Supporting Academic Primary Care Teams Serving Refugees: A Qualitative Study

RGH | https://doi.org/10.18297/rgh/vol3/iss1/5 1 Abstract Introduction: Primary care providers continue to experience significant challenges when caring for refugee patients, yet they are often refugees’ initial point of contact with the U.S. health care system. The purpose of this qualitative study is to expand our understanding of the experiences of academic primary care team members during clinical encounters with refugee patients.


Introduction
Over 669,000 refugees have resettled throughout the United States from the years 2008 to 2017, with the largest group (150,249) arriving from Myanmar (formally Burma). [1] Forced to flee their own country because of persecution, war, or violence, refugees legally enter the U.S. through a complex and extensive resettlement process. [2] They are likely to arrive with significant preexisting medical problems, with their health often affected by traumatic events, experiences of discrimination, and limited access to public health services. [3][4][5] Primary care teams are mainly responsible for the provision of health care for newly arrived refugees, as they are refugees' initial point of contact with the U.S. health care system. [6] However, primary care providers continue to experience significant challenges when caring for refugee patients, as they must address patients' complex health and social needs in cross-cultural interactions and operate within health systems that may not be structurally organized or politically favorable towards refugees. [7] This racially, culturally, and linguistically diverse group of patients presents unique challenges that impact providers' ability to deliver the same quality of care as to the general population. [7] These challenges are also echoed by resident physicians, as studies have emphasized lack of knowledge regarding refugee health, access to interpreter services, and encounter time as areas for residency programs to focus quality improvement efforts. [8] However, while previous studies elicit, or identify, the key domains where improvements can be focused, how these domains can be operationalized to improve refugee care remains unclear. [7][8][9] Thus, we continue to lack practical implications for exactly how to improve training and assist academic primary care teams in clinical encounters with refugee patients, including what refugee-focused trainings should cover and accomplish. If these skills are not acquired during residency or other periods of trainings, then providers are often left feeling isolated and under resourced as they figure it out on their own. [9] Finally, from the literature currently available, we have not been offered insight to the internalization of values and change in attitude, an introspective and ongoing process, that occurs as primary care teams gain exposure and experience in refugee health care. [10] The purpose of this qualitative study is to expand our understanding of the experiences of academic primary care team members during clinical encounters with refugee patients. We focused specifically on relational capacity building, change in attitudes, acquisition of knowledge, and solutions to barriers in the patient-provider clinical encounter. We hope these findings will be useful to other academic refugee clinics as they identify areas to target quality improvement efforts, as well as provide guidance for the development of formal clinical protocols, trainings, and tools.

Methods
This qualitative study was designed under a Participatory Action Research theoretical framework and methodology, meaning a systematic collection and analysis of data for the purpose of taking action and making change. [11] The project protocol was approved as not human subject research by the Aurora Health Care institutional review board.

Study Setting
While providing care to refugee patients for many years, the three primary care clinic sites involved in this study drastically increased their refugee patient population in 2015, after reaching an agreement with the State of Wisconsin to begin performing government required initial refugee health screenings as part of the resettlement process. Primary care clinical encounters with refugee patients occurred at three separate locations in Milwaukee, Wisconsin: two family medicine residency clinics and an associated community health center. As part of an effort to increase cultural competence, several brief department-wide educational seminars were held over the course of 6 months in 2016 prior to the start of this study. The seminars mainly consisted of medical interpreters, resettled refugees themselves, narrating their own refugee experience and difficult journey to the U.S. All clinic staff were invited to these seminars, with all study participants attending at least one seminar except resident physicians who were not yet part of the residency program.

Study Participants
Study participants (n=10), as described in Table 1, were primary care team members in the department of family medicine, and included three faculty physicians, three resident physicians, a nurse practitioner, a pharmacist, a nurse, and a medical assistant. Study participant selection criteria included having worked with refugee patients for at least one year within the last three years, at one of the two family medicine residency clinics and/or an associated community health center.

