Trauma, immigration, and sexual health among Latina women: Implications for maternal–child well‐being and reproductive justice

ABSTRACT Latina immigrant women are vulnerable to traumatic stress and sexual health disparities. Without autonomy over their reproductive health and related decision‐making, reproductive justice is elusive. We analyzed behavioral health data from 175 Latina immigrant participants (M age = 35; range = 18–64) of the International Latino Research Partnership (ILRP) study. We used descriptive and inferential statistics to compare immigrant mothers of minor children to those without, regarding their psychological and reproductive health, and correlates of past exposure to sexual trauma. Over one third (38%) of ILRP participants had minor children, and 58% had citizenship in their host country. The rate for sexual assault was 30 and 61%, respectively, for physical assault; these rates were similarly high for women with and without minor children. Women who reported sexual assault scored significantly higher for depression, posttraumatic stress disorder, and substance‐abuse screens. Odds of experiencing sexual assault was highest for women who experienced physical assault (odds ratio = 10.74), and for those from the Northern Triangle (odds ratio = 8.41). Subgroups of Latina migrant mothers are vulnerable to traumatic stress and related sexual and mental health risks. Given these findings, we frame the implications in a reproductive justice framework and consider consequences for caregiver–child well‐being.


INTRODUCTION
Reproductive justice entails women and girls having autonomy over decisions affecting their own well-being and that of their families and communities (Smith, 2005). Coined first by African American women in a 1994 convention on reproductive health, the term reproductive justice conceptualizes the combining of reproductive rights and social justice, and is a repudiation of a centuries-old history of practices meant to limit reproductive options for women of color, indigenous people, immigrants, and other marginalized communities (Copelon & Petchesky, 1995). The critical values of reproductive justice are "(1) the right not to have a child; (2) the right to have a child; and (3) the right to parent children in safe and healthy environments" (Ross & Salinger, 2017, p. 65). The framing of these issues from a reproductive justice lens readjusts our focus onto public policies that have an economic and health impact, particularly on vulnerable women, and that are rarely explored in the traditional debates on abortion, reproductive choice, and caregiver well-being. This current study examines factors that may diminish reproductive justice for Latina immigrants with children and how this might impact their well-being.

Immigrant women and reproductive justice
Marginalized women of color experience systematic barriers for accessing contraception, comprehensive sex education, sexually transmitted infection prevention and care, alternative birth options, adequate prenatal and pregnancy care, domestic violence assistance, adequate wages to support their families, safe homes, healthcare, mental health services, and more. For Latina immigrant and refugee women, policies that influence migration and mental health/healthcare access and address poverty are of vital importance for reproductive justice, maternal health, and caregiver-child well-being. The National Women's Law Center ( 2017) highlighted that Latina immigrant women seeking to escape sexual trauma and community violence are stereotyped as opportunistic; as simply seeking to give birth and gain employment in the United States. Such characterizations have been used to justify the codification of immigration policies that effectively limit access to healthcare and foster discrimination and fear (National Women's Law Center, 2017). In fact, some federal health insurance policies directly prevent access to seek reproductive health services, preventive care, or other treatment for immigrant and refugee women and their families (National Women's Law Center, 2017).

Immigrant women's experiences of violence, trauma, and marginalization
Experiences of trauma, violence, and poor mental health complicate the circumstances and barriers for attaining reproductive justice among Latina immigrant women. They may be at increased risk for intimate partner violence (IPV), as compared to other minority groups, placing them at greater risk for posttraumatic stress disorder (PTSD), lifelong health problems, and disparities (Galano, Grogan-Kaylor, Stein, Clark, & Graham-Bermann, 2016). Kaltman, Green, Mete, Shara, and Miranda (2010) found that Latina immigrant women who had experienced IPV from rape and physical violence were at highest risk for PTSD and depression comorbidity. Because IPV can compromise the caregivers' ability for attunement and empathy with their child, it can also decrease their capacity of appraising and responding to danger in the environment. Thus, these become vulnerabilities for caregiverinfant well-being (Lieberman & Van Horn, 1998;Lieberman, Van Horn, & Ippen, 2005) and can lead to impaired parenting practices. IPV can disrupt the caregiver-child relationship and is especially toxic for young children who are exposed to violence, as it damages the perception of the caregiver (mother) as a reliable and safe protector (Lieberman & Van Horn, 2008). In addition, disenfranchised Latina immigrant mothers face the consequences of gender, class, racial oppression, and consequently poverty, resulting in poor access to protective systems of social support and mental health services which could otherwise mitigate developing posttraumatic stress and revictimization (Crenshaw, 2005;Galano et al., 2016). Undocumented Latina immigrants are more likely to have higher rates of IPV and involvement with the child welfare system, as compared to women who are permanent residents or citizens (Ogbonnaya, Finno-Velasquez, & Kohl, 2015). Risk is highest for IPV during pregnancy and after childbirth, and immigrant status can increase dependency on a partner, leaving women less likely to report abuse and to seek asylum or legal aid (Coutinho et al., 2015) out of fear of deportation and family separation as well as stigma.

