What makes a medical procedure scientific? What makes it quackery?
Unlike many of my friends, I’m a conventional medicine guy. I don’t have any patience for homeopathy and reflexology and the like because they have no scientific backing. And as a writer about science, I’m convinced that the scientific method—which in the case of medicine centers on random controlled testing of therapies, drugs, and procedures—is the best tool we have for determining what treatments are effective and what are not. There’s really no rational argument you can make that would lead to any other conclusion.
In a randomized controlled clinical trial, a drug or treatment is tested by administering it, and a placebo (an inactive agent), to a sample group of subjects on a random basis. That is, the choice of whether a given subject receives the drug or the placebo is done on a random basis. Furthermore, neither the researchers nor the subjects know which subjects have received the drug and which the placebo. In this way, the efficacy of the drug or procedure can be measured statistically against what would have happened to the subjects had they not received the treatment.
Conventional medicine’s claim to superiority over alternative medicine—acupuncture, shiatsu, healing, and other such systems—is that its protocols are backed up by just such scientific trials.
Now, if I were to take this to the extreme, I would be an uncompromising advocate of evidence-based medicine (for a rundown of this approach’s advantages, see Ivan Oransky’s article in the current issue of The New Republic). But, frankly, like most people, I want a human being for a doctor, not a computer. Evidence-based medicine is based on the assumption that if you punch all the data on the patient into a computer, it will produce the proven best treatment. The human doctor becomes nearly superfluous. Yet we all sense that part of what makes a doctor good is his (or her) acquaintance with patients, and the intuitions that come from that acquaintance, combined with experience.
So I’m an advocate of scientific method, but I don’t discount the personal and non-scientific side of medicine. That’s why, at yesterday’s conference of the Israel Society for the History and Philosophy of Science, I chose to attend a session on “Boundaries and Categories in the Alternative Medicine Discourse.”
In the session’s first talk, Dr. Dani Filc of Ben-Gurion University (presenting work he has done with Dr. Nadav Davidovitch of Tel Aviv University), said that this boundary is a relatively new one. Prior to World War II, he explained, conventional medicine was based on the individual case study, not on the large statistical samples we today equate with scientific medicine. In fact, the earliest forms of the controlled trial of treatments were used by homeopaths in their efforts to provide solid grounding for their theories. Then, Filc said, homeopathy and other non-conventional theories were considered non-scientific precisely because they sought to base themselves on impersonal measures rather than on the intimate knowledge of the individual patient that the well-trained and experienced medical practitioner saw as the sine qua non of modern medicine.
In other words, there has been a change in the standard of what science is. This in turn has affected the way both conventional and non-conventional practitioners see their work. Statistical methods are nearly all that is important to advocates of the new school of evidence-based medicine, which seeks to make health care more effective and efficient by nearly eliminating the physician’s intuitions and prejudices from his work. Many alternative therapists have recast their fields from being alternatives to conventional methods to being supplements to it. At the same time, some conventional practitioners are resisting the pure empiricism of evidence-based medicine, and many alternative therapists claim that the worth of their treatments cannot be measured randomized controlled trials precisely because they are tailored to the individual patient.
As I understood Filc’s talk, the point here is not that we have no objective way of distinguishing between effective and non-effective treatments. Scientific procedure does in fact provide us with important tools for doing so. However, following the development of scientific practice keeps us humble. The scientific gold standard of today would have been foreign to researchers a century ago, and we can assume that a century from now scientific procedures will be different from what they are now.
In other words, as we retain rigorous standards of what counts as knowledge and what is merely wishful thinking, we need to remain humble about our certainties. Not everything goes, and we must examine our hypotheses carefully using the best means at our disposal. But we should not assume that we have, unlike our predecessors, reached the end of history and the end of science. The best ways we have of getting at the facts today may be superseded tomorrow.
For another look at how we evaluate treatments of a very different kind, see my next post, The Demolition Drug.