Linguistic and Cultural Aspects of Tuberculosis Screening and Management for Refugees and Immigrants
We are here to focus on tuberculosis screening. However, I am not working in TB control per se. I am a general internist who provides primary care in an inner-city clinic where I see a large number of people who have recently immigrated from Africa and Asia. The relevance of that experience for TB screening and the management of active cases is based on my familiarity with communication and management issues relevant to diagnosis and treatment of disease across the linguistic and cultural differences between biomedicine and non-western cultures.
The Refugee Clinic is now 14 years old. It serves the Asian and African refugee communities in Seattle, Washington. The clinic began in 1982 after the large influx of Southeast Asian refugees into Seattle, and has continued to care for each subsequent wave of arrivals. We now see about 6,000 patients a year in Harborview Medical Center, a tertiary care referral center and teaching hospital in the University of Washington system. Our clinic provides general internal medicine, family medicine, psychiatric services, and oriental medicine for patients from 13 distinct linguistic groups.
This morning I would like to discuss some strategies we have learned for managing chronic problems like TB treatment or prophylaxis in a setting like ours. I will focus on management strategies and frequent areas of miscommunication that require special attention. I will then discuss how and why translation and patient education is such a complicated process. Finally, I will describe a system we have developed to help us negotiate cultural differences.
Although I am focusing here on non-English speaking patients, the discussion will raise the question; how are these issues any different than those facing our English speaking patients? The answer of course is that they are no different, they only have a different tone or flavor, but the general issues are the same. Of course, the reason for this is that medicine is itself a linguistic and cultural system with assumed values and priorities, and a single way of talking about illness. This assumed difference between medical culture and everyday life lies behind much of this discussion.
The most common cross-cultural miscommunications in our clinic can be grouped according to the following categories:
- The (illness) or "diagnosis" is unheard of.
- The diagnosis is known, but has been re-interpreted.
- The diagnosis is known, but management is not agreed upon.
- The diagnosis and management are agreed upon, but the relative costs (financial or social) of diagnosis are unacceptable.
Diagnosis Unheard Of
In 1990 we conducted a structured interviews of a small sample of patients in our clinic, 20 adult respondents chosen at random from one of five ethnic groups: Lao, Mien, Chinese, Vietnamese, and Cambodian (Thai, 1990). We were interested to know what sense refugee patients made of common biomedical diagnostic terms, words like cold, parasite, tuberculosis, epilepsy, hypertension, and heart attack. In this survey we asked people if they could name the major symptoms of an illness, and beyond that if they had any biomedical notions associated with causation or disease process. The Mien, Lao, and Cambodian patients could identify TB appropriately in about 1/3 of the cases. The ethnic Chinese and Vietnamese gave appropriate responses 47% and 52% of the time respectively. The pathology and etiology of TB could be elaborated on by about 20% of the Chinese and Vietnamese, 13% of the Cambodians, and only about 5% of the Mien and Lao. Similarly, in a survey of 102 patients in Wardha District in India, 2/3 had never heard of the disease tuberculosis (Barnhoorn and Adriaanse, 1992). In this situation the physician is confronted with creating a new illness category for a patient, and describing what the experience of having this illness would be like. The patient must then make sense of the diagnosis within their experience of illnesses and according to their understanding of disease categories.
The Diagnosis is Known But has been Re-Interpreted
The use of the word TB becomes more complicated when the word is known in their culture. The biomedical term may be unfamiliar, the symptoms are often well known, but named and understood according to a different taxonomy of illness. Nichter and Nichter (1994) report that in Oriental Mindoro in the Philippines some of the early stages of latent TB are understood and discussed as mahina ang baga which literally translates as "weak lungs." Weak lungs is a well known concept for many Filipinos in this region, but one that carries none of the specificity that TB suggests. Apparently, physicians often use this term out of sensitivity to the harsh social stigma attached to TB, and often in an effort to explain their preventive treatment plans to someone who is ppd +, or who has an abnormal chest x-ray and who requires preventive therapy. On the other hand, patients understand "weak lungs" as an illness associated with recurrent acute respiratory tract infections, especially in children, for patients in Mindoro the phrase describes loss of weight, lack of appetite, low grade "inside" fever, prolonged cough and cold. TB in contrast, is associated with many of these same symptoms, but is a very serious illness that must be diagnosed by a doctor, and generally involves coughing up blood. Weak lungs can be an antecedent to TB, but is a separate category of illness. In a setting like this the use of the phrase TB is a known diagnostic category but it has been redefined in a very limited way.
