Does Meditation Actually Reduce Stress? What the Research Says

What You Need to Know

Why the Claims About Meditation Outrun the Evidence

Meditation apps, wellness marketing, and casual health advice often present meditation as a near-universal antidote to stress. That framing isn't baseless — there is a real evidence base behind mindfulness-based interventions — but it does overstate what the research actually shows, particularly for the specific claim that meditation reduces stress. The research below is more interesting, and more useful, than the marketing version: it shows where the effect is solid, where it's weaker than assumed, and where the mechanism is more complicated than "calm down."

What the Best Evidence Actually Shows

1. Solid evidence for anxiety and depression, weaker evidence for stress itself

Goyal M, Singh S, Sibinga EM, et al., 2014 — JAMA Internal Medicine

A systematic review and meta-analysis of 47 randomized controlled trials (3,515 participants) examining meditation programs — most using mindfulness-based approaches like MBSR — for psychological stress and well-being.

Finding: the review found moderate evidence that mindfulness meditation programs improve anxiety, depression, and pain, but only low evidence of an effect on stress/distress and quality of life. Critically, the review found no evidence that meditation programs were more effective than active treatments — medications, exercise, or other behavioral therapies — on any outcome studied.

2. A broader meta-analysis found moderate effects across the board

Khoury B, Lecomte T, Fortin G, et al., 2013 — Clinical Psychology Review

A comprehensive meta-analysis of 209 studies on mindfulness-based therapy (MBSR, MBCT, and related programs) across clinical and non-clinical populations.

Finding: mindfulness-based therapy showed moderate effect sizes for improving anxiety, depression, and stress, and was comparable in effectiveness to cognitive behavioral therapy and other established treatments — a more favorable picture for stress specifically than Goyal et al. found, reflecting how much the result depends on which studies and outcome measures are pooled.

3. Meditation can blunt the stress hormone response without changing how stressed you feel

Creswell JD, Pacilio LE, Lindsay EK, Brown KW, 2014 — Psychoneuroendocrinology

Participants completed 3 days of either brief mindfulness meditation training or relaxation training, then underwent the Trier Social Stress Test — a standardized public-speaking and mental-math task designed to reliably provoke a stress response — while cortisol and self-reported stress were measured.

Finding: the mindfulness group showed a blunted cortisol response to the stress test compared to the relaxation-training group — but there was no difference in self-reported subjective stress between groups. Physiological stress reactivity and how stressed someone feels don't always move together, and meditation's effect may show up more reliably in one than the other.

4. Effects hold up, at a smaller size, in people without a diagnosis

Khoury B, Sharma M, Rush SE, Fournier C, 2015 — Journal of Psychosomatic Research

A meta-analysis specifically restricted to studies of MBSR in healthy, non-clinical adult populations — people without a diagnosed psychiatric or medical condition.

Finding: MBSR produced small-to-moderate improvements in stress, anxiety, and depression symptoms even among healthy participants, though the effect sizes were smaller than those typically found in clinical populations dealing with a diagnosed condition.

What This Means in Practice

• The strongest, most consistent evidence is for anxiety and depression, not stress in the abstract — if reducing anxious or depressive symptoms is the goal, the case for meditation is more solid than if the goal is vaguely "feeling less stressed."
• Meditation isn't shown to beat other options — it's one evidence-based tool among several (exercise, therapy, medication), not a superior default. Pick what you'll actually sustain.
• Physiological and subjective stress can move independently. A meditation practice might be lowering your cortisol reactivity even on days it doesn't feel like it's "working" subjectively — and vice versa.
• Structured practice, not occasional use, drives the evidence. Most trials used 8-week programs with regular guided sessions — a five-minute app session before bed is a reasonable habit, but it's not the dose most of the research above is describing.

FAQ

Does mindfulness meditation actually reduce stress?

The evidence is more mixed than popular coverage suggests. A large JAMA Internal Medicine meta-analysis (Goyal et al., 2014) found moderate evidence that mindfulness meditation programs improve anxiety and depression, but only low evidence of an effect on stress and general distress specifically. A broader meta-analysis (Khoury et al., 2013) found mindfulness-based therapy moderately effective across anxiety, depression, and stress outcomes combined.

Is meditation better than exercise or therapy for stress?

No good evidence supports that. Goyal et al. (2014) found no evidence that meditation programs were more effective than active treatments — medication, exercise, or other behavioral therapies — for any outcome measured. Meditation appears to be one effective option among several, not a superior one.

Can meditation reduce stress hormones like cortisol?

Sometimes, but not always in the way people expect. Creswell et al. (2014) found that three days of mindfulness training blunted the cortisol spike from a standardized social stress test compared to a relaxation-training control group — but did not reduce participants' self-reported feelings of stress. Physiological and subjective stress responses don't always move together.

How much meditation do you need before it helps?

Meaningful effects have shown up with programs as short as a few days of brief training (Creswell et al., 2014) up through the standard 8-week Mindfulness-Based Stress Reduction (MBSR) course used in most clinical trials. Most of the strongest evidence comes from structured 8-week programs with daily practice, not occasional use.

Does meditation work for people without a diagnosed mental health condition?

A meta-analysis focused on healthy, non-clinical populations (Khoury et al., 2015) found MBSR produced small-to-moderate improvements in stress, anxiety, and depression symptoms even in people without a diagnosis — smaller effects than in clinical populations, but still measurable.

The Bottom Line

Meditation has a real evidence base — it's just a more modest and more specific one than the wellness-industry version. The clearest support is for anxiety and depression; the support for stress specifically is real but weaker and more inconsistent across studies. It's not shown to outperform exercise or therapy, and it can improve physiological stress markers even on days it doesn't feel like it's helping. None of that makes meditation not worth doing — it makes it worth doing with realistic expectations, and as one tool alongside others rather than a stress-proofing shortcut.

Health Partner's Mind module includes daily mood check-ins alongside journaling, so you can track how you're actually feeling over weeks — not just whether a meditation session felt calm in the moment. Download Health Partner to start tracking.

References

1. Goyal M, Singh S, Sibinga EM, et al. Meditation programs for psychological stress and well-being: a systematic review and meta-analysis. JAMA Intern Med. 2014;174(3):357–368. doi:10.1001/jamainternmed.2013.13018
2. Khoury B, Lecomte T, Fortin G, et al. Mindfulness-based therapy: a comprehensive meta-analysis. Clin Psychol Rev. 2013;33(6):763–771. doi:10.1016/j.cpr.2013.05.005
3. Creswell JD, Pacilio LE, Lindsay EK, Brown KW. Brief mindfulness meditation training alters psychological and neuroendocrine responses to social evaluative stress. Psychoneuroendocrinology. 2014;44:1–12. doi:10.1016/j.psyneuen.2014.02.007
4. Khoury B, Sharma M, Rush SE, Fournier C. Mindfulness-based stress reduction for healthy individuals: a meta-analysis. J Psychosom Res. 2015;78(6):519–528.

Also on the blog