Adverse migration experiences, border community stress, and the mental health of asylum-seeking women in transit at the Mexico-U.S. border

Abstract

We evaluated the association between adverse migration experiences, border community stress, and mental health symptoms in a community-recruited sample of asylum-seeking women. As part of a mixed-methods, community-academic partnership, we analyzed cross-sectional questionnaires conducted with asylum-seeking women at the Tijuana-San Diego border (N=151) between Feb-July 2024. Symptoms of anxiety (56.6%), moderate-to-severe depression (45.9%), and post-traumatic stress (52%) were common, with 70.1% experiencing symptoms of at least one of these. Most (82.1%) faced adverse experiences during migration journey and high levels of current border community stress. When we examined the relationship between each type of adverse migration experience and mental health symptoms, lack of safe shelter during migration was marginally associated with experiencing current generalized anxiety symptoms, whereas other types of adverse experiences had positive, but null, associations. Higher border community stress was associated with experiencing generalized anxiety, depression, and PTSD symptoms. Asylum-seeking women at the Mexico-U.S. border face significant mental health risks during migration and while waiting to enter the U.S. Structural and multilevel interventions to mitigate these are needed, including changes to asylum deterrence policies and scale-up of humanitarian services.

1.1. Introduction

Population mobility is on the rise, and women comprise approximately half of the estimated 123.2 million displaced people globally in 2024 due to persecution, conflict, violence, human rights violations and other events seriously disturbing public order.¹ Globally, the number of asylum-seekers seeking international protection has been on the rise and is currently estimated at 8.4 million,¹ Approximately half of refugees and asylum-seekers globally are women or girls.² Asylum-seeking women face a disproportionate burden of trauma and multiple forms of violence,^3–7 and face high risk of developing mental health problems as well as substantial barriers to mental health services.^8–11

Asylum-seeking women migrate due to a confluence of ‘push’ factors including exposure to gendered and political violence, trauma, and abuse, and mental health problems may develop or be exacerbated by adverse experiences during transit and waiting periods in border communities.^1,12,13 Myriad social-structural vulnerabilities during transit and in border communities can increase the risk of developing mental health problems or result in unmet need for diagnosis, treatment, and prevention. These include gender-based violence (GBV)ⁱ and other human rights violations,^15–19 unsafe shelter,²⁰ economic vulnerability,²⁰ insecure migration status, and barriers to testing, prevention, and care.^19,21 Most flee settings characterized by high rates of GBV.^16,17,22 Trans women are over-represented among asylum seekers due to the high levels of GBV and persecution they face, which are recognized potential grounds for asylum.^23,24 Yet, research addressing forced migrant women’s mental health has largely focused on the resettlement phase¹⁰ or perinatal mental health,^9,25–27 and there remain important gaps in quantitative studies assessing mental health outcomes and how these may be influenced by risks experienced during the transit process among the broader population of asylum-seeking women. For example, limited epidemiologic research has attended to the ways in which experiences during forced migration – including adverse experiences in transit and border contexts (e.g., access to food, water, safe shelter, violence) – may act as or exacerbate other risk factors for mental health problems.

In recent years, migrants from across Latin America have faced particularly long and unpredictable migration durations while traveling to and waiting in Mexican border communities to enter the U.S. At the Mexico-U.S. border, numerous policies and practices have been implemented in recent years to deter the entry of migrants that have resulted in unprecedented populations of migrants waiting in Mexico for long and unpredictable periods under unsafe humanitarian conditions.^28–30 Over the past 5 years, these asylum deterrence policies have included Title 42, a public health policy that was used during the COVID-19 pandemic to expel migrants at land ports of entry between 2021–2023; the Migrant Protection Protocols (known as ‘Remain in Mexico’), enacted in January 2019, which required asylum-seekers who arrived at the Mexico-U.S. border to wait in Mexico until their U.S. immigration court hearings, and was terminated in 2021 (though briefly reinstated after that); and new limitations on asylum seeking processes introduced in May 2023 (known as the ‘Asylum ban’) that greatly restricted asylum eligibility claims at the Mexico-U.S. border.

