In medical encounters, there are at least two cultures, two voices, in the two languages being expressed. The aim of this paper is to explore cross-cultural communication between refugees, specifically Vietnamese refugees who have lived in the United States for more than 20 years, and health practitioners who assist refugees. After approximately 50 hours of participant observation at the Dallas County Health Services Refugee Clinic and six semi-structured interviews, I analyzed reoccurring themes that were associated between communication and its effect on patient’s health outcome. Reoccurring themes include substitution, omission, editorialization with the use of an interpreter, and nonverbal communication expressed by both populations. Communication was negotiated between both parties and nonverbal communication has potential to be beneficial in fostering effective communication.
Table of Contents:
…Another funny story about blood-pressure medicine. [Patient] was told take one pill every day. “Come back in a month and we [clinic staff] check your blood pressure.” [He] come back, [his blood pressure is] still high. “You taking medicine?” He’d agree. “You take one pill a day?” He’d agree. Come back one week and it is still high. So we [clinic staff] made a home visit and there were thirty pills laying on the table. The interpreter said, “Take one pill every day,” but didn’t specify put it in the mouth. (Health practitioner interviewee discussing her experiences with cross-cultural communication, July 2010).
Doctor-patient relationships are already difficult even when both parties speak English, but add cross cultural differences and language barriers, and the situation becomes more complex (Schouten & Meeuwesen, 2006). In the passage above, knowledge differences cause a difficulty between members of different cultures and have the potential to lead to cross-culture miscommunication (Mahi, 1993).
Communication between health practitioners and patients is attracting an increasing amount of attention, yet it continues to be limited (Ong, De Haes, Hoos, & Lammes, 1995) A number of factors hinder such services, including an increase in the number of languages spoken, costs associated with providing such services, lack of knowledge on the part of heath care providers of legal requirements for providing language services, and lax enforcement of federal and state laws, which has allowed many health care providers to neglect the issue (Youdelman & Perkins, 2002).
It was not until 1973 Rehabilitation Act, combined with the 1964 Office of Civil Rights Act that federal agencies mandated that health practitioners have interpreters available when patients cannot communicate their concerns (Abbe, 2010). Currently, some states’ laws addressing language access in health care provide detailed guidance, while others note the importance of language access but are unclear about their legal obligations to provide language services (Chen, Youdelman, & Brooks, 2007).
The purpose of this study is to explore the experience of cross-cultural communication among health practitioners and refugees. More specifically this study seeks to explore how communication between these two populations affects patient health outcomes and how the patients’ experiences with medical care have changed over time. The participants are Vietnamese refugees who have lived in the United States for more than 20 years and currently reside in Dallas, Texas and health practitioners who provide medical care to refugees.
Health practitioners in the United States are frequently confronted by different cultural background because they live in a multicultural society (Schouten & Meeuwesen, 2006). Scholars of medical communication argue that effective communication is necessary for any successful outcome between health practitioners and patients (Ong, et al., 1995). Too frequently, however, cultural and linguistic barriers lead to miscommunications, which have the potential to affect patient trust, understanding, diagnosis, and the treatment plan (Lee, 1997). Pauwels (1995) explains that linguistic differences can affect the communication that takes place in an intercultural setting; for example, between health providers and patients from different language or cultural backgrounds. These communication differences are caused not only by differences in the structures of language but also by socio-linguistic rules that govern how we speak. For example, differences include the way information is given, jokes are told, who has the right to speak, and so forth (Hymes, 1972; Pauwels, 1995).
According to the 2000 census, approximately 18% of U.S. citizens speak languages other than English. Nearly one-third of Texas’s residents are qualified as limited English proficiency (Green, et al., 2005; U.S. Census, 2000). Interpreter services are instated because they are more likely to prevent adverse outcomes, such as less screening and immunizations, treatment errors, non-compliance with medication or treatment plans, missed appointments and patient dissatisfaction (Chang, Feller, & Nimmagadda, 2009). Evelyn Lee explains that interpreters are expected to act as the messenger and provide a line-by-line translation (Lee, 1997). One health practitioner explains her experiences when working with interpreters.
Health Practitioner Interviewee: “Uh…sometimes they stumble over words just like everybody else. Is the translation appropriate?”
