Health effects and user perceptions of the US Customs and Border Patrol One™ mobile application: A qualitative analysis among asylum seekers at the Mexico-US border

Highlights

• •Use of the US customs and border protection (CBP) One™ mobile app is mandated for asylum seekers prior to entering the US.
• •Millions of migrants are using the app, despite limited understanding of usability.
• •In this qualitative study, asylum seekers reported confusing interfaces, technical flaws, and racial biases.
• •Users associated CBP One™ with negative physical and mental health effects.

Abstract

Background

The number of migrants at the Mexico-US border has increased to historic levels, and frequently changing immigration policy impacts this population as they await entry into the US. This study evaluated the usability and health effects of the Customs and Border Protection (CBP) One™ mobile application among asylum seekers near the US port of entry in Reynosa, Mexico.

Methods

We conducted semi-structured qualitative interviews with 20 asylum seekers in Reynosa, Mexico, in February 2023. Our objective was to explore the subjective experiences of migrants, usability of CBP One™, and presence of perceived health effects from using the application. Interviews were conducted until saturation occurred, transcribed verbatim into Word, coded in NVivo using a validated, team-based coding methodology, and analyzed according to internal domains, external domains, and health effects regarding CBP One™.

Results

Twenty participants originated from eight countries throughout Latin America and the Caribbean. In total, 18 subthemes were identified among internal, external, and effects domains. Internal themes included a confusing application interface (80%), technical malfunction (60%), and perceived racial bias from the photo-capture features (15%). External themes challenging CBP One™ use included unavailable appointment slots (80%), inequity and inaccessibility (35%), and inadequate internet (25%). Most perceived effects were negative (85%), including worsening mental health effects (40%), exacerbation of pre-existing physical conditions (35%), and forgoing health expenditures to pay for internet (25%).

Conclusions

Our findings suggest that asylum seekers at the Reynosa port of entry perceive CBP One™ negatively, with detrimental effects towards their mental and physical health. This study highlights how immigration policy can influence health and suggests that more creative and humane approaches are needed for people seeking asylum at the Mexico-US border.

1. Introduction

The US and Mexico face a complex humanitarian situation at their shared border. The number of refugees, asylum seekers, and migrants approaching the Mexico-US border has increased to historic levels. While the precise number of people seeking entry from the Mexico side of the border is unknown, the US Customs and Border Protection (CBP) reported over 300,000 encounters with people irregularly crossing the border in December 2023 (U.S. Customs and Border Protection (CBP) 2023). Further complicating the challenging social and economic factors which push people to migrate, the US has implemented immigration policies which prevent asylum seekers from entering the US during their asylum process. These policies, including the Migrant Protection Protocols (MPP), impacted the medical and public health conditions of migrants, mainly through the growth of tent encampments near ports of entry (Garrett and Sementelli, 2022). Within these congregate settlements, migrants reside in makeshift tents or shelters, facing increased exposure to violence, poor living conditions, and disease with limited health access (Morris and Saadi, 2022). In response, local and international policies are being piloted to curb increases in migration, such as limits on entry in a given time period and increased militarization of security personnel on both sides of the border. One such policy is the Customs and Border Protection App: CBP One™, a mobile phone application which allows asylum seekers to schedule asylum appointments (American Immigration Council 2023). Originally created in 2020, CBP One™ was designed to facilitate asylum appointments and border crossings in a more organized way and help to reduce the time that migrants reside in unsafe camps. However, in January 2023, CBP mandated that asylum seekers must use CBP One™ to request appointments for passage into the US, and those who did not use the application would risk being deported upon entry (Houston Immigration Legal Services Collaborative 2023). There are additional underlying structural restrictions, including that there are only eight ports of entry where asylum seekers can be processed, with drastically variable censuses and resources for processing. Such variability has led to unpredictable waiting times for asylum claims to be heard, in most cases lasting many months.

At present, there is very limited data on the usability and effects, both positive and deleterious, of the CBP One™ app. Through the application, an asylum seeker is supposed to be able to make an individual account, enter demographic information, take and upload a face photo for biometric data, and request an asylum appointment within the following weeks (U.S. Customs and Border Protection 2024). Asylum seekers must have the most updated version of the app before presenting at the border. Appointment slots refresh daily, such that if they are full one day, there should be appointment openings within 24 h (U.S. Customs and Border Protection 2023). However, early reports from advocacy groups show that the application is difficult to use and may not be fulfilling its intended aims (American Immigration Council 2023). Multiple sources have described technical glitches causing migrants to lose already scheduled appointments and an overly complex user interface, creating confusion among users (Pinto, 2023). Issues have also appeared from its implementation. There are limited information sources to answer questions about CBP One, allowing scammers to promote misinformation and take advantage of migrants (Parness, 2023). There was early criticism from groups including the American Immigration Council about the application's design and implementation which were not heeded. Concerns emerged that migrants were not consulted as the intended end users when creating or testing the app, and that it was expediently developed without transparency and despite objections to its inherent flaws (American Immigration Council 2023). The mandated use of CBP One™ as the sole way to request asylum has been challenged in court and faces ongoing litigation. However, given the complexity of legal challenges and its already widespread use, it is likely that millions of people will use CBP One™ in the coming years (National Public Radio 2023). Despite its prolific use, there is limited information on user perceptions of the CBP One™ mobile application. There is even less information on the effects of using the app, particularly on the physical and mental health of asylum seekers and other migrants.