Tijuana, Mexico is a city of 1.9 million which is the busiest land border crossing in the world and a common transit destination for migrants seeking entry into the U.S., particularly from Latin America. The city has faced unprecedented humanitarian challenges (e.g., medical care, shelter, legal) amidst limited migrant-receiving infrastructure and the effects of restrictive U.S. migration policies on the increasingly prolonged and precarious nature of the migration process in recent years. From May 2010–2021 alone, an estimated ~300,000 adults and children – equivalent to roughly 15% of the city’s population - migrated to or through Tijuana.³¹

Prior research suggests that asylum-seekers and other populations of migrants in transit through Mexico and Central America face grave risks during the migration journey that may increase the risk of developing mental health problems, including high exposure to sexual and physical violence and trauma.^32–35 In Tijuana, qualitative research has documented threats to the mental health of asylum-seeking women in this region posed by high exposure to GBV and structural violence, socio-economic vulnerability, and precarious migration experiences.^12,32–36 Moreover, research among migrants in Tijuana shelters during the COVID-19 pandemic found adverse experiences during migration - such as exposure to violence and interrupted travel – to be associated with mental health symptoms.³⁷ Yet, empirical data on these experiences among asylum-seeking women since the pandemic remain lacking. To our knowledge, no published studies have assessed the prevalence or risk factors for mental health problems in asylum-seeking women at the Mexico-U.S. border – with a particular dearth of research conducted amidst ongoing asylum deterrence policies.

This study was underpinned by a ‘Risk Environment’^38,39 framework, which posits that features of the physical, social, policy and economic environments in which people are embedded play a crucial role in health-related risks faced by marginalized populations, and thus represent critical prevention targets.^38,39 This framework has rarely been applied to research with migrants or in relation to mental health outcomes. In Tijuana, the confluence of violence, drug trafficking, deportation, social isolation, sex work, and drug use in certain neighborhoods near the U.S. border pose ‘risk environments’ for HIV, STIs, and related comorbidities among marginalized populations.^40–44 Features of the ‘risk environment’ both during migration and in border communities such as migration and border policies and their enforcement, limited access to safe shelter, housing, food, and water, physical and sexual violence, discrimination, and gender inequities are likely to strongly influence health inequities faced by asylum-seeking women.⁴⁵ Yet, to our knowledge, no published studies have previously used this framework to understand the contextual risks to mental health faced by asylum-seeking women during the migration process and while waiting in border communities.

Given existing gaps in data on the mental health needs of asylum-seeking women in relation to adverse experiences during migration and in border contexts and the importance of this information for informing appropriate prevention and treatment interventions, we aimed to evaluate the association between adverse migration experiences, border community stress, and symptoms of generalized anxiety, depression, and post-traumatic stress in a community-recruited sample of asylum-seeking women in the Mexico-U.S. border city of Tijuana.

1.2. Methods

This study was conducted as part of a larger academic-community collaboration known as the Migration and Inequities in Health among Refugee & Asylum-Seeking Women (MIHRA) study. MIHRA began in July 2022 as an assessment of the maternal health needs of asylum-seeking women and expanded in February 2024 to include a broader focus on HIV/STI prevention and sexual and reproductive health. The study was led by a community-academic partnership of investigators, staff, and partners from a humanitarian organization serving migrants in the U.S. and Mexico. Ethical approvals were obtained by institutional review boards in the U.S. and Mexico. Activities were informed by close collaborations with our community partners as well as a Community Advisory Board of local health and migrant-serving community organizations (e.g., clinics, shelters). The project involves both qualitative and quantitative components.

Participants: For the present study, we analyzed pilot data gathered in Tijuana via a cross-sectional pilot study with Spanish-speaking asylum-seeking women that was conducted as part of the larger MIHRA project between Feb 2024-July 2024. A target sample size of 150 was identified as this was consistent with methodological literature suggesting this to be sufficient for a pilot study⁴⁶ and was identified as the minimum sample needed to assess primary outcomes related to HIV/STI risks. Eligible participants were 1) international migrants or internal Mexican migrants seeking safety in the U.S.; 2) self-identified women (cis or trans); 3) reported sex with at least one partner in the past 3 months; 4) able to speak Spanish; 5) aged >18 years old; and 6) able to provide informed consent.

Participants were recruited in Tijuana, Mexico through extensive community outreach activities including via posters, flyers, and direct referral by community partners. Potential participants were provided with detailed information about the study and could complete eligibility screening and express interest in participation either by completing a self-guided form (using a QR code provided on the recruitment posters) or during in-person with the support of a field staff member. Following eligibility screening, trained research staff guided participants through an informed consent process prior to enrollment.

Questionnaire visits: Interviewer-administered questionnaires were conducted with participants in-person and were administered by community-based outreach and clinical staff in RedCAP. All staff were trained in trauma-informed care and psychological first aid and had local community and/or research experience working with migrants or other marginalized populations in Tijuana.