Healthcare policy, reproductive health, and mental health services
Health reform initiatives actively bar immigrant women from receiving safety net health benefits (National Latina Institute for Reproductive Health, n.d.). While the Patient Protection and Affordable Care Act (2010) expanded coverage for preventive healthcare such as family planning and screening for sexually transmitted diseases and domestic violence, it excluded undocumented immigrants from these benefits. Other policies legislated within the last 10 years have restricted access to abortion services in community clinics and caused reductions in access to affordable reproductive health services and contraception care (Nash, Gold, Ansari-Thomas, Cappello, & Mohammed, 2017). These laws disproportionately impact undocumented women, but also other low-income women who rely more heavily on community clinics to receive reproductive health services.
In the United States, there is a multiple-payer system (Medicaid, Medicare, private insurance, and safety net programs) with great diversity in access, the copayments billed to the patient, and in which many insurance plans limit the total number of mental health visits (Saloner, Bandara, Bachhuber, & Barry, 2017). Undocumented immigrants and authorized immigrants with less than 5 years of legal residence cannot receive federal insurance subsidies or enroll in Medicaid (a social healthcare program for low-income individuals). Most Latina immigrants, even those who have docu-mented status, receive mental healthcare in the United States through community-based clinics where a primary care doctor treats them or refers them if they accept public insurance or the clinic receives federal subsidies; but they still confront colossal barriers in the process of care (Agency for Healthcare Research and Quality, 2015). Spain has a single-payer system in which copayments for mental healthcare visits are not common and an unlimited number of mental health visits are allowed (Dezetter et al., 2013). In both countries, the main entrance to the mental healthcare system is through primary care centers, and significant barriers to specialty mental healthcare exist . These policies, whether part of immigration, social welfare, or healthcare systems, determine what services and health-promoting resources immigrant women can access for themselves and their children.

RELATIONSHIP OF MATERNAL WELL-BEING AND REPRODUCTIVE JUSTICE
Maternal well-being and reproductive justice are interrelated and have direct implications for the development of an infant's capacity for emotional connection, regulation, and expression, as influenced by the quality of the child-caregiver relationship (Balbernie, 2002). Mothers in a deeply traumatized state have more difficulty performing the basic tasks of attunement and mirroring necessary for healthy attachment with their child, who thus experiences relational trauma (Schore, 2013). Greater understanding of these mechanisms along with increased screening for Latino immigrant families, especially pregnant and new mothers and their infants, are essential to mitigate the effects of transgenerational trauma (Phipps & Degges-White, 2014). Traumatic stress, violence exposure, and associations with reproductive justice, policy, and healthcare have been understudied.
In this article, we use an ecological model (Bronfenbrenner, 1979) to represent the multiple distal and proximal risk and protective factors which determine the quality of a caregiverinfant relationship and the multifactorial factors that impact these families. We examine the risks to these relationships, which we frame within a larger sociopolitical context (see Figure 1). Latina immigrant women are particularly vulnerable to trauma and sexual health disparities, yet there has been limited research describing the implications of these experiences for their psychological and reproductive health or for maternal-child well-being. The ecological model guided our approach in selection of measures that investigate immigrant women's individual characteristics and immigrant status within relational contexts and community in two separate countries that historically have had large influxes of immigrants, even as policies vary. We use a combination of F I G U R E 1 Ecological model of caregiver-child well-being approaches to understand the reproductive health of immigrant Latinas. Our empirical analysis of study data best captures individual-level variables that are associated with risk for IPV, sexual trauma, and maternal mental health problems among Latina immigrant mothers. It is supplemented by a composite case vignette from our research and clinical experience that augments our more limited measures of the exo-and macrosystems.
This article reports on findings from a clinical research study of Latino immigrants in the United States and Spain (International Latino Research Partnership; Alegría et al., 2019), which is unique in its investigation of trauma experiences and psychological outcomes in the context of global migration . We use survey methods to examine factors that have an impact on Latina immigrant women's reproductive and emotional health, including the prevalence of exposure to interpersonal and sexual violence. We evaluate whether there are differential risk factors for mothers of minor children, as compared to women without children under the age of 18, among Latina immigrant mothers. What are the variations in risk factors, if any? We identify factors associated with sexual trauma and mental health, and specific recommendations for intervention, prevention, and policy efforts using an illustrative vignette that reflects our ecological model and is a composite from our clinical experience, written to protect the privacy of any one individual. We frame results in a reproductive justice framework by considering caregiver-child well-being, reproductive health, and implications for clinical practice and policy for supporting Latina immigrant women and families.