Understanding the effects of CBP One, both positive and negative, will be critical to optimizing the application's intended uses while mitigating negative consequences. In this qualitative study, we explored the user perceptions of the CBP One™ mobile application and health effects from using the app among migrants seeking asylum at the port of entry in Reynosa, Mexico, located in the Rio Grande Valley on the Southeast Texas border. Migrants were selected as the object of qualitative interviews, given the lack of data regarding their perceptions, despite being the targeted end users.

2. Material and methods

2.1. Study design

This study employed a nested qualitative design within a larger effort examining migrant health at the Mexico-US border. Since there is limited evidence on the health impacts of CBP One™, and even more widely on the effects of mobile technologies on migrant health, this study employed a phenomenological approach with an exploratory survey based in grounded theory (Byrne, 2001). Phenomenology is a qualitative research method that explores how people experience the world, objects within it, and any derivative meaning from that experience. It is useful when exploring understudied concepts, particularly the impact of novel technologies on the reality of human experience. Using this approach, we aimed to understand the impact of this mobile technology on health by investigating the perceptions and experiences of end users. Qualitative design was considered most appropriate to allow for in-depth explorations of complex topics including health, migration, and technology, alongside differentiation of experiences between users with diverse geographic, migratory, and life experiences. A semi-structured interview script was developed following a literature review on the use of mobile health technologies (mHealth) among immigrant and other vulnerable populations (Duclos et al., 2017; Talhouk et al., 2020; Brown et al., 2023). The interview script was revised by content experts including physicians (n = 2), legal immigration experts (n = 2), humanitarian workers (n = 3), and asylum seekers (n = 2), including clarification of themes and purging of repetitive items. The final script lasted approximately 20 min and was designed to be applied by Spanish-speaking researchers with experience in semi-structured qualitative methods. The final version included questions on users’ experiences with CBP One™, positives of application use, challenges to use, and observed effects on health and wellbeing, either personal or within their community, from using the app.

2.2. Study setting

This study was conducted in Reynosa, Mexico, located in the Rio Grande Valley on the Southeast Texas border, and one of eight major ports where asylum seekers are processed for entry into the US. Reynosa was selected for the study site as it was one of the only ports of entry which consistently and actively processed migrants for crossing at the time that CBP One™ was mandated in January 2023 (Amnesty International 2023). Migrants traveled to Reynosa because it was a major port of entry, resulting in a migrant population that was large, unpredictable in number, and semi-transient. The conditions of Reynosa were reported to be some of the harshest for migrants, including living in makeshift camps and shelters and within regions under the control of cartel and other dangerous groups. The shelter capacity of the city was drastically lower than what was necessary to accommodate the influx of migrants, and as a consequence, informal camps came into being in a public plaza near the porty of entry bridge, on city streets, and by the banks of the Rio Grande River. Existing shelters, notably Senda de Vida, added tent camps to permanent structures to accommodate increasing numbers of residents. Congested living conditions became common within shelters and camps, where upwards of fifty migrants slept in one structure with single-packed, bunk style sleeping quarters, or in tent clusters situated dangerously close to bathroom and waste disposal facilities (Jordan, 2021). Food, water, and medical care were provided by aid groups from US and Mexico-based charitable organizations and nonprofits. These groups included but were not limited to Global Response Medicine (GRM), the Sidewalk School, Senda de Vida, Casa Lulu, Solidarity Engineering, and Médicos Sin Fronteras (MSF). Most migrants living in these camps and shelters were exposed to environmental elements, with variable access to potable water and food, much of which was provided by these external aid organizations. Reynosa was also unique from other US ports of entry in that migrants were distributed across various camps and shelters throughout the city, rather than all residing in one area as is the case with neighboring Matamoros. This distribution added additional variance to how migrants experienced CBP One™. At the time, the GRM was one organization providing health care to migrants within Reynosa camps and shelters, which were otherwise restricted to public access for reasons of security.

