Mindfulness for Anxiety: What Helps and What to Know

By Tim Yu, Founder & Editor · Published 22 October 2026

Checked against the primary texts · How we check · Report a correction

Mindfulness for anxiety can help, modestly, but it is not a cure and not a replacement for treatment. Its strongest result comes from one large trial in which an eight-week, in-person Mindfulness-Based Stress Reduction (MBSR) course did about as well as the antidepressant escitalopram for adults with diagnosed anxiety disorders. Reviews comparing mindfulness with CBT are more mixed, UK guidance advises against it as a routine option for social anxiety disorder, and for some people turning attention inward makes anxiety worse.

This page is written for understanding, not as therapy or medical advice. The NHS advises seeing a GP if you think you might have generalised anxiety disorder, which is more likely if you feel anxious most of the time and it affects your everyday life. That comes before anything below. For what mindfulness itself is, start with what is mindfulness.

What the Evidence Shows, and What It Was Compared Against

A study’s comparison group decides what its result means. “Better than a waiting list” only shows that doing something beats doing nothing; “as good as a first-line treatment” is a far stronger claim, and few trials test it.

The TAME trial: MBSR versus escitalopram

The key head-to-head study is the TAME trial (Hoge and colleagues, JAMA Psychiatry, published online in November 2022). It randomised 276 adults with a diagnosed anxiety disorder (generalised anxiety disorder, social anxiety disorder, panic disorder or agoraphobia) at three US academic medical centres. Half took an eight-week MBSR course and half took escitalopram, flexibly dosed at 10–20 mg. A clinician who did not know which treatment each person had received rated symptom severity.

Among the 208 who completed the trial, severity scores fell by 1.35 points with MBSR and 1.43 with escitalopram. The difference fell within the margin the researchers had set in advance, so MBSR counted as noninferior (not meaningfully worse). Analyses that included people who dropped out pointed the same way. MBSR was also easier to tolerate. Ten people (8%) stopped escitalopram because of side effects and nobody left MBSR for that reason, and 78.6% of the escitalopram group reported at least one study-related adverse event, against 15.4% in the MBSR group. A later analysis of the patients’ own ratings found no significant difference at week 8, although escitalopram was ahead on some measures at week 4 (Hu et al., 2024).

The limits, most of them named by the authors:

• It was in person, with teachers available every week, and it was a lot of practice: weekly 2.5-hour classes, a day-long retreat in week 5 or 6, and about 45 minutes of home practice a day. The authors say the in-person delivery limits any extrapolation to mindfulness apps or programmes delivered over the internet.
• The two arms were not matched for time and attention. A course means many hours with a teacher and a group. A prescription does not.
• There was no placebo arm. The authors do not list this as a limitation; they note that the escitalopram group improved about as much as in earlier trials that showed escitalopram beats placebo.
• The sample was narrow: mostly women (75%), relatively highly educated, mean age 33, at three urban academic centres. People with bipolar disorder, psychosis, OCD, current PTSD or substance use disorders were among those excluded.

In a second phase run by videoconference, video MBSR did not meet the noninferiority test against video escitalopram, which also did more for panic symptoms and was rated higher for satisfaction. In a non-randomised comparison with the earlier in-person phase, video MBSR did about as well as in-person MBSR (Hoge et al., Journal of Affective Disorders, 2025). TAME is a strong result for one format, not proof that mindfulness in general matches medication.

What the reviews found

A 2010 meta-analysis (Hofmann et al.) is often quoted for large effects, but those were mostly before-and-after changes within the treated groups. The authors’ fail-safe N analysis judged their controlled comparisons unreliable because there were few studies, with one exception: anxiety severity against active-treatment controls.

What about mindfulness apps?

TAME’s results do not extend to apps, so they need their own evidence. A 2024 meta-analysis of 45 randomised trials, reported in 43 papers (Linardon and colleagues, Clinical Psychology Review) found a small effect on anxiety symptoms (g = 0.28) in favour of mindfulness apps over control groups. Twenty-two of the trials used a waiting list as the control and 22 an active control, such as information resources or a non-therapeutic app. Against therapies intended to help, the difference was not significant, though only four trials made that comparison. Most trials studied students and the general public, not people with a diagnosed anxiety disorder, and effects were measured only at the end of the programmes because few trials followed people up. Our guide to Buddhist apps explains how wellness apps such as Calm and Headspace differ from Buddhist ones.

So how good is it?

