Complementary and Alternative Medicine Use ... - Wiley Online Library

SHELLEY R. ADLER

Department of Anthropology, History, and Social Medicine University of California at San Francisco

Complementary and Alternative Medicine Use among Women with Breast Cancer The legacy of 19th-century social theories applied to the study of nonmainstream treatment use continues to affect contemporary research into complementary and alternative medicine (CAM). Quantitatively based studies of CAM use have been hindered by the lack of an adequate lexicon, inaccurate characterizations of the people who use CAM, and underestimates of the prevalence of usage. Results from a qualitative prospective cohort study challenge previous stereotypes by indicating that CAM usage does not increase dramatically with the initial diagnosis of cancer and that younger women are more likely to use CAM than older women. Qualitative research methods are uniquely appropriate for obtaining accurate information about health practices that, despite growing acceptance in some areas of society, are still viewed as outside of the mainstream, [complementary and alternative medicine, breast cancer, qualitative health research]

T

he present debate over the merit of "alternative" approaches to cancer therapy is not entirely new: much of the discourse represents the late-20thcentury version of the historical conflict between orthodox and heterodox healing systems in the United States (Baer 1989). The existing tensions among the various healing traditions in our medically pluralistic society have increased with the development of patient empowerment and medical consumerism, as well as recent changes in the organization of health care delivery, including managed care (Adler et al. 1998). Additionally, the publication of research findings regarding the prevalence of usage and the cost of nonbiomedical treatments (e.g., Eisenberg et al. 1993) has served to focus both popular and scientific attention on the subject. The renewed interest in alternative therapies on the part of biomedical science represents a belated awareness of steadily persisting parallel health belief systems more than a sudden increase in usage of these treatments. For many physicians the current debate is no longer over whether alternative medicine is wholly quackery or

Medical Anthropology Quarterly 13(2): 214-222. Copyright © 1999 American Anthropological Association.

21

COMPLEMENTARY AND ALTERNATIVE MEDICINE USE

215

panacea, but rather the degree to which various alternative therapies should be sanctioned by biomedicine. Not surprisingly, the lack of a consensual language with which to discuss alternative therapies contributes to the problems of conducting research into this contested, value-laden, and politically charged area (Cassidy 1995). The precise definition of complementary and alternative medicine has been the topic of contentious debate in recent years as nonbiomedical treatments have increasingly drawn the attention of both researchers and the lay public. Individual investigators have more often than not avoided the central issues by crafting definitions that suit the purposes of their own particular projects, an arbitrary practice that complicates the process of comparing research findings, discussion participants continue to attribute vastly different meanings and connotations to even the most commonly used terms—complementary, alternative, and traditional. In April 1995 the Office of Alternative Medicine (recently renamed the National Center for Complementary and Alternative Medicine) of the National Institutes of Health held a conference on research methodology. The following definition of complementary and alternative medicine was developed for purposes of identification and research: Complementary and alternative medicine (CAM) is a broad domain of healing resources that encompasses all health systems, modalities, and practices and their accompanying theories and beliefs, other than those intrinsic to the politically dominant health system of a particular society or culture in a given historical period. CAM includes all such practices and ideas self-defined by their users as preventing or treating illness or promoting health and well being. Boundaries within CAM and between the CAM domain and the domain of the dominant system are not always sharp or fixed. [Panel on Definition and Description 1997:50] According to this definition, therefore, CAM in the present-day United States is the "broad domain of all healthcare resources to which people have recourse other than those intrinsic to biomedicine and the specific theoretical and practice models of biomedicine" (Panel on Definition and Description 1997:51). By situating CAM in a sociopolitical context, the Panel's definition comes closest to doing justice to this large, diverse, and dynamic field. In addition, because of its lack of temporal or geographic referents, the term can be used to describe healing systems and practices in any location and at any point in time. This definition of CAM has been selected to inform the present ongoing study of women's choices and use of nonbiomedical treatments for breast cancer. Three topics relevant to CAM research are addressed: the major methodological flaws in prior scientific research into CAM usage, initial findings from the Women's Breast Cancer Treatment Choices Study; and the essential contribution of qualitative methodologies to CAM research. CAM Research in Context Previous research into CAM treatment usage for cancer has encountered a few problematic areas, most notably: unclear and inconsistent definitions of what constitutes CAM; nonempirically derived characterizations of the individuals and populations who use CAM; and underestimates of the prevalence of CAM use. For purposes of expedience, as Sharma (1993) observes, most investigators have limited