2.3. Data collection and participants

Collaborating with GRM, bilingual researchers conducted semi-structured interviews with migrants residing across four migrant camps and shelters in Reynosa, Mexico, in February 2023 (Fig. 1). Migrants were eligible to participate if they were seeking asylum in the US, over age 18, able to give verbal informed consent, and Spanish-speaking. Those under 18 years and who otherwise did not meet inclusion criteria were excluded from the study. Participants were selected with a snowball sampling convenience method to intentionally reflect a diverse range of sex, age, country of origin, and migratory time. Recruitment occurred in two ways. First, study researchers would approach community leaders within the camp or shelter and ask how to best invite participants, receiving recommendations on willing participants. Second, patients who presented to the GRM clinic who were not in acute distress were invited to participate in an interview following their medical consultation by a clinician. The consent process emphasized that participation in the research study was voluntary and had no impact or bearing on the quality of care they would receive.

Fig. 1.

Aerial view of migrant settlement camps in Reynosa, Mexico.

Figure legend: These photos represent drone footage from four migrant shelters and camps in Reynosa, Mexico including a and b) Shelter Senda de Vida 2, and c and d) Camp Rio. Photos provided by Erin Hughes through Solidarity Engineering.

Participants were introduced to the study as an opportunity to share their experiences and opinions about the CBP One™ application and with healthcare in the migrant camp or shelter more broadly. All participants underwent a process of verbal informed consent before initiating the interview and given the chance to ask questions before agreeing to participate. Interviews were conducted one-on-one in a private location, including unoccupied space outside medical clinics and community areas, according to patient preference. All interviews were conducted in Spanish, recorded with the EasyVoiceRecorder app on iPhone, immediately uploaded to an encrypted DropBox folder, and deleted from the iPhone. The research team met periodically to identify emerging themes and determine when saturation had occurred, defined as the cessation of emergence of new themes (Saunders et al., 2018). Researchers identified that saturation occurred when additional interviews yielded similar observations and experiences to what was already identified. At that point, three additional interviews were conducted with increased scrutiny to verify saturation. The final total of 20 interviews was consistent with the saturation point of other studies using empiric qualitative data (Hennink and Kaiser, 2022).

2.4. Data analysis

While data were inherently deidentified, participants were assigned a code to further guarantee anonymity. Recorded interviews were transcribed by hand by bilingual researchers in Microsoft Word and translated into English. Transcripts were stored on the encrypted DropBox only available to study researchers. Following transcription, researchers used an inductive approach with thematic content analysis through independent immersion in two transcripts for the emergence of major themes. Following theme development, a codebook was drafted and validated by two study researchers using independent and cooperative techniques according to a team-based coding methodology (MacQueen et al., 1998). This process involved two co-investigators using a repetitive process of increasing inter-coder agreement by independently analyzing the same transcript, followed by collaborative discussion to refine the codebook until at least 0.7 Cohen's Kappa measure was achieved. We calculated Cohen's Kappa using a standard formula:

$$\kappa =\frac{p_o-p_e}{1-p_e}$$

where p_o is the observed agreement, and p_e is the hypothetical probability of chance agreement. This calculation is a validated approach to evaluating inter-coder reliability, with a measure of >0.7 indicating substantial agreement between two independent coders on how themes from the free-text transcripts should be coded. Using the validated codebook, transcripts were coded in NVivo12 according to major themes by two study researchers including a researcher with experience working with migrants in Reynosa. We used a phenomenological analytic approach with theme frequencies and quote selection to reflect diverse participants (Carel, 2011). Emerging themes were grouped into three categories: internal domains, external domains, and effects. Internal domains reflected inherent features of the CBP One™ app, while external domains grouped responses related to external factors which affected the application and users’ experiences with it. Effects included the mental and physical health consequences that migrants perceived from using CBP One™. Once themes were identified and grouped, member checking was used among clinician authors and health providers through GRM clinics to verify results based in pragmatic experiences. Member checking was not undertaken with asylum seekers, given that personalized data was intentionally excluded to guarantee anonymity, thereby precluding follow-up. All researchers underwent a reflexivity exercise to determine potential biases as it relates to the data and findings. This study was approved by the University of Michigan Institutional Review Board (HUM00186322) and adhered to the Standards for Reporting Qualitative Research (SRQR) guidelines for qualitative research (O'Brien et al., 2014).

3. Results

3.1. Demographics and CBP One™ usage

Twenty asylum seekers across three migrant shelters and one tent encampment in Reynosa, Mexico participated in the interview. Participants were diverse, representing a range of ages (median [IQR] = 36 [(Alencar, 2023-43] years), eight countries of origin, and time ranging in Reynosa from one month to two years (Table 1). Most participants had heard of CBP One™ (n = 19, 95%) and had experience using it (n = 18, 90%), despite the application being a recently implemented technology. An asylum seeker from Haiti reported, “We know we need to use the program now. Before people would just make a line one by one [to request asylum] and that way you knew when it was time for your appointment, but not anymore. Now we are waiting for the app to say its time” [33yo M, Haiti]. Among the 19 participants who had heard of the application, they all recognized that it should be used daily to attempt to make an appointment and did so consistently (n = 18). Despite its widespread and frequent use, only four asylum seekers had successfully secured an appointment (20%).

