How much of modern medicine is “evidence-based”? (briefly revisited)

I once blogged about an article attempting to address the very question in the title of this post, and I’ve also discussed in depth how messy the process of evidence-based medicine can be and why that provides an opening for purveyors of “alternative medicine” (my preferred term to describe it being “non-evidence-based medicine”) to respond to complaints about the lack of evidence supporting their favored woo with a hearty but fallacious tu quoque.

One of the favorite claims of purveyors of non-evidence-based “alternative” medicine is that modern scientific medicine is actually not very evidence-based. Various credulous supporters of such woo will often throw about claims that less than 50%, and perhaps as little as 15%, of modern medical therapies are based on valid scientific evidence. I had always wondered where that figure came from.

Now I know. I’m grateful to Steve Novella for telling me. It turns out that the 15% figure dates back nearly a half-century ago, all the way back to 1961 and wasn’t even looking at how much of medical practice was truly “evidence-based.”

Not surprisingly, Steve then goes on to show that the true figures are very much higher than the canard being thrown around by fans of unscientific, non-evidence-based therapies. Of course, even if the figure were 15% as is often claimed by some advocates of so-called “alternative” medical therapies, that figure would still be at least an order of magnitude greater than the evidence base supporting their favorite woo. That’s one difference between scientific, evidence-based medicine and woo. The other difference is that scientific medicine is always striving to increase the percentage of its therapies and practices that are based on science and well-designed clinical trials, abandoning therapies that are found not to work or not to work as well as newer therapies, while it is incredibly rare for an “alternative” therapy ever to be abandoned by its practitioners even when it is shown conclusively not to be efficacious.