ranging estimates of the prevalence of IPV-V and IPV-P among immigrants in the U.S. (Table 3). Overall, findings suggest that the degree to which immigrant status is correlated with IPV-V and IPV-P varies depending on how IPV-V and IPV-P were measured and the country of origin of the sample population.
Summary of Critical Findings.
Note. 1. (Shibusawa & Yick, 2007), 2. (Barkho et al., 2011), 3. (Kimber et al., 2015), 4. (Munisamy, 2010), 5. (Sorenson & Telles, 1991), 6. (Gupta et al., 2009).
IPV Victimization among Immigrants
When examining IPV-V among immigrants, our findings suggest that some groups of immigrant women experience rates of IPV that are higher than, or at least comparable to, estimated rates among U.S.-born women. The prevalence of lifetime IPV-V among some immigrant groups in the U.S. is as high as 93% (Barkho et al., 2011). This prevalence is high when compared to a lifetime IPV-V prevalence of approximately 36.4% among U.S.-born women and 33.6% among U.S.-born men (Smith et al., 2018). While Barkho et al. (2011) and Smith et al. (2018) both assessed lifetime IPV-V, Barkho et al. (2011) defined IPV-V as any controlling or threatening behavior, or physical abuse by a partner, and Smith et al. (2018) defined IPV-V as sexual or physical violence, stalking, or psychological aggression. “Controlling behavior,” as defined by Barkho et al. (2011) can be widely interpreted and broad interpretations may help explain the elevated prevalence estimate derived from this study. Additionally, Barkho et al. (2011) used a convenience sample size of 55 women and Smith et al. (2018) used a nationwide random sample of 5,758 women and 4,323 men. Though the article by Smith et al. (2018) was not included in this review as it did not meet eligibility criteria, the comparison to Barkho et al. (2011) illustrates the general variation in prevalence rates of IPV-V found in this review: studies typically reported higher estimates when they measured lifetime IPV-V rather than past-year IPV-V, used convenience samples rather than random samples, had fewer than 1,000 participants, used some form of the Conflicts Tactics Scale, measured more than one form of IPV, and used more than one item to measure IPV-V. Further, women typically experienced higher rates of IPV-V than men.
There are several possible explanations for the wide ranges of estimates and for the general patterns observed. First, there was variation in the timeframes in which IPV-V was measured (e.g., Barkho et al., 2011; Munisamy, 2010). In general, reported lifetime IPV was typically higher than past-year IPV, likely because the measure captures a wider timeframe and larger sample (e.g., Barkho et al., 2011; Hass et al., 2000). This pattern is evident both across studies and within studies that measured both lifetime and past-year IPV.
Second, studies used a variety of instruments to capture IPV-V rates including versions of the Conflict Tactics Scale, the Abuse Assessment Screen (e.g., Ingram, 2007), and qualitative coding of asylum applications (e.g., Cuneo et al., 2021). Although some of the scales used to assess IPV-V were validated in English, many of the scales were translated into different languages, and construct validity was not assessed after translation (e.g., Shuman et al., 2021). Thus, it is possible that errors in translation or back-translation biased estimates. One exception is the Conflicts Tactics Scale, which has been translated and validated for use among non-U.S., non-English speaking populations (Aldarondo et al., 2002; Straus et al., 1996). Additionally, many of the questions derived from scales designed for U.S. populations may not have adequately captured all of the experiences of IPV that individuals from other cultures may have experienced. It is also possible that some immigrants from countries with different cultural and gender norms may not interpret violent behavior in the same way that it is interpreted in the U.S. For example, the CDC defines IPV as “physical violence, sexual violence, stalking, and psychological aggression by a current or former intimate partner” (Breiding et al., 2015). However, sexual violence within a marriage may not be considered to be IPV by all immigrant participants, especially among immigrants from highly patriarchal countries (Midlarsky et al., 2006). Thus, it is possible that some of the scales used mischaracterized IPV experiences or did not capture IPV experiences based on a U.S.-centric understanding of IPV.
Third, studies that specifically measured multiple forms of IPV and used more than one item to assess prevalence typically reported higher rates than studies that only measured one type of IPV or used a small number of items (e.g., Ingram, 2007; Millett et al., 2015). Estimates based on only physical or sexual violence will underestimate the true prevalence of IPV (as defined by the CDC) as participants may not have had designated space to report other forms of abuse. This may also be true in studies with few items assessing occurrences of IPV. For example, one study used only six items to assess physical and sexual IPV-V (Kimber et al., 2015). Affirmative answers to these questions may vary depending on how the participant defines IPV and may not allow room to gather information on emotional or financial abuse. Alternatively, studies that asked a number of questions about various abusive behaviors may present clearer, more defined, opportunities for participants to report instances of any type of violence and will result in increased prevalence estimates, as found in one multi-country study by Fulu et al. (2013). For example, when items ask specifically about a partner forcing a victim to have sex when they did not want to, this might elicit a different response than an item that asks whether the victim experienced sexual violence. Presenting clear examples of specific behaviors that indicate IPV-V could help to capture more accurate prevalence rates of IPV-V and get around issues of definitions varying across individuals and cultures.
