In a previous post I described a lecture given by a faculty member to first-year medical students on my campus introducing us to integrative medicine (IM). Here I describe his lecture to the second-year class on legal and ethical aspects of complementary and alternative medicine (CAM).

Dr. P began his lecture by describing CAM using the now-familiar NCCAM classification. He gave the NCCAM definition of CAM as “a group of diverse medical and healthcare systems, practices, and products that are not presently considered to be part of conventional medicine.” To illustrate how this definition can lead to surprises, he asked us if the therapeutic use of maggots is CAM or conventional. Although it sounds rather CAM-ish, maggot therapy is used at some surgical centers for wound debridement, he told us, and therefore is part of “conventional medicine.”

I continue to be surprised that thoughtful fans of CAM use this garbage pail definition (anything that doesn’t fit into conventional medicine), because it means that (1) no quackery can be easily excluded from the CAM tent and (2) the only common thread among the sundry modalities is a lack of acceptance from the mainstream medical community. One of Dr. P’s case studies of ethical challenges in CAM use involved a patient’s family that insisted on treating his persistent vegetative state with hyperbaric oxygen therapy — the utility of which Dr. P was highly doubtful — and later Dr. P mentioned intravenous chelation therapy as an example of a CAM treatment that he considered high-risk (compared to, say, parsley supplements). However, both hyperbaric oxygen and IV chelation have uncontroversial indications. Why are they called CAM when used in certain diseases? Presumably only because they are so poorly supported or implausible in those contexts as to be shunned by the medical community, the same community that apparently has adopted as weird (dare I say, unconventional) a therapy as maggot debridement. That maggots won acceptance leads me to be a priori skeptical of any therapy currently defined as CAM, particularly an old one (it still hasn’t been proven). Dr. P appears more optimistic than I, because he seems to interpret the same story as a reason to stay open-minded about most CAM therapies (it still might be proven).

Like the last time, Dr. P stressed that advocacy for use of CAM/IM was not a goal of his lecture. I infer from many of the points and tangential comments he made throughout the lecture, however, that he has a goal of normalizing CAM/IM for us. A frequent refrain was that some ethical precept or legal consideration regarding CAM use was “just the same” as for conventional medicine. Often I agreed with him — all therapeutic options should meet the same standards of efficacy and safety, and many ethical or legal concerns are common for either conventional or alternative approaches — but some of his comparisons seemed questionable:

• For example, after acknowledging that “generally recognized as safe” substances may have unexpected side effects when taken as high-dose supplements (he gave the example of ephedra), Dr. P suggested that in many cases the physician can monitor a patient for side effects while using the supplement, just like we do for patients using pharmaceuticals. Outside of a research context, I would argue that most such monitoring is done for previously characterized adverse effects (e.g., known effects of statins or neuroleptics) and as such can be done more effectively and efficiently than screening for unknown reactions from a mystery drug.
• At one point he mentioned that we accept the 10,000 yearly deaths from NSAIDs in exchange for its benefits; again, it seems to me that the benefits of aspirin are well-defined and understood, in contrast to CAM.
• Dr. P also compared use of CAM to off-label use of pharmaceuticals, implying that both practices are, in a sense, unconventional. Is this a fair comparison?
• He further mentioned that some unethical CAM providers may be more concerned about the health of their income stream than the health of their patients, as evidenced by recommending products they sell. He pointed to similar behavior on the “conventional” side with the example of internists giving cosmetic Botox injections to supplement their low reimbursement rates. I thought this was a bizarre example and a completely false analogy; we could argue about appropriateness, but no one is mislead about health benefits of such a procedure.
• Finally, Dr. P talked about how an IM doc would refer patients to qualified CAM practitioners, much like all physicians refer to appropriate specialists when a patient needs care beyond the scope of their practice. He mentioned, for example, that he will refer a patient to a licensed acupuncturist or a credentialed Reiki master if the patient wants those therapies. Professor Edzard Ernst’s infamous quote regarding regulation of CAM comes to mind.