216

MEDICAL ANTHROPOLOGY QUARTERLY

the domain of CAM cancer treatments that they study in two ways: the treatments have been restricted to a specific number of therapies derived from the literature, from pilot studies, or intuitively (e.g., Eidinger and Schapira 1984; Eisenberg et al. 1993; Faw et al. 1977; Louis Harris and Associates 1987; Yates et al. 1993) and are usually those with the greatest structural similarity to biomedicine (i.e., practitioner dependent and clinic oriented). It is not clear in several of the studies to what extent the investigators' catalog of CAM treatments reflects informants' actual knowledge or usage. Because of lack of familiarity with the primary modes of CAM healing and a bias toward practitioner-initiated care, most investigators ignore or intentionally exclude nonbiomedical treatments that are forms of self-care. Other healing methods, such as prayer or exercise, also have been excluded from most previous investigations because of the difficulty of ascertaining their role through the quantitative data collection techniques used (e.g., Eisenberg et al. 1993). Historically, folk medical and ethnomedical research in the United States has been conducted among marginal or peripheral communities, to the exclusion of more mainstream groups. In large part this reflects a long-standing scientific misconception that it was among marginal rather than mainstream groups that populations of users of folk medicine could be found for study. From at least the beginning of the present century, the notion has been prevalent—in both the health professions and much of nonmedical academia—that folk and popular systems of health beliefs and behaviors are gradually dying out with the advent of modern, Western medicine. An evolutionary model of the development of health systems has been to a large extent responsible for this perception of the waning popularity of nonbiomedical health practices. "Progress," implicitly defined as "increasing similarity to the culture of the scholar" (Hufford 1983:307), was viewed as a natural and inevitable process. Discarded, obsolete ideas drifted down and were preserved in the sediment of the lower layers of culture (gesunkenes Kulturgut) (Hufford 1988:228; Hultkrantz 1960:158-159). The legacy of 19th-century social theories, which incorporated positivist assumptions of the unilinear evolutionary process of human thought, remains remarkably influential today. Viewed in this context, the persistent fascination with recent reports of the high prevalence of CAM usage (Eisenberg et al. 1993) and the "conventional" characteristics of CAM users (Cassileth et al. 1984; Eisenberg etal. 1993; McGuire 1988) becomes clear: these research findings challenge the stillprevalent notion of the user of nonbiomedical therapies as marginalized. Evolutionist and survivalist assumptions have sustained a series of stereotypes of participants in nonbiomedical health systems as being socially marginal. As folklorist Bonnie Blair O'Connor explains, the stereotypes of marginality typically include one or more of the following features: [G]eographic remoteness or isolation (the image of rural Appalachia is frequently invoked); recent immigration or minimal acculturation to core American culture; ethnic minority membership or strong ethnic self-identification or group affiliation; poverty or low socioeconomic status; low formal educational attainment; mental or emotional imbalance: or desperation induced by grave illness or poor outcomes of conventional therapeutic efforts. [O'Connor 1995:17]