Table 1.

Demographics of interviewed asylum seekers using CBP One™ in Reynosa, Mexico.

	Count	Frequency (%)
Sex
Male	10	50
Female	10	50
Average age (SD)	37.8 (9.19)
Country of origin
Haiti	8	40
Honduras	4	20
Mexico	2	10
Venezuela	2	10
Guatemala	1	5
Nicaragua	1	5
Colombia	1	5
Dominican Republic	1	5
Time living in Reynosa camp or shelter
<1 month	2	10
1–2 months	10	50
2–4 months	6	30
>4 months	2	10
Average	3 months
Median	1 month
Experiences with CBP One™
Has heard of CBP One™	19	95
Has used CBP One™	18	90
Secured asylum appointment with CBP One™	4	20

3.2. Internal domains

Participant defined inherent characteristics of CBP One™ which increased challenges to use. Seventeen participants (85%) described internal domains, which were categorized into six subthemes: confusion using the app interface, technical malfunctions, biometric biases, racial preferences, scheduling in other cities, and family separations. The most mentioned issue was confusion using the CBP One™ app interface (n = 16, 80%). Confusion resulted from inherent app features, navigating between multiple pages to complete registration, and a lack of confirmation if an appointment was scheduled successfully. Some acknowledged that their own limited experiences with technology and Internet may be contributing to these challenges, including that Facebook was the most advanced webpage most had used (Table 2). However, no designated person from CBP or other government organization was available to answer questions or manage uncertainties when asylum seekers struggled with the app. In many cases, migrants relied on one another to navigate CBP One™.

Table 2.

Internal domains and subcategories regarding usability of Customs and Border Protection (CBP) One™ mobile application.

Theme (%)	Frequency (%)	Representative quote
Internal domains of CBP One™ (n = 17, 85%)
Confusion using app interface	16 (80)	“It is so confusing. They said there's a program called CBP, which is like the US Google. In the morning you are supposed to put some information, but it is very complicated for us. For us immigrants, Facebook is the most complicated thing we've used.”
		33yo M, Haiti
Technical malfunction	14 (60)	“Another problem is that the page collapses because so many people are trying to use it. With it being an application made in the US, I would think it would be able to handle all the traffic, but clearly there are issues with everyone trying to use it.”
		31yo F, Guatemala
Biometric biases	3 (15)	“Yesterday the app was so slow with taking my photo. It kept rejecting. I tried and tried but it didn't want to take my photo or accept my information. That stressed me so much and eventually the whole thing just blocked me out.”
		49yo F, Venezuela
Scheduling appointments in other cities	3 (15)	“I needed help to figure out how to use CBP One, but once I did, I could only get an appointment in another city. So, I need to find a way to travel there and join my family in the US.” .”
		41yo M, Haiti
Separation of families from differentiation of individual and group appointments	2 (10)	“The appointment I made is for one person and I'm with my family. So tomorrow I need to retry to make an appointment for my wife. If I can, she'll join me, and we'll be together. Maybe it won't happen, but they can also be canceled. So, I'll try to find a spot for our complete family and cancel the individual appointment.”
		42yo M, Haiti
Racial preference	2 (10)	“Sometimes we think that for Hispanics the process is more complicated. They say there will be appointments, but no one has gotten permission to make one.”
		36yo M, Honduras
Lack of prioritization for special circumstances	2 (10)	“The worst is the app doesn't take into account special situations like children or parents of families. A mother with three kids should get priority because it is so bad here, sleeping in the cold and in tents where they are getting sick.”
		46yo F, Venezuela

A second inherent issue was technical malfunctions. Even when participants could navigate the interface, 60% (n = 14) reported experiencing technical problems including system crashes, random application closures, screen freezing, pages refreshing at inappropriate times, and previously available appointments disappearing after being selected. Most found that following even one technical glitch, previously available appointments would disappear by the time they refreshed. While possible that these glitches could be due to connectivity issues, migrants attributed it to the application, as it occurred even in instances with access to quality Internet.

Two subthemes emerged regarding participant identity as it related to CBP One™: biometric biases and racial preferences. As part of the scheduling system, users were required to upload a face photo using the smartphone photo capture feature, which would be validated as acceptable by the system. Three asylum seekers with darker skin, two Haitian and one Venezuelan, reported difficulty with having photos accepted. They perceived that this challenge disproportionately affected migrants with darker skin. Similarly, two participants from Latin America (10%) expressed that the app had a racial preference for certain ethnic groups.

