do cochrane reviews rate the evidence for cbt as low or moderate certainty at best?
“The Cochrane reviews, which are systematic attempt to assess how strong the evidence for particular approaches to disease of all sorts are, say that at best the evidence for CBT is of low or medium confidence”
a guest, a historian of psychiatry said on Lex Fridman Podcast, 2026-09-21, at 09:13. hear it in context → the speaker cited: Cochrane reviews.
first, the part that holds
the speaker is right about the cochrane reviews we read. across five cochrane reviews of cbt, covering anxiety in children, anxiety in older adults, chronic pain, tinnitus and insomnia in cancer, nothing was rated high certainty. the best rating anywhere was moderate, and most outcomes were low or very low.
what we found when we went and looked
we read five cochrane reviews of cbt. in every one, the strongest rating any result got was moderate. children's anxiety: cbt beat waitlist on remission, 49.4% vs 17.8%, rated moderate quality. chronic pain: results were moderate to low, and the summary says no outcome was rated high. tinnitus: moderate at best. older adults with anxiety: low and very low. insomnia in cancer: one moderate result (time to fall asleep) and the rest low or very low. a separate review of 494 cbt reviews (not a cochrane review) rated 71% of them low or critically low quality. what the claim leaves out: in these reviews 'low or moderate certainty' describes how sure the reviewers are, not whether cbt works. most of the reviews still found benefits, usually small to modest.
assertion by assertion
| assertion | verdict | what the record says |
|---|---|---|
| cochrane reviews conclude that, at best, the evidence for cbt is of low or moderate certainty | confirmed | in all five cochrane cbt reviews we read, the top rating was moderate. children's anxiety: moderate vs waitlist, low vs attention control. chronic pain: 'ranged from moderate to low', with no high-quality outcome. tinnitus: moderate vs audiological care, low or very low elsewhere. older adults with anxiety: low or very low throughout. insomnia in cancer: moderate for one sleep-diary outcome only. limit: cochrane has more cbt reviews than these five, and we did not read all of them. [1][2][3][4][5] |
| cochrane reviews are systematic attempts to judge how strong the evidence is | confirmed | each review we read gives every outcome a certainty grade (high, moderate, low or very low) and explains what lowered it, mainly risk of bias and too little data. [2][3][5] |
| popular framing: low or moderate certainty means cbt doesn't work | misattributed | the certainty grade measures how confident reviewers are in a result, not whether there is an effect. children's anxiety: cbt beat waitlist, 49.4% vs 17.8% remission (moderate quality, number needed to treat about 3). chronic pain: small benefits over usual care for pain and distress (moderate). tinnitus: cbt probably reduces tinnitus impact compared with audiological care (moderate). what is weak or missing: evidence that cbt beats other active treatments or usual care, and evidence that benefits last. [1][2][3][4] |
| the wider body of cbt reviews is mostly of poor quality | confirmed | a review of 494 systematic reviews of cbt (not a cochrane review) rated 351 of them (71%) low or critically low quality. pooling the 71 higher-quality reviews showed a modest benefit to quality of life across conditions (smd 0.23). this rates the quality of the reviews themselves, not grade certainty of the evidence, so it is supporting context only. [6] |
both directions
for the speaker: every cochrane cbt review we read tops out at moderate certainty, and many outcomes are low or very low. that matches the speaker's 'low or medium' exactly. cochrane's summary of the chronic pain review states that no outcome was rated high quality. for older adults with anxiety, the gains after treatment had mostly faded by three to six months (low certainty). in children, the authors found little to no evidence that cbt beats usual care or other treatments. the wider set of cbt reviews is mostly of poor quality: 71% of 494 were rated low or critically low.
against: the quote can leave the impression that cbt has not been shown to work, and these reviews don't say that. in children, cbt beat waitlist on remission, 49.4% vs 17.8%, with a number needed to treat of about 3 (moderate quality). in chronic pain, cbt beat usual care on pain and distress, again at moderate quality, across 59 studies. the low grades come largely from risk of bias and small samples. psychotherapy trials can't hide from patients which treatment they got, so a high grade is hard to reach in this field.
