Meditation for Anxiety: What Trials Show and How It Compares

Meditation for anxiety has support from randomized trials, but the size of the benefit depends on who is studied and what meditation is compared against. A central study is the TAME trial in JAMA Psychiatry, which randomized 276 adults with a diagnosed anxiety disorder to 8 weeks of MBSR or the antidepressant escitalopram and found MBSR noninferior^[1].

Those results concern diagnosed anxiety disorders, a group that includes generalized anxiety disorder and panic disorder within the broader category of anxiety. Meditation research also includes many trials of adults in nonclinical settings^[6], and the two bodies of evidence should not be read as one.

A 2014 review for JAMA Internal Medicine limited itself to trials with active control groups designed to account for placebo effects^[2]. It included 47 trials and 3,515 participants and found moderate evidence that mindfulness meditation programs improved anxiety, with effect sizes of 0.38 at 8 weeks and 0.22 at 3 to 6 months^[2].

The authors described this as a small to moderate reduction in psychological stress^[2]. The same review found no evidence that meditation programs were better than any active treatment, including drugs, exercise and other behavioral therapies^[2].

TAME (Treatments for Anxiety: Meditation and Escitalopram) recruited adults at 3 urban academic medical centers in the U.S. between June 2018 and February 2020^[1]. Escitalopram was flexibly dosed from 10 to 20 mg, and blinded clinical interviewers rated anxiety severity^[1].

Of those randomized, 208 completed the trial; their mean age was 33 and 75% were female^[1]. After 8 weeks, scores on the Clinical Global Impression of Severity scale fell by 1.35 points with MBSR and 1.43 points with escitalopram^[1].

The bigger gap was in side effects. At least one study-related adverse event occurred in 78.6% of participants randomized to escitalopram and 15.4% of those randomized to MBSR^[1].

Ten escitalopram patients (8%) dropped out because of adverse events, compared with none in the MBSR group^[1]. The most common escitalopram side effects were insomnia or sleep disturbance (41%), nausea (35%) and fatigue (26%)^[1].

A noninferiority result means MBSR was not meaningfully worse, not that it was better. The trial also tested one standardized program taught by qualified instructors, which differs from casual or self-taught practice. Prescription options are covered in more detail in our guide to anxiety medication.

The most direct comparison with cognitive behavioral therapy (CBT) comes from a 2021 meta-analysis in Scientific Reports, which pooled 23 randomized trials with 1,815 adults who had DSM-5 anxiety disorders, testing acceptance and commitment therapy (ACT), mindfulness-based cognitive therapy (MBCT) and MBSR^[3].

Added to usual treatment, all three produced short-term reductions in clinician-rated and patient-rated anxiety compared with usual treatment alone^[3]. Against CBT, ACT and MBCT showed comparable effects, while MBSR showed significantly smaller ones^[3].

Most of the 23 studies had an unclear risk of bias, and the authors concluded that effects beyond placebo mechanisms remain unclear^[3]. The distinction matters: MBCT blends meditation with cognitive therapy techniques, so its showing against CBT does not carry over to meditation practiced alone.

A larger NCCIH-supported analysis from 2018 covered more than 12,000 participants with diagnosed psychiatric disorders such as anxiety or depression^[5]. It found mindfulness-based approaches better than no treatment and about as effective as evidence-based therapies^[5]. Our overview of therapy for anxiety covers the CBT trial evidence.

Meditation covers a range of practices aimed at calming the mind. Some keep focus on a single sensation, such as the breath, a sound, an image or a mantra, which is a repeated word or phrase^[5]. Others train mindfulness, an open awareness of the present moment^[5].

Mindfulness meditation

Most anxiety trials test mindfulness, usually in structured courses. MBSR teaches breath awareness, a body scan that moves attention through one body part at a time, and mindful movement^[1].

Mindfulness-based cognitive therapy

MBCT pairs mindfulness practice with cognitive therapy. In the 2021 meta-analysis, only 3 of the 23 trials tested MBCT, so its CBT comparison rests on few studies^[3].

Mantra and focused-attention meditation

Mantra-based forms, including transcendental meditation, sit outside most of the anxiety-disorder evidence summarized here. The 2021 review was limited to ACT, MBCT and MBSR, and TAME tested MBSR^[3][1].

Breath-focused exercises overlap with meditation but have their own trial literature, covered in our post on deep breathing for anxiety.

Most trial programs run about two months. TAME measured its primary outcome at 8 weeks, and the trials in the 2021 anxiety-disorder meta-analysis had a median treatment length of 10 weeks^[1][3].

Those programs also asked for substantial practice. Standard MBSR includes 45 minutes of daily home practice on top of weekly 2.5-hour classes^[1]. The trials reviewed here did not test how short a daily practice can be and still help, so claims about a minimum number of minutes per day are not backed by this evidence.

Durability is the weakest part of the record. In the 2014 review, the effect on anxiety shrank from 0.38 at 8 weeks to 0.22 at 3 to 6 months^[2]. In the 2021 review, analyses up to 6 and 12 months found no significant differences from usual treatment or CBT^[3].

NCCIH likewise notes that the few studies following participants for longer than 2 months found no long-term effects of the mindfulness-based practices^[5].

Guided recordings and smartphone apps deliver meditation without a live instructor. A 2024 meta-analysis in Clinical Psychology Review pooled 45 randomized trials of mindfulness apps^[4].

Apps produced a small reduction in anxiety symptoms compared with control groups (g = 0.28, 6,082 participants), with a number needed to treat of about 11^[4]. The effect held when the analysis was limited to larger trials and those at lower risk of bias^[4].

Compared with active therapeutic alternatives, apps showed no significant difference for anxiety (g = 0.10), though few trials made that comparison^[4]. The authors called for higher-quality studies with longer follow-ups^[4]. Usage trends appear in our mental health apps statistics.

It can for some people. A 2020 review of 83 studies with 6,703 participants found that 55 studies reported negative experiences related to meditation, and that about 8% of participants had a negative effect^[5].

Anxiety and depression were the most commonly reported negative effects, and the 8% rate is similar to that reported for psychological therapies^[5]. In 3 studies of MBSR with 521 participants, mindfulness practice was not more harmful than receiving no treatment^[5].

TAME found the same pattern. Increased anxiety was the only possibly treatment-related adverse event reported by 5% or more of the MBSR group, affecting 13 participants (11%)^[1]. No serious adverse events occurred in either arm^[1].

Few studies have been designed to detect harm, so NCCIH says definite statements about safety are not yet possible^[5]. It also advises against using meditation to replace conventional care or to postpone seeing a health care provider^[5]. For quick, in-the-moment techniques with a different evidence base, see grounding techniques for anxiety.

Meditation has become one of the most common complementary health practices in the U.S. According to the National Health Interview Survey, the share of adults who meditated rose from 7.5% in 2002 to 17.3% in 2022, making it the most popular of seven approaches surveyed that year^[5].

An earlier CDC analysis tracked a sharp rise over five years: adult meditation use more than tripled from 4.1% in 2012 to 14.2% in 2017^[7].

In 2017, women were more likely than men to meditate (16.3% vs. 11.8%), and adults aged 45 to 64 had the highest use at 15.9%^[7]. These surveys count any meditation, for any reason, so they do not show how many people meditate specifically for anxiety. For how common anxiety itself is, see the anxiety statistics pillar.