Complementary and alternative medicine (CAM) encompasses a bewildering array of healing philosophies, beliefs, and practices with little in common except for their unorthodox nature. The National Center for Complementary and Alternative Medicine at the National Institutes of Health (NIH) defines CAM as “those treatments and healthcare practices not widely taught in medical schools, not generally used in hospitals, and not usually reimbursed by medical insurance companies.”1 Some CAM practices are consonant with accepted biomedical theory and have demonstrated therapeutic value under controlled testing procedures (e.g., certain herbal remedies),2 whereas others have a more dubious theoretical basis whose healing benefits have been called into question (e.g., therapeutic touch).3
The public's embrace of these unconventional therapies is shown by recent studies of health care utilization patterns.4,5,6 Eisenberg et al.5 estimated that 42% of Americans used some form of unconventional therapy in 1997, for which they were willing to pay $27 billion in out-of-pocket expenses. From 1990 to 1997, the number of visits to CAM practitioners increased from 427 million to 629 million, which exceeded the total number of visits to all primary care physicians in the United States.5 Those who seek unconventional therapies appear to do so because they find them more in line with their own values, beliefs, and philosophical orientations toward health and disease.6 Clearly, CAM is offering something that many patients want but are not getting from conventional medical services.
The growing popularity of CAM is beginning to have an impact on medical education. In their 1997–1998 survey of all 125 U.S. medical schools, Wetzel et al.7 found that 64% of the 117 responding schools were teaching CAM topics either as stand-alone elective courses or as part of required courses. This is almost twice the number of institutions found offering CAM instruction in a 1995 survey (34% of 97 responding schools),8 which underscores the rapid acceptance of unconventional therapies in U.S. medical schools. Canadian medical schools have likewise incorporated CAM into their curricula—as of 1998, 81% were teaching CAM topics.9
Collectively, these surveys indicate that CAM has established a significant presence in the undergraduate medical curriculum. However, they provide few details about what is being taught, the amount of time devoted to specific topics, or the intent of such training (i.e., general background information versus application of specific therapies). Of particular interest is whether CAM topics are being presented in the context of uncritical advocacy or scientific evidence of therapeutic value.
We assessed the current state of CAM instruction in U.S. medical schools, with a particular emphasis on the issues raised above. To do so, we contacted individuals from a published listing of CAM course directors affiliated with medical schools accredited by the Liaison Committee on Medical Education (LCME)10 and requested information about the courses they taught. This enabled us to obtain more detailed information about course specifics than was reported in previous surveys.7,8
In August 2000, we mailed questionnaires to 123 CAM course directors at 74 U.S. medical schools. The names and addresses were drawn from a recent listing of CAM instructors compiled and maintained by The Richard and Hinda Rosenthal Center for Complementary and Alternative Medicine at Columbia University College of Physicians and Surgeons.10 Because our primary interest was in undergraduate medical education, we excluded individuals whose CAM instruction was limited to continuing medical education classes or postgraduate training programs. Follow-up mailings were sent to all non-responders in September 2000. Those who failed to respond to this second mailing were contacted by phone or e-mail in October 2000 and urged to return the survey. Questionnaires were collected through January 2001.
The two-page instrument consisted of nine questions with a check-box or fill-in-the-blank format, and one space at the end for written comments or clarifications. Questions were asked about the year of medical school in which the course was taught (first, second, third, or fourth), whether the course was required or elective, whether it was taught by a single instructor or multiple instructors, the total hours scheduled for the course (<20, 20–60, 61–100, >100), the name of the sponsoring unit, whether CAM practitioners or prescribers were involved in teaching, the instructional formats used (instructor lectures, guest seminars, group discussions, case studies, clerkship, Web-based, or other), the principal course objective (broad survey of CAM concepts, scientific evaluation of CAM effectiveness, practical training in CAM techniques, or other), and the specific topics covered and the time devoted to each (selected from a checklist of 19 CAM topics with spaces for additional topics to be provided by the respondent if needed). The questionnaire was designed to be completed in five to ten minutes.
