Can back pain be treated psychologically? Is psychotherapy for back pain — mainly cognitive behavioural therapy, CBT — actually based on evidence?
Several months ago, Scott Gavura (pharmacist and blogger for ScienceBasedMedicine.org), asked for my take on a research summary from a regional medical association’s newsletter, which claimed that CBT is indeed full-fledged evidence-based medicine. Best practices. The right stuff for the job.
News to me! Everything I’d ever seen on this topic looked kinda underwhelming.
But that wasn’t a proper answer. This post is the proper answer, the result of months of diligent effort to fully answer the question — for Scott, for myself, and now for you — and to create a solid new chapter out of it for my back pain book.
(This also inspired a broader initiative to look at other mind-over-pain cures. Thanks, Scott! That’s a lot of inspiration.)
Do you have any idea how hard it is to find stock photograph of ‘psychotherapy’ without very conspicuous clipboards or notepads?
PainSci Member Login
Submit your email to unlock member content. If you can’t remember or access your registration email, please contact me. ~ Paul Ingraham, Publisher
Found! 🙂 Member content on this page only has just been unlocked. To unlock member content on all other pages, see the confirmation email just sent. (If it doesn’t turn up in your inbox, check your spam folder! Email can also sometimes take a few minutes. If it never turns up, just contact me.)
Found, but… you are not an active PainSci member, so this content cannot be unlocked. See your confirmation email for more information about your account.
Not found! 🙁 Sorry, but that email address is not in the PainSci database.
⚠️ Sorry, too many lookups right now. This is a rare error. It should go away if you try again in a little bit.
Privacy & Security of this form This login is private and secure: the information you submit is encrypted, used only to search for matching accounts, and then discarded.
This post is for PainSci members only. There’s also a ramblier-than-usual audio version for members. The full post is in the “fairly huge” category for the blog, but it’s not a beast — about 1500 words, a trifling 5 minutes or so for the faster readers.
To member-ize yourself, JOIN NOW for USD $6/month or $60/year. Payment will unlock this post, many others, and much else. Membership fees make PainSci possible.
If you’re specifically keen on back pain, but you’re not a member, you can also get it by just buying my back pain book.
Cancelling membership is easy & stops future charges. Payment data handled by Stripe.com. More about security/privacy & refunds. PainScience.com is a small publishing company in Vancouver, Canada, since 2001.
The salamander’s domain is for people who are serious about this subject matter. If you are serious — mostly professionals, of course, but many keen patients also sign-up — please support this kind of user-friendly, science-centric journalism.
Coping or curing: what is the goal of cognitive behavioural therapy for back pain patients?
Cognitive behavioural therapy, the infamous CBT, is the elephant in the mind-game room: the overwhelming mainstream favourite, the champion of psychotherapy for all kinds of chronic pain.1
Many patients scoff at CBT for back pain, for the obvious reason that a psychological solution seems to presume a psychological cause, entirely or at least mainly. At best that seems uncharitable to most patients, and at worst it triggers outrage by pushing the “all in your head button.” I have made the case in this book that back pain is definitely affected by the mind, but it is not fundamentally caused by it — in other words, it’s not a “psychosomatic” problem.
With CBT, it’s sometimes hard to tell if it’s about treating a cause or “just” helping people cope.
Usually it seems more like the latter, the official idea being to help people “live a full life with confidence in managing that pain.”2 But many a psychologist crosses the line by hinting that better coping can also lead to lower pain levels … which in turn makes it easier to cope … and so on. Wisely or not, consciously or not, mental health professionals often end up aspiring to treat pain indirectly via that virtuous cycle. And that sends a not-very-subtle message to patients: if you can use your mind to get out of this mess, then your mind must have gotten you into it in the first place.
And some psychologists go further, skip the complex rationalization, and assume that CBT treats the underlying psychological cause for back pain. They may do so with without spelling it out. If it was spelled out, it would go something like this: “You need CBT to treat the stress and anxiety that is the root cause of your back pain.” I don’t think this assumption is common, but it definitely happens. That might sound a bit outrageous, but it harmonizes well with the pervasive demonization of stress and anxiety in the modern world. Stress and anxiety is now blamed from everything from insomnia to cancer — and so why not back pain too?