The situation becomes even more complicated because INH is advertised by a term that suggests that it is a vitamin for "weak lungs." When people experience recurrent respiratory tract infections they may take a week or two of INH to "strengthen" their lungs. The implications for drug resistance or confusion around the true purpose of INH is obvious.
The situation in the Philippines is a common one. A biomedical diagnosis suggests a related symptom complex that is easily associated with a different taxonomy of illness. A variation on this process of finding correlates can also happen where a biomedical term is taken and assigned other symptoms. In my experience it is not often the case that the symptoms of TB are unrecognized. Throughout the former colonies of France and Great Britain where their has been substantial contact with western culture the term is known. What usually happens is that the term is taken, the symptoms of TB are identified and then these symptoms and the associated term is re-interpreted in light of existing taxonomic categories. In other words, new indigenous categories of illness are classified using the biomedical term as an organizing principal.
The Mindoro example illustrates this process. Hard work, change of temperature, chronic fatigue, malnutrition, and weight loss are all felt to be conditions that can contribute to illness. When a laborer works to exhaustion in the hot sun and the sweat then dries on his back, if he chills afterwards, these are the changes in hot and cold that can lead to a serious illness called pulmonya, if this develops further it may become TB (Nichter, 1994). Here humoral theory and other known folk etiologies of illness are used to define, explain, and categorize tuberculosis.
During my fieldwork in rural Nepal among the Tamang a local healer responded to one of my inquiries about common causes of illness in his village. "Oh yes," he said, "there is TeBe there are seven other types you don't know about." He went on to describe different types of illnesses related the process of "wasting" and "drying out" that he felt TeBe was an example of.
In situations like this physicians need to know something of the local semantics of illness to know what key words will convey to the patient.
Agreement About Diagnosis but Disagreement about Management
Many people from many cultures, including our own, find the idea of treating subclinical or latent disease suspicious. If there are no symptoms then what are you treating? We must admit that our decision to treat certain people with INH and not others is based upon probabilistic forecasting. Many non-western health traditions see individuals as unique, our treatment based on arbitrary cutoffs at 10mm or age 35 but without knowledgeable reference to a particular individuals "constitution" seems to attend to an abstraction and not to the living person. Many of their own health traditions are highly individualized taking into account a persons birth date, living situation, symptoms, and health history into the diagnosis and management of illness.
Another common situation in which the diagnosis of tuberculosis is not questioned, but the treatment strategy is unacceptable is in the case of medication side-effects. This again on the surface does not appear to be a cultural clash as much as it seems to be a problem of unacceptable side-effects e.g. rash, nausea, or diarrhea. The cultural clash comes around the risk/benefit decision. From the clinical point of view if we saw markedly elevated LFTs that continued to climb, or noted the INH induced a persistent nausea and vomiting we would undoubtedly agree to stop INH prophylaxis. For medicine these symptoms present an unacceptable risk. When the side-effects are mild, unreported, or unheard of by the clinician it s a different story. In this case the public health benefit or treatment goals out-weigh these "minor side-effect." In other words, for the patient the presence of subtle side-effects may be an important early warning that something is not right but these symptoms may mean nothing in a western medical review of systems.
