As U.S. society becomes increasingly diverse and multicultural, health care providers are being encouraged to learn to recognize and respond to cultural differences. Over the past decade, many health care institutions, and virtually every health profession, have made cultural competency training a requirement, with the goal of providing accessible and appropriate care to all.1–4 Such training primarily targets racial and ethnic minority groups, promoting awareness of cultural difference among diverse and vulnerable populations, and teaching tactics for modifying clinical practice in order to decrease presumed “cultural barriers” to quality health care for such patients.2,5 In this discourse, Latino patients are often perceived to be particularly culture-driven. Clinicians are encouraged to develop an awareness of Latino cultural traits and to take them into account in their clinical interactions with such patients.6,7
Despite its popularity, cultural competency remains a rather vaguely defined goal; explicit criteria for its accomplishment and assessment have only just begun to be established.8,9 Traditional approaches to cross-cultural training have relied on presenting lists of beliefs and practices attributed to certain cultural groups. This approach has been criticized for potentially promoting stereotyping.1 More recently, patient-centered approaches to cultural competency are being developed, which emphasize learning skills for open, respectful communication, and encourage dealing with each patient as an individual rather than a carrier of a presumed set of cultural traits.10,11
How notions of culture are actually understood and applied by clinicians, and how they influence clinical interactions, are of growing interest,12–14 but these questions have not yet been carefully examined. In this report we consider these issues, based on an exploratory ethnographic study conducted in clinics providing prenatal genetic services to Latinas in southern Texas. We examine the range of concepts of Latino culture expressed by the clinicians we interviewed, compare these concepts to patients’ discussions about their health care encounters, and consider how clinicians’ ideas about cultural characteristics were manifested in the clinical consultations we observed.
Sample and data collection
In this report we present data from an ethnographic study, conducted between 2000 and 2002, designed to explore clinician and patient perspectives in the context of decision making about amniocentesis. We interviewed a convenience sample of 50 clinicians at 21 clinics in southern Texas, including 15 genetic specialty clinics and six nonspecialty prenatal clinics. All the clinicians were involved in discussing prenatal genetic testing options with “at-risk” Latina patients, and included a cross-section of practitioners: physicians, nurses, social workers, medical assistants, and certified genetic counselors. We asked the clinicians individually, usually by telephone, if they would be willing to be interviewed. Of the 55 clinicians we approached, two declined and three were not interviewed due to scheduling problems. Most interviews were conducted in the clinicians’ offices, focusing on their strategies for offering prenatal diagnostic testing and their perceptions of issues they encounter when caring for Latina patients.
We also interviewed a convenience sample of 40 patients from three of the prenatal genetics specialty clinics from which we drew our clinician sample. All were self-identified Latinas who had been offered amniocentesis following an abnormal blood screening test (alpha fetoprotein). At the time of their initial consultation at these clinics, patients were invited to participate in our study. Only seven of the 79 patients asked to participate declined. Thirty-two of the remaining 72 patients were not interviewed due to scheduling problems, or because our sample size goals had been reached. Interviews were conducted within two to four weeks of the consultation, usually in patients’ homes, focusing on their understandings of fetal abnormalities and testing procedures, the basis of their amniocentesis decisions, and general impressions about the genetics consultation.
Additionally, we conducted observations of 101 genetic counseling sessions at eight genetic specialty clinics whose clinicians we had interviewed. We kept careful field notes of these observations, designed to record clinician–patient interactions as fully as possible, with special emphasis on clinicians’ communication strategies for conveying technical information, and on indications of patients’ levels of understanding. Some demographic characteristics of the clinician sample, patient sample, and clinical observations are summarized in Tables 1, 2, and 3, respectively.
The interviews followed standardized sets of open-ended questions, and participants were encouraged to answer as expansively as they cared to. Interviews averaged about two hours each, and were tape recorded and transcribed. All study participants gave their informed consent to be observed and/or interviewed following IRB-approved protocols.
We conducted content analysis and established a method for standardizing and displaying interview data. We summarized each interview, then made tables of blocks of text (quotations and summations) for each research participant, which displayed the main topical areas of the interviews. Further, higher-level classification and display of the data were based on thematic patterns identified in these initial tables. We also made databases for all three data sets. These databases consisted of demographic data and open-coded variables of the main questions covered in the interviews and observations.
In order to minimize investigator bias in our analysis, throughout the project all phases of data processing and analysis were cross-checked in conference sessions. In these sessions the research team discussed each case, reviewed emerging findings, honed analysis strategies, and reached consensus about the application of coding categories. At the same time, we also conducted spot checks for consistency in coding and classification procedures, and addressed and resolved any anomalies or discrepancies in coding.