Although this model of interpreting is ideal in theory, interpreters may discover that not all words have a direct one-to-one translation in meaning and connotation (Flores, et al., 2003). In the following passage, my hired interpreter struggled as she tried to translate the word “expectation” and had difficulty making a direct translation.
Interviewer: “…When seeing a western doctor…um…you think your expectations are met?”
Interpreter: “What’s expectation in Vietnamese? Uh…damn it.”
Interviewer: “Sorry, I can change the word for you.”
Interpreter: “No, no that’s alright. I suppose to know that word.” (As she interprets to the refugee interviewee, the word expectation is still said in English).
Furthermore, recent models highlight the need for interpreters to act as cultural liaisons or culture brokers, having the necessary cultural as well as linguistic and medical training (Lee, 1997).
Lack of the Ideal Interpreter
Health Practitioner Interviewee:”Yeah you just go by (points to the waiting room) over there on this side, I will say Arabic, Arabic, anybody? …So you just find people. You do the best you can….” (July 2010)
The participant’s comment above demonstrates her reliance on anybody that can translate. Other research finds that health facilities that do not employ formal interpreters sometimes reply on ad hoc interpreters, or untrained individuals, such as family members, friends of family, and non-clinical employees, such as people in waiting rooms (Flores, et al., 2003). The use of informal or ad hoc interpreters, coupled with clinician lack of training on how to use interpreters, has been found to correlate with errors in interpretation (Flores, et al., 2003).
Four commonly known interpreter errors often observed in a clinical setting are omission, addition, substitution, and editorialization (Flores, et al., 2003). Interpreters who leave out a word when translating between a health practitioner and patient would be omitting a word or phrase. Opposite of omission, addition is when interpreters add a word or phrase that was not expressed between health practitioner and patient. Substitution, as I mentioned before, is used when words cannot be crossed from one language to another without changing the meaning or the word does not exist in the other language. Finally, editorialization, similar to substitution, is used when interpreters provide their own perspective (Flores, et al., 2003).
Effective communication between health practitioners and patients is not just verbally expressed (Ong, et al., 1995). Ong contends that non-verbal communication leaks unintentional messages, and is characterized as tone of voice, gaze, posture, laughter, facial expressions, touch, and physical distance (Ong, et al., 1995). Used positively, health practitioners could portray characteristics that are encouraging, relaxed, and friendly with smiles, gentle touches, and eye contact; while remaining aware that these gestures differ across cultures (Perez & Luquis, 2008).
In the Heart of Refugee Health
At the Dallas County Health Services Refugee Clinic, multiple languages always flowed from the surrounding complexes and it was not uncommon to see people of all ethnicities chatting and hanging around whatever shade was provided from the few trees across the clinic’s parking lot. It was unbearable heat to stand in during the summer, and I assume that was why most patients carried umbrellas or fanned themselves as they entered the clinic. The waiting room seemed fairly spacious, but most of the time it was overly crowded by large families, playful children, and sometimes interpreters accompanying their patients. With foldable chairs as the seating arrangement, they were easily moved and added anywhere within the building when needed. Tables were used as working stations for most health practitioners, except for three who had cubicles located in the room furthest to the back of the clinic. Most patients would not enter the back room unless they were being interviewed about their medical history, and they frequently saw the nurse practitioner unless they were receiving a physical examination. It was, however, frequent that the patients would get vaccine or immunization either in the waiting room or hallway of the clinic.
The Dallas County Health Services Refugee Clinic has moved from a cramped two bedroom apartment to a reconstructed office space located on the first floor of the Sunchase Square apartment complex (“Clinic for Refugees New, Bigger,” 2009). According to the staff, the clinic is the heart of the refugee community and it certainly appears that way. The building is surrounded by multiple apartment complexes that refugee resettlement organizations, like the Refugee Services of Texas, work with regularly to place refugees.
This North Dallas clinic administers tuberculosis and HIV/STD screenings, physical and parasite examinations, and vaccines, which are essential to entering school and the workforce (“Clinic for Refugees New, Bigger,” 2009). Clinic staff and volunteer agencies work together to medically assist patients on Mondays, Tuesdays, and Thursdays from 6:30am-4:00pm and help fill-out and sign completed Green Cards and I693 Medical Examination and Vaccination Records on Mondays and Tuesday from 1:00pm-3:00pm. The clinic has 11 staff members and, depending on the day, most of them are constantly moving about collecting and exchanging information. Waiting room chairs were usually occupied by refugees, especially in the morning and there are occasional lines made by the patients, which flow into the clinic next door. Appointments are routinely made by refugee agencies, like the Catholic Charities of Dallas and International Rescue Community. These agencies ensure medical attention to newly arrived refugees on Mondays and/or Tuesdays and assist returning patients on Thursdays.