Read together, structured programmes reduce anxiety more than doing nothing. Against nonspecific comparison activities, Goyal’s review found a small to moderate advantage in mixed groups of adults, but the sources limited to diagnosed generalised anxiety disorder found no significant advantage over active controls: Hoge 2013 on its main outcome, and the Williams 2023 meta-analysis, whose four independent studies include that trial. In one large trial, an in-person course was noninferior to (not meaningfully worse than) a first-line drug. Two questions remain open. Haller’s review found MBSR weaker than CBT, but on few trials and before TAME, while Goldberg found no significant difference between mindfulness and evidence-based treatments for anxiety. And Haller’s team writes that effects “exceeding those of placebo mechanisms remain unclear”. The general case is on our benefits of mindfulness page.

Why it might help is less settled. A 2015 review of mediation studies of MBSR and MBCT (Gu and colleagues, Clinical Psychology Review), covering mental health outcomes generally rather than anxiety alone, found the strongest evidence for changes in cognitive and emotional reactivity, and moderate evidence for mindfulness itself, rumination and worry. The authors add that methodological weaknesses in most of the studies rule out firm conclusions.

Which Kind of Anxiety Matters

“Anxiety” covers ordinary worry as well as diagnosable conditions, and the evidence differs between them. The NHS says anxiety is more likely to be generalised anxiety disorder if you feel anxious most of the time and it affects your everyday life.

Social anxiety disorder is where the guidance is clearest. NICE’s guideline CG159 (2013) says not to routinely offer mindfulness-based interventions, naming MBSR and MBCT, for social anxiety disorder, and recommends individual CBT designed for the condition. A 2021 meta-analysis (Liu et al.) found mindfulness programmes better than no treatment for social anxiety but less effective than CBT. The MBSR curriculum guide itself flags social anxiety (“difficulty with being in a classroom situation”) as a concern to check before accepting someone onto a course. TAME did include people with social anxiety disorder, but its main result was across all four disorders together, and the video phase found social anxiety improved more with in-person care.

For panic disorder, NICE’s guideline is built around CBT. In TAME’s video phase, escitalopram did more for panic symptoms than video MBSR.

The Recognised First-Line Help

Guidelines for anxiety disorders do not put mindfulness first:

• Generalised anxiety disorder (NICE CG113). Stepped care: education and active monitoring, then low-intensity help such as self-help based on CBT principles or psychoeducational groups. With marked functional impairment, or no improvement, the person chooses CBT, applied relaxation or an SSRI, since NICE found no evidence that either therapy or medication is better. Specialist care comes last. Its recommendations do not mention mindfulness.
• Panic disorder (NICE CG113). Self-help comes first for mild to moderate cases. For moderate to severe cases, NICE says to consider CBT, or an antidepressant if the problem is long-standing or therapy has not helped. NICE also advises that anyone who goes to A&E with a panic attack should have checks to rule out an acute physical problem.
• Social anxiety disorder (NICE CG159). Individual CBT developed for social anxiety.
• In the NHS. See a GP. In England adults can refer themselves to NHS Talking Therapies, and a GP will usually suggest talking therapy before medicine.
• In the US (NIMH). Treatment for generalised anxiety disorder usually involves psychotherapy, medication or both. NIMH says a healthy lifestyle “alone cannot replace treatment”. It adds that exercise, mindfulness and meditation can reduce symptoms and “enhance the effects of psychotherapy”.

Mindfulness fits best alongside treatment, or as a practice for everyday anxious moments, and should not delay an assessment. If you are in treatment, ask your clinician whether a course would suit you now.

The Practices the Studied Programmes Teach

None of these practices from the tested programmes is a treatment on its own.

• MBSR’s formal practices. The course tested in TAME teaches the body scan (with 45-minute recordings for the first two weeks), sitting meditation that the 2017 curriculum guide has attend to “breath, body, sounds, and thoughts and emotions” before opening to choiceless awareness, and gentle yoga and walking meditation. Our MBSR page describes each, and the body scan guide gives the method.
• The three-minute breathing space (MBCT). A short three-step practice from Mindfulness-Based Cognitive Therapy (MBCT), which the site for Mark Williams and Danny Penman’s book calls an “emergency meditation” for when your thoughts threaten to spiral out of control. The steps are set out in mindfulness for stress.

How to Practise When You Are Anxious

Each suggestion below comes from a named source. Treat them as things to try, not as fixes.