Data Collection
Semi-structured, in-person, individual interviews were conducted over a 6-month time interval. The interviews were audio-recorded and later transcribed. The interviews lasted on average 60 minutes and occurred in a private setting. Field notes were recorded during the interview while memoing occurred after each interview session. The development of the interview guide was based on a literature review and expert guidance on clinical experience from two co-principal investigators, both senior faculty members representing either the academic or community setting. The interview guide was revised after half the interviews were conducted in order to provide more in-depth exploration of emerging themes such as the acquisition of knowledge, change in attitudes, and solutions to barriers in the patient-provider clinical encounter.

Data Analysis
Transcripts were openly coded to identify emergent themes and subthemes, while applying an immersion/crystallization and consensus approach to data analysis. [12] This approach entailed repeated cycles of reading transcripts in entirety and selectively coding portions of the data as was relevant to the acquisition of knowledge, change in attitudes, and solutions to barriers in the patient-provider clinical encounter. Clustering of codes and axial coding was then performed to identify codes that cut across all data sets. [12] Codes were verified and discussed with 2nd and 3rd coders, raising additional questions for which to interrogate the data. Upon further review and discussion regarding emergent themes, the coders achieved consensus on the data. The MAXqda software program was utilized to organize and store coded data.

Results
The interview data generated three major themes and corresponding subthemes: Building relationships (over time, earning trust, role of culture in the relationship); Markers of success in clinical encounters (improving communication, adaptation); Knowledge of or lack of clinical resources (focused trainings, formal debriefs, access to resources, unwritten languages). Major themes and subthemes were either spontaneously mentioned or verified throughout the interview process, reaching saturation across all study participants. Table 2 provides examples from each major domain in the form of direct quotations from study participant responses.

Building Relationships
Over Time "I think you just got to sit together for a while. And it is time, just time, and being present in the room… It just takes a long time. I can't emphasize that enough, the amount of time spent just sitting with people." "Within the encounter, time is a huge factor. I don't think our health care system recognizes that interpretation takes additional time. It's already too condensed without language barriers. To address the complex needs of these populations, you need a little extra time built in. That is a system issue, but we need more time in that time frame than we have with them." "I mean it [limited clinical encounter time] is either a problem all the time or not, but not especially with refugees." Earning Trust "The other one that is tough to fake, is just time, and having a calm presence and not seeming like you want to get somebody out of the room. It works to ask people what other questions they have and making sure they know that we are there for them in a time of need if they need us. Those types of things not everyone does well through interpreters, so if you make an extra effort with clarifying, and reclarifying, and doing the teach-backs… the types of things that you are taught in medical school as good adherence work for medications. It works the same for a follow-up visit as for taking a new medicine, or going to get a new test, or new procedure done. 'Hey, do you have any questions? or, What else can I do for you?' And, actually meaning it and taking the time to actually address it. It builds up trust. That is more difficult across cultural or language barriers because we make up a lot of assumptions to our white English-speaking patients that we otherwise wouldn't if it's across a language barrier, but it is just as important for both groups. It ends up having dividends down the road, whether you realize it or not."

Role of Culture
"We had a situation where we had a baby that had just an astronomically high lead level. And they did it repeatedly, it was like in the 30s. And we couldn't figure out why… and then somebody looked it up, and it was a culture that uses kohl… K-O-H-L… that culture, And I don't even remember what the culture was now… but they apply it after the baby is born in a belief that it wards off evil spirits… well the kohl is loaded… and they do that for a period of time… like the first month or so of the baby's life, they keep reapplying it around the eyes. And the kohl is just loaded with lead. So we had no idea to even ask that question… is this something you did for your child? And they had no reason to say, oh this is what I have done with my child, because everybody does that… I think there are things that we don't even know we are missing, and that aren't being brought up to us because we don't have enough knowledge about the culture. Now, we look at that right away if they are coming from… it's like oh, this baby has a high lead level, I wonder if it is one of the cultures where they use kohl, and then we can ask that question." "I'll be honest, at first, after my first experience, my gut reaction was 'oh my goodness, this is going to take forever, I'm not going to know what to say…' it became a very exhausting ordeal. Now, I have a different perspective or attitude because I have realized, especially with those who have been in our system long enough… this is among the most grateful of patient populations, and it is very fulfilling to me as a provider." "We all make assumptions, so I try to overcome that by at least recognizing my own biases… in how I would care for somebody. And I try to overcome that in a variety of ways… asking for more questions, more teach-backs." Table 2. Themes, Subthemes, and Supporting Quotations.