When asked which women are most likely to decline prenatal testing, more than half of the clinicians said “Hispanics” or “Latinas.” Moreover, they were remarkably consistent in the way they described Latino culture, with nearly all readily naming specific traits as typical of Latinas and affecting their willingness to accept testing. These views were commonly expressed by each of the various professional groups we interviewed, with no group being more or less inclined to discuss them. They characterized Latinas as religious, fatalistic, family-centered, fearful, and superstitious. They explained, for example, that Latinas are likely to decline testing because they are mostly devout Catholics who would be unwilling to consider abortion, or because their belief in “God’s will” makes them feel they must accept whatever God sends them. Table 4 summarizes the Latino cultural traits most frequently mentioned by these clinicians.
How do clinicians’ assumptions about the cultural influences on Latinas’ acceptance of testing compare to patients’ discussions about their own decision making? Table 5 displays selected patient responses pertinent to the clinicians’ view of Latinas, contrasting responses of patients who accepted amniocentesis with responses of those who declined.
One common idea among clinicians was that Latinas often decline amniocentesis out of fear. They described Latinas as being generally fearful of medical issues, fearful of needles, and fearful of the amniocentesis procedure itself. In the words of one registered nurse:
Most of the people who decline are of Hispanic origin that I have seen…. I think a lot of that is fear … just the thought of a needle going though the abdomen and near the baby, I think that scares a lot of them.
Another prevalent perception was that Latinas are likely to embrace culturally specific superstitions and folk beliefs that can present barriers to accepting the procedure. Nearly half of the clinicians portrayed Latinas as having peculiar notions about the causes of birth defects, such as believing a lunar eclipse, brujeria (witchcraft), or mal de ojo (evil eye) can negatively affect the health of a developing fetus. The comments of a medical assistant who regularly offers amniocentesis to Latinas illustrate this perspective:
The Hispanic culture has a lot of old wives’ tales … brujeria, witchcraft, and they focus on a lot of old home remedies. They’re more prone to think it’s related to something else, and to try to resolve it with things from their own upbringing.
The popularity of such images of Latino culture notwithstanding, they are not supported by much of what we encountered in our study. For example, the Latinas we observed did not, in fact, decline amniocentesis more often than other patients. With about 60% accepting the test, they were well within the 57% to 77% range of acceptance reported for the general population.15,16 None of those we interviewed invoked folk concepts such as brujeria or mal de ojo, and although many reported fearing the procedure, more of those who reported such fear accepted than declined the test.
How are other familiar images of Latino culture reflected in the amniocentesis decision process described by these patients? How might clinicians’ assumptions about Latino culture affect the care given this population? In order to explore these questions, the following sections compare clinicians’ concepts of Latino culture to the views patients expressed in their interviews, and to our observations of their consultations. In this analysis we will focus on the two cultural characteristics clinicians most frequently mentioned as affecting Latina patients’ willingness to accept amniocentesis: family-centeredness and religiousness.
The clinicians often characterized Latinas as coming from large extended families, being especially family-oriented, and highly influenced by the opinions and advice of their husbands, mothers, or other family members. For example, a social worker commented:
A lot of influence comes from the immediate and extended family members…. They go back and relate everything to the family, so I think that many times it [the amniocentesis decision] is more of a family decision and not a personal, individual choice.
In most consultations we observed, the father of the child or another relative accompanied the patient. In principle, encouraging family participation in decision making may seem a culturally appropriate and useful tactic. However, in several cases, this strategy seemed to go too far, to the point of subverting the patient’s role as the decisionmaker in the consultation.
At times, the clinician ended up consulting with the family member, usually the husband, rather than with the patient. This was especially true in cases where the family member spoke English and the patient did not. For example, in several cases, the clinician and husband had lengthy conversations in English, with both asking and answering questions, but with neither addressing the Spanish-speaking patient directly. In one such case, when the doctor asked the husband to tell the patient what they had been saying, the husband replied, “No, she understands…. I just wanted to hear it for myself.” At this point, without additional explanation, the doctor, who spoke very little Spanish, turned to the patient and simply asked, “Quieres amniocentesis?” [Do you want amniocentesis?] The patient replied, “Si.”