217

I add three characteristics to this list of common stereotypes: being a woman; being elderly; and being gullible (see Pepper 1984 for classic examples of the belief in these stereotypes). Scientific research, too, has contributed to the marginal stereotype of the CAM user by focusing on nonrepresentative groups of people (e.g., Brown 1975; Durant 1991; Ingelfinger 1976). Convenience sampling has resulted in study populations that are poorly differentiated in terms of ethnicity and age. Prevalence studies have included people who use CAM for a variety of different conditions—from short-term discomforts to chronic or life-threatening illness—thus confounding the participants' reasons for and satisfaction with the choice of CAM treatments. Additionally, clinic-based samples, which continue to comprise the most common study populations, can introduce recruitment bias in terms of the types of patients that physicians allow researchers to approach (e.g., "compliant" patients). Finally, there is a "contextual bias" that can arise when informants are interviewed by biomedical professionals in a clinic or hospital, locations at which "an adherent is least likely to feel at ease discussing beliefs and practices deemed 'deviant' by the larger society" (McGuire 1988). Despite the apparent trend toward greater acceptance or at least tolerance of some CAM treatments, the majority of people (70 percent of patients, according to Eisenberg et al. 1993) do not feel comfortable discussing their CAM use with their physicians. Another problem in CAM research has been the difficulty of ascertaining accurate prevalence rates. Estimates of the percentage of people using CAM cancer treatments in a variety of populations range from 7 to 45 percent (e.g., Arkko et al. 1980; Berger 1989; Cassileth et al. 1984; Downer et al. 1994; Eidinger and Schapira 1984; Eisenberg et al. 1993; Gjemdal 1979; Lerner and Kennedy 1992; Morant et al. 1991; Zouwe et al. 1994). Most estimates of the prevalence of CAM cancer treatment usage in the United States are below 24 percent. This broad range of estimates appears to be directly related to each project's chosen methodology. The continuum of research findings from low to high prevalence of usage is paralleled by a methodological continuum ranging from quantitative telephone survey research to qualitative, face-to-face interviews. Additionally, and not surprisingly, studies that utilize narrow definitions of CAM culled from the biomedical literature report lower rates of usage than investigations that rely on broader, emically derived definitions. Thus, problems with unclear definitions, nonrepresentative study populations, and underestimated prevalence rates have hindered progress toward a better understanding of the nature of CAM selection and use.

Methods In an effort to fill some of the gaps in our knowledge of who uses CAM treatments, which treatments are selected, and for what purposes, a qualitative prospective cohort study is being conducted with women from six ethnic1 groups and two age cohorts in San Francisco, California. The population-based study sample consists of 86 women who have received a histological diagnosis of breast cancer. The participants were identified through the California Tumor Registry and consist of San Francisco residents diagnosed at San Francisco hospitals between May 1995 and May 1996.

218


In order to interview informants as soon as possible after diagnosis while avoiding the selection biases of a clinical sample, participants were recruited using a population-based rapid case-finding technique. A team of case finders from the Northern California Cancer Center was sent to the medical records offices of every hospital in San Francisco County every two to three weeks for 13 months. The team assembled a list of all women meeting the inclusion criteria by reading new pathology reports, supplemented by hospital records. Next, the patients' physicians were telephoned to determine whether there were any medical contraindications to contacting the patients.2 When there were no medical contraindications, which was the situation in all but 15 of the cases, the women were sent introductory letters, followed one week later by a recruitment telephone call. Eligibility requirements for study participants included the following: ages 35-49 or 60-74, inclusive; self-identification as African, Chinese, European, Filipina, Japanese, or Latina/Hispanic American; residence in San Francisco, California, at the time of diagnosis; ability to be interviewed in English, Spanish, or Chinese (Cantonese or Mandarin); histological diagnosis of breast cancer (in situ or invasive); diagnosed at a hospital within San Francisco County; diagnosed during the 13-month period of case ascertainment (May 1995-May 1996). Excluded from the study were women with recurring breast cancer and patients whose physicians refused to grant permission for them to be contacted. Eighty-six women were enrolled, 87 percent of those known eligible.3 Data are collected through a series of four in-depth, semi structured interviews with women recently diagnosed with breast cancer. Informants are initially interviewed within a few months of diagnosis (two to four months on average) and again after 6, 18, and 30 months. The face-to-face interviews take place in participants' homes or at a location of their choosing. The study utilizes a semi structured interview guide that is specifically adapted for each of the four interview cycles. The open-ended questions are designed to encourage informants to articulate their personal understandings of their beliefs and practices regarding health and illness. The domains of questions asked during the initial interview include: ideas about health and illness generally (such as personal understandings of good health); the evolution of the individual's awareness of her illness (the discovery and confirmation of the breast cancer); the individual's perception of the nature and extent of her illness; biomedical and/or CAM cancer treatments utilized to date; and experiences with and attitudes toward biomedicine, CAM, and their respective practitioners. All interviews are audiotaped and transcribed verbatim. Cantonese, Mandarin, and Spanish interviews are translated by the interviewers and then transcribed. Qualitative content analysis is proceeding with the use of QSR*NUDIST software. Findings Initial findings indicate a higher prevalence of CAM usage than is usually reported in the biomedical literature. Sixty-nine percent of the informants had used a CAM treatment or modality before they were aware of their breast cancer diagnosis. Since women are interviewed an average of two to four months after their initial diagnosis, their recall regarding treatment usage prior to the diagnosis presumably is excellent. The fact that so many women used CAM treatments before they


