Structural barriers to prenatal care for migrant women at the United States–Mexico border: a qualitative study

Original publication here: https://doi.org/10.1016/j.puhe.2025.105986

Dec 2025

K.M. Gonzalez, W.R. Matias, A.M. Mohareb, D. Sridhar

Abstract

Objectives

Recent United States (US) immigration policies have left thousands of asylum seekers stranded in Mexican border cities with limited access to healthcare. Pregnant women are particularly vulnerable, often facing substantial barriers to receiving adequate prenatal care (PNC). This study aimed to describe the state of PNC delivery at the US–Mexico border through the challenges, experiences, and perspectives of healthcare workers (HCWs).

Study design

Qualitative study.

Methods

Semi-structured interviews were conducted with 10 HCWs (6 nurses and 4 physicians) who volunteered at non-governmental organization (NGO)-operated clinics in Matamoros and Reynosa between 2019 and 2023. Interviews were conducted in June–July 2023 and analyzed using thematic analysis.

Results

Three inductive themes emerged: (1) structural violence; (2) resource limitations; and (3) care fragmentation. Participants reported that PNC frequently fell below international standards. Continuity of care was disrupted by patient transience, limited infrastructure, and reliance on short-term staffing. Additional barriers included shortages of specialized providers, fragmented medical records, institutional racism, and the impacts of cartel-related violence.

Conclusions

This study underscores the structural and systemic barriers shaping maternal healthcare delivery in humanitarian border settings. Improving care for pregnant migrants will require comprehensive policy reform, strategic resource allocation, and rights-based, cross-border approaches. Future research should engage migrant communities and local providers to develop responsive models that bridge the gap between inclusive policy and frontline care.

1. Introduction

Migration from Central America and Mexico has sharply increased in recent years, driven by poverty, violence, and exacerbated by climate change, COVID-19, and economic disruptions.¹ Recent United States (US) policy shifts have fundamentally reshaped the asylum process, creating significant barriers for migrants.²

The Migrant Protection Protocols (MPP) of 2019—also known as the “Remain in Mexico” policy—required asylum seekers to return to Mexico while awaiting their US immigration proceedings.³ Title 42, enacted in March 2020 under the Public Health Service Act of 1944, authorized the US government to expel asylum seekers under the pretext of COVID-19 prevention.⁴^,⁵ Together, these policies restricted border access and left nearly 20,000 asylum seekers stranded in dangerous Mexican border cities.⁶ Although both policies have since been rescinded, public health experts widely condemned them for lacking scientific justification and creating lasting barriers to healthcare, shelter, and other essential services.⁶^,⁷

Pregnant women are especially vulnerable, often lacking adequate prenatal care (PNC), clean water, sanitation, safe shelter, and proper nutrition.⁸ They face heightened risks of violence, including documented cases of miscarriages, kidnappings, rape, and abuse following expulsion.⁹ Timely PNC—the healthcare women receive during pregnancy—is essential for optimizing maternal and neonatal health.¹⁰ Newborns born to mothers without PNC are three times more likely to have low birth weight and five times more likely to die.¹¹ This study defines ‘adequate’ PNC based on the World Health Organization’s 2016 antenatal care (ANC) model, which recommends at least eight provider encounters during pregnancy.¹²

Although reports highlight the challenges faced by pregnant women,⁹^,¹³^,¹⁴ peer-reviewed literature on PNC delivery at the US–Mexico border remains scarce. This study aims to address this gap by describing the state of PNC delivery through the challenges, experiences, and perspectives of healthcare workers (HCWs) amidst the prevailing political climate.

2. Methods

2.1. Study design

We used a qualitative study design, employing semi-structured interviews to explore HCWs’ perspectives on PNC delivery and the factors influencing its provision. We adopted an interpretive standpoint, viewing reality as subjective and ever-changing, emphasizing that multiple perspectives can coexist and influence understanding.¹⁵ This approach guided our recruitment strategy. We focused on HCWs rather than patients to gain broader insight into potential systemic barriers and avoid ethical challenges associated with interviewing migrants. Reporting followed the Consolidated Criteria for Reporting Qualitative Research (COREQ) guidelines,¹⁶ with a completed checklist provided in the Supplementary data.