The increase of Vietnamese refugees in the United States is directly associated with the end of the Vietnam War in 1975 (Ito, 1999). The first groups of Vietnamese refugees arrived in 1975, when South Vietnam fell into communist control to the North. These first wave of refugees were generally professionals, businessmen, and military officers; the second group arrived in the late 1970s and were commonly merchants and farmers displeased with the new political party (Purnell, 2008). By 1979, the Orderly Departure Program allowed Vietnamese people a legal exit to reunite with family in the United States, and most recently, the 1987 Amerasian Homecoming Act allowed entry to former South Vietnamese military officers who served with American servicemen and descendants conceived by U.S. servicemen and Vietnamese women (Purnell, 2008; Do, 1999). By 1990, Texas contained the second largest population of Vietnamese refugees, approximately 67,000 settlers (U.S. Census, 1990). According to the 2000 census, Texas still possesses the second largest amount of Vietnamese refugees, approximately 135,000, particularly in the metropolitan areas of Houston and Dallas. Many Vietnamese participants referred to all four waves of Vietnamese refugees as the old or older refugees, indicating that they have lived in the United States for a reasonable extent of time, often more than 15 years. The majority of health practitioners defined an older refugee as a person that surpassed their one year mark of health care services with the clinic and others associated their status with getting a Green Card or citizenship. In this research, the terms old and older refugee will remain broad and be defined by the Vietnamese refugees’ and health practitioners’ perspectives.
The Health Practitioner
The designation “health practitioner” is used loosely in this paper and classifies anyone who provides medical assistance in a clinical setting. The most frequent health practitioners are nurses, such as a nurse practitioner, a registered nurse, licensed vocational nurses, as well as outreach workers and a phlebotomist. Each practitioner bears some responsibility for medical treatment, examinations, medical reviews known as interviews, or a combination of these tasks depending on the tasks required for the day.
It is vital to note that nursing professions establish formal education programs and standards of practice that bridge biomedicine and alternative healing (Parfitt, 1998). Barbara Parfitt furthers the idea of learning various medical practices by instructing nurses to “do the cultural thing” and “walk in their shoes.” Parfitt’s instruction implies that nurses not only need to learn other cultures’ medical practices, but embrace a culture in its entirety (Parfitt, 1998, p. 129).
The research reported here is based on fieldwork conducted in the Dallas County Health Services Refugee Clinic. Qualitative methods involved approximately 50 hours of participant-observation of the clinic’s waiting and exam room, observation of approximately 10 medical encounters of practitioner and patient interactions, and ethnographic interviews with clinic staff and Vietnamese refugees. In total, six participants completed the semi-structured interview. Five were conducted in person and one through email. Three of the participants were female health practitioners who worked or volunteered at the Dallas County Health Services Refugee Clinic. The one online interview was conducted from a male employee that worked for a clinic other than the Dallas County Health Services Refugee Clinic, and the two refugees were either recruited from the clinic or the Dallas community. Four out of five of the in-person interviews were conducted in English and one was conducted with the use of a Vietnamese interpreter. Each interview was audio-recorded for later transcription and then coded based on reoccurring patterns. During qualitative data analysis there were various times that patterns of nonverbal communication overlapped and are later distinguished in my findings. The following findings explore the interviewees’ perspectives and actions involving cross-culture communication.
“Given the nature and importance of healthcare services, healthcare providers have a special obligation to ensure language access for their patients” (Chen, Youdelman, & Brooks, 2007; 363).
The Need for Formal Interpreters
While there is a phone-line service available for phone-translation in the Dallas County Health Services Refugee Clinic, the majority of the practitioners agree that it is a difficult service at times. For example, a registered nurse shares her experience.