Three final internal domains of the CBP One app which caused challenges reflected a lack of nuance in how appointments were scheduled. First, participants were sometimes only able to schedule appointments at ports of entry different from Reynosa (n = 3, 15%). These included Brownsville, situated an hour from Reynosa, but also Nogales, Arizona and Calexico, California, traveling to which required a plane or multi-day ground transport. Since appointments were so scarce, some asylum seekers elected to travel the hundreds of miles necessary for these appointments, while others waited to schedule through the Reynosa entry of Hidalgo. Second, a similar issue occurred when appointments were only available for individuals and not entire families (n = 2, 10%). There were limited numbers of individual and group appointments each day to register through the app, and families found it more difficult to find an available time which accommodated each member. This resulted in the difficult family decision of whether to wait longer in the precarious conditions facing many in Reynosa to find a group spot or to separate into smaller groups to optimize chances of finding an appointment. In many cases, this led to family separations, where parents or partners were forced to choose between separating from loved ones to cross the border or forgo the appointment to risk finding another for the entire family in the future. Finally, there was a lack of nuance on prioritization for special circumstances, including children, older adults, and those with disabilities. Similarly, there was no system to account for how long an asylum seeker had been waiting for their appointment, resulting in huge variation in waiting time from a few days to months.

3.3. External domains

Participants mentioned multiple contextual factors which affected their experiences using CBP One™, beyond inherent features of the application. These findings were categorized into five subthemes: unavailable appointments, inaccessibility and inequity in access, inadequate cellular service, confusion from rapidly changing immigration processes, and conflicting priorities to proper use. The most mentioned theme was unavailable appointment slots (n = 16, 80%). Likely due to the increased number of migrants seeking asylum far exceeding the number of appointments available, new appointments hardly appeared and would be quickly taken when they did. Importantly, the policy for appointment scheduling, including number of spots, allocation methods, and prioritization, has continued changing. Many waited months for an appointment but were unable to schedule despite waiting weeks (Table 3).

Table 3.

External domains and subcategories regarding usability of Customs and Border Protection (CBP) One™ mobile application.

Theme	Frequency (%)	Representative quote
External domains of CBP One™ (n = 16, 80%)
Unavailable appointment slots from too many users	16 (80)	“Once you find an appointment date, it disappears in like five seconds and you lose your spot. No more date. No one is able to cross because there are no appointments.”
		36yo M, Haiti
Inaccessibility and inequity in access	7 (35)	“There are people who haven't been able to get an appointment because they don't have a cell phone or don't know how to read nor write.”
		27yo F, Haiti
Inadequate cellular service and Internet	5 (25)	“[To use the application] we have to be in the street, fearful that something will happen to us, because sometimes the application doesn't get service from inside our house. The data is very limited.”
		42yo F, Dominican Republic
Conflicting priorities leading to improper use	5 (25)	“I haven't been able to use the app before I migrated because of my situation [of violence] and now am lost about how to do so.”
		60yo M, Nicaragua
Changing process causing confusion	3 (15)	“I had an appointment to cross for November 26 of last year before the app was required. Things can be delayed for people like me traveling alone, so we were waiting and waiting and finally our day was close, but then the application arrived and everything from before was suspended.”
		31yo F, Guatemala

A second external factor was inaccessibility and inequity in access. Personal characteristics affected a user's ability to appropriately engage with CBP One™, including education and literacy, smartphone access, and language barriers particularly for non-English or Spanish-speakers since the application was only available in those languages at the time of interview. Most asylum seekers reported that smartphone use was high among their communities, however many could recall cases of those without smartphones who were unable to access CBP One™. Those without phones would need to buy one just to register with CBP One™ or borrow the phone of a camp or shelter supervisor, aid worker, or friend. Given vastly varied experiences with technology in one's home country, participants perceived an advantage for those who had already used smartphones or experiences with advanced technologies through their occupation, training, or education.

A third external factor which affected migrants’ abilities to properly use CBP One™ was inadequacy of cellular service, internet, or WiFi connections. Most participants believed that Reynosa lacked quality cellular service, made worse within camp and shelter settlements. For security reasons, many shelters were walled behind metal gates or concrete walls, which further limited connectivity to the app since it required internet access. At the time of interview, the migrant camps and shelters did not offer WiFi, and migrants often found they needed to leave the safety of these areas to connect to the app. Some reported that the city offered internet connectivity, but its availability and use were inconsistent. Since most appointments were scheduled early in the morning while still dark, participants felt apprehensive to leave the safety of protected areas to access CBP One™.

