Eight weeks of mindfulness matched a first-line anxiety drug in a randomised trial

A randomised trial that pitted an eight-week mindfulness course against a standard first-line antidepressant for anxiety disorders found the two produced nearly identical results: a mean reduction of 1.35 points on the primary severity scale for mindfulness-based stress reduction and 1.43 points for escitalopram, a difference of −0.07 points (95% CI −0.38 to 0.23; P = .65) [s1]. The trial's prespecified noninferiority criterion was met [s1].

This is one of the few places in the mental-health literature where a popular non-drug intervention has been tested head-to-head against a drug under conditions that allow a real comparison, and the detail of how it was done is what gives the result its weight.

How the trial was built

The Treatments for Anxiety: Meditation and Escitalopram trial used a noninferiority design with a prespecified margin of −0.495 points on the Clinical Global Impression of Severity scale [s1]. Patients were recruited between June 2018 and February 2020 at three urban academic medical centres in the United States [s1]. Of 430 individuals assessed for inclusion, 276 adults with a diagnosed anxiety disorder were recruited, and 208 completed the trial [s1].

Participants were randomised 1:1 to eight weeks of a weekly mindfulness-based stress reduction course or to escitalopram, flexibly dosed from 10 to 20 mg [s1]. Outcome assessments were performed by a blinded clinical interviewer at baseline, at the week 8 endpoint, and at follow-up visits at 12 and 24 weeks [s1].

The primary noninferiority sample comprised 208 patients — 102 in the mindfulness arm and 106 in the escitalopram arm — with a mean age of 33 years (SD 13) and 156 participants (75%) female [s1]. On race and ethnicity, 32 participants (15%) were African American, 41 (20%) Asian, 18 (9%) Hispanic/Latino, 122 (59%) White, and 13 (6%) of another race or ethnicity [s1].

Baseline severity was closely matched between arms: mean CGI-S of 4.44 (SD 0.79) for the mindfulness group and 4.51 (SD 0.78) for the escitalopram group in the per-protocol sample, and 4.49 (SD 0.77) against 4.54 (SD 0.83) in the randomised sample [s1].

What noninferiority does and does not mean

A noninferiority trial asks whether a new treatment is not worse than an established one by more than a prespecified amount. That amount is a judgment call made before the data are seen, and here it was −0.495 CGI-S points [s1]. The lower bound of the observed confidence interval fell within that margin, so the trial's conclusion follows from its own design [s1].

What such a trial cannot do is prove equivalence. It bounds how much worse the new treatment might be. It also cannot show that either treatment beats no treatment, because there was no placebo or waiting-list arm — both groups received an active intervention. And a blinded interviewer does not make a blinded participant: someone attending a weekly meditation course knows they are not taking a tablet, and expectancy is not controlled away by blinding the assessor.

The wider meditation literature, which is more mixed

The trial sits inside a body of evidence that is considerably more equivocal. A systematic review and meta-analysis in JAMA Internal Medicine screened 18,753 citations and included 47 trials with 3,515 participants, restricting inclusion to randomised trials with active controls for placebo effects [s2].

Mindfulness meditation programs showed moderate evidence of improved anxiety, with an effect size of 0.38 (95% CI 0.12–0.64) at eight weeks and 0.22 (95% CI 0.02–0.43) at three to six months [s2]. For depression the figures were 0.30 (95% CI 0.00–0.59) at eight weeks and 0.23 (95% CI 0.05–0.42) at three to six months, and for pain 0.33 (95% CI 0.03–0.62) [s2]. Evidence for improved stress/distress and mental health-related quality of life was rated low [s2].

On a longer list of outcomes the review found either low evidence of no effect or insufficient evidence of any effect at all: positive mood, attention, substance use, eating habits, sleep and weight [s2].

And the finding that speaks directly to the trial above: the reviewers found no evidence that meditation programs were better than any active treatment, including drugs, exercise and other behavioural therapies [s2].

Reading the two together

The two results are consistent, and the way they are consistent is the point. A meta-analysis found no evidence that meditation outperforms active treatments [s2]. A trial then found that it did not underperform one either, by more than a prespecified margin [s1]. "Comparable" is the word both support; "superior" is supported by neither.

The effect sizes in the meta-analysis are small to moderate, and the review's authors framed the clinical implication modestly: clinicians should be aware that meditation programs can result in small to moderate reductions in multiple negative dimensions of psychological stress, and should be prepared to discuss the role such a program could have [s2]. They called for stronger study designs [s2].

The limits worth carrying away

The trial is 208 completers at three American academic centres, three-quarters female, with a mean age of 33 [s1]. Whether the result holds in older adults, in primary care rather than academic settings, or in people who have already failed a treatment, is not something this trial answers. The meta-analysis predates the trial by nearly a decade and its search ended in November 2012 [s2].

Neither source supports a recommendation for or against either treatment for any particular person. That is a clinical decision, and this article — which describes trial results and is not medical advice — is not the place it should be made.

Sources

• Mindfulness-Based Stress Reduction vs Escitalopram for the Treatment of Adults With Anxiety Disorders: A Randomized Clinical Trial — JAMA Psychiatry, 2022-11-09
• Meditation programs for psychological stress and well-being: a systematic review and meta-analysis — JAMA Internal Medicine, 2014-01-06