“…Every place is responsible to have languages available for people that want that preferred language. The way we handle it is…finding somebody that speaks the same language, or, we do have a language line but it never fails that when you call that language line, on the day your client is here they don’t have anybody available that speaks the language.” (Personal interview, July 2010)
Ad hoc interpreters, often friends, family members, and occasional strangers in the waiting room are used as interpreters within the clinic. Often, the ad hoc interpreters would be needed for each procedure or task of the patient unless an outreach worker who spoke the language was available to assist. During my observation, as hoc interpreters substituted, omitted, and edited information during medical interviews or examinations. In one scenario, a nurse practitioner was informing a patient that they need to arrange a dentist appointment. The interpreter translated the practitioner’s statement and the patient responded back in sentences to the interpreter. However, when the interpreter translated for the patient, he only replied with “okay.” The interpreter stated a single word that clarified the patient’s understanding, but nothing else about their conversation.
During my interview with a Vietnamese refugee, my interpreter would often participate in a conversation in Vietnamese; this could go on for two minutes, but the interpreter would reply for the research participant in one or two sentences. Knowing much was omitted, I wondered if this continued because my interpreter felt she only needed to express what she found relevant to the interviewer and interviewee. Most comments that were made following my interview questions were intended to give a sense of understanding or familiarity with the participants; this was not expressed through omission.
Editoralization also occurred in the clinic. In the following scenario, a registered nurse asked about a patient’s past health record during a medical interview. As she communicated through the interpreter, the interpreter answered the question for the patient without translating. As the interpreter was a close friend or family member, it may be that speaking for the patient is culturally acceptable.
Many of these editorial interviews were caused by interpreter assumptions about the patients. But often times, it was not just the interpreters that made assumptions, but also the practitioners that overlooked the patients’ insight and assumed the appropriate action without further discussion. Yet, when asked about shared decision making between practitioners and patients, the majority of my interviewees from both populations felt that there was a shared decision making while discussing health care, even when there was a language barrier or only feeling partially understood.
A minority of health practitioners who expressed that there was little to no shared decision making between practitioners and patients were also the practitioners that assumed their patient’s knowledge of healthcare and treatments, which often lead to miscommunication. Much like the earlier case between the outreach worker and her client, what may be common language usage in a Western healthcare setting, such as “take one pill a day,” may not necessarily translate literally as the same things to a refugee who has never taken pills. This kind of situation was commonly categorized as miscommunication or lack of communication by both populations and further expressed as a factor of noncompliance among the health practitioners.
Health Practitioner Interviewee: “’Shot-baby, no baby.’ They think I give them shots, three months they won’t get pregnant.” (Personal interview, July 2010)
This outreach worker explains that she tries to inform the refugee patients not to get pregnant in the next three months because of the shot. Noncompliant patients, according to the practitioners, were classified as patients that did not return for follow-up appointments and/or follow instructions on taking medication. Less than half of health practitioners felt that this noncompliance had a direct correlation to miscommunication. When the refugee participants were questioned, one stated that if she did not like the medicine or think the medication was for her, then she would not take it. This participant’s “noncompliance” stemmed from knowing what was best for her health.
Nonverbal Communication in Action
During observation in the clinic, a registered nurse stepped out from the hallway and entered the waiting room. She called out a name and when the patient arrived, the nurse put her right hand over her left forearm and said, “Una mas.” The practitioner and patients exchanged laughs and few words in English and headed into another room of the clinic.
Symbols are commonly used as a form nonverbal of communication within the clinic between patients or practitioners. Unlike other nonverbal behaviors suggested by Ong, et al., (1995), symbols are intentional messages that I specifically categorized by hand and arm gestures and usually followed by or associated with a small phrase, such as “Two, two shots.” One health practitioner begins to talk about symbols, which she refers to as sign language and often expressed with her hands.
Health Practitioner Interviewee: “Yeah, a lot of sign language.”
Health Practitioner Interviewee: “I’ll go….” [She lifts her right index and middle finger into the air and places them on her left shoulder; she repeatedly touches her left shoulder, and says,] “Two, two shots.” (Personal interview, July 2010)
Based on observation it appears that these symbols were applied by the practitioners more than the patients; however, it was a common way that patients would also express their voice. A majority of the staff within the clinic commented that they had used symbols at some point in the profession within the clinic, and I found myself using arm gestures and small phrases when I was instructed by a nurse practitioner to send a patient from one room to another. This appears to be a common way to communicate cross-culturally when verbal interaction is not fully understood.