2.2. Sampling and participant recruitment

We employed maximum variation sampling, a purposive sampling strategy aimed at capturing diverse perspectives and experiences.¹⁷ Participants were recruited from Global Response Medicine (GRM), an international humanitarian non-governmental organization (NGO) that provided free clinical and specialty care to asylum seekers at the US–Mexico border from 2019 to 2023 through clinics in Matamoros and Reynosa.¹⁸

GRM operates on a volunteer-based system. Eligible participants were HCWs—including physicians, nurse practitioners, midwives, and nurses—who had provided PNC at GRM’s Matamoros or Reynosa clinics at least once and consented to participate. No further exclusion criteria were specified. We leveraged our professional contacts within GRM and Massachusetts General Hospital (MGH), a partner hospital. HCWs were recruited through direct email invitations facilitated by these contacts, as well as snowball sampling¹⁷ from initial respondents.

We sampled ten participants with relevant experience to ensure high-quality data.¹⁹ No invited participants declined or dropped out.

2.3. Data collection

We developed a semi-structured interview guide (see Supplementary data) informed by investigator experiences at the border, guidelines for semi-structured interviews,²⁰ and a review of existing literature.²¹ The guide was pilot tested through internal review by the research team²¹ and field-tested with two potential participants.²⁰

Between June and July 2023, we conducted interviews with eligible participants after confirming consent. Interviews were conducted by KMG, a Master of Public Health candidate trained in qualitative methods, under the supervision of experienced qualitative researchers. Participants were informed of the interviewer’s background in global health. An optional informational call was offered to discuss study objectives, address questions, and schedule a suitable interview time.

To protect confidentiality, names were not recorded, identifying data were removed during transcription, and identity codes were applied to all transcripts. To maintain privacy, all interviews were conducted remotely via Microsoft Teams, without observers present, at times selected by participants. At the start of each interview, participants were informed of the voluntary nature of their involvement. Participant characteristics (e.g., role, organization, and experience providing care to pregnant patients) were then collected verbally and documented in field notes.

All interviews were conducted in English and lasted 22–59 min, with a mean duration of 43 min. They were audio-recorded and transcribed verbatim. We reached data saturation when no new themes emerged.²² No repeat interviews were conducted, and transcripts were not returned to participants for review.

2.4. Data analysis

We employed a thematic approach to analyze the data, aiming to reveal patterns and themes within the dataset.²³ This approach was chosen for its foundational status in qualitative research and its ability to uncover unforeseen insights, particularly given the limited peer-reviewed literature on this topic. An inductive approach allowed themes to emerge directly from the dataset.²³ We adopted Braun and Clarke’s 6-phase process for thematic analysis.²⁴

KMG performed primary coding with regular input and oversight from DS. We read through the transcripts for familiarization and performed line-by-line coding on the first three transcripts to develop initial codes. These codes were iteratively organized into a detailed coding tree, which grouped related codes into categories and subsequently overarching themes. Themes were cross-referenced with coded extracts from all transcripts.²⁵ During the writing of the results, we refined the codes by comparing evolving interpretations against the raw data to ensure accurate representation, as per Charmaz’s constant comparative approach.²⁶ This method also guided probing in later interviews.

We used NVivo 14 for data management and analysis,²⁷ and Miro to visualize thematic relationships.²⁸ No formal participant checking was conducted.

2.5. Reflexivity

To enhance rigor, we adopted reflexivity practices.²⁹ We maintained a detailed journal documenting methodological decisions, potential biases, and interpretations of the data. Given participants’ awareness of our border experience, we presented ourselves as informed but intentionally neutral to encourage fuller responses.³⁰ We also employed reflexivity to account for prior professional relationships with several participants, recognizing that existing familiarity could influence the data collection process. We actively reflected on our positionality as US- and UK-based public health researchers with previous experience in humanitarian contexts.

3. Results

Participant characteristics are summarized in Table 1. We recruited 10 HCWs, including 6 nurses and 4 physicians. Nine had provided direct PNC at GRM’s Matamoros or Reynosa clinics. Although P2 did not provide direct care, their insights into border-based service delivery were included following transcript review. Findings are organized through three overarching themes: (1) structural violence; (2) resource limitations; and (3) care fragmentation. Additional participant quotes are provided in the Supplementary data.

3.1. Theme 1: Structural violence

Structural violence refers to social systems—such as legal, political, and economic structures—that systematically disadvantage certain groups and restrict access to basic rights and resources.³¹

3.1.1. Denial of care

Participants described how migrants were frequently turned away from hospitals without evaluation, received incomplete diagnostics, or were denied ambulance transport and admission. Routine PNC through state systems were often inaccessible due to financial barriers, and even those who delivered in hospitals were discharged quickly without postpartum care.

“Some people just flat out get turned away. Or there were situations where we would send someone to the hospital … and they didn’t even do the workup … At the public hospital, my understanding is [that] care should be free, but a lot of times you have to provide your own supplies and purchase [your own] medicines. – P1

3.1.2. Discrimination and language barriers in healthcare access

Language barriers and racial discrimination, particularly against Haitian migrants, were cited as major obstacles within Mexican healthcare institutions.