METHODS
Descriptive and inferential statistics were conducted using quantitative data derived from a mental health and substanceabuse screening interview and baseline data collected from study participants between September 2014 and May 2016 as part of the International Latino Research Partnership study (ILRP; Alegría et al., 2019) funded by the National Institute on Drug Abuse. A primary focus of the ILRP was to unite research institutions and community clinics in the United States and Spain to conduct cross-national comparative research investigating Latino migrants' behavioral health service needs. The ILRP study included behavioral health services research focused on rapid screening and referral for 341 men and women in the United States and Spain as well as testing the feasibility, acceptability, and efficacy of integrated behavioral health services in primary care clinics and community-based organizations including a transdiagnostic therapy intervention for anxiety, depression, PTSD, mild to moderate substance-use problems, and HIV prevention for migrant Latinos (Fortuna, Ramos, Fuentes, Falgas, & Alegría, 2017). For this analysis, a subset of the 175 women were included; the remaining 166 male participants were excluded. ILRP participants were recruited in waiting rooms of primary care clinics and community agencies in Boston, Massachusetts, and Madrid and Barcelona, Spain. Local Spanish-speaking research staff in each participating site obtained informed consent from participants and conducted a 90-min, audio-recorded interview that included demographic questions and a battery of measures validated in English and Spanish (discussed later). All research procedures were approved by the Human Subjects Internal Review Boards of each of the participating research institutions.

Sample
The study sample was recruited from primary care clinics and community-based organizations that offer social services to immigrant populations in the United States and Spain. Both countries receive Latino immigrants; however, the pathways differ as most immigrants to the United States come from Central America and the Caribbean. Spain has the second-largest Latino immigrant population, with pople from Argentina, Bolivia, Chile, Colombia, Costa Rica, Cuba, Dominican Republic, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Puerto Rico, Uruguay, and Venezuela. There are key differences upon arrival for immigrants in Spain versus those in the United States. First, Spanish is the official spoken language in Spain. However, cultural differences between Latin Americans and Europeans exist, unemployment and underemployment rates are highest among immigrant populations in Spain, and even while pathways to citizenship are supported, there remains significant anti-immigrant sentiment (Batsaikhan, Darvas, & Raposo, 2018).
Inclusion criteria included positive screening for elevated mental health symptoms and/or substance-use problems (descriptive statistics for the short version of the Alcohol Use Disorders Identification Test (AUDIT; Babor, Higgins-Biddle, Saunders, & Monteiro, 2001), the Drug Abuse Screening Test (DAST-10; Skinner, 1982), and a selection of questions from the Benzodiazepine Dependence Questionnaire (BDEPQ; Minaya, Fresan, Cortes-Lopez, Nanni, & Ugalde, 2011) are provided in the Appendix, but are beyond the scope of this analysis). Participants completed a baseline assessment for the ILRP intervention study (n = 175 women: 44 in Boston, 49 in Madrid, 82 in Barcelona; males excluded from this analysis). There were a higher proportion of participants from Spain, where there were two sites compared to one in the United States, with the highest recruitment in Barcelona. Experiences for women in the two contexts have shown similar outcomes Ramos et al., 2017). Most women were first-generation immigrants (born in a country other than the interview site), comprising 99% of the Spain sample and 82% of the Massachusetts sample. Boston participants were born in Central American countries (18% El Salvador, 11% Guatemala), second-generation immigrant or migrant United States (18%), Puerto Rico (20%), Dominican Republic (18%), Colombia (9%), and Mexico (5%); no participants were born in Spain. Most participant women in Spain were of South American origin (Argentina 2%, Bolivia 14%, Brazil 3%, Chile 4%, Colombia 7%, Ecuador 22%, Paraguay 4%, Peru 14%, Venezuela 19%), with only 6% from Central American countries (El Salvador, Guatemala, or Honduras), and fewer from Mexico (5%), Cuba (>1%), and Spain (>1%). Women in Barcelona were younger, migrated at a younger age, and had a lower education level, on average, than did women in Madrid or Boston. An important difference is that Spain has universal healthcare coverage that includes provisions for immigrants. Immigrants from Central America may have combined economic and refuge-seeking reasons for migration. There were no site differences on trauma-exposure measures (described later). The percentage of participants speaking Spanish as their primary language was 93% in Boston, 96% in Madrid, and 94% in Barcelona. On average, women had lived 10 years (SD = 8.25) in their resident country; however, time spent in the United States for immigrant Latinas was longer (M = 16.3, SD = 11.2), range = 1 to 43 years, as compared to those in Spain (M = 8.7, SD = 6.5), range = <1 to 40 years.