RGH
Supporting Academic Primary Care Teams Serving Refugees: A Qualitative Study

Markers of Success in Clinical Encounters
Improving Communication "I think the biggest things, again, are can you put a smile on their face? Can you tell those subtle, nonverbal clues that say we are connecting across all the differences we have? That, and are we getting to the point that they… that we have that level of trust, that we have that level of openness and communication, that they're comfortable, that they can sit there in that chair and they feel comfortable with me as the healthcare provider." "I think that not knowing if they understand the plan when they are leaving… and maybe that is an interpreter issue. I had one patient that took their latent TB medicine wrong. We had to start over with the treatment, because it was a treatment failure. I felt like, wow, I was really clear. And he was a young, smart guy, who thought he had it. And somehow there was no communication happening there… I had the illusion that I was communicating, but I was not communicating. And that is really frustrating, but maybe that is an interpreter question." "But I think I would define success more as them understanding their health risks. Like if your blood pressure is not at goal, I want you to understand this puts you at high risk that you probably want to avoid. If they can internalize that and not just feel like I am trying to get to this number… and the rest of it is a mystery. To really have them be able to say, I want to be well in the future, and that's why I want to get this blood pressure down." "When patients are able to say back to me what they understand is going on and what the follow up plan is, and we are all on the same page without me prompting… that is a successful moment." "Well, that questions are answered… if they have questions, that they feel free to ask them. And that questions are answered with an interpreter, so that when they walk out of the office, they understand what they are supposed to do… you know labs, where they need to go, where they need to get their medication, how they should take it, and hopefully, if they are not comfortable with something, that that has been brought up… so that concerns that they have, they feel free to have brought those concerns up and we are able to address them in a way that meets their needs…. And they don't walk out of here, nodding their heads, saying Ok, ok… and they have no idea where they are supposed to go or where they get these pills, or how they are supposed to take them, that should all be taken care of during the visit because we have an interpreter… so that is the time to… and that they feel comfortable even after they leave here, if they are not sure, they can call us. I guess that questions are answers, and that they feel respected enough that they can ask the questions, if have them, and not feel like I've already asked 3 questions, now they are going to think I'm stupid if I ask another one. So that we meet their needs." "Communication is not just words, so trying to understand our body language too. Body language is different. We subconsciously read peoples body language. I think it is sometimes hard to read body language of others, who grew in a place where body language is different." "Communication is the ultimate, important thing in medicine and our patient encounters. So regardless of whether you are working with refugees or not, communication… these are lessons that translate to other encounters." Adaptation "Your definition of success changes depending on where they are in your relationship. I think that any success is just the act of being there, together. Any encounter with that person… little successes along the way are when you try something, you are able to connect in some way and then there is positive result that comes out of it. I think especially with refugee patients, the longer you can have a relationship, the better you can meet their health needs and desires… because you are able to... like with anybody, the more you meet the better you can understand that person…. And where they are coming from… and you can figure out how you can best use the system to help them." "I have found myself in the position multiple times of calling, following up electronically as best I can as well, and then calling to advocate again." "I have to think carefully about all the steps in my patient's care, that I don't have to explain to people who have grown up in my system. I think about more carefully in terms of who I'm going to have them see as a specialist or another primary care provider while I'm gone… when I'm referring them for a cardiac stress test, or physical therapy. Just because of the barriers and the locations that they are going to have to navigate…. And where interpreters are going to be available. I know on St. Luke's campus they will for sure be available, so if they see a specialist there versus at a private clinic… they may not have the same access to interpreters or know how to call up the interpreters. I think about that a little bit differently… but I'm always thinking about that, to some degree..."