Fourth, studies that had samples comprised entirely of women (or reported prevalence estimates separated by gender) reported higher prevalence estimates of IPV-V than mixed gender or samples of all men. For example, the lower estimate in the range of lifetime prevalence of IPV-V (13.9%) is derived from men within a mixed-gender sample while the upper estimate (93%) is derived from a sample of all women (Barkho et al., 2011; Shibusawa & Yick, 2007). The lower estimate in the range of past-year prevalence of IPV-V (3.8%) is from a combined sample of men and women, while the upper estimate (46.9%) is derived from a sample of all women (Kimber et al., 2015; Munisamy, 2010). This is consistent with literature, which finds that prevalence rates of IPV-V are higher among women than men (Violence against Women Prevalence Estimates, 2018, 2021).
Fifth, in terms of sampling methods, studies used either convenience sampling or randomized sampling methods, and rates were generally higher in studies using convenience samples compared to random samples (e.g., Millett et al., 2015; Munisamy, 2010). Convenience sampling was often used in health facilities or in immigrant service organizations that have formal reporting opportunities (e.g., Hass et al., 2000). Individuals seeking services probably differ from the general population in that they are likely experiencing more severe forms of IPV (Duterte et al., 2008) and thus are more likely to identify what they are experiencing as abuse (E. Kim & Hogge, 2015). They are also already seeking help which indicates they may be more willing to report their abuse in study surveys. Furthermore, studies that used convenience samples had smaller sample sizes overall, likely resulting in a larger margin of error than nationally representative studies.
One of the aims of this review was to examine the prevalence of IPV-V among immigrants from any country of origin in order to determine whether immigration status may be conceptualized as a risk factor for IPV-V among immigrants in the U.S. While the variety of methodologies and IPV definitions likely contributed to the wide prevalence estimates, findings suggest that the degree to which immigration status is correlated with IPV varies depending on the country or region of origin. When examining regional patterns, the highest estimate of any type of lifetime IPV-V (93%) was derived from a convenience sample of women immigrants from Iraq (Barkho et al., 2011), and the lowest estimate of any type of lifetime IPV-V (0%) was derived from a convenience sample of men from El Salvador, Guatemala, and Honduras (Cuneo et al., 2021). In general, studies with immigrants from Asia reported higher lifetime estimates of IPV-V, and lower past-year estimates when compared to studies with immigrants from Central and South America. As previously mentioned, IPV may be conceptualized differently among populations with different beliefs about topics such as gender, marriage, or power. Among immigrants from regions where patriarchy is the norm, IPV is likely to be more prevalent (Sikweyiya et al., 2020; Tonsing & Tonsing, 2019). For instance, the WHO estimates that Southern Asia has some of the highest rates of IPV globally (Violence against Women Prevalence Estimates, 2018, 2021) and a systematic review by Pande et al. (2017) found that a main driver of IPV in Southern Asia was patriarchy. This current review found that immigrants from Southern Asia consistently report high rates of IPV. It is also possible that regional variations are due to differences in reporting. Highly patriarchal regions may have lower levels of reporting because IPV is viewed as normal and acceptable (Li et al., 2020; Midlarsky et al., 2006; Moynihan et al., 2008; Sabri et al., 2018). Future qualitative research could be undertaken to help clarify the contextual risk factors among certain groups. Contextual factors identified through future research could also be compared across countries and regions in order to gain a richer understanding of contextual risk factors to help guide the development of prevention and intervention protocols.
It is also worth noting that estimates may vary because underreporting is likely (Marrs Fuchsel & Brummett, 2020; Smith et al., 2018). Underreporting of IPV experiences due to shame, fear, or misunderstanding would underestimate the prevalence of IPV in immigrant victim populations (Palermo et al., 2014).