Following pre-test counseling, participants also completed point-of-care testing for HIV (INSTI, Biolytical) and syphilis (One Step rapid test, InTec); reactive tests were repeated, and those testing positive were referred to local clinics for follow-up care. Neisseria gonorrhea and Chlamydia trachomatis were detected via self-collected urine sample (Aptima) and processed at Genómica Laboratorio (Unihealth Mx). Participants received post-test counseling and were scheduled to return receive their STI test results within 48–72 hours. All participants received appropriate post-test counselling regarding their results.

All study participants received an honorarium of $25 USD for their time, valued expertise and contributions, and participation, as well as extensive active referrals to needed humanitarian, health, and other services facilitated by study staff.

1.2.1. Measures

Questionnaires included theory-based, validated measures as well as community-developed measures previously used in this study context. All measures were conceptually underpinned by the study’s ‘risk environment’ framework and were translated and pilot-tested prior to use.

Outcomes:

Symptoms of generalized anxiety were assessed based on responses to the Generalized Anxiety Disorder-7 (GAD-7), which asked ‘During the last 2 weeks, how often have you been bothered by the following problems?’⁴⁷ Scores (range: 0–21) were examined as a continuous measure and dichotomized using a cutoff of ≥8 (moderate to severe) vs. 0–7 (minimal-to-mild).⁴⁸ Symptoms of depression were assessed via the Patient Health Questionnaire-9 (PHQ-9), which asked ‘During the last 2 weeks, how often have you been bothered by the following problems?.’⁴⁹ Scores (range: 0–27) were examined continuously and dichotomized based on a cutoff of ≥10 (moderate-to-severe) vs. 0–9 (minimal-to-mild). Symptoms of post-traumatic stress were based on the Primary Care PTSD Screen for DSM-5 (PC-PTSD-5).⁵⁰ This began with asking whether participants had ever experienced a traumatic event (‘Sometimes events that are unusual or especially frightening, horrible, or traumatic can happen to people, some examples being a serious accident, an earthquake, seeing someone who has been hurt, or being exposed to violence. Have you ever experienced this type of situation?’), then asking those who responded affirmatively a series of 5 items to assess how this affected them over the past month. These measures have been found to have high validity and reliability in Spanish-speaking populations^51–53 and Latin American migrants.⁵⁴ Scores were dichotomized based on a cutoff of 3 to indicate the presence of probable PTSD symptoms.⁵⁰

Primary migration-related exposures of interest:

Adverse experiences during the most recent migration journey were assessed by the following question: “Since leaving the last place where you lived for a full year, until arriving in Tijuana, have any of the following situations happened to you?” This community-derived measure has been developed and used previously by colleagues to reflect the most common adverse migration experiences reported in this setting. Responses were coded as ‘experienced extreme temperatures (yes to ‘Endured extreme cold or heat’), not enough food (‘Did not have enough food’), No safe place to sleep (‘Did not have a safe place to sleep’), not enough water/liquid to drink (‘Did not have enough clean water or other liquid to drink’), and physical/sexual violence (‘You or another family member or friend that you traveled with suffered physical violence’, and/or ‘You or another family member or friend that you traveled with suffered sexual violence’). Physical and sexual violence were assessed as a combined measure due to the limited number of responses We also developed a combined exposure assessing the number of types of adverse migration experiences experienced during the migration journey (on a scale of 1–5).

Recent border community stress was assessed via an adapted version of the Border Community & Immigration Stress Scale (BCISS)(α = 0.91).⁵⁵ This was based on the following question, ‘in the last two months, how stressed have you felt about the following situations?’ As the BCISS was originally validated for use with those living in border communities, including those who are living in the U.S., we removed items from the original BCISS which did not reflect the experiences of in-transit migrants (e.g., ‘fear of losing your job to new immigrants’ and ‘your family having difficulties combining Latino & American culture’). The adapted scale included 9 items including ‘needed to speak in a language that is not my native tongue,’ ‘being far from my family,’ ‘not knowing if I will be given asylum in the U.S.,’ and response options included ‘not stressed at all’, ‘a little stressed’, or ‘very stressed’ (for a full list of items, see Table 2).

Table 2.

