David Eisenberg, M.D. Speaks to
WHCCAMP on May 15, 2001

This section is compiled by Frank M. Painter, D.C.
Send all comments or additions to:    Frankp@chiro.org

Panel Session I: Overview of CAM Research Challenges

DR. GORDON: This presentation is going to be a little bit different, in that, we are going to have slides. David is going to spend 25 minutes going over the area of research in CAM and making his recommendations, and then we will have 20 minutes to have a dialogue with him about the recommendations.

Presenter: David Eisenberg, M.D.

DR. EISENBERG: Jim, thank you.

I am honored to be here, and delighted to see many dear friends, and to meet several other people whom I have looked forward to meeting for some time. So it is a pleasure to be here with you.

I am particularly grateful to the staff and to the Commission for giving me some time to present. I thought it would be useful before you read the recommendations, which are written somewhat succinctly. There are 15 of them. I have given you a summary of them in the form of an executive summary, and I have asked for copies of these to be distributed, the extra ones that we brought.

The presentation is really to give you the background behind the specific recommendations. I have also handed out to the Commissioners a bit of reference material that describes every study I will be referring to so you can go back to the original source.

If I could have the lights down just a bit.

[Slide presentation.]

DR. EISENBERG: I don't want to spend so much time on terminology as I do some of the language that is evolving to look at this very complex and provocative field. I am then going to discuss for about five or 10 minutes, trends that I think set the stage for research and research recommendations for the Commission.

I will then talk about some of the relevant research to give you a sense of the state of the science at this moment in time.

To give you a little preview, I want to give you the sense of how the research is relatively new. There has been precious little that has been satisfactory to the more conservative stakeholders of biomedicine, and in particular, we don't have as much in the way of basic science or health service research as I think we need to make proper recommendations.

I am going to give you some suggestions, and then you have my written recommendations.

You are all familiar with the fact that before the nineties, the pejorative language of unorthodox, unconventional, unproven was predominant in the literature, and that switched in the late eighties, early nineties to complementary and then alternative as a result of congressional direction and legislative -- I hate to use the word "fiat," but prerogative. They called the office the Office of Alternative Medicine, and that has stayed with us since 1992.

Now, I will use CAM as a shorthand in my presentation, but I want to point out that most academic institutions these days are really wrestling with the term of "integrative care" or "integrative medicine."

Integrative medicine, at least conceptually, as discussed by a variety of deans and research directors, combines the best of both conventional and evidence-based CAM therapies, emphasizes patient participation, exercise, diet, stress management, and maximizing health, emphasizes the primacy of the patient-provider relationship, and the importance of shared decisionmaking, and emphasizes the contribution of the therapeutic encounter itself and the science to be gleaned from looking at encounter expectation, conditioning, and those sorts of so-called nonspecific effects.

Now, put in context, I want to share with the Commissioners that they should be sensitive to the debate going on at most of the universities as to whether we should continue with the language of alternative or complementary, or wrestle with this concept of integration.

One could argue that it as filled with optimism as unconventional or unorthodox or pejorative, but this is the debate going.

I want to especially bring your attention to the language used by Richard Smith in the British Medical Journal that was devoted entirely to integrated or integrative medicine in January of this year.

"Integrative Medicine," in the editorial, he said, "focuses on health and healing rather than disease and treatment. The patient is seen as a whole, complete with dreams, disappointments, stories, loved one and enemies, not just a biochemical puzzle to be solved. It mightn't be too pretentious (although it might)" -- is that British -- "to say that such a growth of integrated medicine might restore the soul to medicine, the soul being that part of us that is the most important, but the least easy to delineate."

I think Ralph Snyderman said it best. He is the Dean at Duke and CEO of Duke Medical Center. He said, "We are really wrestling with the issue, not so much of integrative medicine, but re-integrative medicine, after we come back to the basic tenets that we all went into health delivery for and which aspects of CAM can be applied judiciously in an evidence-based fashion."

So, I would like that discussion to have a place in the Commission's report, because it is valuable and I think it takes away some of the provocation us-them dichotomization.

I will very quickly, for five minutes, summarize what I call a brief journey in time, since 1992, to set the stage for the research agenda now. You know all this, and I speaking to the choir here.