Of the 123 individuals contacted, 73 returned completed questionnaires and 12 replied that they were not currently teaching CAM topics to medical students. This latter group included individuals who had moved, retired, or otherwise ceased teaching their courses. Because we were interested in the current state of CAM instruction, these 12 respondents were excluded from the analysis, yielding a response rate of 65.8% (73/111). The 73 respondents comprising our sample were affiliated with 53 medical schools, which represents 42.4% of the 125 LCME-accredited medical schools in the United States.
As shown in Table 1, about three fourths of the respondents indicated that they taught an elective CAM course, and about a third taught a required course. Our survey did not specifically ask whether the required courses were devoted entirely to CAM or whether they represented CAM material incorporated into larger, preexisting courses. Most CAM courses (79.5%) were team-taught and were offered in one or more years of the medical curriculum, with the first and fourth years predominating.
As a rule, these courses were fairly brief; half of the respondents indicated fewer than 20 contact hours (see Table 1). However, nearly a fifth of the respondents taught courses that had more than 60 contact hours. Clinical departments sponsored the majority (64.9%) of the courses taught by the respondents, with family medicine alone sponsoring a full third. Non-science units (e.g., medical humanities, dean's office, etc.) provided sponsorship for 16.2% of the courses. Relatively few courses (5.3%) were affiliated with basic science departments.
Virtually all of the respondents used either lectures by instructors or guest seminars to convey relevant information (see Table 2). Group discussions and case studies were likewise heavily employed. Nearly a fourth of the instructors taught CAM material during third- or fourth-year clerkships. A small portion of the instructors (4.1%) used the Internet for teaching. The most frequently mentioned “other” instructional formats were workshops, demonstrations, and office visits to CAM practitioners.
Most of the respondents (61.6%) taught courses whose primary objective was to provide students with a broad survey of CAM and its diversity (see Table 2). Many of the “other” course objectives cited by the respondents also seemed to emphasize a broad overview perspective, such as “expose students to philosophy of integrative medicine,” “holistic medical relationships and mind—body connections,” “demonstrate limits of the biomedical model,” “integration of CAM with conventional techniques,” “expose students to different views of health and disease,” and “explore spiritual issues in clinical practice.” About a fifth of the respondents offered practical training in the application of specific CAM therapies such as acupuncture. Few of the respondents (17.8%) considered a scientific evaluation of CAM's effectiveness to be a principal course objective. Over three fourths of the courses were taught by individuals identified as being CAM practitioners or prescribers of CAM therapies.
Of the 19 CAM topics specified in our survey, seven were taught by more than half of the respondents (see Table 3). These dominant topics were acupuncture, herbs and botanicals, meditation and relaxation, spirituality/faith/prayer, chiropractic, homeopathy, and nutrition and diets. The remaining topics were less common, ranging from 47.9% for ethnomedicine to 12.3% for energy medicine, music therapy, and reflexology. Any topic that was taught by fewer than 10% of the respondents was placed in the “other” category. Examples include legal considerations and business opportunities related to CAM, evidence-based medicine and CAM, and miscellaneous topics not otherwise specified (e.g., anthroposophy, the role of poetry and art, and specific CAM treatments for cancer, allergies, etc.).
Although the amounts of instructional time devoted to these topics varied considerably, the median numbers of contact hours were fairly uniform (see Table 3). Most topics received about two hours of instruction, which is consistent with the finding that most courses were general surveys, which by their nature would include a variety of CAM topics. However, the range of contact hours indicates that some topics required substantial time commitments from their instructors (e.g., 80 hours for acupuncture). These were probably elective courses devoted to single CAM topics. Several of the respondents (8.2%) indicated that the times allocated to specific topics varied and depended on the interests of the students.
Our mail survey of 73 course directors revealed the depth and diversity of CAM instruction currently offered in 53 U.S. medical schools. Unlike previous surveys,7,8 ours was not intended to estimate the prevalence of CAM instruction among all 125 U.S. medical schools, but rather to document the form, content, and purpose of such training in a substantial subset of these schools. Our use of a published listing of CAM course directors enabled us to contact directly those individuals who were uniquely able to provide the necessary information.10 As such, our sample of 73 respondents and the 53 schools they represent hardly constitutes a random sample of all possible CAM instructors in the nation's medical schools. If the individuals on this listing were not representative of CAM instructors elsewhere, or if those who responded were somehow atypical, then our results would be biased and not generalizable beyond this limited sample. To what extent our results might have differed if a more comprehensive sample could have been obtained is not known. Nevertheless, we believe that our results reflect a better-informed data source than had we relied on secondary sources of information, such as academic deans7 or department heads,8 who often lack detailed knowledge of course specifics.