Adverse migration experiences and border community stress among asylum-seeking women, Tijuana, Mexico, 2024 (N=151)

	N (%)*
Adverse migration experiences during most recent journey
Any adverse migration experience	124 (82.1)
Extreme temperatures	95 (62.9)
Not enough food	86 (57.0)
No safe place to sleep	79 (52.3)
Not enough water/liquid to drink	62 (41.3)
Physical or sexual violence	24 (15.9)
Number of adverse migration experiences (median, IQR)	2.0 (1.0, 4.0)
Border Community Stress
Border & Immigration Community Stress Scale (BICSS) (med, IQR)	19 (16–22)
Border & Immigration Community Stress Scale Items
Needing to speak in a language that is not my native tongue
Not stressed at all	83 (55.3)
A little stressed	41 (27.3)
Very stressed	25 (16.7)
Being far from my family
Not stressed at all	14 (9.3)
A little stressed	47 (31.3)
Very stressed	89 (59.3)
Not having work or having problems at work
Not stressed at all	27 (18.0)
A little stressed	32 (21.3)
Very stressed	90 (60.0)
Not having enough money
Not stressed at all	15 (10.1)
A little stressed	34 (22.8)
Very stressed	100 (67.1)
Not knowing if I will be given asylum in the US
Not stressed at all	12 (8.0)
A little stressed	50 (33.3)
Very stressed	88 (58.7)
My own health problems or those of people who travel or live with me
Not stressed at all	51 (34.0)
A little stressed	39 (26.0)
Very stressed	60 (40.0)
Hearing comments like people saying “Go back to your country”
Not stressed at all	69 (46.0)
A little stressed	41 (27.3)
Very stressed	40 (26.7)
The crime or violence in Tijuana
Not stressed at all	37 (24.7)
A little stressed	27 (18.0)
Very stressed	86 (57.3)
Fear of the police, National Guard, or immigration authorities in Tijuana
Not stressed at all	88 (58.7)
A little stressed	32 (21.3)
Very stressed	30 (20.0)

^*All data represent n (%) unless otherwise indicated

Other covariates included region of birth, language proficiency (based on the question, ‘Which of the following languages do you know how to speak and understand well?’), demographics (age, Indigenous ancestry, marital status, education, gender, sexual orientation), and the Alcohol Use Disorders Identification Test (AUDIT) (Score 2–10 calculated as total of all responses). Immigration detention (both in the U.S. and for international migrants, in Mexico) was assessed as a descriptive covariate of interest, given its known physical and psychosocial impacts on migrants.

1.3. Data analysis

We used descriptive statistics to characterize the sample and distribution of outcomes, exposures, and covariates and reported these as counts and percentages for binary variables and medians and interquartile range for continuous variables. We used Chi-square (or Fisher’s exact test for small cell counts) or the Kruskall Wallis test for bivariate comparisons. To address our aim of assessing the association between adverse migration experiences, border community stress, and symptoms of generalized anxiety, depression, and post-traumatic stress, we conducted bivariate and multivariable modified Poisson regression with a log link and robust standard errors to estimate unadjusted and adjusted prevalence ratios (PRs) and 95% confidence intervals (CIs). We developed separate models to examine the relationship between each type of adverse migration experience and symptoms of generalized anxiety, depression, and PTSD; we also examined a potential dose-response relationship between the number of types of adverse migration experiences faced during migration and symptoms of generalized anxiety, depression, and PTSD. For these models, predicted values were truncated at 0–100% to support interpretability. For adjusted estimates, each model controlled for confounders identified a priori based on existing literature and subject area knowledge, including Indigenous ancestry (based on self-identification as Indigenous vs. Black, mixed, white, mestizo, or other), education (years), marital status (married or cohabitating with partner vs. single, separated, divorced, or widowed), and age. Analyses were conducted in R v4.4.1.

1.4. Results

Among 151 participants, most (86.1%) were born in Mexico, with the remainder originating in Central America (8.6%) or elsewhere in Latin America and the Caribbean (e.g., Colombia, Venezuela, Haiti). Almost two-thirds (n=100, 66.2%) arrived to the Mexico-U.S. border from their place of origin in 2024; 27.2% (n=41) arrived in 2023, and 6.7% in 2023 or earlier. A total of 81.5% (n=123) had requested and were waiting to receive an asylum appointment in the U.S., and 2.7% (n=4) already had an asylum appointment date. Nearly all were fluent in Spanish, with small proportions indicating fluency in English (9.9%), Haitian Creole (2.7%), or another language (6.6%) (Table 1). Among participants, 10% had been in immigration detention in Mexico or the U.S. Over one-third (35.1%) were of Indigenous ancestry; over half (52.0%) were married/cohabitating; and the median educational attainment was 10 years (interquartile range (IQR): 9–12). Almost one-third (30.8%) were at moderate or high/severe risk for alcohol dependency. 4.6% tested positive for HIV or an STI. Most identified as cisgender and heterosexual; 8.7% identified as trans women and 11.2% as a minoritized sexual orientation.