The Senate hearings in '92, The New England Journal survey in '93, the Office of Alternative Medicine in '93; '95, the 10 CAM centers, initial centers were funded. By '98, the OAM was established as a National Center.

'98 was a pivotal year. There was an important editorial in The New England Journal of Medicine, Jerry Kassirer and Marcia Angell speaking for many in the biomedical community said it is time to stop giving alternative medicine a free ride.

I think their point was that there was precious little science, and that there was a danger of promoting this field in the absence of science lest it be viewed entirely as pure advocacy.

A month later, JAMA and the AMA journals published a total of 80 articles, 18 randomized trials, some positive, some negative, and their editorial, George Lundberg and Phil Fontanarosa said, "There is no alternative medicine." I read their editorial to mean there is simply medicine, good medicine, and our collective job was to distinguish useful from useless, safe from unsafe.

By 1999, the NIH had funded additional centers. They were beginning to fund Phase III clinical trials, definitive trials, each of which will cost or has cost millions of dollars.

I want to bring your attention to the fact that it is only two years ago that the big pharmaceutical corporations began to sell products in the urban supplement domain.

Here is an example, a partial list of representative household name labels - Bayer, One-A-Day, American Home Products, Warner-Lambert, Johnson & Johnson. You can't go into a drugstore in this country without being a bit confused as to where the OTC drug ends and the herb, vitamin, or supplement begins.

This past year, your commission was founded. I think that was an important political act. The Macy Foundation, arguably one of the most important foundations interested in medical educational reform, devoted one of its conferences to this topic this past year, and I recommend their summary to you. It is referenced in the handout.

There has been a new group that has formed, the U.S. Consortium of Academic Health Centers for Integrative Medicine. Now, this is composed of executive deans and principal investigators currently from 11 universities, medical schools. I see Audi, who is here, and Jim Gordon has been to one of these meetings, and there may be others here who have attended.

This has been a discussion group that has met twice to discuss what is the role of the academic health center, and by "academic health center," I am speaking about university, university hospital, university affiliates, in terms of research, education, and training, and responsible delivery of care.

I want to put this in your heads as the overture to the symphony here, that these academic affiliates, universities, hospitals, medical schools are part of the engine that must guide the recommendations of this commission, and I think this consortium has a very important role to play.

It now consists of 11 medical schools. I think the goal for the next year is to bring it to 25. At that point, we will have 25 out of 125 medical schools represented, and we could begin to discuss this at a very different level vis-a-vis the larger medical community.

Just a few points. Since the beginning of this year, most of you are aware that the NIH had a joint conference with the Royal College of Physicians in London to discuss issues and research challenges involved in complementary and integrative medicine.

The British Medical Journal that I referred to earlier devoted its entire issue to this topic. Many of you may not be aware that the Federation of State Medical Boards, which is in charge of licensure of all M.D. licenses, has put together a committee to discuss practice guidelines in the area of CAM.

Specifically, they and I have served as a consultant to this group. They are looking not only at physicians who deliver CAM services, but, in my view, importantly, physicians who are co-managing patients who are simultaneously being seen with licensed or unlicensed CAM providers. In my view, that is the much bigger challenge.

I also want to bring your attention to another theme - that safety will forever trump efficacy, and in the last year we have become increasingly aware of that which we all predicted, clinically significant adverse drug-herb effects.

Now, there have been a series of articles mainly in the Lancet. This is one from Adrian Fugh-Berman, et al., looking at case reports. These are individual patients who had clinically significant adverse events that are attributed to -- I don't want to say we are absolutely certain -- but attributed to interactions between herbs like ginkgo or dong quai with Coumadin, warfarin, a blood thinner, St. John's wort with an antidepressant Yohimbe, which is an herbal aphrodisiac with antihypertensives.

I think the granddaddy in this set -- and they are all referenced in your handout -- is this study, because this is not a case report, this is a prospective clinical trial. Did Steve Strauss mention this to the Commission, are you all familiar with this? Well, if you are not, it is important.

