Tip · Stress & mind
Meditation 10 min daily
Meta-analyses show small to moderate effects on stress, anxiety and focus. The cortisol findings are inconsistent.
Sitting and breathing for ten minutes in the morning: meditation is the most frequently given advice against stress and has one of the largest data bases in the field. The effects are real, small and limited to certain endpoints. This page shows what remains against good comparison groups — and which popular claim about the brain no longer holds.
In short
Meditation has a measurable effect on anxiety, depressiveness and pain. In the reference meta-analysis, which admitted only studies with an active control group, the effect sizes after 8 weeks were 0.38 for anxiety, 0.30 for depressiveness and 0.33 for pain; after 3–6 months they were still 0.22 and 0.23. Something also moves physiologically: across 39 MBSR studies with 2,059 participants, cortisol at g = −0.38, and still −0.29 against active controls. The catches: in healthy people, the advantage over an equivalent alternative measure shrinks to zero, and for focus there is no positive evidence.
What is behind it
Mindfulness meditation trains a single skill: directing attention deliberately and responding to inner states without immediately judging them. Physiologically, this shows up in the stress axis. In the largest meta-analysis on MBSR and cortisol, which controlled for the time at which samples were taken, cortisol levels fell by g = −0.38 compared with control conditions. In randomized trials with active controls, blood pressure, heart rate, C-reactive protein, triglycerides and TNF-alpha fall in parallel.
How large the effect is depends on the comparison group
Across 142 samples with 12,005 participants, a clean staircase emerges: against no treatment d = 0.55, against minimal treatment d = 0.37, against non-specific active control d = 0.35, against specific active control d = 0.23, against established treatments d = −0.004. The lower end is what matters: at follow-up, the advantage over non-specific active controls held at d = 0.52 and over specific ones at d = 0.29.
In healthy people, the calculation is stricter. In 136 randomized trials with 11,605 participants, all four endpoints improved compared with no intervention: anxiety −0.56, depressiveness −0.53, psychological distress −0.45, well-being 0.33. Against non-specific active control, only depressiveness remained significant; against specific active control, nothing did.
The claim about the brain
The second half of the popular narrative is that the brain visibly remodels itself after eight weeks. The original study compared 16 MBSR participants with 17 people on a waiting list and found more gray matter in the left hippocampus, the posterior cingulate cortex, the temporoparietal junction and the cerebellum. The replication with 218 participants and an active control found none of this. Eight weeks remains the standard duration of the programs from which the psychological effect sizes come — that is the rationale for the number, not the brain.
What is well supported
The most stable finding is that of the paper that admitted only studies with an active control group, that is, comparison conditions that also capture the placebo component. Of 18,753 citations screened, 47 randomized trials with 3,515 participants remained. For anxiety, depressiveness and pain, it rates the evidence as moderate: anxiety 0.38 after 8 weeks and 0.22 after 3–6 months, depressiveness 0.30 and 0.23, pain 0.33. Small to moderate effects that last for months.
What the studies show
The reference paper with exclusively active control groups
Meta-analysis in clinical populations; of 18,753 citations screened, 47 randomized trials with 3,515 participants remained. Moderate evidence for anxiety (0.38 after 8 weeks, 0.22 after 3–6 months), depressiveness (0.30 and 0.23) and pain (0.33). No evidence that meditation performs better than any active treatment, including medication and exercise.
The better the comparison group, the smaller the effect
Meta-analysis of 142 samples with 12,005 participants, control groups coded in five levels: d = 0.55, 0.37, 0.35, 0.23 and −0.004 from waiting list to established treatment. A re-analysis of 60 comparisons with 5,627 participants showed that the superiority disappeared when researcher allegiance was balanced or absent.
In healthy people, nothing remains against good controls
136 randomized trials with 11,605 participants from 29 countries, non-clinical settings, preregistered, Cochrane risk-of-bias tool 2.0. Compared with no intervention, anxiety (−0.56), depressiveness (−0.53), psychological distress (−0.45) and well-being (0.33) improved. Against specific active control, nothing remained significant.
Cortisol falls, but not consistently
The first meta-analysis to bring together MBSR and cortisol while controlling for sampling time — crucial for cortisol because of its daily fluctuation. 39 studies, 2,059 participants, between groups g = −0.38, against active controls −0.29. Randomized trials found g = −0.37, uncontrolled ones, by contrast, −0.06.
No structural brain changes after eight weeks
Two combined three-arm randomized trials with 218 meditation-naive participants, structural MRI before and after: waiting list (n = 70), MBSR (n = 75), effort-matched active control (n = 73). No differences were found in volume, density or cortical thickness, neither in the whole-brain comparison nor in the predefined target regions.
Where the data stop
Against a well-designed active comparison condition, the effect in healthy people shrinks to zero. For attention and focus, one of the reasons meditation is traded as a biohack, there is no positive evidence: the reference meta-analysis places attention in the category of low evidence of no effect or insufficient evidence, together with positive mood, sleep and weight. And part of the advantage over other treatments is due to researcher allegiance: when the authors’ attachment to the method was balanced or absent, nothing remained of g = 0.13.
The cortisol findings are also less consistent than the pooled figure suggests, and the data on apps are thin, with 8 studies and 574 participants.