“The Haitian population [is] really discriminated against and not provided care [for]. [Hospitals are] just churning Haitians away or [providing] basic care and then turning them away. – P7

3.1.3. Cartel violence and safety concerns

Fear of cartel violence deterred migrants from accessing local services. Participants noted that unsafe conditions—such as pregnant women being unable to walk unattended—compromised their sense of safety, which may adversely affect maternal health.

“I learned about the drug cartels there … They’re patrolling the streets and it’s hard to know who is who. So, for a pregnant woman to walk the streets unattended, [it] can be dangerous.” – P4

3.1.4. Overburdened public health system

Participants emphasized that the Mexican public health system was already under strain and faced additional pressure in responding to the needs of the migrant population.

“They’re already taking care of very destitute groups of Mexicans [and] are also expected to handle this new influx of migrants.” – P6

3.2. Theme 2: Resource limitations

Resource limitations refer to constraints in space, staffing, or supplies that compromise the consistent delivery of quality healthcare.³²

3.2.1. Shortage of specialized providers

Participants noted a significant lack of specialists, particularly obstetricians and gynecologists. Most providers were generalists, and many participants reported working beyond their usual scope of practice, gaining experience through their clinic work at the border.

“This is not my area of expertise at all … I don’t have the kind of comprehensive knowledge of what to expect at a particular stage of pregnancy or what normal [is]. What [is] abnormal? That was all learned throughout my time in the clinics.” – P6

3.2.2. Limited medical supplies and diagnostic equipment

All participants described limitations in available supplies and diagnostic equipment. While basic labs and limited ultrasounds were accessible, budgetary constraints and reliance on external referrals restricted comprehensive screening for infections, congenital anomalies, and fetal growth.

“So that a pregnant woman was getting regular exams, ultrasounds, lab[s] … That kind of care is just not possible to maintain.” – P6

3.2.3. Lack of privacy in makeshift clinics

Participants noted that GRM clinics operated in makeshift spaces—such as courtyards, dining halls, or porches—where limited privacy impeded the ability to address sensitive issues like gender-based violence (GBV) or perform sexually transmitted disease (STD) screenings.

“We weren’t screening for STDs. No, because that means undressing them … There’s really no privacy in those clinics.” – P7

3.3. Theme 3: Care fragmentation

Care fragmentation is characterized by the distribution of care across multiple providers without consistent oversight, leading to disruptions in continuity and coordination.³³

3.3.1. Patient mobility disrupts follow-up care

Participants identified the mobile nature of the migrant population as a major barrier to continuity of care. Migrants frequently moved between cities while awaiting asylum or seeking safety or resources, sometimes traveling over 50 miles to access services.

“That person either crosses over or something happens to them, they disappear. You don’t know what happens to them … That’s not an element [that] can be accomplished there.” – P5

3.3.2. Short-term staffing reduces provider continuity

The use of short-term, rotating volunteers made it difficult for patients to see the same provider across visits. Participants noted that this disrupted continuity of care and hindered relationship-building and consistent follow-up.

“If we wanted to see [a patient] the following week, chances are they were going to see a different [provider] … You’re writing notes in their little prenatal book, hoping that the next person can pick up where you left off.” – P4

3.3.3. Fragmented medical records hinder clinical decision-making

Participants explained that inconsistent medical records hindered effective follow-up and care planning. Key clinical data—such as pregnancy history, group B streptococcus (GBS) status, and prior complications—were often unavailable. Several participants expressed concern about tracking preeclampsia, which they noted was particularly common among Haitians, who comprised the majority of GRM’s patients.

“Part of prenatal care is pregnancy history. If someone has a history of pregnancy loss, that’s a very important indicator for the current pregnancy.” – P5

Although not described as a standalone barrier, participants reflected on the emotional and ethical toll of delivering care in this context. Communication challenges—particularly the absence of trained interpreters—limited their ability to assess patients and discuss sensitive topics. Several described feelings of helplessness when unable to meet patients’ needs or provide the level of care they believed was ethically appropriate.

4. Discussion

This qualitative study explored PNC delivery along the US–Mexico border from the perspective of HCWs. Three interrelated themes emerged as critical to understanding the barriers to care: structural violence, resource limitations, and care fragmentation. Despite dedicated efforts of organizations such as GRM, the quality and accessibility of PNC for migrants often fall below international standards.