According to our results, the “typical” CAM course offered by these respondents was a team-taught elective course sponsored by a clinical department. It was most likely to be taught in the first or fourth year of medical school, have fewer than 20 contact hours, and use more than one instructional format. These general course descriptors are similar to those reported for CAM courses in earlier surveys,7,8 which suggests that our respondents represented a fairly typical cross-section of CAM instructors, despite the potential bias of our non-random sample.
What is unique to our findings, however, is the greater detail with which we can describe the specific CAM topics being taught, the relative importance of each, and the intent of such training. Regarding the latter aspect, we find it troubling that so few of the respondents (17.8%) appear to have emphasized a critical perspective in evaluating CAM treatments and claims of therapeutic efficacy. In fact, only 8.2% of the respondents specifically mentioned that they included topics about evidence-based medicine in their courses. This may reflect the fact that most of the courses (78.1%) were taught by practitioners or prescribers of unconventional therapies. Although an instructor's use of a CAM therapy does not necessarily imply uncritical advocacy, it does imply that he or she believes a particular CAM treatment modality to have genuine merit. In this situation, then, the instructor may be less inclined to impart a critical perspective based on accepted standards of scientific evidence. Whether our findings truly reflect the state of CAM instruction nationwide is uncertain, but the apparent lack of a critical approach by most of our respondents is cause for concern.
The range of CAM topics being taught was broad, representing a continuum of unconventional beliefs and practices. The most prevalent topics (>50%) included therapies with nearly mainstream status (e.g., chiropractic), as well as more esoteric practices with little or no scientific support (e.g., homeopathy). Even the less prevalent topics (<50%) included examples of validated therapies (e.g., hypnosis), as well as those lacking experimental verification (e.g., therapeutic touch). Based on these data, there appears to be no overriding principle governing the selection of topics taught or their relative importance in the curriculum. We suspect that the inclusion of a given CAM topic reflects the background and interests of the instructor more than its demonstrated value as a therapy. To be fair, however, any survey course intended to introduce medical students to the diversity of CAM would logically include a broad array of “popular” CAM topics for discussion. And, because most of the respondents teach general survey courses, this may partly account for the observed diversity of CAM topics. Similar to our findings, Carlston et al.8 reported that relaxation techniques, acupuncture, biofeedback, chiropractic, and hypnosis were the top five CAM topics being taught in 33 U.S. medical schools in 1995.
One of the major rationales for teaching medical students about CAM is that consumers are using unconventional therapies in ever-greater numbers and their physicians must be able to safely treat them with an awareness of the possible harm some of these therapies pose.11 Knowledge of CAM is, therefore, essential. Eisenberg et al.5 found that most patients (>60%) who use unconventional therapies do not disclose this fact to their physicians. Moreover, of those patients taking prescription medications, nearly a fifth are also taking an herbal product or a highdose vitamin.5 The potential for an adverse drug reaction is significant, but it is unlikely that most physicians would be alerted to this concurrent drug use or recognize the health risk even if they were aware of it. For example, in a survey of 500 surgical patients, 51% took some form of alternative medicine (i.e., herbs, vitamins, dietary supplements, or homeopathic substances) in the two weeks before surgery.12 The authors' review of the literature revealed that 27% of these patients consumed substances that potentially could interact with anesthetics to inhibit coagulation, affect blood pressure, cause sedation, produce cardiac effects, or alter electrolytes. Physicians should be trained in the fundamentals of CAM so they can elicit the relevant information from their patients, assess the potential risks, and guide the treatment accordingly. Patients will inevitably question their doctors about the appropriateness of certain CAM therapies, and physicians need to be sufficiently versed to advise them intelligently.