Table 1.

Descriptive characteristics among asylum-seeking women, Tijuana, Mexico, 2024 (N=151)

	N (%)*
Mental Health Outcomes
Generalized Anxiety Disorder Symptoms (GAD-7) (score >=8)	82 (56.6)
Generalized Anxiety Disorder Symptoms (GAD-7), continuous (median, IQR)	8 (4, 12)
Depressive Symptoms (PHQ-9), Moderate-to-Severe (Score >=10)	68 (45.9)
Depressive Symptoms (PHQ-9), continuous (median, IQR)	8 (4, 13)
Post Traumatic Stress Disorder (Pc-PTSD-5) (Score 3–5)	77 (52.0)
Post Traumatic Stress Disorder (Pc-PTSD-5), continuous (median, IQR)	3 (0, 4)
Moderate to severe symptoms of depression, anxiety, or PTSD (GAD-7, PHQ-9, or Pc-PTSD-5)	106 (70.1)
Other covariates
Region of birth
Mexico	130 (86.1)
Central America	13 (8.6)
Other	8 (5.3)
Language proficiency
Spanish	150 (99.3)
English	15 (9.9)
Haitian Creole	4 (2.7)
Other	10 (6.6)
Age in years (median, IQR)	32 (26, 37)
Ever in Mexican or U.S. immigration detention	15 (9.9)
Indigenous Ancestry	53 (35.1)
Marital status
Married or cohabitating	78 (52.0)
Single	94 (48.0)
Years of education (med, IQR)	10 (9–12)
Alcohol dependency (AUDIT score)
Low risk (0–2 points)	103 (69.1)
Moderate risk (3–5 points)	33 (22.1)
High to severe risk (6–12 points)	13 (8.7)
Gender identity
Cisgender woman	137 (91.3)
Transgender woman	13 (8.7)
Sexual orientation
Heterosexual	103 (88.8)
Minoritized sexual orientation (lesbian, gay, bisexual, asexual)	13 (11.2)

^*All data represent n (%) unless otherwise indicated

We identified high prevalence of symptoms of anxiety disorder (56.6%), moderate-to-severe depressive symptoms (45.9%), and post-traumatic stress (52%). Almost three-quarters (70.1%) of participants experienced symptoms at least one of these mental health problems.

Adverse experiences during the most recent migration journey:

Most (82.1%) had been exposed to adverse experiences during their most recent migration journey, the most common of which were extreme temperatures (62.9 %), inadequate food (57.0%), lack of safe shelter (52.3%), or insufficient water or other liquids to drink (41.3%) (Table 2). 15.9% reported that themselves or another member of their traveling party had experienced physical or sexual violence during migration. Most participants experienced more than one of these adverse migration experiences (median: 2.0, inter-quartile range (IQR): 1.0, 4.0).

In bivariate analysis, exposure to adverse experiences during the most recent migration journey was associated with increased symptoms of anxiety, depression, PTSD, or any of these mental health conditions in the prior two weeks; these effects were mostly null, with the exception of the association between lack of safe places to sleep and anxiety symptoms (Table 3). In adjusted models, similar trends were observed, with lack of safe places to sleep retaining a marginally significant association with generalized anxiety symptoms (Adjusted Prevalence Ratio (aPR): 1.48, 95% confidence interval (CI): 0.94–2.35) (Fig 1). We identified a dose-response gradient in the relationship between the number of adverse migration experiences reported and the predicted prevalence of generalized anxiety, depression, and PTSD, although this was only marginally significant. In adjusted models, exposure to each additional type of adverse migration experience was associated with a 12.3% higher relative prevalence of symptoms of generalized anxiety (aPR=1.12, 95% CI: 0.98–1.29), an 8.0% higher relative prevalence of symptoms of depression (aPR=1.08, 95% CI=0.93–1.26), an 8.1% higher relative prevalence of symptoms of PTSD (aPR=1.08, 95% CI: 0.94–1.25), and a 4.3% higher relative prevalence of symptoms of any of these (aPR=1.04, 95% CI: 0.91–1.18), though none of these were statistically significant (Fig 2).

Table 3.