Knowledge of or Lack of Clinical Resources
Focused Trainings "It's good knowledge to have, and it has changed how people view the refugees when they come in, to know a little more about what they have gone through in their own country when they have gotten here." "Even before my first time, that I worked the initial refugee visit, I had one of the other residents beforehand, just kind of… just sat me down and he gave me 10 minutes' worth of information and helpful hints to get through the visit, but otherwise I didn't have any sort of training to go into that visit. It was kind of following the template and figuring out as I went. Even training people for those initial visits would be helpful." "we did have [training] sessions originally for all of the residents when we started with refugee health screenings… And then we haven't revisited that as much yet. We were kind of going in the direction of doing more education for people on an individual basis with a power point and discussions when they have their first refugee health screening clinic." "There is a mental health training happening in Milwaukee with a second day dedicated for training for refugee patients… the registration notice came to be a month beforehand, and I already have days full of patients and I can't cancel on everyone in order to attend this. There are these events that happen, but often we are not able to attend because of conflicts in schedules." "it's very hard for us to understand. So, we often don't go there because we don't know how to ask and if we do ask, how are we going to address it. From mental health especially… but in many ways it also affects physical health, in terms of physical torture… we are not really sure what to do. There's a gap. And it's tough."

Knowledge of or Lack of Clinical Resources
Formal Debriefs "I think having opportunities for periodic… almost M&M type, discussions, where you debrief on challenging cases…With the focus on refugee care… then having critical players, interdisciplinary players involved. So, you would have nursing, MA, case worker, or social worker, pharmacy, and the physician… so an interdisciplinary case review or case conference. And also, psychologists or psychiatrists… cross cutting fields, for debriefing and problem solving." Access to Resources "I feel like I underutilize what is available to me in EPIC, and through the library website from EPIC, that takes you to all those… there is a ton available. It's just a matter of organizing the information where I can access it quickly." Unwritten Languages "one of the things that I learned early on, is that Rohingya does not have a written language, it only has a verbal language. So even when I had an in-person interpreter… I'm like 'Oh, this is great! I get someone in person, and I don't need to get anybody on the phone… so let me type some of these couple key things and then you can write it down in Rohingya form, and they can take it home.' The Interpreter is like, 'I can't write that down.' 'What do you mean you can't write it down? Don't you know how to write?' 'No, Rohingya doesn't have a written language.' 'OHHH, well I guess that would make it hard for you to write it down.' So, that can be huge too…one, there is literacy in general, and then health literacy, but if you're dealing with a language that just simply doesn't have a written version, now all of the stuff that you're trying to communicate, honestly has to be communicated orally, but then needs to be retained in some way. So, obviously, we do a lot of teach back, repeat back, all of those kinds of things, but I can't imagine if that were me in that situation… If I went to this visit, and got all this stuff, and went to the pharmacy and got these bottles, and I'm relying two weeks down the road on my memory of all those things to make sure that I don't get the this wrong. So, I think that's a very significant challenge." "Handouts, that are essentially pictured-based handouts, rather than text-based handouts… so you know, that generic… Aurora handout that talks about low blood sugar… not that it's a bad handout… it's at a 5th grade reading level, or whatever, but it's still all text, which isn't going to work for my patient that doesn't read English, or whatever. The same kind of handout that has… and they are kind of cheesy, funny looking photos of this generic 'cartoonized' person with all the different symptoms of low blood sugar… so what does it look like to be sweating? What is it look like to be dizzy or shaky? What is a look like to be irritable? Honestly, those things are in text underneath, but again if you are dealing with a population that isn't going to be able to read those single words… at least you can review that that handout again with an interpreter, and go back and forth, and hopefully the pictures are simple enough. Pictures are worth a thousand words. The patient can take that home and recall, and even if you can't have any kind of a printed Rohingya handout because there is no such thing as a printed language."

Building Relationships
All participants found it important to build relationships with their refugee patients. From the data, it emerged that there were three significant ways that they reported that these relationships shifted or evolved: over time, with earning trust, and through a changing internal attitude towards culture's role in the patient-provider relationship.
All providers referred to the need for this relationship to build over time, whether over the span of a single encounter or years of experience gained by working with a refugee patient population. Over time, providers felt they eventually could earn trust, frequently commenting on patients becoming comfortable enough to disclose sensitive personal information about mental health, trauma, and torture. Trainee physicians particularly emphasized limited clinical encounter time as a major barrier to building relationships with refugee patients, more so than other patients. In comparison, more experienced faculty physicians generally acknowledged encounter time constraints as a barrier they faced with all patients but did not consider the refugee population as distinct.
Furthermore, all study participants remarked on the critical role of culture within the relationship. Culture was discussed as part of recognizing differences crucial to refugee health. For example, knowing to ask about kohl as a source of lead poisoning or cessation counseling for chewing betel nut to prevent oral cancer. Participants also recognized the importance of checking one's own prejudice and bias while being aware of power imbalance and being sensitive or appropriate in order not to offend patients and be respectful of their cultural norms, traditions and customs. Developing a better understanding of the refugee experience and difficult journey to the U.S enhanced empathy and patience. Participants saw working with refugee patients as a learning experience that provided opportunity for personal growth and further learning.