This is a study that was done by Piscitelli and his group who took college age volunteers, gave them indinivir, which is part of the AIDS cocktail. It is a viral inhibitor that is part of the very potent drug therapy used to keep people with HIV alive and well, and in this instance, they were watched, they were observed while they reached a steady state of their indinivir level, and then they were given over-the-counter St. John's wort, and their indinivir level decreased significantly and, in many instances, below the therapeutic window that would safeguard somebody with HIV, or to put it bluntly, people with an HIV infection who were taking indinivir, who then exposed themselves to St. John's wort may not be in danger of a recurrence of the HIV virus because of the interaction of the herb with the drug, making the drug more easily to metabolize or metabolize more quickly I should say, and therefore, less available.

Two other smoking guns. Patients with heart transplants who had been clinically stable reported in the Lancet, I believe, who then took St. John's wort, they were on cyclosporin to help keep their hearts from experiencing a rejection immunologically, once they started St. John's wort, they rejected their hearts, and the time course was close enough that the authors thought this was a causal relationship.

Now, both of these patients lived because this was picked up in time, but I don't think we need a long list of smoking guns like this to convince us that safety, drug-herb interaction, quality assurance of herbal products to the American consumer must be a very high priority.

As you all know, the business of herbs and supplements is enormous. These are marketing data estimating the worldwide sale of herbs and vitamins at $46 billion. We are the biggest single consumer at 11 billion, followed by about 10 billion for Europe, and then Asia makes up most of the remainder.

Now, this is a study that just came out last month, Robert Linden of Harvard School of Public Health, and now the Kennedy School of Government. He and his group did a series of surveys that looked at Americans' views of herbs. I thought this was a very important paper, and I brought a copy if you want me to make it available to the Commission.

These were three random surveys, each with more than 1,000 respondents, between '96 and '99, that found that among average Americans, 48 percent had used at least one dietary supplement regularly, 44 percent users believe that their medical doctor knew little or not much at all about these products.

These were the statistics that really surprised me - 72 percent would continue to use their herb or supplement even if a government study was negative. I want you to think about that. I don't have the answer to that. I don't know if it means they don't believe research or they believe that these are safe and effective, and they won't believe a study to tell them otherwise, or if this is just aberrant sort of, the way people who believe in any therapy will persist in spite of the evidence.

On the other side, the majority were not aware of DSHEA and the current environment with which regulation occurs or does not occur for herbs, vitamins, and supplements, but in the abstract -- and this is just a hypothetical question -- they were asked, and 81 percent would require evidence of efficacy, safety, and FDA approval prior to allowing for the sale of an herbal supplement.

So, to suggest that there is not political will to go back and talk about a higher degree of quality assurance and safety of herbs is not founded by the data. Congress may think that, but these surveys argue, I think very potently, in the other direction.

Now, all of you are familiar with the JAMA article, so maybe I will just skip over the summary slides. You know that CAM use increased dramatically from '90 to '97, to 82 million adults, 42 percent. Number of visits increased to nearly 629 million we estimate. That was more than 200 million more visits to CAM providers than to primary care medical doctors.

Expenditures for CAM services increased by 45 percent to $21 billion. These are all in the handout, and you probably all know this article. Out-of-pocket expenditures exceeded $27 billion, disclosure still I think unacceptably low with fewer than 40 percent of CAM therapies discussed with physicians.

This was the predictor of some of these bad apples I have talked about, that we estimated conservatively that 15 million adults or 1 in 5 prescription drug takers was simultaneously taking herbs or high-dose vitamin supplements, therefore setting the stage for these clinically significant drug-herb interactions.

Lest you think this is just a baby-boomer phenomenon, among the elderly roughly a third used CAM regularly to treat their most severe illness, and 1 in 5 see a CAM provider per year. This is from the Journal of the American Geriatric Society.

Interestingly, the top two CAM therapies used by adults over 65 are chiropractic and herbs, and particularly for herbs, I am concerned about adverse events of too weak kidneys and livers.

Again, just as backdrop, I think it is important to say we lack definitive information about minority patterns of use of CAM therapies. We need to have oversampling of our Spanish-speaking population, African-Americans, Asian-Americans.

The NIH recently has put together a committee that I have worked with to give NHIS, National Health Inventory Survey, to perform a survey of random households in 2002, but these questions will only scratch the surface, and I think you should vociferously interest the Surgeon General, the NIH, and the Centers that are appropriate, CDC and elsewhere, to look at detailed patterns of use by the minority populations.