Maternal status and sociodemographics
Women were categorized as having minor-age children if they reported having one or more children under the age of 18 years. Some of these women may also have adult children, but are still categorized in this group if they have at least one child under 18. Self-report sociodemographic information included current age, country of origin, country of residence, citizenship status, race, language, education level, number of children ages 17 and under, and family economic status (i.e., live very well or comfortably; live check-to-check or poor), highest level of education (i.e., less than high school; completed high school/GED, or vocational school), and having or not having a primary sexual partner.

Mental health and sexual health profile
All measures included in this article have been validated and/or used extensively with Latinas who are Spanish speakers, in addition to English-speaking populations. This includes the AUDIT (Babor et al. 2001), which has been developed, validated, and used by the World Health Organization in multiple languages and cultural contexts. Other measures were used in the NLAAS, the largest epidemiological study of Latinos in the contiguous mainland United States. The NLAAS included a focus on testing measures for cultural relevance by formulating the research problem with attention to the fundamental cultural and contextual differences of Latinos and Asians, as compared to the mainstream population (Alegría, Vila et al., 2004). Reported psychometrics were calculated from the full sample of ILRP study participants. Depression symptoms were measured using the 9item Patient Health Questionnaire (PHQ-9; Huang, Chung, Kroenke, Delucchi, & Spitzer, 2006;Kroenke, Spitzer, & Williams, 2001

Sexual risk behaviors
We categorized the sexual partner status in the last 12 months into four categories: only having a main partner, only having a casual partner, having both a main and a casual partner, and having neither a main or a casual partner. Given the sexual partner status in the last year, we asked participants about the number of sexual experiences and whether a male and/or female condom was used during their sexual encounter. Finally, we categorized sexual behavior status during the last 12 months into three categories: unprotected sexual behavior if ever engaged in sex without a condom during the last 12 months, protected sexual behavior if condom was used all the time, and no sexual behavior if no sexual partners were reported for the last 12 months.

Traumatic stress exposure and symptoms
Trauma exposure was assessed with the Brief Trauma Questionnaire (BTQ; Schnurr, Vielhauer, Weathers, & Findler, 1999), a 10-item self-report measure that includes participant perception of risk of death/serious injury and constructed as a binary variable indicating whether the respondent had experienced any traumatic event. This 10item checklist of traumatic events administered by trained interviewers was used to assess physical assault and sexual trauma exposure. In addition to questions about specific events, participants were further asked about whether they felt their lives were in danger or if they had experienced serious injuries because of that trauma. Exposure to a specific trauma event was coded as 1 if women indicated that they experienced the event and it resulted in serious injuries and/or felt life-threatening; otherwise, it was coded as 0. Physical assault was constructed using two items from the checklist related to physical abuse: (a) Before age 18 were you physically punished or beaten by a parent or caregiver to the point of injury, and (b) Were you ever attacked, beaten, or mugged by someone other than a parent or caretaker, such as a stranger, friend, or family member (which could indicate interpersonal violence). Physical assault was coded as 1 if either of the items was endorsed, and 0 if neither was endorsed. Sexual trauma exposure was measured using a single question asking whether they had ever been forced or pressured to have unwanted sexual contact, coded as 1 if Yes and 0 if No.