Bivariate analysis of adverse trip experiences and mental health symptoms among asylum-seeking women, Tijuana, Mexico, 2024 (N=151)

	Generalized anxiety symptoms (GAD-7)1 Score <8 = No Score >= 8 = Yes			Moderate-to-severe depressive symptoms (PHQ-9)2 Score <10 = No Score >= 10 = Yes			Post-traumatic stress (Pc-PTSD-5)3 Score <3 = No Score 3–5 = Yes			Any symptoms of moderate/severe anxiety, depression, or PTSD4
	No n (%)	Yes n (%)	Prevalence Ratio (95% CI)	No n (%)	Yes n (%)	Prevalence Ratio (95% CI)	No n (%)	Yes n (%)	Prevalence Ratio (95% CI)	No n (%)	Yes n (%)	Prevalence Ratio (95% CI)
Any adverse migration experiences during most recent journey	80 (79.4)	68 (82.9)	1.11 (0.65, 2.06)	64 (80.0)	57 (83.8)	1.16 (0.63, 2.33)	56 (78.9)	65 (84.4)	1.21 (0.68, 2.35)	35 (81.4)	87 (82.1)	1.01 (0.63, 1.71)
Extreme temperatures	35 (55.6)	57 (69.5)	1.31 (0.83, 2.14)	49 (61.3)	45 (66.2)	1.12 (0.84, 2.24)	40 (56.3)	54 (70.1)	1.35 (0.84, 2.24)	26 (60.5)	68 (64.2)	1.05 (0.71, 1.57)
Not enough food	31 (49.2)	50 (61.0)	1.23 (0.80, 1.94)	44 (55.0)	39 (57.4)	1.05 (0.65, 1.72)	39 (54.9)	45 (58.4)	1.07 (0.68, 1.70)	23 (53.5)	61 (57.5)	1.05 (0.72, 1.55)
No safe place to sleep	25 (39.7)	51 (62.2)	1.49 (0.96, 2.36)	39 (48.8)	40 (58.8)	1.25 (0.77, 2.04)	45 (63.4)	60 (77.9)	1.26 (0.81, 2.00)	19 (44.2)	60 (56.6)	1.16 (0.79, 1.70)
Not enough water/liquid to drink	20 (31.7)	37 (45.7)	1.28 (0.83, 1.99)	28 (35.4)	31 (45.6)	1.25 (0.77, 2.01)	24 (34.3)	36 (46.8)	1.27 (0.81, 1.99)	14 (32.6)	46 (43.8)	1.14 (0.77, 1.68)
Physical or sexual violence	6 (9.5)	16 (19.5)	1.36 (0.76, 2.28)	11 (13.8)	12 (17.6)	1.16 (0.59, 2.09)	9 (12.7)	14 (18.2)	1.21 (0.65, 2.09)	4 (9.3)	19 (17.9)	1.20 (0.71, 1.92)
Number of adverse migration experiences (median, IQR)	2.0 (1.0, 3.0)	3.0 (1.0–4.0)	1.13 (0.99, 1.29)	2.0 (1.0–3.3)	2.5 (1.0, 4.0)	1.07 (0.92, 1.24)	2.0 (1.0, 3.0)	3.0 (1.0, 4.0)	1.09 (0.95, 1.26)	2.0 (1.0, 3.0)	2.0 (1.0, 4.0)	1.04 (0.93, 1.18)

¹n=145,

²n=148,

³n=148,

⁴n=149

Figure 1: Unadjusted and adjusted PRs and 95% CIs for adverse migration experiences and current mental health symptoms among in-transit migrant women, Tijuana, Mexico (N=151), 2024.

NOTE: Modified Poisson regression with robust standard errors was used to estimate prevalence ratios (PRs) and 95% confidence intervals (CIs) for each exposure–outcome association. Adjusted models controlled for age, Indigenous ancestry, marital status, and years of education. Points represent prevalence ratios and horizontal bars represent 95% confidence intervals.

Figure 2: Predicted prevalence of number of types of adverse experiences faced during migration and current mental health symptoms among in-transit migrant women, Tijuana, Mexico (N=151), 2024.

NOTE: Modified Poisson regression with robust standard errors was used to estimate prevalence ratios (PRs) and 95% confidence intervals (CIs) for the total number of adverse migration experiences reported (0–5) and each mental health outcome. Adjusted models controlled for age, Indigenous ancestry, marital status, and years of education. Points represent predicted prevalence and horizontal bars represent 95% confidence intervals.

