Detailed guides to painful problems, treatments & more

The unflattering evidence on CBT for pain

 •  • by Paul Ingraham
Get posts in your inbox:
Weekly nuggets of pain science news and insight, usually 100-300 words, with the occasional longer post. The blog is the “director’s commentary” on the core content of PainScience.com: a library of major articles and books about common painful problems and popular treatments. See the blog archives or updates for the whole site.

A big 2020 review of psychotherapy for chronic pain, mostly cognitive behavioural therapy (CBT), is resoundingly negative (see Williams et al). Not that you can actually tell from the conclusion, which is superficially positive:

“We found sufficient evidence across a large evidence base (59 studies, over 5000 participants) that CBT has … beneficial effects for reducing pain.”

I deliberately omitted a couple important words before “beneficial” there. Just how beneficial were these beneficial effects? “Small” or “very small,” even “slight”!

The longer I do this job, the more I think that the weak positive can be a particularly informative and trustworthy result, either true and underwhelming, or wrong in the direction of overstatement. You can trust that the reported benefits are minor at best. So this a classic case of damning with faint praise (which is the fate of most pain treatments). No one wants a dozen sessions of psychotherapy for a “slight” benefit. That is not a good return on investment. Ain’t nobody got time for that!

While nothing is ever the “last word” in this business, I think this paper comes close. It’s the most recent of several versions and “by far the biggest,” and mostly about cognitive behavioural therapy (CBT), which is by far the dominant mainstream option. (There wasn’t enough data about all other therapies combined to fill a bathtub, let alone a statistical pool with a deep end.)

I doubt I’ve ever seen a scientific review that didn’t seem a bit embarrassed by what a sorry mess the evidence is. In this case, “a large number of underpowered trials with poor evaluation practices raise concerns about research waste.” But there was so much data on CBT that there were enough good-apple studies to work with for a reasonable conclusion.

CBT’s good reputation for great results is not backed up by the data.

I discuss this paper in more detail in my full review of cognitive behavioural therapy, where I also have some nice things to say about psychotherapy — despite this evidence.

P.S. There have been a lot of posts on the psychology of pain lately — a strong theme. Don’t worry, I will return to less squishy subject matter! For instance, there's a post about injuries just about ready to go for next time: “Why don’t all injuries hurt? Five theories.”