Cultural, contextual, and psychosocial factors
We included measures used in the NLAAS ) that evaluate experiences of discrimination, sense of belonging, and the immigration experience. The threeitem Discrimination scale assessed whether these experiences were due to being Latino or not speaking English, Cronbach's = .71. The immigration experience was assessed via participant report of citizenship status in the country of residence, age of immigration, and number of years lived in the host country. These questions were derived from the Mexican American Prevalence and Services Survey (Vega, Kolody, Valle, & Hough, 1986). The six-item Acculturation Stress Scale measures the level of stress in the adaption to the new culture, range = 0-18, with a higher score indicating a higher level of stress due to acculturation, Cronbach's = .70.

Composite vignette
The authors each have at least 15 to 20 years of experience in serving immigrant populations and/or conducting research with Latino immigrants (Alegría, Álvarez, & DiMarzio, 2017;Alegría, NeMoyer, Falgas Bague, Wang, & Alvarez, 2018). All authors reviewed the vignette for face validity and accurate representation of clinical issues for Latina immigrants. Vignettes are short stories about a hypothetical person, traditionally used within research (quantitative or qualitative) to gather participant perspective on sensitive topics (Gourlay et al., 2014). However, in this case, the vignette serves to contextualize our quantitative results and to highlight in our discussion section critical issues related to reproductive justice, specifically the association of traumatic stress and Latina immigrant mental health, on reproductive health and caregiver-infant well-being. (See vignette in section 5.1: Illustrative clinical example.)

Analyses
Given the small frequency of missing data (3% or less in most variables), the analytic sample includes only the sample with nonmissing data in all reported variables used in the analyses. Descriptive and Fisher's exact statistics were used to compare women with and without minor children (ages 0-17 years) on psychosocial, psychological, and health measures. We conducted both Fisher's exact and chi-square tests, and report on Fisher's for this article as it is a comparatively more conservative approach with small sample sizes or highly unequal cell distribution. Building on these descriptive analyses, logistic regression models were fit to test for correlates of sexual trauma history, beginning with maternal status, exposure to sexual risk, migration information and demographics, discrimination and acculturative stress, and psychological measures. To address the problem of collinearity between psychosocial, psychological, and health measures, we used a backward elimination approach to remove predictors whose association was not statistically significant, p > 0.25. The results of final models were presented as odds ratios (OR) with 95% confidence intervals using Stata: Release 14 (StataCorp, 2015).

RESULTS
Over one third of the sample (36%) included mothers of minor children younger than 18 years of age (Appendix Table A1). Women with minor children tended to be older adults (age range = 35-49) whereas women without minor children tended to be younger (ages 18-34). Women with minor children were also more likely to currently be citizens in the country of residence, to have spent more years in the host country, and to have lower education levels, as compared to women with no minor children.
Almost 30% (28.6%) of women reported exposure to sexual trauma (Table 1). Women from the Northern Triangle were more likely to have experienced sexual trauma, as compared to their counterparts from other regions. Report of sexual trauma was highly associated with having reported physical assault. Screening positive for mental health disorders (increased likelihood of elevated symptoms of depression, generalized anxiety, PTSD, and drug abuse including benzodiazepines) were women who had experienced sexual trauma. They also had been in the host country for a longer time and scored higher on the acculturation stress measure.
Results from the logistic regression (