Markers of Success in Clinical Encounters
Providers were asked to explore what makes clinical encounters with refugee patient populations successful. From the data, it surfaced that providers have primary goals for clinical encounters other than traditional numeric health outcomes. Analyses also revealed that providers adapt their clinical approach in various ways to help refugee patients be successful.
Across all provider role types and levels of experience, providers' perceptions of success in clinical encounters with

RGH
Supporting Academic Primary Care Teams Serving Refugees: A Qualitative Study refugee patients were primarily associated with varying themes of improving communication, as opposed to numeric patient health metrics such as blood pressure measurements to guide hypertension management or hemoglobin A1c levels to assess diabetes control. For example, improved communication included effective language interpretation; awareness of nonverbal cues; ability to ask and answer questions; and patient understanding of medical condition, medication/treatment regimens, and the U.S. healthcare system. Many respondents also alluded to adapting their expectations of success for refugee compared to non-refugee patients and discussed how those expectations change as the provider-patient relationship evolves. Moreover, providers commented on how they adapted their approach to refugee patients in order to achieve success, such as increasing the frequency of clinical visits by shortening follow-up time intervals as a method to maintain oversight of refugee patients. Other adaptations included increased hands on coordination of care through personal referrals to a specific specialist and calling ahead to the emergency room with more detailed provider hand-off notes.

Knowledge of or Lack of Clinical Resources
Study respondents identified barriers to current clinical resources and potential areas that would benefit from the development of additional clinical resources. From the data, several ways to support providers in clinical encounters with refugee patients emerged: focused trainings, formal debriefs, access to resources, and additional support with unwritten languages.
All study participants had previously attended at least one refugee-related seminar. They commented on the experience as a positive one, mentioning their own improved empathy and patience toward refugee patients, with some even noticing changes in their colleagues. However, one study respondent did express concern that the improved sense of empathy and patience throughout the clinic may be only temporary, and that over time, people would likely revert to prior attitudes and behaviors.
Resident physicians, especially, recalled feeling overwhelmed and underprepared for their initial refugee patient encounter, with some commenting on learning from their experiences and some asking for more structured and standardized trainings. Attending physicians also commented on the difficulty in accessing additional trainings, largely due to busy schedules. Finally, all providers mentioned their own discomfort and lack of expertise surrounding mental health, torture, and trauma of refugee patients, highlighting the necessity for more guidance in this area.
Resident physicians in training perceived that formalized debriefs with their colleagues or attending physicians were a useful method to reflect on and improve from clinical encounters with refugee patients. Comparatively, more experienced providers tended to comment more about the utility of debriefing with providers across disciplines, such as with interpreters who were recognized for their role as cultural brokers in general.
Medical residents were more unaware of specific clinical resources to use with refugee patients compared to more experienced providers, who all spoke of their own database of favorite resources for refugees they had personally gathered over time. Resident physicians generally noted that they underutilized available resources due to inefficient accessibility given the time constraints of the clinical encounter. They acknowledged that a centralized database of refugee-related resources would be useful. However, one more experienced respondent did comment on the possible negative impact of a large amount of available resources, as it can also be overwhelming and inefficient to sift through if dealing with time constraints. Lastly, given that a large portion of newly arrived refugees in the Milwaukee area primarily spoke Rohingya, barriers from unwritten languages frequently appeared in all study participants' responses as an area clearly lacking clinical resources.