what this cannot say
we read five cochrane reviews of cbt, not all of them. there may be a cochrane cbt review with a high-certainty outcome that we did not find, and that would make 'at best' too strong. the speaker did not say which reviews they meant. we did not find a cochrane review of standard cbt against control for adult depression, the condition people most often have in mind, so the claim can't be tested there. grade ratings are judgments that depend on the outcome and the comparison. the 494-review study rated the quality of the reviews themselves, not the certainty of the evidence. we did not compare cbt's grades with the grades given to medication reviews, so this check says nothing about how cbt stacks up against other treatments. two full reviews on cochranelibrary.com would not load, so for those we relied on cochrane's own plain-language summaries on cochrane.org.
sources, each one fetched and read
- Cochrane plain-language summary of a 2020 Cochrane review, 'Cognitive behavioural therapy for anxiety disorders in children and adolescents', CD013162.pub2 What we read there: 'CBT probably increases post-treatment remission of primary anxiety diagnoses' (49.4% vs 17.8%), moderate quality evidence, number needed to treat 3; 'little to no evidence across outcomes that CBT is superior to usual care or alternative treatments.' Fetched 2026-10-18 (HTTP 200). https://www.cochrane.org/evidence/CD013162_cognitive-behavioural-therapy-anxiety-disorders-children-and-adolescents
- Cochrane plain-language summary of a 2020 Cochrane review, 'Psychological therapies for the management of chronic pain (excluding headache) in adults', CD007407.pub4 What we read there: small benefits for CBT over treatment as usual at treatment end for pain and distress (moderate quality), disability (low quality); evidence quality for CBT outcomes ranged from moderate to low; no outcome rated high quality. Fetched 2026-10-18 (HTTP 200). http://www.cochrane.org/CD007407/SYMPT_psychological-therapy-adults-longstanding-distressing-pain-and-disability
- Cochrane plain-language summary of a 2020 Cochrane review, 'Cognitive behavioural therapy for tinnitus', CD012614.pub2 What we read there: 'CBT probably reduces the impact of tinnitus on quality of life' vs audiological care (moderate certainty); vs no intervention, low certainty; anxiety very low to low certainty. Fetched 2026-10-18 (HTTP 200). https://www.cochrane.org/evidence/CD012614_cognitive-behavioural-therapy-adults-tinnitus
- Cochrane plain-language summary of a 2024 Cochrane review, 'Cognitive behavioural therapy and third-wave approaches for anxiety and related disorders in older people', CD007674.pub3 What we read there: 'CBT may result in a small-to-moderate reduction of anxiety post-treatment' (low-certainty evidence); little to no difference at three to six months (low certainty); clinical recovery very low certainty. Fetched 2026-10-18 (HTTP 200). https://www.cochrane.org/evidence/CD007674_cognitive-behavioural-therapy-and-third-wave-approaches-anxiety-and-related-disorders-older-people
- Cochrane plain-language summary of a 2025 Cochrane review, 'Cognitive behavioural therapy for insomnia in people with cancer', CD015176.pub2 What we read there: insomnia severity very low certainty, sleep quality low certainty; moderate certainty only for sleep onset latency; 'very low-certainty to moderate-certainty evidence' overall. Fetched 2026-10-18 (HTTP 200). https://www.cochrane.org/evidence/CD015176_does-cognitive-behavioural-therapy-insomnia-cbt-i-help-people-cancer
- 'The evidence for cognitive behavioural therapy in any condition, population or context: a meta-review of systematic reviews and panoramic meta-analysis', Psychological Medicine, 2021 (PMC7856415) What we read there: 494 systematic reviews, 221,128 participants; 351 of 494 reviews rated low or critically-low quality on AMSTAR-2; 'CBT produced a modest benefit across-conditions' on health-related quality of life (SMD 0.23). Fetched 2026-10-18 (HTTP 200). https://pmc.ncbi.nlm.nih.gov/articles/PMC7856415/
speakers are described by role, not named. one sentence is quoted under fair use, with attribution and a link to the original at the timestamp; we do not re-host audio or video. nothing on this page is a reason to start, stop or change a medication. if you have the file behind anything marked unverifiable, the research board is where people can answer. adam