Fourth, the consequence of introducing CBP One™ as a new immigration process without any instruction or support mechanisms created confusion in Reynosa. Specifically, most of those interviewed had started their migration journeys or been living in Reynosa since before CBP One™ was mandated. They experienced a shock when in the middle of navigating through the old asylum process, they now needed to engage with an entirely new system which did nothing to account for time already spent waiting. While some ports of entry may have implemented support systems to help end users navigate the app, this was not the case in Reynosa. For those with questions, there was no person or verified source to reference for help with navigating CBP One™. Most often, community members or aid workers more familiar with the interface would have to assist.

The final external factor which emerged were conflicting priorities to CBP One's™ proper use, manifested in two major ways. First, migrants were unable to use CBP One™ to schedule asylum appointments before migrating, as was intended for special populations including those from Cuba, Haiti, Nicaragua, and Venezuela. Most left their homes hastily fleeing persecution including physical or psychological violence, intimidations, threats, and death of family. No participants attempted to use CBP One™ before arriving in Reynosa, likely because they were unfamiliar with the application given its recent rollout. The second conflicting priority was cost associated with using the app. While CBP One™ itself was free to users, its data requirements without access to free connectivity meant that asylum seekers purchased cellular data even for just the chance to schedule appointments. With many living in poverty, unemployed, and without financial support, participants reported difficulty affording the daily cost associated with CBP One™ use.

3.4. Perceived effects of CBP One™ usage

Interviews revealed perceived effects of using CBP One™, particularly impact on the physical and mental health of participants. While there were some positive effects of the application (n = 3, 15%), the majority of responses were negative towards CBP One™, particularly for health (n = 11, 55%). Among positive effects, participants reported that CBP One™ provided a legal avenue to cross into the US (n = 3, 15%). Many emphasized their priority was to cross legally and CBP One™ gave an avenue to do so, despite its flaws. Others appreciated CBP One™ because it added some order to an otherwise chaotic process. They emphasized that despite confusion, each person was able to manage the process individually from their personal phone, replacing the previously disorganized lines which emerged from the previous registration system.

Nevertheless, most participants perceived negative effects from CBP One™ usage, categorized into four themes: psychological health effects, physical health effects, forgoing health expenditures, and disproportionate impacts on vulnerable populations including children. Psychological health effects corresponded to symptoms of mental health conditions including anxiety, depression, post-traumatic stress disorder, personality or other mood disorders. In terms of psychological effects, participants reported anxiety and hopelessness related to uncertainty using the application. The most mentioned symptom was sleep disturbances, caused by waking up at 3:00am to register for an appointment daily (Table 4). “Depression,” “hopelessness,” and “unpredictability” were phrases frequently used to describe the mental health burden of using CBP One™ (n = 8, 40%): “Everyone is depressed. They all want to go home but they are giving everything they can to be here. There are many people who don't have appointments and they are becoming bored, depressed, and sick because those mental effects can have many consequences” [35yo M, Colombia].

Table 4.

Perceived effects of using CBP One™ mobile application among asylum seekers in Reynosa, Mexico.

Theme	Frequency (%)	Representative quote
Positive effects of CBP One™ (n = 3, 15%)
Provides a legal avenue to cross into the United States	3 (15)	“[CBP One] helps with the issue of not being able to cross into the United States because we are looking to ask for asylum in the most legal way possible.”
		35yo M, Honduras
Adds a modicum of order to a chaotic process	2 (10)	“I saw how disorganized the process was and it made me uncomfortable. I thought I didn't belong here because I am generally a calm person. With the application's arrival things will be better, because everyone will have their own cell phone. Those that don't know how to do it can be helped.”
		42yo M, Haiti
Health effects from CBP One™ (n = 11, 55%)
Psychological / mental health	8 (40)	“Mentally, I have been very bad. When we arrived each of us had a specific date and anxiously waited for the day to come. Now that they've introduced the CBP One application, everyone has become depressed because we are trapped here, looking every day to find an appointment. And I don't know why but nobody is able to. Some say it's the network, others the service, and I don't know. But yes, very discouraged from all that is happening.”
		31yo F, Guatemala
Physical health	7 (35)	“Truly, I don't sleep. I don't sleep because like many people I wake up at 3am to take the photo and can't sleep after that. And because I have stomach problems, the nerves make my condition worse and I get attacks where I have terrible vomiting, which causes my nerves to be even worse. Seriously this application is so bad.”
		42yo F, Dominican Republic
Health expenditures	5 (25)	“I think every day I'm spending 200 pesos to buy Internet data for my phone. And sometimes a mother only has 200 pesos at all, and she spends it on internet and not for her child, so her child suffers from hunger. Also, there are single mothers here without support from the child's father or a sponsor to send her money. They have no money. So, she needs to use a towel in place of a diaper for her child and will wash it to reuse. This happens a lot.”
		27yo F, Haiti
Disproportionate impact on vulnerable populations	3 (15)	“I feel horrible because you see that many people are waiting a long time here. It is so much harder for certain groups, especially the children. It hurts to see them here.”
		34yo F, Mexico

Physical health effects included somatic symptoms relating to app use, including in relation to pre-existing conditions or new onset of physical effects. For physical health, participants with pre-existing conditions reported that the application increased uncertainty and exacerbated their physical symptoms (n = 7, 35%). One patient with angina experienced more frequent chest pain, and another attributed her worsening achalasia symptoms to the app.

