Dec 11 2020

Skeptical of Skepticism regarding Medical Skepticism

In a recent article in Medpage Today, Vinay Prasad offered his critiques of what he calls “medical skepticism”. Essentially he is talking about Science-Based Medicine and all my colleagues who engage in related activities. I am always open to criticism and love to engage about these topics. Unfortunately, Prasad’s criticism’s were based largely on his ignorance of what it is, exactly, that we do, wrapped around some huge logical fallacies. They are also arguments we have dealt with on numerous occasions before, so he could have saved time by just reading some of the very literature he felt knowledgeable enough to criticize. (And as an aside, the “skeptical of skeptics” meme is way overdone and ready to be retired.)

If I had to give a paraphrasing executive summary of Prasad’s article it would be this – medical skeptics should stop focusing on what they think is important, and should instead focus on what I think is important, even though I don’t really understand what it is that they do. In fact there is so much wrong with Prasad’s article it’s hard to know where to begin, but let’s start with some basic framing. Part of what Prasad is criticizing is our science communication (scicom), but again he seems to be unaware that scicom is a field unto itself, and so he is making some basic false assumptions, without being aware that he is doing so. This false assumption leads Prasad to conclude that medical experts should restrict themselves to the big problems within their area of medical expertise, without seeming to realize that scicom itself is an area of expertise.

Before I go further it is important to understand what we in the Science-Based Medicine and broader skeptical community do, and what our expertise actually is. First, we are science communicators, and this involves studying science communication itself. The big lessons of the last few decades, backed by actual research, is that the old “knowledge deficit” paradigm is mostly incorrect (not completely) and definitely insufficient. In most contexts you cannot change the way people think or behave by just giving them facts. You have to also engage with what they already believe and the complex motivations and patterns of belief that underlie them. Scicom involves, therefore, not just addressing scientific literacy but also critical thinking skills and media literacy. And in order to do this you need to understand the complex relationship between science and pseudoscience, and cognitive biases, conspiracy thinking, science-denial, and a host of other “critical thinking” skill sets.

Those of us who focus on scicom within medicine have further been studying how all this applies not just to science in general but specifically to professions that apply science in a clinical setting. Science-based medicine (SBM) is essentially the culmination of all this. This involves understanding how to properly apply statistics, the effects of publication bias, citation bias, and other perverse incentives that shape the medical literature, the nuances of placebo effects, p-hacking, how to think about when it is proper to adopt a practice, how to think about subjective outcomes, and many other details. What Prasad fails to understand is that this is a specialty unto itself. You might naively think that all medical scientists should (and therefore mostly do) understand all this, but that is demonstrably incorrect. P-hacking, for example, is rampant in the literature and most researchers seem to be unaware they are doing it, or at least the magnitude of the resulting effect. I lecture frequently on SBM, as you might imagine, and I often ask people who are not only academics but experts in their relevant field basic questions, such as “have you heard of funnel plots”, and I will often get blank stares.

Another fundamental pillar of what we do is an understanding of the relationship between science and pseudoscience, with specific reference to the “demarcation problem“. There is, in fact, no clean separation between pristine science and rank pseudoscience. That would be a false dichotomy. Rather, there are practices that are more scientific and others that tend toward pseudoscience. When you look at any individual discipline, or claim, or practice, it can lie anywhere along the spectrum from science to pseudoscience. This further means that a lot of pseudoscientific practices and arguments find their way into mainstream science. Perhaps they are just more subtle, but they are there. All scientists, therefore, need to have a thorough understanding of pseudoscience in order to ward against its intrusion and to be vigilant against such practices themselves. To adapt a common saying – those who are ignorant of pseudoscience are doomed to repeat its failings. One of the criticisms of SBM against “ivory tower” science, in fact, is this false dichotomy – the thought that they do not need to engage with or understand pseudoscience because it is beneath them. It turns out, not so much.

Prasad makes all of these rookie mistakes in his criticism, and demonstrates he understands none of this. This is a classic error – to criticize an entire field you don’t understand. Prasad doesn’t even seem to recognize that it is a field, let alone its fundamental principles and supporting literature. Prasad’s framing, therefore, is a logical fallacy, the false dichotomy of saying that we should focus on alternative medicine less and problems within mainstream medicine more. But this is not how we frame what we do. We chose the name “science-based medicine” very deliberately. We are not anti-CAM. We promote one scientific standard within all of medicine – there is not alternative medicine and mainstream medicine (that is their framing), there is only medicine. The standards we promote apply up and down the spectrum of pseudoscience, without any demarcation point.