The bulk of the lecture was devoted to how a physician should respond to a patient request for or use of CAM. Dr. P began by asserting that a physician should neither categorically refuse nor automatically agree to cooperate with CAM administration. Each request should be evaluated in terms of risks and benefits, with consideration given to the patient’s beliefs, cultural values, therapeutic goals, and severity of illness. In doing so, physicians should uphold the ethical principles of autonomy, beneficence, non-maleficence, and justice. Dr. P spoke well on enabling patient autonomy, which he said involved correcting misinformation as much as supporting unconventional choices. When a patient comes to Dr. P with a bag of supplements, Dr. P will go through them and ask non-judgmentally for each, “Why do you take this?” He is then able to gently correct misconceptions or guide towards better information sources, and he has more luck paring down the list when he is knowledgeable about and accepting of the ones more likely to be useful. (Glucosamine? Sure, that may help. But I’m not so sure about this shark cartilage.) Beneficence and non-maleficence speak to risk-benefit considerations; here was the line about deaths from NSAIDs. Justice is about societal fairness and access to care. Dr. P expressed frustration that proven CAM interventions like acupuncture for osteoarthritis (he referenced twice a “definitive” study by Brian Berman, discussed on this blog here and here) are often not reimbursed by Medicare.

Dr. P described three important factors to consider when evaluating a patient request for CAM: safety, efficacy, and curability. “Curability” refers to the patient’s clinical state and prognosis. Dr. P recommends greater tolerance for ineffective or even unsafe interventions if the patient is unlikely to suffer ill effects, perhaps due to comatose state or imminent death. He pointed out that physicians often provide interventions at the end of life that are more for the family’s sake than the patient’s, such as futile resuscitation attempts, and the same leeway should be granted for last-ditch CAM efforts when no standard therapies are available. In the aforementioned (real-life) case of the coma patient treated with hyperbaric oxygen, the family was wealthy enough to easily afford the treatment and was willing to reimburse hospital resources (nursing, ambulance) spent shuttling the patient to and from the hyperbaric quack (for a planned 30+ treatments). As expected, Dr. P spoke eloquently on the need to tease out family dynamics, goals for treatment, unreasonable expectations, etc., and I agree with him that theses thorny ethical issues regarding chronic disease or end-of-life care are independent of CAM use.

On “safety” Dr. P said that CAM (or any) interventions could be classified as proven safe (within reason of course, not in any absolute sense), not proven safe, or proven not safe. He started by telling us that patient’s requests and beliefs never excuse the physician from the professional duty not to harm. Then, with the caveat that some of us may disagree with the following statement, Dr. P read from his slide, “A treatment proven to be safe should be administered out of respect for the patient and/or family autonomy, and to promote an open and cooperative relationship” (my emphasis). Although Dr. P did not mention any such consequence, I believe that this mandate would necessarily lead to physician acceptance of the use of homeopathy and Reiki, even if we all agreed they had no specific efficacy. In my opinion, a more broad-based (dare I say, holistic) consideration of adverse effects of CAM may instead conclude that physicians have a professional and societal responsibility to discourage magical thinking… As for interventions “not proven safe,” here is where Dr. P suggested that cautious use with careful monitoring was appropriate (see my third paragraph). Interventions that are “proven not safe” should be avoided except as allowed by issues of “curability” (previous paragraph).