Due to CBP One™ associated costs, five participants forewent necessary health related expenditures in favor of app use (25%). These included not filling medication prescriptions, refusing hospital admission due to possible medical bills, and withholding nutrition or clothing for themselves or children. While medical care in theory should be free of charge in Mexico, this was not the reality of most interviewees as they experienced financial barriers to receiving care. Many participants stressed over potential catastrophic health expenditures in emergency situations and were unsure how they would finance these costs should they arise.

Finally, three participants emphasized the app's impact on vulnerable populations, particularly children (15%). Since CBP One™ did not prioritize families, parents worried about developmental effects for children living in precarious conditions posed by the camps and shelters. They cited a lack of housing, nutrition, education, and safety, alongside violence and trauma as negative consequences of extended migration waiting time when caregivers were unable to schedule appointments through the app.

4. Discussion

In this qualitative study evaluating the usability and health effects of CBP One™ among migrants in Reynosa, Mexico, one of the Mexico-US border ports of entry, we found multiple internal and external factors limited its successful use and that users associated the application with negative impacts on their physical and mental health. The application's design and technical flaws caused confusing interfaces, technical malfunction, and perceived biometric biases which prevented migrants from scheduling appointments. Biometric biases were especially notable among our sample, as there is a higher concentration of Haitians in Reynosa compared with other ports of entry, and importantly that this observation may have disproportionate impacts on the thousands of African migrants as well (Jordan, 2024). External elements which limited CBP One's use included unavailable appointments, inaccessibility and inequity for those without phones or specific language proficiency, and living in Reynosa without security or Internet access. Participants identified adverse effects on their physical and mental health through exacerbation of preexisting conditions, sleep deprivation, and feelings of hopelessness, anxiety, and depression. While some hoped the application would streamline asylum processing, the inaccessible platforms, interface malfunction, data expenditures, apparent racial biases, and health effects all negatively impacted people's experiences.

Our findings on users’ negative perceptions towards CBP One™ are shared by other reports. Since its implementation, CBP One™ has faced concerns over technical glitches which frustrate users and prevent individuals from obtaining appointments (Kocher, 2023). Advocacy groups criticized its impacts on accessibility and equity, including that those with better-quality phones may be the only ones able to schedule appointments (Dib and Deslandes, 2023). Recent reports revealed questions about the app's development, including for worrisome features (Pinto, 2023; Morrissey, 2023). For example, the photo capture was promised to be optional to advocacy groups who protested the feature, but later made mandatory without explanation (American Immigration Council, 2023). Multiple sources have confirmed that migrants with darker skin tones are having a harder time registering and validating their uploaded photos (Pinto, 2021). Additional questions were raised in terms of the legality of requiring a smartphone for asylum, which is a legal and international right, alongside privacy implications for the data being collected and stored, including GPS capabilities (del Bosque, 2023). Findings from our study reflect these reports, particularly with equity, technical glitches, and biometric biases.

Our study introduces a novel element: that features of CBP One™ may have detrimental health effects for migrants using the app, a finding which has not been previously reported. The impact of digitalization process of asylum is likely to have significant impact on migrants, including how their claims are accessed, interpreted, and processed. While there may be some positive aspects to this process including information dissemination and integration into new communities, our sample highlighted several negative user and health consequences of this process (Alencar, 2023). Participants related aspects of the application's design to their wellbeing through exacerbation of preexisting conditions, symptoms consistent with anxiety and depressed mood, forgoing health expenditures to purchase Internet data, and expressing concern for vulnerable populations including children, pregnant women, and those with disabilities. It is important to consider in light of our findings whether the mental health effects reported by our sample are due to the CBP One™ app itself or a flawed immigration system which is using it as an interface for engagement.