Border community stress:

Participants experienced high levels of border community stress within the prior two months (median score: 19, IQR: 16–22). The most frequently reported items that participants reported being ‘very stressed’ about related to not having enough money (67.1%), not having work/problems at work (60.0%), being far from family (59.3%), uncertainty regarding their asylum process (58.7%), and crime and violence in Tijuana (57.3%). Additionally, 26.7% reported being very stressed by receiving discriminatory or xenophobic comments, and 40% by their health problems or those of those traveling/living with them. In bivariate analysis, border community stress scores in the prior two months was significantly associated with experiencing symptoms of generalized anxiety, depression, PTSD, or any of these mental health conditions in the prior two weeks (Table 4). In adjusted models, border community stress scores were independently and positively associated with experiencing current symptoms of anxiety (aPR: 1.10, 95% CI: 1.04, 1.16), depression (aPR: 1.10, 95% CI: 1.04, 1.17), PTSD (aPR: 1.08, 95% CI: 1.10, 1.14), or any of these mental health conditions (aPR: 1.06, 95% CI: 1.01, 1.11) (Fig 3).

Table 4.

Bivariate analysis of border community stress (BCISS) and mental health symptoms among asylum-seeking women, Tijuana, Mexico, 2024 (N=151)

	Generalized anxiety symptoms (GAD-7)1 Score <8 = No Score >= 8 = Yes			Moderate-to-severe depressive symptoms (PHQ-9)2 Score <10 = No Score >= 10 = Yes			Post-traumatic stress (Pc-PTSD-5)3 Score <3 = No Score 3–5 = Yes			Any symptoms of moderate/severe anxiety, depression, or PTSD4
	No Med [IQR]	Yes Med [IQR]	Prevalence Ratio (95% CI)	No Med [IQR]	Yes Med [IQR]	Prevalence Ratio (95% CI)	No Med [IQR]	Yes Med [IQR]	Prevalence Ratio (95% CI)	No Med [IQR]	Yes Med [IQR]	Prevalence Ratio (95% CI)
Border Community & Immigration Stress Scale (BCISS)	17 [15, 20]	21 [18, 24]	1.10 (1.04, 1.16)	18 [15, 21]	21 [18, 24]	1.10 (1.04, 1.17)	18 [15, 20]	21 [17.75, 24]	1.08 (1.03, 1.15)	17 (15, 19)	20 (17, 23)	1.06 (1.01, 1.11)

¹n=145,

²n=148,

³n=148,

⁴n=149

Figure 3: Forest plot, unadjusted and adjusted PRs and 95% CIs for border community stress and current mental health symptoms among in-transit migrant women, Tijuana, Mexico (N=151), 2024.

NOTE: Modified Poisson regression with robust standard errors was used to estimate prevalence ratios (PRs) and 95% confidence intervals (CIs) for each exposure–outcome association. Adjusted models controlled for age, Indigenous ancestry, marital status, and years of education. Points represent prevalence ratios and horizontal bars represent 95% confidence intervals

1.5. Discussion

In this cross-sectional study among Latin American asylum-seeking women at the Mexico-U.S. border, more than three-quarters faced adverse experiences during their most recent migration journey and high levels of border community stress was reported, which was associated with increased risk of experiencing current symptoms of anxiety, depression, or PTSD. Findings suggest that adverse experiences during migration – particularly lack of safe shelter during migration – and high levels of border community stress while waiting to enter the U.S. may threaten the mental health of asylum-seeking women at the Mexico-U.S. border. These findings add to a limited body of literature on asylum-seeking women’s mental health, particularly in the Mexico-U.S. border context.^10,56,57

This study documented high prevalence of symptoms of anxiety, moderate-to-severe depressive symptoms, ands symptoms of post-traumatic stress among asylum-seeking women at the Mexico-U.S. border, with almost three-quarters of participants experiencing at least one of these. These figures provide some of the first community-based epidemiologic data on these outcomes among asylum-seeking women in the Mexico-U.S. border region. For example, a 2020 systematic review and meta-analysis of the prevalence of mental illness in refugees and asylum seekers (measured via clinical interview using a validated diagnostic assessment measure) measured a pooled prevalence of anxiety disorders at 11%, depression at 31.5%, and PTSD at 31.5%, with heterogeneity by host and receiving context.⁵⁸ No studies from Latin America were identified, and only one from the U.S. Our findings are within the range reported by a recent study which assessed the prevalence of common mental disorders (depression and/or anxiety, assessed via the CES-D-7 and GAD-2) among female migrants in Tijuana shelters at 62%.³⁷

In this study, we identified consistent associations between exposure to a unique community-developed measure of border community and immigration stress and increased relative prevalence of current mental health problems, including 28%, 22%, and 20% elevated relative prevalence of experiencing symptoms of anxiety, depression, or PTSD, respectively, for each 1-point increase in stress. The most prevalent stressors included economic, family, immigration, crime, and violence-related concerns. These findings are novel, as few studies have used community-developed, standardized and validated measures to assess dimensions of immigration stress within border environments, or their impacts on migrants’ health.⁵⁵ Studies have used other standardized measures to assess immigration stress (e.g., Hispanic Stress Inventory-Version 2), showing inconsistent relationships with mental health outcomes;⁵⁹ however, these do not address specific border contexts. Future work evaluating the relationship between border-specific measures of immigration stress such as the BCISS and migrant women’s health in complex border contexts such as the Mexico-U.S. border is needed.

