I am troubled by the approaches currently taken by most U.S. medical schools in their courses on what is commonly called “complementary and alternative medicine” (CAM). Although CAM therapies are anomalous practices whose claims for efficacy are either unproved or disproved, my research indicates that most medical schools do not present CAM material in a form that encourages critiques and analyses of claims. In this article I briefly present what I believe are the reasons for the unwarranted acceptance of CAM, give highlights of my findings about CAM in the curricula of U.S. medical schools, and describe a course that I teach at the Stanford University School of Medicine that approaches CAM with skepticism and critical thinking. I conclude with thoughts about needed reforms in CAM curricula.
UNWARRANTED ACCEPTANCE OF CAM
The past 30 years have seen increased interest in CAM. But its level of acceptance is not warranted, as many CAM claims have been convincingly disproved or remain unproved. One reason for CAM's high level of acceptance is the movement's attempt to alter standards for validity. Some CAM advocates propose expanding such standards to include the acceptance of historical traditions, subjectivity, ritual, personal experience, and transcendental experience.1 Others claim that their systems supply emotional support and cultural meaning to illness and that such support may improve course of specific diseases. Although some advocates accept common standards and participate in Cochvane Collaboration review for evidence, others reject accepted approaches of controlled clinical trials and other testing methods for traditional medicines and therapies as being not appropriate for certain areas of CAM.2
Even though some CAM concepts represent a sharp deviation from the accepted ways of evaluating medical therapies, an unidentified number of physicians and social scientists in the academic community support them. Indeed, many CAM claims have reached a high level of acceptance despite insufficient scientific support. As a result, government and large foundations generously fund the study of disproved methods and the integration of other methods.3 Unfortunately, some governmental panels have reached erroneous conclusions, while claiming to have based those conclusions on clinical trial evidence. A case in point is the practice of acupuncture, which was approved for certain uses by a consensus conference of the National Institutes of Health,4 although the predominance of evidence from the scientific literature demonstrates acupuncture's ineffectiveness except as a placebo or conditioning agent.5 Another example is the Agency for Health Care Policy and Research's recommendation of manipulation for back pain,6 even though the evidence for the usefulness of this therapy is largely negative. With inadequate approaches that fail to uphold criteria for validity and plausibility, so-called “evidenced-based” medicine remains fluid and loses its value to help physicians discern what is truly useful.
Other expert students of the field are speaking out with their opinions about CAM's lack of scientific validity. In a recent article published after the present article was written, Knipschild7 reviewed briefly his conclusions after developing and reporting numerous systematic reviews and meta-analyses of “alternative” methods. While opining that although findings are dubious, there is room for more rigorous trials of acupuncture (“Among the better trials some are positive but many others are negative”) and effectively dismissing homeopathy (“ridiculous”) and whole herb therapies (he reports that the data are inconsistent), he concluded, “I still believe that, in general, alternative treatments do not work.” CAM advocates coined the terms “alternative,” “complementary,” and “integrative” to increase CAM's acceptance. The terms obscure the fact that the methods they describe are, in fact, unproved or disproved.8 At the same time, advocates use slogans and myths to diminish the authority of biomedicine's scientific tradition. “Doctors do not know anything about nutrition” demeans standard medicine and medical education. “Cut, burn, and poison” demonizes cancer care. The myth that “only 10-15% of medicine is proved” reduces medicine's credibility and elevates the perceived worth of CAM methods.
CAM IN THE CURRICULUM
Because of the concerns stated above and my observation that many courses that had CAM components were being taught either from an advocacy view or from a “neutral” view—in neither case considering content validity—I surveyed U.S. medical schools in 1995-1997 to learn of their approaches to CAM in their curricula. Representatives of all 125 schools surveyed responded, either to the questionnaire or by phone.
The findings of the survey were not encouraging. Of the 56 schools that had some form of relevant course offering, only nine had invited critical lecturers on occasion; their courses were otherwise generally supportive of CAM. Two course directors claimed to present information “neutrally,” but did not teach critical methods or invite critical lecturers. Only four courses either presented a critical orientation or offered critical arguments in a way that significantly investigated advocacy arguments. (The complete findings of this survey are available upon request.)
Since the time of that survey, courses in medical schools concerning CAM have increased from 38 credit courses to 150 such courses in 70 schools.9 I suspect, however, that the level of skepticism and investigation of CAM claims has not risen from the level disclosed in the 1995-1997 survey. In fact, my own informal observations in November 2000 of 15 randomly selected courses at various medical schools revealed that none had added critical analysis to the course content.
In 1996 the Office of Alternative Medicine (OAM)—now the National Center for Complementary and Alternative Medicine (NCCAM)—convened a three-day meeting on professional school CAM education in the United States. The meeting reviewed various courses and methods of teaching. All sectarian systems and methods were implicitly accepted as effective. Teaching-method evaluation was not addressed, and concern about validity was mentioned only in one address by a dean of a medical school. Courses on acupuncture and chiropractic were accepted in the same context as other courses. OAM organizers did not invite the directors of critically-oriented courses to speak. I do not believe that NCCAM's approach to professional education shows much improvement from that demonstrated at the 1996 meeting.