In short, although most psychotherapy providers think or behave like they are helping with the consequences of back pain, which could conceivably indirectly address the cause… but a few unwisely believe that they are directly attacking the cause of back pain.
Are any of them right? However CBT supposedly works, does it work? What does the data say?
Sorry, but the CBT data says “hell no”
It’s easy to find clinicians who have the opinion that CBT is evidence-based care for back pain. I do not agree. I see the usual scattershot mess of inconclusive clinical anecdotes, and I have studied the literature thoroughly and found no basis for those opinions. For example…
A colleague of mine from my days at ScienceBasedMedicine.org forwarded me a newsletter post from a regional medical association to its members touting the power of CBT for back pain and looking for all the world like a credible source of evidence-based medical information. But they had cherry-picked a handful of positive-looking sources, all with major issues, taken them at face value, and then — this boggles my mind — cherry-picked even more aggressively by just copying and pasting the single best-sounding result into their “bottom line”!
And then they padded their footnotes with a few more really weak citations, and even a couple irrelevant citations. Boo! I’ve seen more rigorous undergrad term papers.
CBT for back pain specifically has not actually been studied much, despite the fact that both parts of that equation are a Big Deal. CBT is big business, and there’s lots of CBT research. And back pain is also studied to death. And yet those research circles barely overlap! As well studied as CBT may be in general, trials of CBT specifically for non-specific back pain… neglected!
Yet again we run into the limitations of modern musculoskeletal medicine. We are stuck mostly extrapolating from other sources, like CBT for chronic pain in general.
"Moody & frustrated" due to pain & exhaustion is the main reason most back pain patients might need any kind of psychotherapy — not to solve the pain, but to help cope with it.
The closest thing we have to direct evidence on psychotherapy for back pain: two big bad studies
There are two unavoidable “positive” studies on this topic, routinely cited by anyone who stands to benefit from promoting psychological therapy for back pain:
- Cherkin et al in 2016 comparing CBT to mindfulness,3
- Ashar et al in 2022, a study of Pain Reprocessing Therapy (which is not entirely CBT-ish, but somewhat).4
They are “unavoidable” because both have strongly positive results, and both were published by good journals (JAMA and JAMA Psychiatry), so they are impossible to ignore. They also share some major problems:
- High risk of bias. (Very high in the case of Ashar et al. It’s worth zooming in on, which I do in the next section, which focusses on that paper.)
- Results so good they just can’t be accepted without replication. (And yet they are too different to back each other up. They are each, independently, “just one study.”)
- The same major technical flaw … a flaw so serious that Dr. Edzard Ernst (a true expert in research methods) called it “unethical pseudo-science.”5
The 2016 trial showed that both meditation and CBT were effective for back pain. It’s often cited in praise of meditation, but the results actually flattered CBT a little more. Good news? Unfortunately, my bottom line is that I simply don’t trust the study.
The 2022 trial is even more impressively sketchy, though, with particularly extreme conflicts of interest, a second bizarre and serious methodological flaw, and results that were unusually far into too-good-to-be-true territory. The weren’t just good, they were great, they were bloody amazing — far better than we have come to expect for any kind of treatment for any kind of serious chronic pain (let alone a psychological therapy). And so I trust this study even less.
Either of these untrustworthy trials could be right, but there’s just no way I’m buying that without replication. Their only value for now is to inspire other researchers to check their work.
Less direct and relevant evidence?
A huge 2020 review of psychotherapy for chronic pain in general was quite negative — and of course back pain is the most prevalent kind of chronic pain.6 So it’s not a half bad sub for more focused research. CBT is the most studied type of psychological treatment for pain by a wide margin, so there's plenty of that general data … and it shows a “slight” benefit only, which is effectively synonymous with a negative result.
No one wants a dozen sessions of psychotherapy for a “slight” benefit. That is just not a good return on investment. Ain’t nobody got time for that.
If CBT and psychotherapy for pain in general is underwhelming, it’s probably underwhelming for back pain specifically. It’s unlikely that back pain — the canonical example of stubborn musculoskeletal pain — is going to be significantly more responsive to psychotherapy than other kinds of pain. Possible, but unlikely. And that is generally what the research shows.7 Results tend to be either negative or only weakly positive.
