Mar 31 2008
I’m on the record multiple times as saying that I reject the entire concept and nomenclature of “alternative medicine” as being distinct from “conventional” medicine as a false dichotomy, when in reality there should be just “medicine.” Indeed, if there is one major theme to which this blog is dedicated it’s that medicine should be as much as possible science-based, a concept that takes into account not just clinical trials, which are prone to all sorts of false-positive results in the case of modalities that have no plausibility from a scientific perspective. In essence, I advocate treating “alternative” medicine the same as “conventional” medicine subjecting it to the same scientific process to determine whether it has efficacy or not, after which medicine that is effective is retained and used and medicine that fails the test is discarded. Where it comes from, the “alternative” or the “conventional” medical realm, matters little to me. All that matters is that it is based on sound science and that it has been demonstrated to have efficacy significantly greater than that of a placebo.
Given that, you’d think I’d be all in favor of subjecting alternative medicine, be it woo or more credible, to rigorous scientific testing. In many cases, you’d be right. My sole caveat is that, when testing alt-med, priority should be given to modalities that have at least a modicum of scientific plausibility, even if a bit tenuous. Herbal remedies would thus be at the front of my line to be tested, while obvious woo whose core principle on which it is based is so utterly ridiculous and scientifically implausible (like homeopathy, for instance) would be relegated to the back of line, if it’s ever tested at all. More implausible modalities that might work (albeit by a method that has nothing to do with the “life energy” manipulation that is claimed for it) like acupuncture would be somewhere in the middle. It’s a matter of resource prioritization, in which it makes little sense to test blatant woo before more plausible therapies are examined. Indeed, it’s arguable whether blatant woo like homeopathy should even have resources wasted testing it at all, given its extreme scientific improbability. Finally, regardless of what modality is being tested in scientific and/or clinical trials, it has to be done according to the highest ethical standards, on adults fully cognizant of or able to be taught about the questions being asked, the issues involved, and the potential risks who are thus able to give truly informed consent.
Given Dr. Atwood’s recent post about the dubious ethics of certain CAM trials funded by the NIH, I thought I’d discuss one such trial that I first came across about about a year and a half ago that demonstrates many of the ethical problems that come with testing unscientific medicine. In this case, the unscientific medicine is homeopathy, and the research subjects are children in a Third World Country. The study was trial of homeopathy published in October 2006 in The Journal of Alternative and Complementary Medicine:
Homeopathic Combination Remedy in the Treatment of Acute Childhood Diarrhea in Honduras
Jacobs J, Guthrie BL, Montes GA, Jacobs LE, Mickey-Colman N, Wilson AR, DiGiacomo R.
Department of Epidemiology, School of Public Health and Community Medicine, University of Washington, Seattle, WA.
BACKGROUND: Despite the widespread availability of oral rehydration therapy, diarrheal illness remains a major cause of morbidity and mortality around the world. Previous studies have shown individualized homeopathic therapy to be effective in treating childhood diarrhea, but this approach requires specialized training. OBJECTIVE: A homeopathic combination medicine, if effective, could be used by health personnel on a widespread basis. METHODS: A double-blind randomized controlled trial was conducted in Honduras to evaluate the effectiveness of a homeopathic combination therapy to treat acute diarrhea in children. A total of 292 children with acute diarrhea was recruited; 145 were randomized to the experimental group and 147 to the placebo group. Tablets containing a combined preparation of the five most common single homeopathic remedies used to treat diarrhea or placebo were administered by a parent after each unformed stool. Children were followed up daily for 7 days or until symptoms resolved, whichever occurred first. Time until resolution of symptoms, daily rate of unformed stools, and total number of unformed stools were compared between the two groups. RESULTS: There was no significant difference in the likelihood of resolution of diarrheal symptoms between the treatment and placebo groups (hazard ratio = 1.02, 95% confidence interval: 0.79-1.32), with a median time until resolution of 3 days for both groups. Children in the treatment group had an average of 2.6 unformed stools per day compared to 2.8 among those in the placebo group; this difference was not significant (p = 0.43). The median number of unformed stools was 7 among children in the treatment group and 8 among those in the placebo group (p = 0.41). DISCUSSION: The homeopathic combination therapy tested in this study did not significantly reduce the duration or severity of acute diarrhea in Honduran children. Further study is needed to develop affordable and effective methods of using homeopathy to reduce the global burden of childhood diarrhea.
I will give the Journal props for one thing in this one instance: It actually published a completely negative study of homeopathy. If you peruse the archive of its past issues, you will find precious few instances of totally negative studies on favored “complementary and alternative” therapies (or so-called “CAM” therapies) like homeopathy being published. Basically, the study above found that the homeopathic “remedies” used above didn’t provide any detectable benefit for diarrheal illness in these unfortunate children in Honduras. Of course, given the utter scientifically implausibility of homeopathy, that’s pretty much a “well, duh!” conclusion. Indeed, it would have been far more surprising if the investigators had found an an obvious treatment effect due to homeopathy that was due to anything more than placebo effect or poor experimental design. However, this study is a beautiful case study of how, no matter how negative the data is, CAM advocates will try to spin it as being something other than it is, as the authors do here:
Although individually prescribed homeopathic therapy by a trained practitioner reduced the duration and severity of acute diarrhea in three previous studies, the combination homeopathic medicine used in this study showed no evidence of efficacy. There was also no evidence that efficacy differed between subgroups defined by demographic or baseline clinical characteristics or category of pathogenic agent detected.