The assumption that Latinas are male-dominated, and expected to defer to their husbands’ opinions in making important decisions, was an idea discussed by many of the clinicians. Again, making such an assumption may create a self-fulfilling prophecy. Consider, for example, the following case from our observations. An obstetrics–gynecology resident appeared to be so certain that Latinas would defer to males that he insisted that one patient go home and consult with her husband, despite her repeatedly stating she was ready to accept the test. He later commented:
Oh, no, I would never do it before she talks to her husband because if she feels pressured, and we do it today, and then she miscarries, having never even spoken to her husband? I’d never do that.
While all of the patients we interviewed did indeed say they had discussed the amniocentesis with husbands and family members, rather than seeking permission, they described this as seeking information and advice. In fact, all the women reported that their husbands or families were supportive of their decisions, equally so whether they had accepted or declined. The following are typical of patients’ remarks about the role of family members in their decision making:
My husband told me, “You do what you want to do, I’m just here to support you.”
Both brothers and my sister-in-law were supportive and just said, “It’s up to you.”
My boyfriend, my mom, his mom, and my grandma … they told me it was going to hurt [laughs]. But they told me just to do what I thought was right.
Another view widely held by the clinicians we interviewed is that Latinos are a particularly religious group. Indeed, nearly three-fourths of the clinicians we interviewed asserted that Latinos have strong Catholic religious beliefs, and that this affects their willingness to accept testing. However, this view was not consistent with our patient interviews. While most of the patients were indeed Catholic, only a handful of them declined testing, and those attending church more than once a month accepted amniocentesis more often than those attending with less frequency.
Nearly half the clinicians characterized Latinos as fatalists who believe that God’s will determines one’s fate. They often cited a common Spanish phrase “Si Dios quiere” (If God wills it) as explaining the perceived Latino tendency to decline testing. People who believe God is responsible for their health, it is reasoned, will be passive about taking action to promote their health, and thus are likely to decline testing. The comments of a social worker sum up this view:
Faith and religion are just so very important to them, quite honestly, and the majority of my patients say, “I’m going to leave it to God’s hands. It’s God’s will. God sent me this baby, and it’s for a reason.”
Most patients did make references to God and prayer when discussing their experiences in being offered amniocentesis, saying that they had prayed about their amniocentesis decision. They made comments such as “If God sends us a child like that, we have to accept it”; “It will be whatever God wants”; “I decided to put it all in God’s hands.”
It is noteworthy that all of these comments were made by women who had accepted the test. The idea that God controls one’s fate clearly did not keep these women from accepting the test. Instead, “Si Dios quiere” was often invoked as a way of accepting the miscarriage risk engendered by the amniocentesis procedure itself. As one woman put it, “If God wants me to lose it, I’m going to lose it; if God doesn’t want that, my baby will be born fine.”
When asked what they had prayed for, most said they prayed that their baby would be healthy and everything would come out all right, and the vast majority of these were accepters. It seems that belief in the power of prayer and that God determines one’s fate, is consistent with accepting the test, contrary to what many of the clinicians had presumed.
Another important way that the clinicians thought Latinas’ religiousness may inhibit their use of prenatal testing was through religion-based attitudes about abortion. More than half the clinicians said that because many Latinas are Catholic, they are likely to be antiabortion, and therefore unwilling to terminate an affected pregnancy. A genetic counselor summed up this view:
I think that as a culture they generally, uh, don’t believe in terminating pregnancies. It goes against their faith and most Hispanics are Catholic, so that’s an issue to begin with, so most women who are Hispanic that I meet with would say that you were accepting what God is going to give.
However, in our patient interviews we see that many women who are against abortion did in fact accept the test. Most who had accepted also reported that they would not consider an abortion. Indeed, nearly everyone, accepters and decliners alike, said they would want the child regardless of its condition. This, incidentally, is consistent with findings reported by Browner and Preloran for a patient sample they interviewed in California.17 Thus it seems clear that any shared ideas about the acceptability of abortion that may exist among these women did not interfere with their willingness to have an amniocentesis.
Let us consider how this perception may affect clinical practice. Being able to choose to terminate an affected pregnancy was a primary reason given by the clinicians for providing access to prenatal testing in the first place. However, in the consultations we observed, the option of termination was brought up only about a quarter of the time, usually only in passing, and kept to vague, euphemistic terms. Could such omissions be a side effect of efforts to be culturally appropriate? Indeed, when asked about strategies for enhancing communication with Latina patients, many clinicians said they provide less information to these patients, by either simplifying the content of the consultation, or by excluding topics they think might be confusing or off-putting to Latinas, such as abortion or the nature of the anomalies in question. (For further discussion of this point, see Hunt et al.18).