DISCUSSION
Gender-based and IPV, socioeconomic factors and policies and practices that limit protections from violence, are an important part of the reproductive justice discourse for many Latina immigrant women. Although immigrant women in our sample with and without minor-age children had experienced substantial trauma and social challenges, we found a trend toward more anxiety in women with minor children. These women were less educated and more likely to have resided longer as immigrants in the United States or Spain, as compared to their counterparts without children under the age of 18 years. Experiences of violence can contribute to a higher prevalence of depression and anxiety in the population (Fortuna, Porche, & Alegría, 2008).
Our findings from the ILRP study demonstrate that exposure to trauma is common especially among subgroups of immigrant women from Latin America residing in the United States and Spain. We found a higher likelihood of reports of sexual trauma among women from Central America's the Northern Triangle, as compared to other Latina immigrant women in the study. Controlling for age, women who have lived for a longer time in the host country also show a relatively higher prevalence of exposure to sexual trauma, which may be related to being older and/or having a longer period of risk for exposure both pre-and postmigration. The prevalence of reported childhood sexual abuse in the three Central American countries of the Northern Triangle has been found to range from 5 to 8% (Gómez & Speizer, 2009). Women from Guatemala and El Salvador who have experienced childhood sexual abuse have increased likelihood of being in violent relationships, as compared to women who do not have these experiences (Arias, 2004;Gómez & Speizer, 2009).
Exposure to sexual trauma poses a sexual health risk. Our findings from the ILRP study suggest that women without a sexual trauma history are more likely to engage in protected sex with a condom, as compared to women with sexual trauma histories, again pointing to the potential saliency of this risk factor. The ILRP study included an intervention for the treatment of mental health and substance-use problems as well as offering HIV risk prevention. Many women participating in the intervention study described fear that their  partner would accuse them of mistrust or would assume she was not being monogamous if she tried to negotiate condom use. In some cases, this incited the potential risk of violence. Women shared that these unsafe situations limited their ability to reduce their risk for HIV and sexually transmitted infection and/or engage in regular HIV testing. In our practice and research with immigrant Latinas, we have found that experiences of sexual violence and IPV frequently co-occur and result in cumulative risk and barriers for healthcare as well as safety for the woman and her children. Many women endure separations from very young children, leaving their children with relatives in their country of origin before seeking a safe reunification later, sometimes many years later. Sexual abuse and sexual violence histories may also lead to stigmatization and isolation (Denov, 2015). Women may experience educational and economic challenges because of the cumulative effects of gender-based violence and discrimination and the stress of parenting under circumstances of disadvantage and oppression, all which have a potentially negative impact on caregiver-infant well-being.

Illustrative clinical example
Using the following vignette, which is a composite of our clinical experiences and research, we illustrate common themes relevant to our study findings and as faced by immigrant women and mothers, their circumstances, and how they try to overcome these risks for themselves and their children.
Luisa is a 16-year-old girl who recently arrived from El Salvador to be reunited with her mother after receiving death threats from a local gang in her community. Her mother had arranged for her to be smuggled through Mexico by a coyote smuggler with other migrants. Upon entering Mexico, she was separated from the group and found herself lost, although near the U.S. border. Gang members kidnapped her and held her in a room in an undisclosed building for several days, during which time she was raped along with several other young girls. She was released and, although sexually traumatized, made her way to the border, crossing into the hands of U.S. Immigration and Customs Enforcement. Upon being detained and then sent to a large city in the Northeast to live with her mother, she discovered she was 3 months pregnant, clearly because of the rape. Note that Luisa did not disclose the rape to anyone during the border crossing nor to her mother upon initial reunification.
At age 16, despite arriving to the United States without authorization, Luisa could not access adequate prenatal care in her new city. Without a social security number, she was not able to receive in-school childcare and instead relied on her mother so that she could complete school. Luisa felt extremely traumatized by the birth of the child, as it was a not only a reminder of the rape but also of her own abandonment as an infant. Due to language and insurance barriers, this new parent was unable to access postpartum support outside of routine "well-baby" visits within a local clinic. In addition, she has been socially isolated given her newcomer status, non-English speaking skills, lack of access to adolescent programming for teen mothers, and lack of an established peer-community support. Luisa faced significant barriers in accessing ongoing reproductive healthcare, including family planning, mental health services, and any early intervention services for her baby. She had difficulty navigating HIV prevention services and was fearful of negotiating safer sex practices with her partners.