219

knew they had breast cancer challenges the stereotype of the "desperate" patient who is willing to try anything after receiving such a diagnosis. Interestingly, the percentage of informants that reported using CAM treatments specifically targeted to their breast cancer in the first few months after diagnosis is 72 percent. Although this figure is much higher than previously reported in the literature, it does not represent a significant increase from prediagnosis CAM usage by the study participants. At least in the first few months after initial diagnosis, therefore, most participants appear to maintain their prediagnosis health behaviors with regard to CAM. The study findings also challenge assumptions that older women are more likely to engage in general CAM treatment use. In terms of the two age cohorts, 58 percent of women aged 60-74 used CAM treatments in contrast to 78 percent of women aged 35-49. (There was no significant postdiagnosis change by cohort.) The lower prevalence of usage among women in the older age cohort runs counter to the conventional stereotype of CAM users as older and more gullible. These results confirm findings that people who use CAM tend to be younger and have relatively more resources with which to explore and obtain nonbiomedical treatments (Cassileth et al. 1984; Eisenberg etal. 1993). Contributions of Qualitative Research Methodologies to the Study of CAM Qualitative science is ideally and uniquely situated to investigate the wide variety of CAM systems, including modalities, practices, theories, and beliefs. As Claire Cassidy explains, qualitative research methodologies display a high degree of "model fit"; that is, they comprise "research designfs] and techniques [that] fit the explanatory model/s of the study population/s" (Cassidy 1995:35). By sharing what amounts to a common perspective with many CAM systems, qualitative inquiry avoids many of the problems that can arise with mismatched methodological techniques. Two of the particular strengths of qualitative methodologies applied to CAM research are the emphasis on emic perspectives and a holistic style of inquiry. In a discussion of the premise of folkloristics that also applies to qualitative research, David Hufford explains that "ordinary people tend to be underestimated and . . . their knowledge tends to be discredited by authorities" (Hufford 1990). By promoting participants' views and understandings, qualitative researchers can interpret CAM phenomena in terms of the meanings that people bring to them. Through the use of unstructured or semi structured interviews, informants are given the opportunity and encouraged to describe their beliefs and practices using their own logic and terminology. This emphasis on the emic perspective is the only way to obtain an accurate understanding of individuals' reasons for the selection and use of CAM modalities. Second, and perhaps most strikingly, the holistic nature of qualitative inquiry blends well with the holistic model of many CAM healing systems (Cassidy 1995). Just as many CAM systems do not include the traditional biomedical hierarchy of doctor-patient, for example, in qualitative inquiry the researcher-participant dyad is constructed to promote a more egalitarian relationship. Participants' "subjective" views are valued as expert and informants' conceptualizations and models are integrated into research design. The primacy of meanings, relationships, and processes