From the patient's point of view a treatment should make you feel better not worse. For example, many older Vietnamese patients will complain about INH making them feel "hot", sometimes this means warm, sometimes it means flushed, often this complaint refers to a constellation of symptoms (hot breath, rapid pulse, dry cough) taken to indicate too much "heat" in the body. To a prudent person this should mean that their system is out of balance and requires a dietary or medicinal intervention. If the TB clinician minimizes the complaint, or po oh-poohs their concern, the trust that is supposed to comprise the therapeutic relationship is jeopardized.
Consider the following example, Cambodian patients will often complain of being "ahh kaamlang" (without strength) a condition by itself or in association with other illnesses. This complaint refers to more than just being out of strength, it means losing balance in a way that has left them without adequate "internal strength" to sustain their health. In a study of "non-compliance" among Cambodian patients in our clinic, Shimada (1995) found that Cambodian patients were often concerned that medication was too strong for them (43%) concluded when someone feels "ahh kaamlang" the solution was to adjust the medication or discontinue it altogether.
As one of our patients explained:
Patient: Its like if I feel like I don't have the internal strength (kamlaang), then I'm not gonna take the medicine.
Interviewer: And what, what would happen if you did take the medicine when you didn't have energy or strength?
Patient: I never, if I, like I feel like I have a severe dizziness, then I have to not take it. But after I have a little bit to eat, like rice soup, stuff like that, and give some strength, then I take the medicine...its like if you have low internal strength (kamlaang), then you take the medicine, then its your gonna have even less internal strength (aah kamlaang).
Interviewer: Is there the same problem with Cambodian medicine?
Patient: Cambodian medicine is okay. If you don't have strength then you can take it, it's fine. But I'm afraid of Western medicine, it's more strong.
In some cases people are more sophisticated and are well aware of the TB diagnosis, but take issue with our treatment algorithms. Most of the patients in our clinic come from cultures that hold the physician in very high regard. While they are inclined to trust physicians and to trust their diagnoses, they may feel that treatment decisions are made with reference to regular Americans and not with their community in mind. For example, Brudney and Dobkin (1991) report that in the developing world the average annual risk of infection with mycobacterium tuberculosis is 2% to 3% so that by age 20 virtually everyone is PPD positive. Given the prevalence of recent infections and the scarcity of diagnostic resources the strategy in many of these countries is to track and treat active infections only and not to treat with INH the inactive, but PPD positive cases. People that immigrate from a health system operating according to this logic to another where the opposite strategy makes sense, given our epidemiologic context, fail to appreciate this epidemiologic shift. If they are aware of our different treatment strategy they may feel that biomedicine is operating differently here, and it must have more to do with their minority/immigrant status, and less to do with concern about their health or the health of their community. Add to this the use of BCG in many developing countries and admonitions to be careful with PPD testing if one has BCG and the story becomes even more confusing for patients. They don't understand why we are inclined to ignore their BCG history and give them INH.
Disagreement about Diagnosis and Management, Because the Social Costs Are Unacceptable
In our clinic, more than disagreements about diagnosis or appropriate management is a fear of the diagnosis because of the social isolation that results. In some cases there is a frank denial of the TB diagnosis because it will create unacceptable financial and social costs. Norman Bethune, a surgeon to the liberation forces of China in 1939 reflected on these costs when he said, "and as a doctor I suffered from two very different diseases. I was only beginning to make my way as a surgeon when I came down with a bad case of tuberculosis...My second sickness...well that wasn't so simple. I caught it from no one and I caught it from everything. I got it as a boy, as a man, as a doctor. It was much worse than tuberculosis. It wasn't like curing an infection. I couldn't get rid of it by drugs. And many of the things I saw as a doctor only made it worse...But there came a time when I no longer gave into it. That was after I came to understand that tuberculosis was not merely a disease of the body but a social crime..." (Waitzkin, 1983).