Findings align with prior research documenting systemic barriers to healthcare access for migrant populations at the US–Mexico border, including care denial, discrimination, and institutional neglect.⁶^,³⁴ Although Mexico’s revised General Health Law guarantees universal access to public health services,³⁵ implementation remains inconsistent, and migrants continue to face substantial access barriers.³⁶ Pregnant women are especially vulnerable to neglect and abuse.⁶^,⁹^,³⁴ Asylum seekers more broadly are perceived as overburdening Mexico’s strained public health system, leading to discrimination from both healthcare providers and the public.³⁴^,³⁷ Participants further emphasized that pervasive cartel violence—particularly in border cities—deters migrants, especially pregnant women, from seeking necessary care, increasing stress and the risk of adverse pregnancy outcomes.⁶^,⁹^,³⁴^,³⁸^,³⁹

Our findings also underscore structural challenges in humanitarian health delivery. Participants described shortages of specialized staff, diagnostic equipment, and medications—findings that echo prior research on NGOs at the border, including GRM and Médecins Sans Frontières (MSF), which operate under critical resource constraints.³⁴ GRM’s clinics in Reynosa were mobile, and while mobile clinics have expanded PNC access in other low-resource settings,⁴⁰ their effectiveness depends on sustained staffing, infrastructure, and service scope.⁴¹

With these settings, privacy-related limitations—particularly in makeshift clinical spaces—emerged as barriers to addressing sensitive issues such as GBV and conducting essential screenings. Similar concerns have been documented in mobile clinics in the US,⁴² South Africa,⁴³ and Brazil.⁴⁴ Participants also noted that fragmented or incomplete medical records hindered clinical history tracking and risk assessment, echoing broader systemic challenges in humanitarian contexts where inter-organizational data sharing remains limited.⁴⁵ Comprehensive medical, social, and reproductive histories are critical for early detection and management of high-risk pregnancies, including hypertensive disorders such as preeclampsia,⁴⁶ which participants noted were especially prevalent among Haitian populations—who accounted for the majority of GRM’s patient population.⁴⁷

While our findings affirm existing literature, they also provide novel insight into the delivery of PNC in border humanitarian contexts. For example, participants emphasized how rotating staff and the transient nature of patients compromised relationship-building and the ability to provide consistent longitudinal care throughout the prenatal period. These disruptions extend beyond general access barriers and point to gaps in long-term care.

This study has several limitations. Its small sample size and recruitment from a single NGO may limit generalizability. Intermittent, rather than continuous, volunteer participation may not fully reflect the experiences of full-time frontline clinicians. The focus on US-based HCWs excluded valuable perspectives from local Mexican providers and migrants, which future studies should prioritize. Additionally, the rapidly evolving policy landscape and fluctuating border conditions may affect the long-term relevance of our findings.

4.1. Conclusion

This study reveals an escalating humanitarian crisis at the US–Mexico border, where basic PNC remains out of reach for pregnant migrants. Structural violence, resource limitations, and care fragmentation undermine the quality and accessibility of maternal healthcare. Despite the efforts of humanitarian organizations, entrenched systemic barriers persist. Addressing these challenges will require comprehensive policy reform, strategic resource allocation, and coordinated binational action grounded in human rights. Strengthening integration between NGOs and formal healthcare systems could enhance service coordination, equity, and outcomes for underserved border communities.⁴⁸ Future research should incorporate the perspectives of Mexican healthcare providers and migrant communities, using community-based participatory approaches to develop rights-based, responsive models of care that bridge the gap between inclusive policy frameworks and on-the-ground healthcare delivery.³⁶

Author statements

Ethical approval

This study was approved by the Usher Masters Research Ethics Group (reference no. UM23247), Usher Institute, University of Edinburgh. Written informed consent was obtained from all participants.

Data availability

The data that has been used is confidential.

Author contributions

KMG contributed to conceptualization, methodology, project administration, investigation, formal analysis, data curation, and writing the original article. WRM contributed to reviewing and editing the article. AMM contributed to methodology, reviewing, and editing the article. DS contributed to reviewing and editing the article, validation, and supervision.

Funding

This work was supported by the US National Institute of Allergy and Infectious Diseases (grant number T32 AI007433, awarded to WRM) and the Harvard University Center for AIDS Research (CFAR), an NIH-funded program (grant number P30 AI060354, awarded to WRM). The funders had no role in study design, data collection and analysis, preparation of the manuscript, or decision to publish.

Competing interests

DS is a member of the editorial board for Public Health. The other authors declare no competing interests.

Acknowledgements

We are grateful to the participants who took part in this study. We acknowledge the collaboration of the GRM team, particularly Brendon Tucker and Clara Lee Arnold. Our appreciation extends to the MGH Center for Global Health, especially Dr. Lindsey Martin and Kristen Giambusso, for their valuable support during the planning stages of this research.