Their herbs and supplements, all the products taken by mouth are probably different, their patterns of access are different, their reimbursement profiles are different. I think we need those data.

This is from an abstract, but it shows that among quartiles of income, interestingly, even those in the lowest quartile, this is from the same national survey, those making under $20,000, while income is a barrier to CAM use, 43 percent of the lowest income quartile has spent, approximately $250 annually, which in terms of expendable income, is still considerable.

I think I am speaking to the choir here, but it would be a shame if, as CAM therapies are introduced by third-party payers, those without insurance and those with less expendable income are excluded.

Let me go more quickly to get to the research side. You know that current use underrepresents utilization, not if but when insurance increases, and now we come to the research.

I have given in your handout examples of I think the best examples we have to date, most of them from the last five years, where the evidence suggests that there is some efficacy of CAM therapies for particular conditions, and I have culled the literature and given you about 20 to look at that come from major journals, and they are either well-designed and peer- reviewed clinical trials, or they are consensus conferences from the NIH or other federal agencies.

I think some of these you are familiar with, but there is at least to some degree, some evidence that chiropractic is useful for acute low back pain, mind-body for pain and insomnia, acupuncture for nausea, dental pain, moxibustion, just one study in the treatment of foot first or breech presentation, psychosocial support for cancer.

Two studies, meta-analyses suggesting homeopathy as distinct from placebo, St. John's wort for depression, ginkgo for Alzheimer's, Chinese herbs for irritably bowel, saw palmetto for benign prostatic hypertrophy. You are aware of these. These are sort of the examples of good research suggesting positive effectiveness. This last slide is just five more from this year alone.

I would be remiss if I didn't mention that there is, not an equally large, but I think almost as large group of studies suggesting lack of evidence in these same journals. Some of them get the same amount of press as the positive ones, and this is my point - that all this research, which is really five or six years old, is now in the medicine literature.

The scientific community is engaged, positive and negative studies are being reported, so this means there is now traction, and I think this is a good sign that the field is beginning to mature as a research field that is worthy of our collective attention.

Many of you know that I leave from here this afternoon to go to San Francisco for the First International Conference on CAM Integrative Medicine Research. That meeting cosponsored by Harvard and UCSF is a research meeting only, only original research will be presented.

We received 327 abstracts, we didn't know if we would get 50. We accepted 223, 23 of them are being presented orally, the other 200 by poster, and the meeting is sold out with 400 registrants and standing room only.

So, I think this means that the national and international community want to come to a research meeting to discuss basic science, clinical research, research methodology, and policy-related science.

So, that is what is coming about over the next few days. Again, I think this ushers in another chapter within the discussion of CAM and its appropriate insertion into the health care system.

So, this is really I think the new slide of my presentation. In thinking about the evidence we have so far, if you think about the way the trajectory of conventional orthodox medical and biomedical research versus complementary and integrative medicine, much of, not all, but much of conventional orthodox medicine proceeds from basic science to in vitro animal model, to clinical trials, Phases I, II, and III, cost effectiveness, health service research, policy, reimbursement, politics issues come up later, after the science is done, and then there is a change in the delivery.

But if we reflect honestly about the field of CAM, it was a series of epidemiologic surveys, mine included, that got the attention of a whole group of stakeholders. It was epidemiology and popular demand that launched the OAM, that fed NCCAM, that informs much of the Commission's work and mission.

Political support is already engaged, I need not tell you that, you think about it all the time as commissioners, but there have been few clinical trials, most of them Phase II, and now a few in Phase III, osteoarthritis of the knee, chondroitin sulfate, St. John's wart for depression. You know these.

There are very few, I don't know of any Phase I trial of a CAM entity, a safety trial, as there is in drug studies. There is precious little -- these are in italics and yellow, but you can't see -- precious little in the way of cost effectiveness and health service research, and that is where I want to spend the last five or six minutes of my time. Can I still go for another five or six, Jim?

DR. GORDON: Yes.

DR. EISENBERG: And even less in the basic sciences. One of the themes that I have put front and center in the list of written recommendations is that research in this area has to have basic science, clinical research, and health service cost effectiveness research, and all have to be fully resourced.