Steve Nachman (1993) found that this was the worst aspect of the disease for the detainees of the INS camp at Krome, Florida for Haitian boat people with TB. For the Haitians TB is translated into creole a variety of ways, maladie de poitrine, maladi touse, teve, tibekiloz. Nachman explains that Haitians traditionally consider TB to be the worst of all afflictions since the sick person must be isolated. Maldi pwatrin (TB) is considered the worst possible diagnosis that presaged certain death. "The diagnosis was terrifying because traditional Haitian culture defines family relationships as warm, close, and enfolding. To be barred from one's own family is nearly unbearable and because illness is a time when one is supposed to want and need companionship" (Nachman, 1993).
I have a patient who is terminally ill. She is a 73 year old Chao Jho speaking Chinese grandmother who presented with fatigue and developed a dry cough and shortness of breath. Her x-rays was suggestive of widely metastatic lung cancer, and the biopsy showed very poorly differentiated non-small cell carcinoma. In keeping with her family's wishes we put off telling her her diagnosis until we were absolutely certain. One evening in clinic, while we were preparing to tell her and I was explaining that she had a very bad lung illness that was not treatable and would probably continue to get worse she began to cry. She asked me in an urgent tone "is it TB?" I said, no it is lung cancer and it is not treatable. To my surprise she appeared somewhat relieved. She was clearly grieved that she had a terminal illness, but she went on to explain that the possibility that she would not be able to spend the remaining months holding her grandchildren, cooking for them, and that her cough might infect them was the worst possible scenario. The cancer was tragic, but it would not impose the sentence of isolation that she felt TB would require.
There are numerous other examples throughout Asia and the Pacific where the presence of TB in the family is a social humiliation, it jeopardizes marriage contracts of the family, it results in the isolation of the family, and can affect business relations. The importance of this is that no amount of individual patient education will help TB patients to deal with the larger community attitudes about TB. The community must be addressed, but the community is often an amorphous one that migrates between cities, coasts, and continents.
Agreement About Diagnosis and Management, but Attempts to Comply Misdirected
When Shimada (1992) surveyed the Cambodian patients in our clinic and interviewed them at length about the medications (including INH) and their strategies for taking them she was amazed to discover that 70% were non-compliant. After reviewing their charts and then speaking with them we discovered that our adult Cambodian patients were non- compliant when narrowly defined as "not adhering to a prescribed medical regimen" but we also discovered that 2/3 of these people were attempting to comply, that is they were unintentionally non-compliant. Of this group of patients that incorrectly took their medications they chose to do so for different reasons, 27% took the medicine when they felt they needed it, another 27% were unaware why the medication had been prescribed. Thirty-seven percent used other peoples medications for their illnesses. The remaining non- compliant behaviors were because people became alarmed about side effects and tried to manage the side-effects by adjusting the dose or timing of the doses. At times patients were concerned about taking too many medications so they decreased the amount (3%), at other times they felt that taking more medicine would hasten the healing process (30%). Some attempted to balance their effects with diet and combined medications in unpredictable ways. Many people did one or more of these behaviors. The point here is that there were attempts, at times heroic attempts, to comply with medical therapy according to our Cambodian patient's understanding of our intention and their own understanding of health, but the end result was still medical failure. The most common explanation about the need to adjust their medication was that it was "tanam klaing" or "strong medicine", so strong in fact that unless dealt with properly, it would make their bodies weaker.
Miscommunication, Misinterpretation, and the Search for Equivalence
There are other kinds of miscommunication, but these four large categories illustrate common linguistic and cultural negotiations that must take place between clinic staff and patients from immigrant communities. It is true that these kinds of miscommunications between lay culture and medical culture can occur between physicians and their English speaking patients also. When attempting to communicate to non-English speakers it becomes particularly problematic. One piece of the solution is the obvious need for a trained interpreter, but the complex issues in interpretation are often lost to the monolingual medical team since that work occur inside the interpreters head. More importantly, if the process of translation is not considered critically then decision might be made by hired interpreters that my contribute to the miscommunication. In an effort to help monolinguals appreciate the importance of interpretation it is useful to look a the process to understand where it can often go awry.