Risk factors
This vignette demonstrates the layers of proximal and distal, acute, and historic trauma that have an impact on many Latina immigrant women and implications for reproductive justice. Luisa had trouble in fully assessing the impact of the T A B L E 2 Logistic regressions for associations between covariates and exposure to sexual trauma and physical assault (N = 154) Note. Odds ratios and 95% confidence intervals were reported. GED = general equivalency diploma. * P < .05. ** P < .01. *** P < .001.
sexual trauma on her life for lack of counseling and other appropriate resources specializing in this needed care. This placed her at higher risk for poor mental health and negative implications for the caregiver-child relationship. The history of experiencing sexual violence, family dynamics that could fuel additional detachment, conflict, heightened lev-els of stress, and overall despair in Luisa are all risks for experiencing attachment/bonding difficulties with her baby. Luisa's recent arrival to the United States amid an increasingly volatile political environment contributes to added stress and anxiety that exacerbate existing barriers to reproductive and mental healthcare services. These barriers, even while living in the host country, limit the possibility of having a healthy and safe environment for both caregiver and minor children. Luisa's story is an example of the multiple accounts by immigrant mothers who present at our clinics with their young children, in search of relief for the pain and shame caused by the unspeakable. There is evidence that the trauma of rape (Breslau et al., 1998;Cohen & Roth, 1987) can have a longlasting impact in women's functioning, even after years of living in the host country, affecting all areas of their lifeincluding parental adjustment. Sexual and gender-based violence can compromise a woman's parenting ability. In addition, the shame and fear of stigmatization and rejection by others can contribute to feelings of isolation for the caregiver.
However, not all caregivers in these circumstances develop symptoms, and not all child-parent relationships are threatened (Zraly, Rubin, & Mukamana, 2013). A caregiver's reflective capacity, sensitivity, and interactions with her child are affected by the severity of the maternal mental health symptoms, her attributions of the child (e.g., as a living reminder of her trauma), intergenerational trauma issues, the child's functioning, and the environmental support and stressors. A study conducted by Schwerdtfeger and Goff (2007) has suggested that trauma history, in general, does not negatively impact expectant mothers' current prenatal attachment with their unborn child. However, a history of experiencing IPV and sexual abuse has been associated with poor attachment and mental health problems for both caregiver-infant and therefore needs to be identified and addressed (Cohen & Roth, 1987;Lyons-Ruth & Block, 1996). This includes evaluating for overprotectiveness toward children (Schwerdtfeger & Goff, 2007), decreased involvement with the infant, restricted maternal affect, and disorganized attachment (Lyons-Ruth, Bronfman, & Atwood, 1999). Our study results, as informed by this case vignette, point to the importance of understanding the role of interpersonal trauma exposure and the importance of Latina immigrant women receiving support.

Implications for practice: A framework for supporting caregiver-infant well-being
There are differences in health policy between Spain and the United States, for example. Spain has a universal healthcare system, but the United States does not. However, recent policy changes in Spain have begun to potentially limit healthcare access for undocumented immigrants (Bernal-Delgado et al., 2018). Sustaining a reproductive social justice lens to address the needs and strengths of these women requires a multipronged, multilayered approach where attention to caregiver-infant health, and therefore to the enhancement and reparation of the child-parent relationship, is central and nonnegotiable. By improving the child-parent relationship, it is possible to improve the caregiver's and the child's functioning; thus, mental health assessment and treatment with this population must include not only a relational, trauma-informed, and developmentally based approach but also encompass a sociocultural lens (Lewis, Noroña, McConnico, & Thomas, 2013). Healthcare policy should facilitate access to quality healthcare, intervention, and services, as described next.
One of the interventions aligned with this approach is Child-Parent Psychotherapy (CPP; Reyes, Stone, Dimmler, & Lieberman, 2017), a trauma-focused, relationship-based intervention for young children 0 to 6 years old and their caregivers who have experienced trauma. Its main goal is the restoration of the child-parent relationship, which is the mechanism of change. Here, the therapist's working relationship with the child-parent dyad is used as the vehicle for treatment under the premise that the quality of the relationship with the therapist will affect the child-caregiver relationship (Reyes et al., 2017). The CPP clinician serves as a translator between caregiver and child, helping the child and the parent to make meaning of trauma-related perceptions and behaviors and to respond to each other in more realistic and positive ways (Kronenberg, 2014). All of this can restore trust and intimacy and improve ways of relating and communicating between the child and the caregiver (Lieberman & Van Horn, 2008). CPP attends to both the parent's and the child's affect regulation and physical and emotional safety, and emphasizes directly addressing and having open communication about traumatic experiences, symptoms, and reminders (Kronenberg, 2014). The model acknowledges how the past affects the present and can help families break cycles of intergenerational neglect and violence by unearthing the "ghosts" and eliciting the "angels" of the caregiver's past (Lieberman, 2007;Lieberman, Padrön, Van Horn, & Harris, 2005). The therapeutic space in CPP provides room not only to bring up painful interpersonal and collective events but also to unearth memories of safety and attunement that promote self-worth. In CPP, clinicians are encouraged to use their reflective capacity to explore implicit biases and the impact of their own history in shaping their values and beliefs. Future research will need to explore how dyadic, relational, and culturally responsive intervention models can positively impact maternal mental health, selfcare, trauma healing, and reproductive health over time.
In general, infant mental health practitioners would best serve clients by creating safe, confidential, and culturally engaging spaces within the clinical setting for reproductive and mental health screening, inviting clientdriven/practitioner-supported conversations which consider the culture, history, and challenges embedded in other areas of clinical presentation (Gilliam, Neustadt, & Gordon, 2009). Approaching these conversations with a spirit of "cultural humility" invites both practitioner and client to create more robust and meaningful conversations about the client's understanding of reproductive and sexual health within the context of their culture and lived experiences. Thus, discussion about options and resources relating to family planning, pregnancy, and other areas of sexual health as well as mental health services can more readily emerge and potentially be offered in a behavioral health integration model in primary care (Decker et al., 2012). Interventions have been developed which provide behavioral health consultation to primary care settings, with great potential to support caregiver and infant mental health (Byatt et al., 2016).

