On Scientific Plausibility
Scientific plausibility permeates discussions and debates about research on complementary, alternative, or integrative health approaches. This is no surprise; many interventions that fall under this rubric are ensconced in belief systems about illness and health—some ancient and some modern—that lack foundations in modern science. In addition, those who support research on these approaches often fail to articulate a scientifically grounded rationale or approach to research. Thus, it is common to see criticism based on scientific plausibility in the scientific literature, news stories, and blogs.
This criticism often suggests that the existence of research implies either belief in scientifically implausible explanations or ignorance of basic scientific principles and concepts. So how do we justify investment of public resources in research on complementary interventions that are associated with pre-scientific or unscientific explanations? Simply, it is both possible and necessary to disconnect scientific interest from unscientific “trappings.” For example, an objective look at the body of accumulated evidence (from patient reports, clinical observations by many good clinicians, and clinical studies) suggests that some people with chronic low-back pain are deriving meaningful benefit from acupuncture, yoga, or procedures involving spinal manipulation. It is entirely possible to be scientifically curious about that body of evidence and investigate it further, while not in any way embracing scientifically unfounded explanations for those practices. For instance, it is not necessary to believe in meridians or qi to study the effects of the procedure of acupuncture on pain, or to explore the hypothesis that acupuncture mediates pain by conditioning or expectancy effects produced by a convincing ritual combined with a counter-irritant.
NCCAM’s strategic plan now includes a framework of four factors we use to sharpen the focus of our research investments. Two of them address important aspects of scientific plausibility. One is “scientific promise”: how strong is the body of evidence supporting the concept? In the case of acupuncture, yoga, or spinal manipulation for chronic back pain, credible signals from a variety of clinical sources provide a sufficient basis for interest in further research. There is no need to bring associated non-scientific explanations into consideration of scientific promise. By contrast, unscientific notions should assume increasingly greater importance when clinical signals are weak, unconvincing, or non-existent.
In all cases the question of whether and how to invest valuable resources in research must also move to consideration of the second factor—“amenability to rigorous scientific inquiry.” Do we have reliable and reproducible tools, methods, diagnostics, outcome measures, quality control processes, etc., to allow us to mount a study that will elucidate a clear and unambiguous answer? In some cases we do. For example, it is very possible (although not necessarily easy) to design a study that employs the most rigorous mainstream clinical research methodology to investigate whether acupuncture, spinal manipulation, or yoga alters a patient’s low-back pain. On the other hand, it is not possible to design studies that will yield clear and unambiguous answers when objective, validated measurement tools, or processes and procedures to ensure and document quality control, are lacking.
Let me be clear that I do not mean to suggest that we can or should launch expensive clinical trials of anything and everything complementary or alternative or integrative just because we have some intriguing anecdotes. The strength, reliability, reproducibility, and other particulars of the signals from clinical observation and preliminary clinical investigations are critically important. Adequate methods and tools are equally important. I also do not mean to suggest in any way that mechanisms are irrelevant to questions of plausibility. One lesson NCCAM has learned from extensive experience is that mechanistic insight into biological effects creates sharper scientific hypotheses and allows one to design better clinical trials to investigate those hypotheses. I simply mean to suggest that it is a mistake to assume that scientific inquiry is equivalent to acceptance of unscientific mechanistic thinking.
NCCAM’s first decade entailed a relatively broad and investigator-initiated approach to funding. This was appropriate to the time and the state of the available scientific evidence. The four factors we now consider evolved out of lessons learned during those years. So with the benefit of hindsight, it is pertinent to note that a number of studies funded during that timeframe would probably not be funded today because they could not pass our current hurdles regarding plausibility. In fact, the portfolio of research NCCAM has actually funded over the past several years demonstrates clearly that both the peer review process and NCCAM are now using these factors to shape our investments in research.
So plausibility matters a great deal, but it is a mistake to equate interest in research with acceptance of unscientific trappings. By the same token, I urge those who support research on these interventions to carefully parse rationale from “trappings” and give due recognition to the validity of concerns about scientific plausibility.