There is substantial evidence documenting how immigration policies can adversely affect asylum seeker health (Juárez et al., 2019). Past studies demonstrate that other immigration policies, including prolonged detention, habitation in informal and dangerous encampments, deportation to unsafe and unfamiliar areas without adequate processing, and extended delays in hearings, can have deleterious health impacts. (Hvidtfeldt et al., 2020; Human Rights Watch 2021). Recently, migrants residing within shelters and camps at major ports of entry have reported increased fear, anxiety, and uncertainty related to their health and immigration processes, even before the implementation of CBP One™ (Reynolds et al., 2022). Additional stressors include the lack of work opportunities and violence in border towns, to which migrants are particularly susceptible. While the health effects of using the application cannot be separated from the inherent structural factors causing stress that may exacerbate perceptions, recognizing CBP One™ and its perceived effects as an inherent part of the immigration process is an important step to mitigate these consequences. With CBP One™, humane treatment of asylum seekers should influence not only how they are treated upon entry, but the systems with which they interact to enter the US (Physicians for Human Rights Asylum Policy Working Group 2021). Nuanced approaches to immigration policy could be helpful to optimize how asylum seekers interact with CBP One™. For example, the Health in All Policies (HiAP) approach recognizes that population health can be affected by policies that do not directly target health systems or care delivery (WHO 2024; Center for Disease Control and Prevention 2024). Adopting an HiAP or similar framework to implement and evaluate CBP One™ could be helpful for mitigating the detrimental effects of its use.

Recent modifications to CBP One™ may have helped to address concerns raised by our sample. Instead of appointment selection being based on random chance, US CBP now claims that appointments are scheduled using a semi-random selection among registered users, with priority based on how long accounts are registered in the system (American Immigration Council, 2023). In theory, the longer an account is active, the longer an applicant has waited and the higher priority their registration becomes. However, it is unclear whether this system was already used at the time of interview and if participants still perceived a lack of prioritization. Additionally, the application is now also available in Haitian Creole, Russian, and Portuguese, and US CBP claims that smartphones can use inherent translation capabilities for when additional languages are needed (U.S. Customs and Border Protection, 2023). Still, crucial features of the app including registration and two-factor authentication are still only available in Spanish and English (Deck, 2023). Finally, appointment times are being released later in the day, at 11am EST instead of 3am, and listing appointments up to thirteen days in advance (U.S. Customs and Border Protection 2023). Though this modification may mitigate sleep disturbances from registering in the middle of the night, it is unlikely to address the anxiety, uncertainty, or hopelessness many participants felt when using CBP One™. While positive modifications have been made, significant issues remain including separation of families, technical glitches, unavailable slots, and biometric biases.

4.1. Strengths and limitations

Our study had several strengths and limitations. We included participants who were using the app prior to crossing the Mexico-US border, which means that we did not limit study participation to those who successfully used the app and already entered the US. Additionally, we employed a purposeful recruitment method to learn from diverse identities and life experiences. Finally, we used validated approaches to data analysis and interpretation, which strengthened the validity of our thematic analysis. Important limitations to this study include the small sample size and the fact that we sampled participants from only one port of entry, located in the Rio Grande Valley on the Southeast Texas border. This limitation means that the reality of using CBP One may be different at other ports of entry. While qualitative interviews allowed for exploration of in-depth user experiences, it limits generalizability to areas and populations beyond our sample. Furthermore, the CBP One™ interface and implementation are subject to further change, given its ongoing legal challenges, and some features implicated in these study results may not be present in future versions. Notwithstanding, the use of CBP One™ is likely to continue, and our findings hold value for understanding future technology-based immigration policies.

5. Conclusions

In this qualitative study of user experiences and health effects of the CBP One™ mobile application at the Mexico-US border, we found that migrants reported inherent flaws, external factors limiting successful use, and negative physical and psychological health impacts from using the app. Many of these criticisms, though situated within an immigration system affected by multiple structural factors which may exacerbated negative health consequences, were perceived as inherent to using the CBP One™ app by our sample. Working with migrants and other stakeholders to improve CBP One™ while protecting the health of asylum seekers will be imperative to guarantee access to the legal right to asylum. Future studies should pilot more creative approaches to appointment scheduling and quantitatively evaluate the health effects of CBP One™, particularly for mental health disorders such as anxiety and depression.

Funding sources

This work was supported by a Global Individual Grant from the University of Michigan, USA

Ethics approval

This study was approved by the Institutional Review Board at the University of Michigan Medical School (HUM00186322). All participants underwent a process of verbal informed consent prior to participation in the study.

Availability of data and materials

The data used and/or analysed during the current study are available from the corresponding author on reasonable request.

CRediT authorship contribution statement

Christopher W. Reynolds: Writing – review & editing, Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Brendon Tucker: Writing – review & editing, Supervision, Data curation. Samuel Bishop: Writing – review & editing, Visualization, Supervision, Investigation, Data curation. Sarah Draugelis: Writing – review & editing, Methodology, Conceptualization. Michele Heisler: Writing – review & editing, Supervision, Methodology, Investigation, Conceptualization. Amir M. Mohareb: Writing – review & editing, Visualization, Validation, Supervision, Project administration, Methodology, Investigation, Formal analysis, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.