Translation is an effort to find equivalence between two languages (Sechrest, 1973). There are several tasks to attend to in the process of translation, several types of equivalence to find. The first and most obvious is vocabulary equivalence, or finding the key word that conveys the meaning in both languages. The next is experiential equivalence. That is making certain that the equivalent experiences are conveyed. For example, the phrase "hair dresser" does not have a Navajo language equivalent, because in traditional Navajo life their is no equivalent cultural experience.
The final type is conceptual equivalence, this means finding the word that links to other words in the culture with unique emotional and experiential associations. This type of equivalence is very difficult to find, the more dissimilar languages and cultural experiences are, the harder it is to find conceptual equivalence. The translation of humor, literature, and especially poetry relies on finding conceptual equivalence. This is routinely impossible, and hence the shrugged explanation "it loses something in translation."
Using TB as an example, the absence of a vocabulary equivalent of TB for the Mien, means the Mien have no word that differentiates TB from any other lung infection. Conversely, for the Tamang the presence of multiple types of TB means that the use of the phrase TB may evoke one of their seven types of wasting illnesses with equally distinct implications for treatment.
In the clinic, during medical translation we often rely on experiential equivalence. Here the symptoms of the illness are assumed to be held in common. While we have already discussed that at times this is problematic, in general there are vocabulary equivalents for straightforward symptoms of common illnesses. However, experiential equivalents associated with a word or phrase are affected by both reference and performance that the word implies. Let me say this again, a word has both referential and performative aspects. For example, when Jucovy (1978) asked dematolo-pathologists about their use of the word "dysplastic" they explained that while they agreed upon its definition, they rarely used it outside of the small circle precisely because of the performative implications of its use. Dysplasia for pathologists referred to features of the cell and implied a limited range of diagnostic performances, but for surgeons it often implied wide resection of potentially "dysplastic" tissue. If this is true among our own clinical sub-cultures, this will be even more pronounced cultural differences.
In the case of TB the referential value of TB for me means collecting sputa, awaiting smear results, reviewing x-rays, beginning therapy, and monitoring it. Difficult steps, but not emotionally laden to speak of. The referent TB for my Chao Jho patient, or a Haitian refugee implies sputa, x- rays, and treatment, but it also implies the experience of embarrassment for their family, and community imposed isolation of the patient. For some groups TB might also imply expensive adjuvant ceremonies and trips to distant places in search of a healer and a cure.
Negotiating these complex performative implications, and mediating the differences between referential fields is the work of the professional medical interpreter. Knowing the implied references and performances of a target community means finding someone culturally competent in both the culture of medicine and target culture. This is why employing a bilingual-bicultural professional to translate their work into the target language will not create in and of itself cross cultural materials, or a truly cross- cultural plan of action. The logic and idioms of the target class and cultural group are not usually adequately incorporated into programs and materials for patient education. More often than not cross-cultural programs are really unilateral programs driven by the logic of public health and medicine. They are not truly bi-cultural efforts. Interpreters of higher socio-economic class, with an interest in linguistic precision may fumble the needed vernacular translations that will be meaningful to patients. Recent literature produced by AIDS activists for the at risk communities are better examples of appropriate language and behaviorally translated materials. Having said that our TB management strategies are to be developed in a bi-cultural manner, and knowing that it impossible to predict the nature of the miscommunication and sets of issues that will arise for individual patients, how can we adequately address linguistic and cultural barriers in TB screening and treatment programs? There is already evidence that the use of culturally appropriate outreach workers enhances compliance and enables improved case finding. Radiate and Menzies (1994) found that compliance among the Cree in the Mistassini Cree territory in Northern Quebec dropped from 78% in 1981-1985, to less than 20% since 1986. The factor identified was a change in TB treatment follow-up from Cree community health workers to non-Cree nurses. Similarly, Etkind et al. (1991) attribute their success among Southeast Asia patients in Boston to the use of bilingual-bicultural outreach workers. Compared to a national average of 61%, the Boston program found that 82% of the Southeast Asians beginning therapy completed it when the follow-up contact was made by bilingual-bicultural outreach workers. It comes as no surprise then that from Honduras, to Mexico, Haiti, and here in the States, the focus is on developing affordable systems that employ bilingual-bicultural outreach (Farmer 1991, Mata 1985, Rubel 1992).
