Is CBT Superior?

Rethinking the "gold standard" in light of new evidence

Cognitive-behavioral therapy (CBT) is based on the idea that our thoughts, feelings, and behaviors are interconnected, and we can change how we feel and behave by changing how we think. It’s long been considered the “gold standard” in psychotherapy and is generally perceived to be more effective and scientific than other talk therapies.

But over the last decade, the superiority of CBT has been challenged (Leichsenring & Steinert, 2017). Psychologists skeptical of CBT have pointed out that the literature is low-quality, fails to control for placebo effects, pits CBT against strawman treatments, and suffers from publication bias. Furthermore, when CBT goes head to head with other therapies, they’re often found to be similar in effectiveness. Defenders of CBT point to its long history of clinical use and research, and the fact that no other therapy has been shown to be better. Who’s right?

A new meta-analysis of CBT for depression (Cuijpers et al., 2023) came out a few weeks ago that may help answer this question. It’s the biggest meta-analysis of psychotherapy for a mental disorder ever conducted, with 409 RCTs spanning four decades with 52,702 patients in total. The results show that while CBT is an effective treatment for depression, it doesn’t seem to be more (or less) effective than other psychotherapies.

The authors found that a patient receiving CBT, compared to being waitlisted or receiving care as usual, had a moderate to large reduction in depression symptoms (g=0.79; 95% CI: 0.70-0.89), which was robust to sensitivity analyses and statistically significant. But when CBT was compared to other therapies, such as psychodynamic, behavioral activation, or interpersonal therapy, the overall difference in effectiveness was small (g=0.06; 95% CI: 0­-0.12) and became insignificant when they performed sensitivity analyses (i.e., only looked at studies with low risk of bias, excluded outliers, or adjusted for publication bias.) In the authors’ own words, “the superiority of CBT over other psychotherapies for depression does not emerge clearly from this meta­-analysis.”

Now, it’s reasonable to assume that even if CBT isn’t better than other therapies for treating depression, it might be better for some mental illnesses. However, other meta-analyses looking at different mental disorders have tended to find null or mixed results. For instance, Magill et a., (2019) compared CBT to other therapies in treating substance abuse disorders and found no significant difference. Tolin (2014) and Linardon et al. (2017) found mixed results for anxiety and eating disorders respectively, with both studies noting the low quality of the available literature. My guess is that if studies were higher quality and all meta-analyses were as rigorous as Cuijpers et al., we would find CBT and other psychotherapies to be near indistinguishable in their efficacy for most mental illnesses. (If readers have a different view, let me know in the comments).

Part of the reason that CBT has a reputation for being the best therapy, is that it was the first psychotherapy to be rigorously studied, and other therapies were tested against it:

Cognitive behavioral therapy was the first form of psychotherapy tested with the most stringent criteria (e.g., randomized trials and active comparator) of evidence-based framework used in the health field (e.g., similar for those used in case of pharmacotherapy). Therefore, it was the first psychotherapy largely identified as evidence-based in most clinical guidelines (along with interpersonal psychotherapy for depression). Consequently, many newer, less thoroughly and/or later tested psychotherapies started to use CBT as the reference treatment, often arguing for their efficacy/effectiveness when finding no difference from CBT (Daniel et al., 2018).

Since CBT got its foot in the “evidence-based” door first, and no other therapy could surpass CBT in effectiveness, everyone defaulted to thinking that CBT was substantively better than its alternatives. Throw in low-quality studies and publication bias, and you have a recipe for CBT superiority.

All that said, even if CBT turns out not to be superior to other therapies, it might still be a good idea to say that it’s the “gold standard” and encourage psychotherapists to use it over other methods. For one thing, its reputation as “scientific” and “evidence-based” may lead people who are skeptical of talk therapy to give it a try, since cognitive-behavioral therapy sounds less wishy-washy than, say, psychodynamic therapy. And since CBT is generally prescribed for a limited number of sessions, often between 5 and 20, it tends to be more cost-effective than therapies that take longer or have no clear endpoint. Still, if CBT and other psychotherapies work about equally well, it means that patients can afford to be much less picky. If someone wants therapy and their choice is between CBT in 6 months or another therapy right now, there’s probably no sense in waiting.