How to do it
The standard program from which almost all effect sizes come lasts 8 weeks. The time actually practiced in it is well above ten minutes: across 43 studies with 1,427 participants, reported home practice was 64 % of the prescribed amount (95 % CI 60–69 %), which the authors put at about 30 minutes per day on six days per week. More practice is associated with better outcomes, but weakly: across 28 studies with 898 participants, r = 0.26 (0.19–0.34).
Ten minutes is therefore below what was practiced in the cited studies. For shorter sessions there are the app studies: negative emotions SMD = −0.232, depressive symptoms −0.367, anxiety symptoms −0.490. Ten minutes is a plausible lower limit, not a tested dose.
Safety
Meditation is low-risk, but not free of side effects, and this is practically never mentioned in popular accounts. The systematic review on this included 83 papers with 6,703 participants. 55 of the 83 studies, that is 65 %, reported at least one adverse event; the overall prevalence was 8.3 %, in experimental studies 3.7 %, in observational studies 33.2 %. The most common were anxiety (33 %), depressiveness (27 %) and cognitive abnormalities (25 %), and this can also affect people without a pre-existing mental health condition. With a history of trauma, a history of psychosis or acute depression, getting started should therefore be discussed, not begun alone from an app. The serious events come predominantly from retreats and sessions lasting several hours. If anxiety, depersonalization or intrusive memories occur, that is a reason to stop and seek advice.
BK-Score Supported, with caveats
| Human evidence | 6 | |
|---|---|---|
| Mechanism | 5 | |
| Safety data | 9 | |
| Hype gap | 5 | |
| Track record of use | 10 |
There are many randomized trials; meta-analyses find small to moderate effects on anxiety, stress and depressiveness – mostly against waiting lists, less often against active comparison groups. For cortisol, the latest meta-analysis shows a small but measurable decrease, with continued large variation between studies.
The score rates the state of knowledge, not the effect. “Safety data 9” means well studied – not harmless.
Subjective assessment by Biohacking Kompakt based on published scoring rules – not a scientific rating and not a medical recommendation. Rules and all ratings (German)
Frequently asked questions about meditation 10 min daily
Are 10 minutes a day really enough?
The effect sizes quoted everywhere come from eight-week programs in which participants practiced on average about 30 minutes daily on six days per week. These figures do not exist for ten minutes a day. What does exist are studies on app-guided mindfulness with shorter sessions, which found smaller but measurable effects on anxiety and depressive symptoms. Ten minutes is a plausible lower limit, not a tested dose.
Does meditation really change my brain after eight weeks?
That was the claim of a widely cited study with 16 participants against a waiting list. The replication with 218 participants and a genuine active comparison group found nothing, neither in volume nor in gray matter density nor in cortical thickness. The claim is thus not confirmed but refuted. Eight weeks nevertheless remains the sensible duration, because the psychological effect sizes come from exactly these programs.
Does meditation lower my cortisol?
Yes, on average and measurably, even compared with active control groups. In the meta-analysis of 39 MBSR studies with 2,059 participants, the difference between groups was g = −0.38, and against active controls g = −0.29. That is a small effect, and the variation between studies remains large and unexplained. Anyone who wants to measure this on themselves needs standardized sampling times; otherwise they are measuring the daily rhythm.
Is meditation better than exercise or therapy?
According to the available data, no. The reference meta-analysis explicitly found no evidence that meditation programs perform better than any active treatment, including medication, exercise and other behavioral therapies. Across 142 samples, the difference compared with established evidence-based treatments was d = −0.004, that is, exactly zero. That does not speak against meditation; it speaks against the ranking.
Does meditation make me more focused?
This is where the evidence is weakest. The large meta-analysis with active control groups places attention in the category where there is either low evidence of no effect or no sufficient evidence at all, together with sleep, positive mood and weight. What is mainly supported is anxiety, depressiveness and pain. Anyone meditating for focus is following a hope, not a finding.
Can meditation do harm?
Rarely, but it happens. In the systematic review of 83 studies with 6,703 participants, the overall prevalence of adverse events was 8.3 percent, most commonly anxiety, depressiveness and cognitive abnormalities. In experimental studies it was 3.7 percent, in observational studies 33.2 percent. The authors note that this can also affect people without a history of mental health conditions.
Related
- In depthAshwagandha
- Same category: Stress & mindGratitude journaling
- Same category: Stress & mindNo phone in the first hour
- Same category: Stress & mindForest bathing (shinrin-yoku)
- Same category: Stress & mindGrounding (earthing)
- Same category: Stress & mindNootropics starter stack
Sources
- Goyal et al., JAMA Internal Medicine 2014
- Goldberg et al., Clinical Psychology Review 2018
- Galante et al., PLoS Medicine 2021
- Johnson et al., Health Psychology 2026
- Kral et al., Science Advances 2022
- Parsons et al., Behaviour Research and Therapy 2017
- Farias et al., Acta Psychiatrica Scandinavica 2020
- Wu et al., Frontiers in Public Health 2021
Open in the database – with all tips, filters and BK-Score (German app)
Information only, not medical advice and not a usage recommendation. With pre-existing conditions and before major changes, consult a physician. Last updated: 2026-10-06.