During an interview with a Vietnamese refugee, I asked a question about my interviewee’s health, but directed my posture to the interpreter. I repeated this pattern several times until I remembered that I am asking these questions to my participant; gradually I started to acknowledge her.
Approximately all interaction involving health practitioners and patients acknowledged each other’s presence through eye contact or by body positioning when their interaction was conducted without the use of an interpreter. However, when interpreters were introduced to the interaction, most times, both populations would shift their attention from each other and focus on the interpreter. Often, individual acknowledgement from both populations was absent in the presence of interpreters and correlated with increased conversation with interpreters instead of each other. Nonetheless, the minority of practitioners that would continue to make eye contact or position their bodies towards their patients, even with the use of an interpreter, seemed to have increased responsiveness from their patients (Mehrabian, 1971).
Laughter within the clinic sometimes silenced the sound of children’s cries. Whether the laughter was heard coming from two practitioners, two patients, or from patient and practitioner discussion, it was undoubtedly present. Often when laughter was shared between patient and practitioner, it followed after someone attempted to speak the other’s native language. The health practitioners would commonly pronounce a name roughly four times until a patient would approach them and sometimes correct their pronunciation of the name; this regularly created laughter between both parties. The minority of practitioners would attempt to say a few words in their patients’ native language even when interpreters were present; most of the time, the phrase was not comprehended by the interpreters or patients and often created laughter instead of conversations. I found that my liaison and I would laugh after my several attempts to pronounce Vietnamese desserts properly and vice versa when she would pronounce American slang. During an interview with a Vietnamese refugee, the interviewee’s husband would laugh when I mistakenly pronounced Vietnamese words. Much like my own experience, I witnessed practitioners’ and patients’ responses and engagement increase through attempted communication and effort to teach each other correct pronunciations from their native language.
“While only 7% of the emotional communication is conveyed verbally; 22% is transferred by voice tone” (Ong, et al., 1995). Considering cultural influence while examining vocal qualifiers, such as pitch, volume, tempo, and tone (Samovar & Porter, 1991), I documented two commonly distinct voice tones projected from the health practitioners when observing the clinic. The first voice tone was sharp in tempo, strict in projection, and often followed with demands or questions. The second tone would also be associated with demands or questions but was usually high-pitched, soft, and accompanied with a smile. The latter often created longer attempts of communication from the refugee population during practitioner and patient interactions.
The refugee interviewees agreed that cultural and language recognition within the Western healthcare has improved since they arrived more than 20 years ago. And while there continues to be interpreter barriers in the healthcare setting, there are current factors that could foster a more gradual progression to effective communication. Communication is negotiated among both of the populations represented in this research, whether it was witnessed through sign language, mutual gaze, or laughter. Mehrabian (1971) states that characteristic differences in levels of self disclosure are revealed by the way people handle their physical environment, and differ depending on culture. If these characteristics are applied to a communication from the clinical and hospital settings while remaining culturally sensitive, I contend that these characteristics have the potential to increase effective communication, especially in situations where there are cultural and language communication barriers. Improved communication could further help and eliminate barriers such as miscommunication and, therefore, provide a better healthcare outcome for patients.
The Notebook Effect
More times than not, my notebook grabbed more attention during my work than I did. Oddly, my slightly worn yellow Wide Ruled notebook made people more curious and standoffish than my own presence. Generally, the practitioners never seemed to mind as much as the patients in the waiting room or interviewee in her own home. But this does not mean that I would not receive the occasional, “Oh, don’t write that down” from the practitioners. Secrecy was not an issue among the practitioners and for that I am thankful. My notebook and I were immediately accepted, even if I was still cautioned where I left it. It took longer with the patients within the clinic and one of my refugee interviewees in terms of gaining trust and understanding that I was not intentionally trying to seek “dirt” on their status. Either way, I thanked them as well, when they finally accepted my notebook and reassurance that I was only there to observe and ask questions about their communication experiences in the United State’s healthcare system.
I felt like the minority for the first time when I entered my interviewee’s home. I could not speak the language and hardly understood a word. I knew enough to say thank you and nothing more. Often when in the clinic, I would describe my voice as limited or compare my situation to a child learning to speak. It was difficult to express myself when I was aware that the probability of being understood by the refugees was very low. Whether in a home visit or clinical setting, I wanted to be heard. I could not imagine what these refugees endure when they first arrive, but I applaud their bravery and their company was joyful.