On the slide for “efficacy” that described the requirement for physicians to “do good” in addition to “do no harm,” Dr. P read a statement that made me sit up with interest: “However, providing some CAM modalities, though not scientifically validated, may have significant benefits for patients by reason of the placebo effect or by improving psychological well-being by demonstrating concern and regard for the patients’ and/or families’ wishes.” But he immediately said that although he included this argument for completeness, he does not actually agree with it; when he uses or recommends a piece of CAM, it is because he truly believes it may have specific benefits. One factor that influences his beliefs and clinical decision making, he next mentioned, is personal experience with a CAM therapy. I must quote him directly: “if I’ve recommended glucosamine 20 times for my patients or 200 times and I’ve seen some benefit, even if a study comes out saying it may not work, that may not change how I’m practicing.” He commented that one can always find flaws in any study’s design (I agree), that one’s biases influence this critical analysis (I agree), and that only rarely does a single article change one’s practice (I agree). Where it seems we disagree is on the value of a single clinician’s uncontrolled observations; I tend to think that such data is hopelessly flawed compared to a consensus view based on the totality of scientific evidence.

Frustratingly, Dr. P segued from this interesting epistemological point to the banal assertion of the importance of listening to patients and engaging in a therapeutic relationship that may lead to healing, not just curing. The next slide implored us to administer any therapy that is proven effective “regardless of its origins.” Again, the maggot story tells me that this is done, but Dr. P seems to feel that CAM is often given short shrift. For example, although several studies have shown efficacy of glucosamine for osteoarthritis (he asserted that many rheumatologists both prescribe and personally use it), Dr. P was annoyed to read press reports of one particular study (GAIT, I presume) that described a limited effect of glucosamine without mentioning that Celebrex had similarly poor effects in some of the study groups. I guess we also agree that mainstream media reporting on science is often misleading!

Dispersed through the lecture were presentations of three cases that illustrated ethical challenges surrounding CAM use: hyperbaric oxygen for a comatose patient, a mother of five who insisted on only alternative therapies for early-stage breast cancer (he tried everything, even called a psychiatrist, to convince her to accept definitive treatment but she refused), and a terminal leukemia patient who asked his advice on an absurd alt-med regimen (massive supplement use, coffee enemas). He stressed that these three patients, like sensational public reports of chemotherapy refuseniks, are very rare exceptions and that the vast majority of his patients use CAM as a safe complement to standard care.

The discussion of legal implications was brief. In Charell v. Gonzalez, a New York court found that “no practitioner of alternative medicine could prevail…as…the term ‘non-convention’ may well necessitate a finding that the doctor who practices such medicine deviates from ‘accepted’ medical standards.” (Interestingly, Dr. P called Gonzalez a “famous” physician and mentioned that after this case he won an NIH grant for a large trial of his anticancer regimen. As of November 2009, Dr. P apparently had not yet heard about the disastrous conclusion of that trial, made public in August 2009 and described here and here.) In Schneider v. Revici, another cancer quack was exonerated because the patient had signed a detailed consent form that marked an “expressed assumption of risk.” Dr. P pointed out that CAM providers are very rarely sued because they tend to have very good relationships with patients. For protection against malpractice charges when using CAM (or any) therapies, he recommended meticulous documentation, clear communication, and a willingness to apologize for failures.

The lecture ended with a touching story about how the husband of the terminal leukemia patient (who went to the Revici clinic against Dr. P’s advice) came back to visit Dr. P after his wife’s death. The husband thanked Dr. P for talking with them frankly and compassionately about their goals and expectations for her final months.

Much of this talk, like the previous one, must have seemed attractive and reasonable to the student with no more than a passing interest in CAM. However, few details were directly relevant to the ultimate question, “Do particular CAM therapies have specific effects?” I am intrigued by how often Dr. P and I agree, such as when he suggested that selection bias may help account for success stories from heroic alternative cancer regimens (i.e., only healthy patients can tolerate them). But if I watch carefully and dig deeply I can pick out issues of contention between us. The forums where we might argue over the evidence for this or that indication, or have the general discussion about interpretation of evidence, are the elective CAM courses presumably attended only by the enthusiastic. Most of my classmates get little more than the vague and rosy exposure that I describe here and here, and in one more future post on third year. I fear that these lectures, while making CAM/IM more palatable, do not adequately equip students for critical analysis of unusual claims.