CBT for back pain’s cousin, neck pain, is equally unimpressive according to Monticone et al in 2015.8
And that’s just all there is. None of these means that the mind is irrelevant to back pain or that people with back pain should never consider psychotherapy. Nor does it even mean that CBT has actually been entirely knocked out of the race. But it’s not looking good, and the widespread belief that “CBT for back pain is good evidence-based care” is not even close to true.
- There are other ways to psychotherapize, of course, but CBT is so dominant that it’s really the only option worth discussing. No other approach can be standardized enough to study! (And that’s kind of a problem with CBT, too, actually.)
- EvidentlyCochrane.net [Internet]. Williams A. Managing chronic pain in adults: the latest evidence on psychological therapies; 2020 October 8 [cited 20 Oct 8].
- Cherkin DC, Sherman KJ, Balderson BH, et al. Effect of Mindfulness-Based Stress Reduction vs Cognitive Behavioral Therapy or Usual Care on Back Pain and Functional Limitations in Adults With Chronic Low Back Pain: A Randomized Clinical Trial. JAMA. 2016;315(12):1240–9. PubMed #27002445 ❐ PainSci #51980 ❐
- Ashar YK, Gordon A, Schubiner H, et al. Effect of Pain Reprocessing Therapy vs Placebo and Usual Care for Patients With Chronic Back Pain: A Randomized Clinical Trial. JAMA Psychiatry. 2022 01;79(1):13–23. PubMed #34586357 ❐ PainSci #51974 ❐
- Ernst E, Smith K. More Harm than Good? The Moral Maze of Complementary and Alternative Medicine. 1st ed. 2018 ed. Springer International Publishing; 2018. p. 76. Writing specifically about the Cherkin study:
In other words, this is yet another trial with the “A + B versus B” design. As with our other examples, because A + B is always more than B (even if A is just a placebo), this study design could never have generated a negative result! The results are therefore entirely compatible with the notion that the two tested treatments, MSBR and CBT, are pure placebos. Add to this to the disappointment many patients in the ‘usual care group’ might have felt for not receiving an additional therapy for their pain, and you have a most plausible explanation for the observed outcomes.
The “A + B versus B” design can only produce positive findings. Any such study allegedly testing the effectiveness of therapy XY and concluding that “it is effective” ought to be categorised as unethical pseudo-science.
- Williams AC, Fisher E, Hearn L, Eccleston C. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database Syst Rev. 2020 Aug;8:CD007407. PubMed #32794606 ❐
Nothing is ever the “last word” in this business, but this paper comes close. It’s the latest of several versions and “by far the biggest,” with data pooled from 9400 trial subjects in dozens of experiments.
A few technical notes: “Mindfulness” was excluded, as was treatment for headache. Only professional face-to-face delivery was considered. The researchers shunned small studies to avoid “a higher risk of unrealistically positive results.” They analyzed 75 studies — more than double the last version — representing 9400 people, two thirds women, mostly middle-aged, mostly with back pain, fibromyalgia, and arthritis.
It was mostly about three related methods, and primarily CBT. There wasn’t enough data about all other therapies combined to fill a bathtub, let alone a statistical pool with a deep end.
- Henschke N, Ostelo RW, van Tulder MW, et al. Behavioural treatment for chronic low-back pain. Cochrane Database Syst Rev. 2010;(7):CD002014. PubMed #20614428 ❐
This review of the results of dozens of scientific studies shows that behavioural therapies for low back pain have generally been failing the “impress me” test. It is possible that behavioural therapy is more effective for a certain kind of patient. However, if so, apparently there are not enough of those kinds of patients, or the effect is not big enough, to have any discernible effect on the average results of experiments. If evidence of a benefit is being “washed out,” it is being washed out rather easily. Behavioural therapy might work, a little, for some, but scraps of efficacy hardly seem worth fighting over.
- Monticone M, Cedraschi C, Ambrosini E, et al. Cognitive-behavioural treatment for subacute and chronic neck pain. Cochrane Database Syst Rev. 2015 May;(5):CD010664. PubMed #26006174 ❐ PainSci #52083 ❐
This flaw was also highlighted by Dr. Harriet Hall regarding Ashar et al in her review of that paper.