THE STANFORD COURSE
In the 1970s student-sponsored lecture series at Stanford University School of Medicine featured speakers whose claims for alternative methods conflicted with rational thought and accepted knowledge. In 1979, after faculty indicated their interest in investigating the validity of CAM claims, I was encouraged to develop a course that I still teach in which students learn how to examine claims factually.
The course, Alternative Medicine—A Scientific Perspective, is modeled after two others existing at that time, one at the Loma Linda University School of Dentistry and one at San Francisco State University School of Public Health. The Stanford course operates on the general principle that science proceeds via a series of statements based on accurate observations and subsequent attempts at proof and disproof. Using this approach, claims for unproved methods are considered suspect. Students are taught to think critically about such claims and are introduced to tools for investigating them.
The course is given for two hours per week for nine weeks. The first segments explore sources of human error that contribute to misperceptions and to drawing incorrect conclusions. Guest psychologists discuss primitive modes of thought, perceptual and cognitive errors, cognitive dissonance, memory faults, and belief formation and perseverance—all associated with unproven medical belief systems. A magician demonstrates principles of misperception, misdirection, legerdemain—such as psychic surgery, and ideomotor action—such as dowsing, pendulum diagnosis, and mind reading. These and other sources of misperceived medical experiences form the bases for testimonials and unproved claims.
Other classes explore phenomena that contribute to placebo experience, including counterirritation, suggestion, expectation, consensus, conditioning and reinforcement, the natural history of illness, and regression toward the mean. Then others explore language distortion, myths, propaganda techniques, and cult-like behavior, the last sometimes presented by former cult members. Other subjects include mathematical and statistical approaches to coincidences, and probability problems, including prior probability and Bayes' theorem. The final sessions of the first half of the course concentrate on reading scientific and medical literature, principles of clinical trials, and the analysis of famous errors in science.
The last half of the course explores specific subjects, such as electromagnetic fields, vitamins and supplements, acupuncture, homeopathy, Laetrile, and other anomalous cancer therapies. Students learn methods for recognizing misinformation in medical literature. In some years, the course includes consideration of the role of dysfunctional and somatiform syndromes in CAM claims. The schedule is flexible, accommodating subjects such as books by Carlos Castaneda, quantum mechanics and consciousness, and the roles of prayer and “touch” therapies. In some years, students make a field trip to a “Whole Life Expo.”
Representatives of CAM systems present talks, followed by students' evaluations using methods learned in the first half of the course. (CAM lecturers understand that they cannot report their presentations on their curriculum vitae or in advertisements.) Students are encouraged to examine a specific method or substance of interest to them. Individual students have chosen on-site visits to acupuncture clinics and other offices, where they may receive treatments or merely observe—they then report their experiences to the class. They are encouraged to relate their subjective experiences and to analyze the material objectively, juxtaposing the two descriptions.
By the end of the course, students are expected to be able to gauge the validity of CAM claims, understand those claims' subjective attraction, and appreciate the values of truly complementary methods. These include relaxation, music, massage, art, poetry, body movement, and forms of meditation. These are methods that enhance adjustment, help reduce symptoms, and add a sense of meaning to the experience of illness.
I have outlined the components of this course above in hopes that others will be encouraged to use similar critical approaches when beginning or revising courses dealing with CAM.
NEED FOR REFORM
To my knowledge, there are no formal standards for teaching about CAM in medical schools. If a survey similar to mine were done today, it would probably reveal that most CAM information taught in medical schools continues to be ideologically or advocacy-based, or is taught without concern for a therapy's or system's validity. Medical schools have allowed a double-standard system for teaching. At one extreme, the standard allows instruction about CAM as a set of viable systems of knowledge that can be integrated into practice. And even if the course teaches only about CAM therapies—which is more common—it provides little or no critical analysis. The other standard, applied to the usual curriculum subjects, demands both basic scientific validity and clinical evidence-based confirmation.
A consortium of medical schools recently agreed to integrate CAM into their curricula through an evidence-based approach.10 However, evidence-based analysis of clinical trials is not sufficient to establish validity.10 Systematic reviews do not incorporate individual reports' errors, inconsistencies, misrepresentations, or subsequent refutations. Nor do they consider plausibility or prior probability based on non-trial data or basic scientific evidence. Such evidence points to the implausibility of “therapeutic touch,” homeopathic principles, and manipulative therapy, all of which require extraordinary clinical evidence prior to acceptance.
While I believe that the consortium's approach is one step in the right direction, it is not enough. It is time for all medical schools to make a concerted effort to formally include in their curricula ways to teach students to analyze and critically assess the content validity of CAM claims.