A number of factors could account for the ineffectiveness of the homeopathic combination therapy. Although the homeopathic remedies included in the combination therapy were those most commonly prescribed in the previous studies, it is possible that these remedies would not have been prescribed individually in this population and/or that a different combination medicine would have been more effective. There is also a possibility that the remedies included in the combination therapy counteracted each other in some way, rendering the individual remedies ineffective. Other factors could be that the therapy was not administered correctly by parents in this study, or that it had lost its potency because of improper storage or handling before it was administered.
I cannot help but chuckle at that last sentence. How does a placebo whose allegedly active ingredient has been diluted to a homeopathic dilution of 30C, (which is the equivalent of a 1:1060), a point where it is incredibly unlikely that even one molecule of “active ingredient” remains, “lose its potency”? In fact, to find a single molecule of the “active substance” would require a container over 30 billion times the size of the earth. “Lose its potency,” indeed. It never had any potency to begin with. In essence, the author is suggesting that, somehow 87% alcohol could inactivate 87% alcohol or water could inactivate water!
Another possibility is that the pathogens infecting children in this study were less susceptible to homeopathic therapy in general or to this specific combination therapy in particular.
Which begs the question of how any pathogens could be susceptible to homeopathy at all!
But as much fun as it is to enjoy the discomfiture of the study authors as they try to explain in one of the woo-iest “scientific” journals there is why they found no treatment effect due to homeopathy without stating the glaringly obvious (that homeopathy is nothing more than an elaborate placebo containing no active ingredient), I’m more interested in the ethics of this study. First off, this study gives off an unpleasant whiff of exploitation. Think about it. Here we have comfortable academics from a wealthy nation heading to a country where diarrheal diseases in children, although still mostly self-limited, still cause considerable morbidity and mortality (world-wide around 1.5 million children die every year of such diseases). True, oral rehydration therapy, which has become more widespread, has greatly decreased the death rate. Nonetheless, in spite of this, these diseases still cause considerable morbidity due to malnutrition, mainly because babies suffering from them are unable to eat properly or absorb the nutrients during the course of the illness in whatever milk, food, or formula that they can manage to eat. In our wealthy nation, we rarely, if ever, see the devastation these diseases routinely wreak in impoverished areas like the ones where the study population lives, as described by the authors:
Subjects were drawn from two municipal clinics in the Metropolitan Health District of Honduras. The clinics serve an impoverished population of approximately 80,000 people, with no municipal water, electric, sewage, or garbage disposal services. Educational opportunities are low and unemployment is common…Among the clinic patients, there was a high incidence of childhood diarrhea; in 2001, there was an estimated incidence of 23 cases of diarrhea per 1000 children 5 years old.
Now, let’s look at the specifics of the homeopathic remedies used:
The homeopathic combination medicine was composed of the five most common single remedies, which were prescribed for the treatment of 80% of cases of childhood diarrhea in previous studies.12,13 These included Arsenicum album, Calcarea carbonica, Chamomilla, Podophyllum, and sulphur. The homeopathic remedies were prepared in the United States using standard methods.18 The combination therapy was prepared by a homeopathic pharmacist by impregnating pellets made of 85% sucrose and 15% lactose with a liquid homeopathic dilution in the 30C potency (the active substances diluted 1:100 in an 87% alcohol solution 30 times for a final concentration of 1 x 10-60). Placebo was prepared in the same manner using 87% alcohol instead of the homeopathic dilution. Treatment and placebo preparations were identical in taste, odor, and appearance.
Arsenic. Podophyllin. I wonder if the IRB knew that these were the ingredients being used. Granted, they are so dilute that there is nothing left, but there was no data presented to show that these components had actually been diluted that much and therefore posed no threat to the study subjects. Think of it this way. If the remedy has any active ingredients, it’s potentially toxic. If it is truly homeopathic, then it’s nothing more than 87% ethanol and sugar. How is it ethical to give this to anyone in a clinical trial, other than in a placebo control arm?
Here’s another consideration. The study subjects were children. The Common Rule that covers all federally funded human subjects research makes it very clear that special care must be taken when vulnerable populations are the subjects of a study. Depending on the age of the child, informed consent must be obtained from the subject himself. If the child is too young, then both parents must assent, unless one is dead or unavailable. In any case, the ethical considerations of research on children are very difficult. Now, in the light of the fact that children are considered vulnerable populations who require special protection above and beyond that given adults and that, more than that, they are not American and live in the slums of a Third World nation, think about how these investigators would have signed up patients in Honduras. It’s obvious from the writing in this paper that they are true believers in homeopathy. They probably told parents that they would be randomized to a placebo or to a remedy that might help their child and weren’t lying when they said it–because they truly believed it. Even so, to sell this trial they would have been spreading misinformation, namely that anyone in the study would receive anything that might actually have a therapeutic benefit.