A recent survey of medical students found that most had a positive attitude toward unconventional therapies and believed them to be useful treatments with little potential for harm.13 Like the public at large, these students' knowledge of CAM was derived mostly from anecdotal and media-based information. Such findings highlight the need for objective, scientifically-based information about CAM in the undergraduate medical curriculum. As medical educators, we must fulfill this need without appearing to adopt a dogmatic stance to defend the medical orthodoxy. Our students should be trained to consider the evidence both for and against a given CAM therapy, critically evaluate the source and quality of supporting data, and appraise the therapy's potential for harm when used alone or in combination with conventional therapies.
Based on our study's findings, we offer a few suggestions that may help achieve these teaching goals.
- Emphasize a critical evaluation of the scientific literature. This suggestion has been advanced before,7,14 but it bears repeating. When evaluating any claim of therapeutic efficacy, unconventional or otherwise, students should be familiar with the rules of causal inference15 and be able to apply them critically to the available evidence.
- Enlist the involvement of basic science departments. Faculty with expertise in experimental design and statistical analysis of data can help impart a critical perspective to the course, thereby fostering a better appreciation of medicine's scientific basis.
- Avoid advocacy of unproven therapies. Proponents of CAM abound, and they may have something of value to teach, but their conviction and enthusiasm alone should not substitute for rigorous evidence. The teaching of CAM is too important a task to be left solely in the hands of CAM enthusiasts. All CAM courses, whether required or elective, should be held to the same academic standards as other courses and should require curriculum committee approval.
We believe that CAM instruction in U.S. medical schools will continue to grow as more faculty and students come to share the same fascination with unconventional therapies as the society at large. Our obligation as medical educators dictates that we teach our students what the best available evidence has to offer, mindful of the scientific basis underpinning modern medicine.
2. Linde K, Mulrow CD. St John's wort for depression. Cochrane Database System Review. 2000;2:CD000448.
3. Rosa L, Rosa E, Sarner L, Barrett S. A close look at therapeutic touch. JAMA. 1998;279:1005–10.
4. Eisenberg DM, Kessler RC, Foster C, Norlock FE, Calkins DR, Delbanco TL. Unconventional medicine in the United States: prevalence, costs, and patterns of use. N Engl J Med. 1993;328:246–52.
5. Eisenberg DM, Davis RB, Ettner SL, et al. Trends in alternative medicine use in the United States, 1990–1997: results of a follow-up national survey. JAMA. 1998;280:1569–75.
6. Astin JA. Why patients use alternative medicine: results of a national survey. JAMA. 1998;279:1548–53.
7. Wetzel MS, Eisenberg DM, Kaptchuk TJ. Courses involving complementary and alternative medicine at U.S. medical schools. JAMA. 1998;280:784–7.
8. Carlston M, Stuart MR, Jonas W. Alternative medicine instruction in medical schools and family practice residency programs. Fam Med. 1997;29:559–62.
9. Ruedy J, Kaufman DM, MacLeod H. Alternative and complementary medicine in Canadian medical schools: a survey. Can Med Assoc J. 1999;160:816–7.
10. Bhattacharya B. M.D. programs in the United States with complementary and alternative medicine education opportunities: an ongoing listing. J Altern Complement Med. 2000;6:77–90.
11. Straus SE. Complementary and alternative medicine: challenges and opportunities for American medicine. Acad Med. 2000;75:572–3.
12. Norred CL, Zamudio S, Palmer SK. Use of complementary and alternative medicines by surgical patients. J Am Assoc Nurse Anesth. 2000;68:13–8.
13. Hopper I, Cohen M. Complementary therapies and the medical profession: a study of medical students' attitudes. Alternative Ther Health Med. 1998;4:68–73.
14. Kligler B, Gordon A, Stuart M, Sierpina V. Suggested curriculum guidelines on complementary and alternative medicine: recommendations of the Society of Teachers of Family Medicine Group on Alternative Medicine. Fam Med. 2000;31:30–3.
15. Rothman KJ, Greenland S. Causation and causal inference. In: Modern Epidemiology. Philadelphia, PA: Lippincott Williams & Wilkins, 1998.