IPV Perpetration among Immigrants
We found the past-year prevalence of IPV-P among immigrants in the included studies to be between 3.0% and 24.8%. The lower bound of this estimate (3.0%) represents a sample of immigrant men who were chosen specifically because they never experienced any form of child maltreatment (Kimber et al., 2015). Given that childhood maltreatment is a risk factor for IPV-P (Manchikanti Gómez, 2011), it is logical that the prevalence of IPV-P would be lower in this group. Excluding this estimate, the reported range of past-year IPV-P among immigrants is between 8.99% (Vaughn et al., 2015) and 24.8% (Gupta et al., 2009). In both of these studies, the participants’ country of origin was unspecified, so regional differences are hard to parse. Three studies did indicate the participants’ country of origin. Two studies contained populations from Mexico, Puerto Rico, the Dominican Republic, other Caribbean countries, and unspecified countries within Central America. These studies reported IPV-P prevalence rates of 12.8% (Sorenson & Telles, 1991) and 17.9% (Gupta et al., 2009). Previous literature suggests that gender roles promoting aspects of “machismo” among Mexican American men and men from South America is a risk factor for IPV-P (Goicolea et al., 2012; Mancera et al., 2017). However, one of these studies (Gupta et al., 2009) also included participants from Cape Verde and other unspecified African countries without differentiating IPV-P prevalence by region. The third study included participants from India and reported an IPV-P prevalence of 30.6% (Munisamy, 2010). This study looked at IPV perpetrated by women against their husbands. Without a similar study (regional or looking specifically at women as perpetrators) to compare prevalence rates, it is unclear whether this estimate is representative of the region. For all studies examining IPV-P, the cross-sectional nature of this data, does not allow us to determine whether individuals began perpetrating IPV before or after immigrating. Nevertheless, these results suggest that IPV-P among immigrants is common.
Limitations of this Review
There are five notable limitations of this review. First, while it was not the goal of this research to produce findings that are generalizable to other countries, estimates here can only be considered within the context of the U.S. Second, most studies did not specify when the IPV-V or IPV-P took place in the context of migration. The studies included in this review used measures to capture IPV within the past-year or during one’s lifetime, but only one also specified whether the violence occurred pre- or post-migration (Baranowski et al., 2019). Knowing when the violence occurred would help to examine whether immigration is a causal risk factor for IPV-V or IPV-P and may help to clarify determinants of the violence as they are likely different before, during, and after immigration. Future studies should seek to include this information, and current measures of IPV should be adapted for use among immigrant populations so that this level of detail can be captured. Nevertheless, it is still important to recognize that past violence predicts future violence. Third, studies did not provide information on the immigration status of the abusive partner. For instance, where the victim may be an immigrant, the perpetrator may be U.S.-born or vice versa. One risk factor for IPV-V is fear of deportation if one reports. This may be more relevant when the perpetrator is U.S.-born and the victim is not, as it creates a power imbalance that could exacerbate IPV. Understanding such dynamics between partners may help better explain the wide prevalence ranges. It is possible that lower prevalence estimates result when underreporting is likely, and that higher prevalence estimates result when the victim is U.S.-born and familiar with the legal system. Fourth, included articles collected data across four decades. Immigration factors vary based on the social and political climates in origin and destination countries. For example, there have been changes in U.S. immigration policy during this 38-year time frame. Using cross-sectional studies, as done here, does not accurately reflect changes in migration patterns. Future studies may seek to address this question using longitudinal studies. Most recently, there is also emerging literature on the effect of the COVID-19 pandemic and related policies on IPV-P and IPV-V on immigrant populations. Future research may seek to expand upon our work to assess patterns in IPV among immigrant populations during a pandemic. Fifth and lastly, the majority of studies included in this review were comprised of small convenience samples. This limits the strength of the study conclusions and indicates a need for larger scale studies comprised of random samples (Table 4). The heterogeneity of measures used to assess IPV-V also limited the ability of the authors to conduct a meta-analysis of prevalence rates.
Implications for Practice, Policy, and Research.
Conclusion
This systematic review identified relatively few studies examining potential relationships between IPV and immigration. Moreover, the estimates of IPV-V and IPV-P vary widely. This is likely, in large part, due to differences in methodological approaches. Therefore, standardized measures, larger random samples, and longitudinal studies are critical to more accurately estimate the scope of the problem. Such standardization would allow future reviews to conduct a meta-analysis or pooled prevalence estimate of IPV-V and IPV-P rates among immigrants. Gaining such knowledge may help to inform immigration and IPV-related policy, practice, and future research.
Supplemental Material
sj-docx-1-tva-10.1177_15248380231165690 – Supplemental material for Intimate Partner Violence and Immigration in the United States: A Systematic Review
Supplemental material, sj-docx-1-tva-10.1177_15248380231165690 for Intimate Partner Violence and Immigration in the United States: A Systematic Review by Abigail M. Morrison, Julia K. Campbell, Laurel Sharpless and Sandra L. Martin in Trauma, Violence, & Abuse
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) received no financial support for the research, authorship, and/or publication of this article.
Supplemental Material
Supplemental material for this article is available online.
Author Biographies
References
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.