Going Beyond Interpretation to Cultural Mediation
Clearly, the solution to identifying and addressing cultural barriers to TB control does not lie in memorizing anecdotes or stereotypes about specific populations. As I just said, it also does not lie in interpreting educational materials and programs alone, because it is very hard to target the practices, values, educational level, and changing dynamics of the various subgroups in the community. It lies in developing a workable, cost- effective system, that may need to target several clinical or public health concerns to be affordable.
Let me give the example of a program we developed in Seattle to facilitate the care of African and Asian refugees that use the Harborview Medical Center.
Structure and function of Community House Calls
Goals: In 1993, the Children's and Refugee Clinics at Harborview Medical Center were funded by the Robert Wood Johnson Foundation to develop a program that would accomplish the following goals (Jackson-Carroll, Graham, and Jackson, 1995):
- Create a common fund of knowledge between ethnic and medical cultures
- Decrease language barriers to care
- Improve cross-cultural health care education of providers and trainees
- Enhance efficient utilization of resources by High Risk/High Needs families.
We did this by creating a network that would allow for communication to flow from the institution to the targeted communities and from the communities to the institution. We targeted five of the neediest refugee groups seen in our clinics, three Ethiopian groups, the Cambodian community, and the Somali community. We contacted their leadership and established a Community Advisory Board. The Board in turn helped us hire an interpreter cultural mediator (ICM) for each community. The Board also identified one or more advisors to serve as a contact person for ethnographic information within that community.
ICM Roles: ICMs are bilingual/bicultural outreach workers who are familiar enough with the biomedical and American cultures that they can act confidently within the health care system, they must be know and trusted by the institutions, and have influence with provider and clinical teams. They have a variety of responsibilities. They case manage an average of 30 families (150 people) at any one time. They interpret in clinics. They represent their communities during conferences and didactic activities with House Staff and in the medical community. They arrange for home visits. They also arrange for community teaching sessions where residents, nurses, or other clinic personnel can address the community on public health or a medical topic.
ICM Roles: The CAs are usually monolingual uneducated community members on public assistance. They are paid a small stipend to be available as community resources to help support the ICMs or to serve as contacts in the community for outreach from the hospital. They are particularly helpful in assessing how our activities are being received in the community since, unlike the ICMs, they are much more reflective of the patients and families that use our services.
EthnoMed: EthnoMed is yet another innovation of the program. EthnoMed is a World Wide Web site that serves as a clearing house for information to clinics and to community groups. This is still under-development but is a nice model for the kinds of information that can be provided to clinicians and to patients around a focused topic like tuberculosis.
TB Programs with Cultural Mediation
A program like House Calls would allow TB control efforts to work in cooperation with the targeted communities in a given region. Regional and individual differences compounded by the process of acculturation make it impossible to predict on a person by person basis which of the above common communication issues will arise, and when these issues will arise. The House Calls Program is a living system that can address communication issues, adjust to points of resistance, monitor medical compliance, find cases in hard to navigate communities, establish a collaborative relationship with TB clinics, and provide other needed services to immigrant communities. Importantly, it will also enable TB nurses to operate a little less like infection control police and a little more like a part of the primary care system. While this may make some sense, it may also seem unworkable because of cost. I think there are strategies for making a system like this affordable:
- One means of creating an affordable system would mean creating collaborative affinities between public health programs targeting the same populations. For example, Hepatitis B, STD's, and TB sharing the cost of such a system.
- More effective would be collaboration between primary care services serving a particular community and the public health system overlapping their effort.