Finally, given the conditions of the areas in which the study population live, it’s not unreasonable to presume that the people living there are mostly not well educated. This brings up the question of whether the parents were in fact able to give truly informed consent. The answer: Almost certainly not. Now, don’t get me wrong. One ethical principle that would be violated in withholding the benefits of a clinical trial from the poorly educated is that such benefits should not be limited only to the well-off or highly educated. However, communicating the risks and benefits of any trial to uneducated people unsophisticated people takes a special effort. It’s a lot of work. Did the investigators make that effort? Maybe I’m being cynical, but I tend to doubt it. In fact, to do this study in the first place, they would by definition of homeopathy have to be giving the parents of the study subjects misinformation. There’d be no way around it.
How this study ever passed muster at the Institutional Review Board at the University of Washington, the home institution of the investigators, I’ll never know. This IRB should have viewed its charge as protecting these Honduran children from being included in an dubious study that could not possibly do study participants any good and at best wouldn’t harm them. I’m guessing that the reason for this is that the members of the IRB had no clue about what homeopathy actually is, why it’s so scientifically ridiculous, and why the above study is highly dubious ethically. Indeed, I wonder if it is a general problem with IRBs that few of their members know exactly what it is that is in so-called “CAM” remedies like homeopathy. One other possible reason in this case is that the investigators probably cited a 13-year old study done using essentially the same methodology in Nicaragua and published in Pediatrics in 1994. This study purported to show a treatment effect, although it was small and arguable whether, even if real, it would be clinically significant. It also used a slightly different bunch of homeopathic remedies, one of which included mercury, of all things. In addition, this study was riddled with methodological difficulties, including failure to control for diet (diet can, as you would imagine, have a considerable impact on diarrhea in these illnesses); the level of dehydration (the treatment group where an effect was allegedly observed suffered from milder dehydration); and previous treatment interventions (patients in the treatment group were more than twice as likely to have received treatment before starting homeopathy). Finally, they did not use optimal recommended methods for quantifying diarrhea output
I truly wonder if the UW IRB would have approved this sort of study twice–or even once–if the study population had been American. I’m not sure which would be worse: If it wouldn’t have–or if it would have.
Finally, looking back at this trial again a year and a half later, I wondered if more such trials had been published in the interim. I was quite relieved to be unable to find any. Indeed, searching PubMed, all I could find were two articles. The first was a systematic review of homeopathy for pediatric and adolescent ailments that concluded that the “evidence from rigorous clinical trials of any type of therapeutic or preventive intervention testing homeopathy for childhood and adolescence ailments is not convincing enough for recommendations in any condition.” The second was a survey that starts out with the assumption that homeopathy is effective in childhood diarrhea and surveys pediatricians about whether they are using homeopathy for such diseases. To my relief, only 3% reported that they were using homeopathic antidiarrheal agents. To my amusement, the authors bemoaned this very low number and proposed means of increasing it:
We conclude that homeopathic antidiarrheal agents have garnered little support among pediatric health care providers. The lack of correlation between adoption and individual innovativeness suggests that adoption will not increase barring intervention. Survey methods have allowed us to identify specific barriers to adoption. Remediation would include improving the evidence base for this innovation to the degree possible, increasing awareness, and providing means of improving experiential knowledge of these agents.
In other words, even though homeopathy doesn’t work for pediatric diarrheal diseases, we propose increasing awareness and getting people “hands-on” experience with it. Of course, the most tortured conclusion is suggesting “improving the evidence base for this innovation as much as possible.” Apparently the authors don’t realize that the evidence base for this “innovation” (how one can call a 200 year old bit of magical thinking based on prescientific ideas on vitalism and sympathetic magic an “innovation” I’ll never know) is as “good” as it’s ever likely to get.
One thing you can be sure of, though, is that homeopathy advocates won’t give up. They will keep trying to carry out ethically dubious trials such as the one described above. Thanks to the existence of the National Center for Complementary and Alternative Medicine, sadly, you and I will likely keep paying for them too.
- Jacobs, J., Guthrie, B.L., Montes, G.A., Jacobs, L.E., Mickey-Colman, N., Wilson, A.R., DiGiacomo, R. (2006). Homeopathic Combination Remedy in the Treatment of Acute Childhood Diarrhea in Honduras. The Journal of Alternative and Complementary Medicine, 12(8), 723-732. DOI: 10.1089/acm.2006.12.723
- Altunç, U., Pittler, M.H., Ernst, E. (2007). Homeopathy for Childhood and Adolescence Ailments: Systematic Review of Randomized Clinical Trials. Mayo Clinic Proceedings, 82(1), 69-75.
- Pappano, D., Conners, G., McIntosh, S., Humiston, S., Roma, D. (2007). Why Pediatric Health Care Providers Are Not Using Homeopathic Antidiarrheal Agents. The Journal of Alternative and Complementary Medicine, 13(10), 1071-1076. DOI: 10.1089/acm.2007.0651
83 Responses to “On the ethics of clinical trials of homeopathy in Third World countries”