1. Keep it short. MBSR’s informal practice is “small doses, many times”: brief moments of tuning into the breath, the body, sounds or what can be seen. In a survey of 342 meditators, unwanted effects were reported more often with sessions over 20 minutes, practice alone, and focused-attention meditation (Cebolla et al., 2017). That was a self-selected sample, so it shows an association, not proof of cause, and most of those effects were transient and did not need medical help. Our meditation timer can hold a three- or five-minute sit.
2. Choose an anchor that steadies you. The breath is only one of the objects MBSR uses; sounds and what you can see are on the same list. If focusing on your chest or breathing raises your alarm, switch. Willoughby Britton’s 2019 review notes that high levels of attention to inner body signals are associated with increased arousal, anxiety and panic, effects that have been reported in mindfulness training. The breath caveat and other anchors are covered on Buddhism and anxiety. Paced techniques such as 4-7-8 or box breathing are different: they deliberately control the breath, while the practices above observe it. The difference is explained in breathwork vs Buddhist breathing, and our breathing pacer will hold a slow rhythm on screen if that is what helps.
3. Move. The NHS notes that some people find a busy mind easier to work with during gentle yoga or walking.
4. Name it. The NHS suggests silently naming what is there, for example “This is anxiety”.
5. Use the breathing space when anxiety rises (MBCT). If its middle step, which gathers attention on the breath, unsettles you, the advice in point 2 applies.
6. Ground yourself through the senses (NHS inform). For moments when anxiety or stress feels overwhelming, NHS inform suggests the 5,4,3,2,1 method: sit comfortably, take a deep breath, then focus on five things you can see, four you can feel, three you can hear, two you can smell and one you can taste. NHS inform presents it as grounding, a way to take your mind off uncomfortable symptoms, rather than as mindfulness training.

For restlessness during a sit, see common meditation problems; for everyday pressure, mindfulness for stress; for anxious nights, mindfulness for sleep.

Can Mindfulness Make Anxiety Worse?

It can, for some people. The MBSR curriculum guide itself warns that feelings of sadness, anger or fear “could seem or become stronger as practice develops”. It also notes that symptoms can sometimes worsen, particularly in the early weeks, calls this a normal part of learning, and encourages participants to take any concerns to their teacher. Where a history of trauma, loss or addiction heightens reactions, it suggests deciding with the teacher whether to modify practice, drop the course or wait for a better time.

How often this happens depends heavily on how researchers ask:

• Most trials barely asked. Goyal’s review found no evidence of harm but notes that few trials reported on it. Haller’s review says adverse events were reported insufficiently.
• Pooled reports. A 2020 systematic review (Farias et al.; 83 studies, 6,703 participants) put the overall rate of adverse events at 8.3%: 3.7% in experimental studies and 33.2% in observational ones. Anxiety was the most commonly reported type.
• Active interviewing. When Britton and colleagues (2021) interviewed 96 people after an eight-week MBCT course (adults with mild to severe depression, not an anxiety sample), 37% described a meditation-related effect that had a negative impact on functioning. Lasting bad effects occurred in 6–14%. The authors conclude that these rates are similar to those of other psychological treatments.
• Which practices? The evidence points both ways. Britton’s 2019 review cites a trial in which body scan and breath awareness produced the largest cortisol stress response of the forms compared. Cebolla’s survey found body-awareness practice linked to unwanted effects less often than focused attention. Breath and body focus are not bad for everyone, but they can intensify anxiety for some.

If a practice reliably makes you more anxious, shorten it, change the anchor, or stop, and tell your teacher or clinician. Stopping is a sensible adjustment, not a failure.

Where the Buddhist Roots Fit

MBSR and MBCT are secular programmes. They grew out of Buddhist meditation, as our profile of MBSR’s founder Jon Kabat-Zinn explains, but they are not Buddhism, and the trials above test the clinical courses, not the Buddhist path (secular mindfulness covers that distinction and its critics). One early image many people now apply to anxiety is the Buddha’s “second arrow” in the Sallatha Sutta (SN 36.6): an untrained person struck by painful physical feeling adds mental distress to it, like being struck by a second arrow. The discourse is about painful physical feeling, so reading it as a lens on anxiety is a modern application. Buddhism and anxiety works that through, the second arrow explains the sutta, and our second-arrow reflection turns it into a short exercise.

Mindfulness, studied honestly, is a modest and useful practice. It is not a cure, and it is not a reason to put off seeing someone who can help.