I will say that we do specifically apply those standards to the lower end of the spectrum, practices that are mostly pseudoscience, for several very good reasons. First, most academics (operating under many of the false assumptions I outlined above) are “shruggies” who don’t see the threat of pseudoscience within their own profession and so ignore claims they perceive as being on the fringe. This is a fatal mistake. If you will excuse a military metaphor, they are not guarding their left flank, leaving them open to conquest – and that conquest is happening, right under their nose. Further, while we are guarding medicine’s left flank, there is a tremendous amount of wisdom to learn by studying extreme pseudoscience, providing lessons that can be applied to all of science. I often liken this to studying an advanced disease, which might help a practitioner recognize the more subtle manifestations of the disease in its early or mild form.

Prasad’s framing is therefore fatally flawed, but let me address some of his specific arguments. His core argument is based on another logical fallacy – the fallacy of relative privation. This is the fallacy of arguing that something is not important enough to address, because something else is more important. This ignores the fact that everything is relative – there is always a bigger problem out there. Further, there are over 7 billion people on this planet. There are enough people to go around – every problem can have its champion. Let a thousand flowers bloom, as they say.

Prasad’s logic, for example, could apply to every orphan drug. Why spend finite research dollars on drugs for rare diseases that will by definition help a few people. We should focus on the most common diseases. There is a kernel of validity here – we should apportion limited resources appropriately based upon need and impact. But this does not have to be absolute, otherwise all medical researchers would focus on the one biggest problem (whatever that is) until it is completely solved. We should spend more money and resources on the bigger problems, sure, but if someone wants to spend their time curing a rare disease, go for it. Again I will point out that curing that rare disease may provide insights that ultimately help cure a much bigger disease – just like studying pseudoscience provides insights that help all of science.

The arguments that Prasad lays out are therefore entirely based on logical fallacies and a fundamental misunderstanding of modern scicom and the role of skepticism in general and science-based medicine in particular. But I also don’t think he has a grasp on the relevant facts. He characterized homeopathy, for example, as a small problem – too small to be worthy of attention (and not just his attention – the attention of others). But homeopathy is a nearly 16 billion dollar industry world wide, and growing. That is a lot of health care dollars that could be spent more productively. Homeopathic products are in our pharmacies, and in some (mostly European) countries are woven into mainstream medicine. He also singles out acupuncture, another massive industry. Is he aware that billions of people live in cultures that take the efficacy of acupuncture for granted?

This definitely seems like a “see no evil” problem (again, an “ivory tower” phenomenon). You have to engage with the culture and with your patients specifically on their attitudes and uses of unscientific medicine, and also look beyond your borders, to see the true scope of the issue. Prasad is also implying a “what’s the harm” fallacy, which he makes more explicit elsewhere in his article. He states, for example:

All things being equal, we should focus our time on interventions that have greater harms — death, perforation, cardiac arrest — than those that offer minor harms — cutaneous bruising. We should focus more on things that are invasive — placed inside or within — that those that are merely superficial.

This is partly true, but it is incomplete. It ignores harm that is not directly physical. Convincing a patient that magic is real, for example, may not puncture their lung, but can have dramatic downstream negative effects. Any physician who has dealt with a patient dying of a serious illness, but ignoring effective therapies in favor of slickly marketed ineffective treatments, has seen this harm. Or – just pick your head up and see what is happening right now with the anti-maskers and now anti-vaxxers foiling our attempts at controlling a deadly pandemic. How many more people will die from COVID-19 because of pseudoscientific medical beliefs in the public? This number is impossible to pin down, but that does not mean it is not real or large.

Seriously – in the middle of a massive fight between science and pseudoscience while a deadly pandemic rages, and is currently killing more than three thousand Americans a day, Prasad wants to play the “what’s the harm” card to argue that opposing pseudoscience is not all that important? Prasad dismisses all this as people wearing copper bracelets.

There is much more, but let me end with this final nugget. Part of Prasad’s advice is this:

Your unique skill set. In deciding what you should spend your time on, it is important to think about what you are uniquely qualified and able to comment on.

I agree. But Prasad simply does not recognize that science-based medicine, science communication, and scientific skepticism are “skill sets”. They are deep and broad skill sets, that we have developed over decades, and are informed by a growing scientific literature. If Prasad had any inkling of this then perhaps he would follow his own advice and only comment on topics he is “qualified and able to comment on.” Or at the very least, engage with the very people you are criticizing – do a little “peer review”.  Alternatively he could have read some of the articles we have already written on this topic. The point of his article ostensibly was to save us time we are currently “wasting”, but all he succeeded in doing was in wasting time having to correct (again) his own misconceptions and lack of knowledge.


No responses yet