Discussion
We found that within the clinical encounter, there were three key aspects to building a successful relationship between a provider and refugee patient. These aspects were: 1) that these relationships were built over time 2) the critical role of earning trust and 3) developing an internal attitude commonly referred to in the literature as 'cultural humility.' [13][14] Additionally, providers' perceptions of success in clinical encounters with refugee patients were primarily associated with varying themes of improved communication. These results reveal important considerations for those developing focused refugee clinical protocols, trainings and resources as explained further below.
Our study found that providers measured success according to improved communication over time rather than by other quantitative patient health outcomes. Existing literature has shown that good provider-patient communication is directly linked to higher patient satisfaction, adherence, and subsequently improved health outcomes. [15][16][17] Therefore, our findings suggest that training in communication and providing appropriate resources could address a critical upstream determinant and ultimately improve refugee health. Future development of refugee-related trainings and

RGH
Supporting Academic Primary Care Teams Serving Refugees: A Qualitative Study resources should focus on supporting providers in their ability to communicate with patients despite cultural and linguistic barriers, including communication surrounding mental health, trauma, and torture. [18][19] While a lack of knowledge regarding refugee health, access to interpreter services, and encounter time as barriers to success have already been emphasized in the literature, our study found that navigating an unwritten language adds an additional obstacle to the clinical encounter. [8] A large portion of newly arrived refugees in the Milwaukee area primarily speak unwritten languages, such as Rohingya. While this theme may seem familiar as it echoes many of the barriers faced by providers who treated newly arrived Hmong refugee populations in the 1980s-90s, our current booming visual/technologic era can offer innovative solutions and resources to address barriers from unwritten languages. [20][21][22] Examples listed by providers in our study to address this barrier include using picture-only educational materials, showing videos on the internet, and even creating new health education videos for patients to watch while waiting to be seen.
Our findings also suggest the potential for interesting applications that provide detail to the existing call in the literature for additional cross-cultural training. [8,23,24,25] The multiple concepts that encompass cultural humility, which were found to be important in our study, provide an opportunity to develop customized trainings that meet providers' varying backgrounds and experiences. Interest in flexible yet tailored training approaches is growing under the rubric of time-variable learning. Adoption of this approach would offer the space for continued growth, regardless of provider role, point in training, and years of experience working with refugee patient populations. For example, residents who have not had prior exposure to a culture's beliefs may be assigned a training that describes these, whereas another resident already familiar with the culture might be assigned a different communication module that addresses the barrier of unwritten language.
There were several limitations to this study. The small study sample size, especially among participants who were not a resident or attending physician, may inadequately represent non-physician experiences working with refugee patients as part of the primary care team. Study participants were selected based on availability and the inclusion criteria of working with refugee patients for one year or longer. An element of response bias should be considered, as study participants may have been more eager to discuss their experiences with refugee patient populations compared to study non-participants. Of note, providers working in these selected refugee clinics also have extensive clinical experience working with other underserved patient populations and thus may not reflect the spectrum of training needs of those not in such an environment. Lastly, the scope of the study was limited to the clinical encounter, meaning solely the time spent with the provider and the patient together during a clinic appointment. The study only reflects the one-sided perspective of the primary care team provider. While the perspectives of medical interpreters and case managers are presented in a different study, future work should consider including refugee patient experiences and perspectives of what constitutes a successful clinical encounter. [26] In conclusion, our findings revealed that providers' perceptions of success in clinical encounters with refugee patients were primarily associated with varying themes of improving communication, as opposed to mainly numeric patient health metrics. We advise that the development of any formalized trainings, tools, and resources to support primary care providers working with refugee patient populations should take these goals of success into consideration. Formalized refugee health trainings should be tailored to the level of provider experience. While targeted didactics that cover health topics that disproportionately and specifically affect refugee patient populations will help resident physicians improve their clinical knowledge, this confines refugee health training to a cultural competency endpoint. [8] By incorporating formalized debriefs to these trainings as part of a cultural humility curriculum, providers will be able to reflect on and improve from their clinical experiences. [27] However, more extensive research is needed on the feasibility of translating these findings into a formal curriculum that is evaluated. Lastly, potential solutions were identified to address existing barriers for primary care providers in clinical encounters with refugee patients, particularly in response to barriers from unwritten languages and inefficient accessibility of clinic resources. However, further research and development is necessary to assess efficacy and utility of these proposed interventions.