In this study, similar to one prior study with migrants in Tijuana shelters,³⁷ we identified elevated proportions of anxiety, depression, and PTSD symptoms among women exposed to adverse migration experiences (e.g., extreme heat or cold, insufficient food or water, violence). In separate models examining the association between each type of adverse migration experience and mental health symptoms, only lack of safe places to sleep during one’s most recent migration journey was associated with marginally increased relative prevalence of generalized anxiety symptoms. These findings address an important gap in the quantitative literature on the relationship between living conditions and mental health among asylum-seeking women, particularly in complex border settings and in Latin America. A systematic review and narrative synthesis of studies published between 2017–2023 found that most recent quantitative studies on the relationship between housing and mental health among asylum-seekers identified significant associations between poorer living conditions and mental health, with collective housing and detention identified as particularly detrimental.⁵⁷ Three themes explained how housing impacts asylum seekers’ mental health: lack of autonomy, feeling unsafe, and lack of support.⁵⁷ This review identified only 3 studies focused on women, and found that most studies on living conditions and mental health among asylum-seekers and refugees addressed post-migration concerns. Only 9 quantitative studies were identified, of which there was a complete dearth of studies conducted in Latin America or the Mexico-U.S. border.⁵⁷ Other types of adverse migration experiences were null, which may be explained by the relatively high overall exposure to adverse migration experiences. We also observed a dose-response gradient suggesting a marginally significant relationship between exposure to multiple adverse migration experiences and experiencing current symptoms of generalized anxiety, depression, and PTSD.

This analysis was limited to cross-sectional data, and future efforts to ethically and safely follow-up with this population via rigorous, longitudinal designs should be explored. As with most research involving marginalized populations and sensitive topics, self-reported measures are subject to potential social desirability and recall bias; our community-based team, training in trauma-informed and non-stigmatizing interviewing, and community collaborations were designed to mitigate this to the degree possible. Participants were recruited via community-based techniques; given the challenges of sampling hard-to-reach populations, results may not be generalizable to all asylum-seekers and may under-represent more marginalized asylum-seekers, such as those experiencing sexual violence during migration. Participant eligibility included being sexually active (sex with >1 partner, last 3 months) as this was a focus of the larger study from which data were drawn. Finally, this study was restricted to Spanish-speakers due to feasibility considerations, as our pilot funding did not provide sufficient funds to culturally tailor our measures and approaches across the full diversity of asylum-seeking women transiting this border region. Further research using probability-based recruitment methods for reaching marginalized and hidden populations (e.g., respondent-driven sampling) and involving multilingual approaches are recommended to better understand the mental health status and needs of the broader population of asylum-seeking women is recommended.

Future Directions:

Findings suggest the urgent need to implement and evaluate structural and multilevel interventions to mitigate the mental health risks posed by adverse migration experiences and border stress among asylum-seeking women. These should include shifts away from asylum deterrence policies and dedication of resources towards scale-up of comprehensive humanitarian services along migration corridors and in Mexican border cities. Efforts to address unmet needs for safe shelter for asylum-seeking women during the migration process as a mental health prevention intervention are especially warranted.

1.6. Conclusions

U.S.-bound asylum-seeking women at the Mexico-U.S. border face significant threats to their mental health during their migration journeys, particularly in relation to adverse migration experiences such as lack of safe spaces to sleep, as well as border stress while waiting to enter the U.S. Structural and multilevel interventions are recommended, including changes to asylum deterrence policies and scale-up of humanitarian services along migration corridors and in Mexican border cities are recommended.

Highlights.

• Asylum-seeking women face threats to mental health across the migration process

• Adverse experiences during migration and border stress linked to mental health symptoms

• Changes to migration policies and increased humanitarian services are needed

Data Availability Statement:

Data for this study are not publicly available for legal and ethical reasons, as this study involves sensitive data collected with a highly criminalized and stigmatized population of marginalized women. Under our current ethical approvals, de-identified data can be made available upon reasonable request and pending ethical approval. Please submit all requests to the corresponding author.

Code Availability Statement:

All custom statistical codes used in data analysis has been provided as part of the submission.