220


is shared by traditional qualitative inquiry and many CAM modalities, practices, theories, and beliefs. Because of its unique strengths, some of the areas in which qualitative research methodology is particularly appropriate for CAM research are physicianpatient and CAM practitioner-client communication, quality-of-life issues in cancer treatment, patient activism, and placebo/nocebo effects. Although CAM is increasingly being explored from a biomedical perspective, it is important to remember that, whether conceptualized as ethnomedicine or folk medicine, CAM has traditionally been studied within the domains of medical anthropology and folkloristics. Qualitative inquiry has a long history of providing the appropriate methodological tools to researchers of complementary and alternative medicine. NOTES

Acknowledgments. This research has been supported by joint funding from the National Cancer Institute and the National Institute on Aging (CA64634) and a small grant from the University of California, San Francisco, Academic Senate Committee on Research. Cancer incidence data used in this research have been collected by the Northern California Cancer Center under contract N01-CN-05224/25482 with the Division of Cancer Prevention and Control, National Cancer Institute, National Institutes of Health, Department of Health and Human Services, and under subcontract 0501-8701/8708-S0207 with the California Public Health Foundation. I am deeply grateful to the women with breast cancer who are generously sharing their thoughts and experiences by participating in this project: their altruism, openness, and courage make this research possible. Correspondence may be addressed to the author at University of California, San Francisco, Department of Anthropology, History, and Social Medicine, 3333 California Street, Suite 485, San Francisco, CA, 94143-0850. 1. The problematic term ethnicity is used here for purposes of expedience and to indicate that a representative sample of women's experiences is reflected in the study's two age cohorts. Although participants self-identified as belonging to different social categories—popularly referred to as "ethnic groups"—the sample is not divided along these lines for the purposes of analysis. Since so little is known about the behaviors involved in selecting CAM treatments for breast cancer, the project's goal is to study a representative sample of the population, not to compare behaviors by race or ethnicity. 2. The University of California at San Francisco's institutional review board required this step. 3. This remarkable response rate was not unexpected. Epidemiological studies using rapid case ascertainment methods also have reported high participation rates. Despite an extremely light-handed recruitment strategy, timing the initial contact of potential informants within a few months of diagnosis seems to have improved, rather than lowered, the response rate. Several of the informants indicated that they chose to participate precisely because they were in the midst of a mental and emotional challenge and, therefore, welcomed the opportunity to express themselves. REFERENCES CITED

Adler, Shelley R., Sarah A. McGraw, and John B. McKinlay 1998 Patient Assertiveness in Ethnically Diverse Older Women with Breast Cancer: Challenging Stereotypes of the Elderly. Journal of Aging Studies 12(4):331-350.


