Conflicts of Interest in Alternative Medicine
When we think of conflicts of interest, we almost automatically think of money. In my area of research, complementary alternative medicine (CAM), there is no money—well, almost none (contrasted with most areas of mainstream medicine). Despite this fact, conflicts of interest are rife in CAM research. I am, of course, talking about a different type of conflict: the one that is created by strong belief and evangelic conviction.
Across the globe, I personally know many individuals who are full-time CAM researchers. They have different personalities, backgrounds, and skills. But they all have, as far as I can see, one characteristic in common: they are strong believers in the benefit of at least some aspects of CAM. On the one hand, this may seem entirely reasonable: if one didn’t believe in CAM, why would one dedicate one’s career to investigating it?
On the other hand, if the vast majority of CAM researchers are made up of CAM believers, things might not be quite right either. In other areas of medical research, the situation is—in my experience—very different. I know many pharmacologists, for instance, who are keenly aware of the dangers of medicines and extremely critical of some of the activities of the pharmaceutical industry. I cannot say that I know many CAM researchers who are truly concerned about the dangers of CAM or of the activities of those individuals or organisations that promote CAM uncritically.
Conflicts of interest are precarious because they tend to cloud judgment and generate bias—the type of bias that creeps in unnoticed and cannot be readily identified when studying a published paper. After some detective work, we might be able to find out, for example, that a certain paper that draws positive conclusions about the homeopathic remedy Traumeel® (Schneider et al. 2008) was coauthored by at least one “expert” who is on the payroll of the manufacturer of that very remedy—even if the paper itself fails to disclose this fact (Schneider et al. 2008). But what about more subtle yet potentially powerful conflicts of interest? I fear that they have far too much impact on CAM.
If the totality of researchers in one field is open to unidirectional bias, one has to worry about the area as a whole. The danger, then, is obvious: the field will collectively lose its balance and make serious and repetitive mistakes without even noticing them. In the absence of criticism “from the inside,” such an area of research can neither prosper nor mature. In my experience, CAM has very little internal criticism, as the following examples suggest.
After the British Chiropractic Association (BCA) sued science writer Simon Singh for libel, the United Kingdom’s General Chiropractic Council (GCC, the regulatory body of its members) was inundated with about 700 complaints from skeptical bloggers about chiropractors who had made similarly bogus therapeutic claims. The GCC reacted by commissioning North American chiropractors to write an “evidence report.” Presumably, the GCC hoped that the report could clarify the evidence regarding the disputed claims. Bronfort et al.’s (2010) published report revealed no strong evidence for “manual therapies.” Despite this seemingly critical stance, the report is still not critical enough, in my opinion. The reason becomes clear upon a reading of its fine print. Even though the authors repeatedly mention the “quality of the evidence,” they fail to formally evaluate it. Thus, poor-quality primary studies are taken at face value, which inevitably leads to false-positive conclusions. Without the very obvious conflict of interest (chiropractors commissioned by the GCC), the report might have been far more critical than it turned out to be.
A similar situation occurs with systematic reviews of chiropractic as a treatment for specific conditions. Such articles are now emerging regularly, and they tend to display interesting discrepancies. For example, a review on the subject of asthma written by four chiropractors concluded that “chiropractic care showed improvement in subjective measures . . .” (Kaminskyj et al. 2010). Meanwhile, my own review, which included a critical assessment of the quality of the primary data, stated that “spinal manipulation is not an effective treatment for asthma” (Ernst 2009). Here I should mention perhaps that I, as an independent academic, have no conflicts of interest and receive no payments from Big Pharma or similar institutions that might have an axe to grind.
Years ago, I had already noted that reviews published by chiropractors tend to arrive at positive conclusions while those by independent experts do not (Ernst and Canter 2006). The explanation for this phenomenon seems to be simple: conflict of interest.
Vickers et al. (1998) and others (Tang, Zhan, and Ernst 1999) have shown that 100 percent of all acupuncture trials originating in China report positive results. Recently, an in-depth analysis of acupuncture articles published between 1991 and 2009 revealed that China is now producing more acupuncture research papers than any other country (Han and Ho 2011). To make matters worse, this analysis also names the journals that publish the bulk of these articles: unsurprisingly, they tend to be the ones I have previously identified as publishing virtually no negative results (Ernst and Pittler 1997). Thus there is reason to fear that we are currently exposed to a mountain of research on acupuncture, much of which might be less than reliable.
Here the explanations might be more complex, and there could be more than one factor at play. Yet I have little doubt which one is the most important: conflict of interest.
What Can Be Done?
My message is clear: non-financial conflicts of interest can be just as powerful as financial ones, and in my area of research they seem to be quite overpowering. This problem will inevitably lead to significant distortions of the truth about the value of alternative medicine. The issue at hand is more than just academic: misleading results in health care endanger our health.
It is relatively easy to identify the problem, yet it is hard to solve it. I don’t pretend to have the ideal solution. All I can suggest is that journal editors consider making their authors’ conflicts of interest transparent and that readers of such papers apply a healthy dose of skepticism. Whenever there are two discrepant opinions (and that is the case more often than not), my advice is to determine which one might be prompted by a conflict of interest. In theory, this sounds fine; in practice, I am afraid, it will not be nearly enough to remedy the problem.
Bronfort, G., M. Haas, R. Evans, B. Leninger, and J. Triano. 2010. Effectiveness of manual therapies: The UK evidence report. Chiropractic and Osteopathy 18(3). doi: 10.1186/1746-13 40-18-3.
Ernst, E. 2009. Spinal manipulation for asthma: A systematic review of randomised clinical trials. Respiratory Medicine 103(12): 1791–95.
Ernst, E., and P. Canter. 2006. A systematic review of systematic reviews of spinal manipulation. Journal of the Royal Society of Medicine 99(4): 192–96.
Ernst, E., and M.H. Pittler. 1997. Alternative therapy bias. Nature 385: 480.
Han, J.S., and Y.S. Ho. 2011. Global trends and performances of acupuncture research. Neuroscience and Behavioural Reviews 35(3): 680–87.
Kaminskyj, A., M. Frazier, K. Johnstone, and B.J. Gleberzon. 2010. Chiropractic care for patients with asthma: A systematic review of the literature. Journal of Canadian Chiropractic Association 54(1): 24–32.
Schneider, C., B. Schneider, J. Hanisch, and R. Van Haselen. 2008. The role of homoeopathic preparation compared with conventional therapy in the treatment of injuries: An observational cohort study. Complementary Therapies in Medicine 16(1): 22–27.
Tang, J.L., S.Y. Zhan, and E. Ernst. 1999. Review of randomised controlled trials of traditional Chinese medicine. BMJ 319 (7203): 160–61.
Vickers, A., N. Goyal, R. Harland, and R. Rees. 1998. Do certain countries produce only positive results? A systematic review of controlled trials. Controlled Clinical Trials 19(2): 159–66.