The medical community will not accept the field changing its behavior in the absence of basic science, mechanistic, plausible mechanistic and rational explanatory models. The health care delivery side, policymakers and third-party payers will not, from a fiduciary standpoint, pay for this in the absence of health care, health service research that shows clinical benefits at reduced cost or cost neutral.

White and R. Ernst reviewed all the literature to look for economic analyses of this field, and he basically summarized it as follows - there is a paucity of rigorous studies that could provide conclusive evidence of differences in cost and outcomes between complementary therapies and orthodox medicine.

Following up on some of the themes I heard George DeVries pick up on this morning, in spite of the fact that there is precious little in the way of health services research, these CAM programs are being offered by managed care organizations throughout the United States.

This is an ad that you can't quite see from the New York Times just a few months ago. It is an HMO, it doesn't matter what the name is, and in addition to orthopedics, neurosurgery, dermatology, and all the other conventional allopathic orthodox specialties, I have highlighted where it says this HMO also provides acupuncture, massage, chiropractic, naturopathy.

To the average American, this is now insinuated within our health delivery system. They don't understand the provocative nature of this debate of how much evidence is there and should it be paid for and who licenses, et cetera.

So, most of you know, and I am sure George has briefed you, that most of the CAM delivery services are really reduced fee-for-service at this point in time. Networks of providers are described by HMOs or insurance carriers. These are efficient, convenient, carved-out services, but they are not yet, in my view, fully integrated.

We can't track the clinical and cost offsets, and that is really I think part of the Holy Grail.

So, I was asked for examples of models that are ongoing to look at how you track this, and I think the critical question is in any condition, if you randomize patients -- and I am a believer here, in this instance, of randomized trials being useful -- if you randomize a population to integrative or CAM therapy options versus usual care, what does it do to your clinical outcome costs and satisfaction?

I will showcase two examples. We could talk more about them in the discussion if you would like. This is a study funded by NCCAM, one of the few health service research projects I am aware of funded by the NIH. It was originally sent to the Agency for Health Care Policy and Research. They did not have the funds to do it, NCCAM picked it up.

The point of this slide is that this is a trial with people in an HMO, Harvard Pilgrim, Harvard Vanguard, with acute back pain, who show up at their doctor's office, and they are invited to participate in a randomized trial.

One in three get usual care, and two in three get a choice of expanded benefits. This is fairly heretical design in the eyes of NIH. We are really looking a monotherapy and fairly causal relationships.

The only thing this study can tell us is if the group that gets expanded benefits, their choice of chiro, acupuncture, massage, or usual care fares better in terms of clinical outcome, cost, satisfaction, that sort of thing.

The hypothesis in the study is that people have access to these benefits and they are paid for as part of this NIH study, will have greater symptom relief, greater improvement and function, greater satisfaction, and decreased utilization of conventional services, visits to the orthopedist, follow-up x-rays, PT, et cetera.

The medical community I think can argue on very solid grounds that there is not much evidence to suggest that this will happen. They say balderdash, these people are not going to get better faster, 85 percent are going to get better in six weeks no matter what.

The CAM community says you give them to us quickly, we turn them around more quickly, we reduce the rate of occurrence, we change the degree of dysfunction over time, but particularly in the acute setting, let us prove our worth.

So, this is when I think a controlled trial is useful. That is what this study is about. I am showing it to you as a model, but embedded in this model are a few things that are key to your discussion.

We drew a circle around our clinical site, our first pilot site, West Roxbury, a suburb of Boston, a working class suburb. There were 347 licensed chiropractors, acupuncturists, massage therapists, basically in walking distance from that medical site.

Which ones do you bring into this research project? Which ones do you, as an HMO medical director, bring into your HMO? Which ones do you, as a physician, refer your patient or your brother-in-law to?

This goes way beyond licensure. This is levels of credentialing, and by whose criteria? Now, we have worked with the chiropractors, acupuncturists, and massage therapists for four or five years. They co-funded the pilot of this. They built the scope of practice that we embedded in our criteria to pick the researchers.

We could spend a lot of time on this, but how do you decide who is the best practitioner among the licensed practitioners? Licensure and name is not sufficient. What about malpractice history? What about a written application, who writes it? What are the criteria for scoring it? Can it be reproducible, so somebody doesn't say you are just picking your friends? What about site visits and medical records? What about billing practices?