221

Arkko, Pertti J., Birgit L. Arkko, Onni Kari-Koskinen, and Pentti J. Taskinen 1980 A Survey of Unproven Cancer Remedies and Their Uses in an Outpatient Clinic for Therapy in England. Social Science and Medicine 14A:511-514. Baer, Hans A. 1989 The American Dominative Medical System as a Reflection of Social Relations in the Larger Society. Social Science and Medicine 28(11): 1103-1112. Berger, R. D. P. 1989 Tumour Patients and Alternative Medicine: An Attempt at Characterizing Users of Alternative Methods in Oncology. Dtsch Med Wschr. 114: 323-330. Brown, Helene 1975 Cancer Quackery: What Can You Do about It? Nursing 5(5):24-26. Cassidy, Claire M. 1995 Social Science Theory and Methods in the Study of Alternative and Complementary Medicine. The Journal of Complementary and Alternative Medicine 1(1): 19-40. Cassileth, Barrie R., Edward J. Lusk, Thomas B. Strouse, and Brenda J. Bodenheimer 1984 Contemporary Unorthodox Treatments in Cancer Medicine: A Study of Patients, Treatments, and Practitioners. Annals of Internal Medicine 101:105-112. Downer, S. M., M. M. Cody, P. McCluskey, P. D. Wilson, S. J. Arnott, T. A. Lister, and M. L. Slevin 1994 Pursuit and Practice of Complementary Therapies by Cancer Patients Receiving Conventional Treatment. British Medical Journal 309:86-89. Durant, John R. 1991 If It Quacks.... Cancer 67:2225-2226. Eidinger, Richard N., and David V. Schapira 1984 Cancer Patients' Insight into Their Treatment, Prognosis, and Unconventional Therapies. Cancer 53:2736-2740. Eisenberg, David M., Ronald C. Kessler, Cindy Foster, Frances E. Norlock, David R. Calkins, and Thomas L. Delbanco 1993 Unconventional Medicine in the United States: Prevalence, Costs, and Patterns of Use. The New England Journal of Medicine 328(4):246-252. Faw, Cathy, Ron Ballentine, Lois Ballentine, and Jan van Eys 1977 Unproved Cancer Remedies: A Survey of Use in Pediatric Outpatients. Journal of the American Medical Association 238:1536-1538. Gjemdal, T. 1979 The Use of Alternative Medicine among Patients. Journal of the Norwegian Medical Association 99:883-886. Hufford, David J. 1983 Folk Healers. In Handbook of American Folklore. Richard M. Dorson, ed. Pp. 306-313. Bloomington: Indiana University Press. 1988 Contemporary Folk Medicine. In Other Healers: Unorthodox Medicine in America. Norman Gevitz, ed. Pp. 228-264. Baltimore and London: Johns Hopkins University Press. 1990 Folklore and Medicine. In Putting Folklore to Use. Michael O. Jones, ed. Pp. 117-135. Lexington: University Press of Kentucky. Hultkrantz, Ake, ed. 1960 International Dictionary of Regional European Ethnology and Folklore, vol. 1. Copenhagen: Rosenkilde and Bagger. Ingelfinger, F. J. 1976 Quenchless Quest for Questionable Cure. The New England Journal of Medicine 295(15):838-839.

222


Lerner, Irving J., and B. J. Kennedy 1992 The Prevalence of Questionable Methods of Cancer Treatment in the United States. CA: A Cancer Journal for Clinicians 42(3): 181-191. Louis Harris and Associates 1987 Health Information and the Use of Questionable Treatments: A Study of the American Public. Washington, DC: U.S. Department of Health and Human Services. McGuire, Meredith B. 1988 Ritual Healing in Suburban America. New Brunswick, NJ: Rutgers University Press. Morant, R., W. F. Jungi, C. Koehli, and H. J. Senn 1991 Why Do Cancer Patients Use Alternative Medicine? Schweiz Med Wschr. 121:1029-1034. O'Connor, Bonnie Blair 1995 Healing Traditions: Alternative Medicine and the Health Professions. Philadelphia: University of Pennsylvania Press. Panel on Definition and Description 1997 Defining and Describing Complementary and Alternative Medicine. Alternative Therapies 3(2):49-57. Pepper, Claude 1984 Quackery: A $10 Billion Scandal—A Report by the Chairman of the Subcommittee on Health and Long-Term Care. Washington, DC: U.S. Government Printing Office. Sharma, Ursula 1993 Contextualizing Alternative Medicine: The Exotic, the Marginal, and the Perfectly Mundane. Anthropology Today 9(4): 15-18. Yates, Patsy M., Geoffrey Beadle, A. Clavarino, Jake M. Najman, Damien Thomson, Gail Williams, Lisbeth Kenney, Sydney Roberts, Bernard Mason, and David Schlecht 1993 Patients with Terminal Cancer Who Use Alternative Therapies: Their Beliefs and Practices. Sociology of Health and Illness 15(2): 199-216. Zouwe, N., F. S. A. M. Dam, N. K. Aaronsen, and G. J. F. P. Hanewald 1994 Alternative Therapies in Cancer: Prevalence and Backgrounds. Ned Tijdschr Geneeskd 138:300-306. Accepted for publication March 23,1998