- In rural areas where resources are spread thin across a wide area the system of collaboration may also have to spread beyond medicine to include the school system e.g. Head Start Programs, or other health and social services.
Managing illness cross-culturally depends on communication, but communication depends on understanding the socially and culturally determined references and performances associated with a diagnosis and its treatment. Because people, their communities, and the process of acculturation is dynamic, there are no simple rules of thumb or tactics that are eternal. In my opinion what the management of TB cross culturally requires is a living system that can adapt with the target population as it adapts to the forces acting on it. It also requires a system that can educate and intervene in the culture of medicine and in the target community on both an individual and community level. The difficulty of TB prevention in particular lies in the fact that the treatment is worse than the disease. If one feels fine, having to take INH, even if you are not bothered by it, is more of an imposition than not taking it. Bilingual- bicultural outreach workers have been shown to be valuable in TB outreach, the network we described today may be a second step toward better cross-cultural TB management.
- Barhnoorn F, Adriaanse H. In search of factors responsible for noncompliance among tuberculosis patients in Wardha District, India. Social Science Med 1992;34(3):291-306.
- Brudney K, Dobkin J. A tale of two cities: tuberculosis control in Nicaragua and New York City. Seminars in Respiratory Infections 1991;6(4):261-227.
- Etkind S, Boutotte J, Ford J, Singleton L, Nardell EA. Treating hard-to-treat tuberculosis patients in Massachusetts. Seminars in Respiratory Infections 1991;6(4):273-282.
- Farmer P, Robin S, Ramilus SL, Kim JY. Tuberculosis, poverty and "compliance": lessons from rural Haiti. Seminars in Respiratory Infections 1991; 6(4):254-260.
- Gaudette LA, Ellis E. Tuberculosis in Canada: a focal disease requiring distant control strategies for different risk groups. Tubercle and Lung Disease 1993;74:244-253.
- Jackson-Carrol LM, Graham E, Jackson JC. Beyond medical interpretation: the role of interpreter cultural mediators (ICMs) in building bridges between ethnic communities and health institutions. Community House Calls. Seattle, WA, 1995.
- Jucovy PM. Developing a critical model for diagnostic language. Proceeding Medcomp '82-first IEEE Computer Society International Conference on Medical Computer Science/Computational Medicine, September 1982:465-489.
- Mata JI. Integrating the client's perspective in planning a tuberculosis education and treatment program in Honduras. Medical Anthropology 1985; Winter:57-64.
- Nachman SR. Wasted Lives: Tuberculosis and Other Health Risks of Being Haitian in a U.S. Detention Camp. Medical Anthropology Quarterly 1993;7(3):227-259/
- Nichter M. Illness semantics and international health: the weak lungs/TB complex in the Philippines. Soc Sci Med 1994;38,5:649-663.
- Nichter M. Acute respiratory illness; popular health culture and mother's knowledge in the Philippines. Medical Anthropology 1994;15:353-375.
- Ridout M, Menzies R. Factors affecting compliance with preventive treatment for tuberculosis at Mistassini Lake, Quebec, Canada. Clin Invest Med 1994;17(1):31-36.
- Rubel AJ, Garro LC. Social and cultural factors in the successful control of tuberculosis. Public Health Reports 1992;107(6):626-636.
- Secrest L, Fay TL, Hafeez Zaida SM. Problems of translation in cross-cultural research. Journal of Cross-Cultural Psychology 1972;3,1:41-56.
- Shimada J, Jackson JC, Goldstein E, Buchwald D. "Strong Medicine": Cambodian views of medicine and medical compliance. Jour Gen Int Med 1995;10:369-374.
- Thai D-Q, Buchwald D, Hooton TA. Medical Knowledge and Familiarity with Western Medicine Among Southeast Asian Refugees. Unpublished Thesis, 1993.
- Waitzkin H. The Second Sickness: Contraditions of Capitalist Health Care. Free Press 1983, New York, NY.