Very occasionally — and I mean that — I forward an interesting skeptical article to my family in the States. This happened recently when I read Dr. David Gorski’s scathing summary of two recent Dr. Oz episodes at Science Based Medicine last week. I did think twice before I hit Send — Dr. Gorski can be a bit overwhelming to the uninitiated, and this article was one of his more animated rants — but he was calling out “Americas Doctor” for promoting faith healing on his show and at the time it just seemed like something that needed to be shared.
One of my relatives has recently become involved in the CAM community, and reacted to the article…well…negatively. In retrospect, I should have seen this coming and simply left her off the cc list, and I’m still not quite sure why I didn’t. I’ve been trying to avoid a confrontation with her for some time now, and up until that point had simply responded to her baiting (or at least, what I’d perceived as her baiting) with deflective humour. Yet this time I was the one doing the baiting, and when the bullets started flying, I returned fire.
Beyond whatever less than noble personal motives I might have had (I’ll take those up in “group“), what triggered my reply was this: she never addressed what the article had to say. The response started with an ad hominem attack against Dr. Gorski, followed by the tired trope about physicians managing illness rather than healing, and then an accusation that doctors are deliberately misleading the public about the evidentiary basis of “western medicine”. Maybe that was a fair response — in all honesty, I probably wasn’t just talking about faith healing when I sent off the link. But the form of her reply is typical of the type of evasive maneuver I regularly see from the CAM community, and I guess I just went on auto-pilot in my response to it.
Anyway, at the risk of making Thanksgiving even less comfortable this year, I’ve excerpted that response below. While I stand by the arguments herein and (apparently) think they’re worth making publicly, I’ll add the caution that what follows is also a textbook example of how not to promote skepticism to someone you care about. Being right doesn’t matter if you say it in a way that no one will listen, and I’m pretty sure I didn’t convince anyone but myself with what you’re about to read. I’ll let you decide whether you think I steered into Dick territory, but I’m giving it due consideration myself.
I think you can count on one hand the number of these sort of things I’ve sent you in the last couple of years. Occasionally I read something that I think my family would find interesting and I send it along. I’m not home during the day, so I didn’t realize quite how credulous Dr. Oz had become…reading this was enlightening to me, and I thought it might be to my family [...] It was really no more than that, and if you don’t want me to send them to you, I’ll gladly stop.
That said, some of the things in your reply I fundamentally disagree with, and I’m going to try to tell you why. If you don’t care, feel free to stop reading now.
Let’s start with “Western Medicine.” This is a term I hear a lot, and only from the alt-med community. It’s intentionally coded language, meant as a pejorative, with roots in postmodern relativism. Maybe you didn’t realize that, maybe you did, but that’s what it means. On the most charitable read, it implies that science is just another way to look at the world, no more or less valid than any other way of looking the world, and has no objective basis. Post-modernists believe that nothing has an objective basis, and that everyone’s subjective experience of reality is equally valid.
More often though, the term ‘Western Medicine’ also implies something more pointed — that science is a tool of imperialism fundamentally opposed to the ‘Eastern Medicine’, which is rooted in an ancient wisdom that has been suppressed by ‘Western’ imperialism. This shouldn’t be surprising — it’s a common theme that runs through much postmodern deconstruction of official histories / narratives. And while deconstruction can be a useful tool to force us to double check our deepest assumptions, esp. as they relate to power structures, when applied to medicine it simply sets up a false dichotomy. ’West’ vs. ‘East’ is a meaningless distinction because geography and history are irrelevant to medicine — the only thing that matters is whether a treatment works.
So the next question becomes, how do we know whether a treatment works? It seems simple — someone has a complaint, they’re given a treatment, the complaint subsides, therefore it works. Unfortunately, there’s a big problem with that: we have no idea that the treatment was the cause of the improvement.
Maybe the complaint was self limiting and would have gone away on its own with or without treatment. Maybe the patient did or took something else unbeknownst to their practitioner that caused the improvement. Maybe the patient was somatizing and the complaint was illusory in the first place. Maybe the improvement was based on the placebo effect. We simply have no idea, and without that, the anecdotes are nothing more than helpful signposts showing where to direct further research.
The main way that researchers get around this problem is the Randomized Controlled Trial — they test the treatment against a placebo, and if it does better they know it works. This, it turns out, is harder than it seems, so a number of modifications have been added over time. Double blinding stops participants and researchers from inadvertently skewing the results. Larger participant pools are used to get around sampling bias. Statistical analysis is used to provide more objective analysis of the resulting data.
While this solved many of the problems within studies, there were still problems with trying to interpret data across studies. Given the sheer volume of research done, and the variability of techniques employed, results across studies are often noisy and hard to interpret. The answer to this is the systematic review (e.g. Cochrane Review), a way of evaluating multiple studies at once, normalizing / weighting for different study methodologies, and coming up with a best read of the data as a whole.
Is this process imperfect? Of course, and this is widely recognized within the research community and discussed openly in the philosophy of science — it’s not some big smoking gun like the alt-med community repeatedly claims. In fact, scientific knowledge never makes absolute claims — by definition, any good research will include an analysis of study limitations, a discussion of questions left unanswered, and critically, a confidence interval indicating how likely the results are to be right. Science is a probabilities game by definition, and any doctor who suggests otherwise is being willfully ignorant, because this is 101 stuff taught in medical school. I’m sure such doctors exist, but despite the vague accusations of the alt-med community, they’re few and far between. If nothing else, the threat of malpractice litigation encourages doctors to be as forthcoming with this information as possible.
Compare all of this to alt-med modalities. How do you measure effectiveness? How do you compare treatments? What’s the process by which the standard of care improves over time? The unfortunate answer is that there’s no answer. After all, why would you want to mess with the wisdom of the ancients? If people have been using it for thousands of years, it must be effective, right? Never mind that the ancients lived until 40 and we live until 80.
What I see repeatedly in alt-med is a reliance on anecdotes, combined with fantastic abuses of scientific research — cherry picking individual studies with no care for the broader body of work, mistaking what happens in test tube studies / preclinical trials for what happens in the real world, and a refusal to change one’s opinions when the facts change.
Now that’s not to say that any specific alt-med treatment doesn’t work. Lots of herbal remedies have been shown to have pharmacologically active ingredients. Aspirin is derived from willow bark, for example. But in most cases, significant research has been done on these treatments, only to find that the effect disappears when you do proper, well-powered randomized controlled trials. The data simply isn’t there, yet the practice continues.
The result is that ‘alt-med’ ends up being anything that either hasn’t been proven to work, or that has been proven not to work. The rest gets incorporated into science-based medicine and is no longer alternative.
So you can hand-wave all you want, call science ‘Western’ as if that were a bad thing, and cast aspersions at science-based researchers for being imperfect. But none of that makes any alt-med treatment work. Only chemistry and physics can do that, and if they are, you can test it and find out.