Invoking what might be called “clinical myths” about Latino culture, the clinicians we interviewed shared many perceptions about Latino culture and its effect on amniocentesis decision making. As noted earlier, they described Latinas as likely to decline amniocentesis because they are religious, fatalistic, male-dominated, family-centered, hold superstitions and folk beliefs, and harbor fears rooted in misunderstandings and hearsay. These characterizations of Latino culture are consistent with how this group is commonly portrayed in the health literature. Claims abound in clinical journals regarding the presumed values, morals, and beliefs of Latinos. However, these cultural stereotypes are typically asserted in the absence of careful empirical examination of the actual prevalence of these characteristics or their salience to health care. (For critical discussions of this point see Hunt et al.19 and Lucas and Barrett.20)
Our study brings us to question the accuracy of how such cultural traits are assumed to influence Latino health behaviors. For example, the Latina patients we observed accepted amniocentesis at a rate comparable to that of non-Latina populations, more of those who expressed fears about amniocentesis accepted the procedure than did not, and none attributed birth anomalies to brujeria, mal de ojo, or other folk explanations commonly cited by clinicians. While these patients did consistently report discussing their decision with family members, their comments indicate that this by no means deterred them from accepting amniocentesis, nor that they are dominated by their husbands. Similarly, although most were Catholic and believed in the importance of God’s will, this did not seem pertinent to their accepting or declining the procedure.
Our analysis brings into focus the importance of avoiding cookbook approaches to cultural competency training. Although we do not know whether the clinicians in our study had formal training in cultural competency, or what type of training they might have had, they clearly have a shared notion of Latino cultural traits and their relevance to amniocentesis decision making. The inconsistency of these assumptions with what we learned in our patient interviews, and the apparent negative impact they have on the content of the consultations we observed, illustrate the pitfalls of relying on stereotypes. Cultural competency training that discourages stereotyping, and instead emphasizes open communication and negotiation with individual patients in light of their specific needs, would seem a more promising approach.1
Taken together with the growing literature documenting the negative impact of racial and ethnic stereotyping by caregivers on the health care of minority patients,21 our analysis leads us to question the practical impact of encouraging clinicians to take cultural factors into account in caring for minority patients, in the absence of comprehensive cross-cultural education. Recently developed training programs that encourage clinicians to become self-reflective in addressing cultural difference and to build patient–clinician alliances seem likely to help clinicians avoid such shortcomings.1,9,13,19,20,22,23
Of course, the current study was designed to be exploratory, and cannot begin to test hypotheses or make generalizations beyond the small group of patients and clinicians we observed. Future research with representative samples and comparison groups of non-Latinos, and of clinicians who have received different kinds of cross-cultural training, would be required to do so. Still, our analysis raises some disturbing questions about the accuracy of clinicians’ perceptions of cultural difference and the practical impact of their attempts to be culturally aware.
Could broad calls to become “culturally sensitive” be encouraging clinicians to provide some patients with less complete information? Could clinicians, by assuming certain cultural characteristics, be inadvertently promoting them? Could this be part of a larger pattern of omission, wherein common notions about ethnic culture lead clinicians to be less direct and less complete in caring for these patients? If such practices should prove to be widespread, this would indeed be a disquieting outcome of earnest efforts to provide culturally sensitive health care.
Increasing exhortations to become “culturally competent,” in the absence of careful training, could have the unintended effect of promoting a rather disturbing image of ethnic minority patients as cultural “others.” We are concerned that the cultural competency movement is barreling forward at an alarming pace, in the absence of careful, empirical assessment of its actual impact on patient care. A careful examination of how such efforts are actually manifested in clinical care would seem an essential step toward promoting true cultural competency.
The clinician interviews reported here were part of a larger study conducted in collaboration with Carole Browner of the University of California, Los Angeles. This research was supported by the National Center for Human Genome Research (1RO1 HG001384-01, Carole H. Browner and Linda M. Hunt as co-principal investigators). The research was also supported by a grant from the San Antonio Area Foundation.
The authors wish to thank the clinicians and patients whose kind cooperation made this research possible. Heide Castañeda, Julia Sargent, and Aaron Whiteford helped with a variety of data analysis and literature review tasks. The authors also wish to thank Carole H. Browner and Mabel Preloran for their integral involvement and thoughtful insights throughout all phases of the project. The authors also thank two anonymous reviewers, whose comments have greatly enhanced our argument.
An earlier version of this report was presented at the annual meetings of the American Anthropological Association in New Orleans, Louisiana, in 2002.