- Abbe, M. (2010, forthcoming). An analysis of cultural competence, cultural difference, and communication strategies in medical care (Unpublished doctoral dissertation). Case Western Reserve University, Ohio.
- Chang, M., Feller, E., & Nimmagadda, J. (2009). Barriers to healthcare access in the southeast Asian community of Rhode Island. Medicine & Health, Rhode Island, 92(9), 310-313.
- Chen, A., Youdelman, M., & Brooks, J. (2007). The legal framework of language access in healthcare settings: Title Vi and beyond. General Internal Medicine, 22, 362-367.
- Clinic for refugees new, bigger. (2009, August) Dallas County Health and Human Services. Retrieved from http://www.dallascounty.org/department/hhservices/services/bioterrorism /documents/AugNews_000.pdf
- Do, H.D. (1999). Issues in the Vietnamese American community. In D. Bayor (Ed.), The new Americans: The Vietnamese Americans (pp. 49-77). Westport, CT: Greenwood Press.
- Flores, G., Laws, M.B., Mayo, S.J., Zuckerman, B., Abreu, M., Medina, L., & Hardt, E.J. (2003). Errors in medical interpretation and their potential clinical consequences in pediatric encounters. Pediatrics, 111(1), 6-14.
- Green, A.R., Ngo-Metzger, Q., Legedza, A.T.R., Massagli, M.P., Phillips, R.S., & Lezzoni, L.I. (2005). Interpreter services, language concordance, and health care quality. Experiences of Asian Americans with limited English proficiency. Journal of General Internal Medicine, 20(11), 1050-1060.
- Hymes, D. & Gumperz, J. (1972). Models of the interaction of language and social life. In Gumperz, J. & Hymes, D (Eds.), Directions in sociolinguistics: The ethnography of Communication (pp. 35-71). New York, NY: Holt, Rinehart and Winston Inc.
- Ito, K.L. (1999). Health culture and the clinical encounter: Vietnamese refugees’ responses to preventive drug treatment of inactive Tuberculosis. Medical Anthropology Quarterly, 13(3), 338-364.
- Lee, E. (1997). Cross-cultural communication: therapeutic use of interpreters. In E. Lee (Ed.), Working with Asian Americans: A guide for clinicians (pp. 477-489). New York, NY: The Guilford Press.
- Mahi, A.S. (1993). The challenges of cross-cultural communication: how to express the cultural elements in translation. (Published Dissertation). Retrieved from http://www.library.yorku.ca/ccm/Home/eResources/YorkTheses.en
- Mehrabian, A. (1971). A language within language. Silent Messages (pp. 89-107). Belmont, CA: Wadsworth Publication.
- Ong, L.M.L., De Haes, J.C.J.M., Hoos, A.M., & Lammes, F.B. (1995). Doctor-patient communication: A review of the literature. Social Science and Medicine, 40(7), 903-918.
- Parfitt, B. (1998). Getting in the culture: a taxonomic analysis of an emerging domain.Working across culture: a study of expatriate nurses working developing countries in primary health care (pp. 89-144). Aldershot, UK: Ashgate
- Pauwels, A. (1995). Communicating in and through different languages. Cross-Cultural Communication in the Health Sciences: Communication with Migrant Patients (pp. 30-59). South Melbourne, Australia: Macmillan Education Australia.
- Perez, M.A., & Luquis, R.R. (2008). Complementary and alternative medicine in culturally competent health education. In M.A. Perez & R.R. Luquis (Eds.), Cultural competence in health education and health promotion (pp. 67-85). San Francisco, CA: A Wiley.
- Purnell, L.D. (2008). Traditional Vietnamese health and healing. Urological Nursing, 28(1), 63-70.
- Samovar, L., & Porter, R.E. (1991). Nonverbal communication: sound and action. In P. Randall & L. Blass (Eds.), Communication between cultures (pp. 175-212). Belmont: CA: Wadsworth Inc.
- Schouten, B., & Meeuwesen, L. (2006). Cultural differences in medical communication: A review of the literature. Patient Education and Counseling, 64, 21-34.
- U.S. Census Bureau. (1990). State and county quick facts.
- U.S. Census Bureau. (2000). State and county quick facts
- Youdelman, M., & Perkins, J. (2002) Providing language services in small health care provider settings: examples from the field. The Commonwealth Fund