What if this person does something that is out of keeping with 9 out of 10 of his colleagues in the community?

These are the credentialing issues that I have also put as front and center as recommendations from the Commission. Above and beyond licensure, we need to have credentialing.

The last model I want to give you as an example of health service research was recently published, just two weeks ago, in the Archives of Internal Medicine. This is a more fastidious randomization of chronic low back pain patients from Group Health in Seattle randomized to these three arms.

This article published by Dan Gherkin [ph] and others really compares two CAM interventions with sort of standard educational intervention. You can't see the slide very well, but the outcomes are what the health service research community looks for, disability scores, and in this instance, among the three groups, massage, acupuncture, and education, massage fared better in terms of disability. In terms of weeks, in terms of medication, number of pills, and cost of drug, massage fared better.

In terms of dollars spent, there was a trend to suggest that massage was the most cost effective.

Now, these conclusions of just one study give you an example of how you can look at CAM therapies, look at clinical outcome, and cost, and the conclusions were that massage appeared to be safe, acupuncture is safe but had little effect, self-care promising, but I think the new news here was that in this one study, provocative as it may be, that massage may, in fact, be a cost effective modality, and this warrants further research.

The only other theme I want to mention -- and I know my time is out, so I will just flag it for you -- we are seeing now integrative care teams crop up in major hospitals and medical groups throughout the country, some private, some public, some in major academic health centers. These are examples of major, high-visibility hospitals that have rolled out integrative care clinics.

I repeat my challenge. We must test whether these integrative care units, when given patients with a chronic or life-threatening illness, provide better clinical outcomes and at what cost.

These are testable hypotheses. These integrative care units at the major universities should be encouraged to develop randomized trials to see if they do better with the patients with cancer or low back pain or intractable headache.

I don't have a chance to go into the model that we are working on in Boston, but you get the gist.

So, my final slide is the stakeholders and some of my recommendations that are written out for you. I encourage you to think about the universities, medical schools, hospitals, and these integrative care centers affiliated with them as part of the driving force.

To date, the managed care and the insurance industry have allowed us to look at some of their outcomes, but they are not investing in research in this area. Similarly, the private sector, although making a lot of money, pharmaceutical companies in particular, they have not invested in research in this area. They need to be engaged. Either they should pay for the research, or perhaps they could be incentivized to match federal dollars to invest in basic science, clinical research, and health service research.

Similarly, don't forget about the Fortune 1000 employers and employer groups whose employees want these therapies, and I think would pay good money to have randomized trials to see if this is cost effective for their employees - will it get the back to work, will they be more efficient and effective employees, will they be more satisfied.

Philanthropy has jump-started almost all of these university programs, Harvard included. The consumers are the ones you know the most about. Let's not forget the NIH to date has been the largest repository of funding, but I encourage you to explore ways in which CDC and the Agency for Healthcare Research and Quality can be engaged.

This is the group that has the most expertise, by the way, in best practices and cost effectiveness research. For some reason, they have not been a major player, and I think they need to be.

You are aware of Congress. I encourage you also to think about the Federation of State Medical Boards, and of course, the FDA.

I will close with a quote. Well, I don't have a chance to summarize this, but I think you get the gist. Let me make one point.

Clinical research in the absence of basic science is insufficient. Health service research doesn't exist. Education, you are going to talk about, but let me also plant the seed that faculty across our universities arguably know perhaps the least about this, and while you are talking about training the trainees, think about faculty development.

When you think about delivery systems, I encourage you to fund research collaboratively between the private and the public sector, that looks not only at the efficacy of individual CAM providers, but do these CAM networks, as George has spoken about, deliver cost value, and do these integrative teams deliver improved clinical outcomes and at what cost, yes or no.

Credentialing is a bugaboo. Quality assurance and safety, I have highlighted. From my view, you cannot create judicious policy informed by research in the absence of getting these things right.

So, I will close there with a quote. "Doing everything for everyone is neither tenable nor desirable. What is done should ideally be inspired by compassion and guided by science, and not merely reflect what the market will bear."

Thank you very much. I am open to your questions.

Return to WHCCAMP Testimony

Return to EISENBERG's CAM ARTICLES

Since 9-01-2001