Limitations
Our binary variable of exposure to physical assault is a composite from the BTQ that asks about physical assault by anyone-including friends, family members, or strangers. Thus, while IPV is included within this measure, we note the lack of a more focused measure of IPV specifically which would also include timing. Similarly, the BTQ item regarding sexual trauma does not indicate timing of the event or the perpetrator's identity. We argue that sexual or physical trauma may occur at any or all stages of migration, and that we need to conduct more research to understand the temporal association between IPV and sexual assault risk. We used a vignette, developed and based on composites from our clinical experience with Latina immigrants, to protect the confidentiality of the women we serve who are in quite vulnerable positions. Both the quantitative results of this article and the exercise of illustrating common themes can inform future research using a mixed-methods study design including interviews with women to highlight their voices to better understand complex issues of trauma, immigration experience, maternal-child well-being, and reproductive health. Data were not available on the actual ages of the children of the women in this study or any other details about the children themselves, thus we cannot describe differences in risk factors or health needs for women with infant children versus their counterparts with school-age children, or the impact of risk factors on the infants. The sample is relatively small and heterogeneous regarding country of origin. The sample included in our analyses is limited to women who screened positive for mental health symptoms via the research study in primary care and who received further assessment as part of enrollment in the treatment study. Results are therefore not generalizable to community samples of women or to those who do not access any healthcare or community organization supports. We do not have further data for women who screened negative. Further research on the impact on mothers of variations in healthcare policy for immigrant women is needed.

Conclusions: Healthcare practice and policy implications for immigrant women
Political legislation in both the United States and Spain have only further potentially limited access to services for immigrant populations since this study was conducted. Redefining women's access to quality and holistic healthcare that considers the transnational experiences and socioculturalpolitical-historical legacies affecting pregnant and parenting immigrant women while reducing disparities in maternal and child health is a critical public health goal. We recommend the following considerations: • Screening Latina immigrant women and all mothers for depression and stress in maternal and child well visits and universally in primary care. This is, in fact, an important and growing practice in working with all mothers and young children in primary care and in combination with the assessment of infant development and mental health.
• Training for clinicians and systems of care in addressing trauma using dyadic maternal-child therapies such as CPP, which provides a culturally responsive and attachment supporting intervention (Lieberman & Van Horn, 2008).
• Eliminating health care and immigration policies that create barriers for immigrant women's access to reproductive, mental health and other healthcare.
• Encompass a sociocultural lens (Lewis et al., 2013) for an exploration of how trauma and sociopolitical context have impacted the caregivers' and family's identity and functioning as well as the child's socioemotional development.
• Facilitate engagement through a collaborative process with the caregiver(s) regarding scope and goals of intervention and for the identification of their perceived resources, needs, values, and meaning-making of inequities and injustices (Lewis et al., 2013, p. 17).
• Be aware of the implicit biases of our professional and organizational cultures.
Otherwise, we risk becoming bystanders when systems fail to address the needs of immigrant Latin American women. Change and reflection starts with the self; however, this cannot be done in isolation. Reflective practice/supervision have been identified as effective strategies in promoting selfreflection, changing practice, and preventing secondary traumatic stress in the work with young families affected by trauma. These recommendations, from a reproductive justice lens, have the potential to change the trajectory and intergenerational consequences of trauma in the lives of women and children.

ACKNOWLEDGMENTS
All research procedures were approved by the Human Subjects Internal Review Boards of each of the participating research institutions. This study is supported by Research Grant R01DA034952-01 a 1, funded by the National Institute on Drug Abuse. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.