Meditation has side effects. A systematic review of 83 studies covering 6,703 participants found an overall adverse-event prevalence of 8.3%, and reported that they can occur in people with no previous history of mental health problems. [farias-2020-adverse]
That is not an argument against practising. It is the thing you need in order to practise sensibly, and it is missing from almost everything written about mindfulness.
What the numbers actually say
The headline figure of 8.3% is less interesting than the spread underneath it.
The same review found 3.7% in experimental studies and 33.2% in observational ones. [farias-2020-adverse] That is a ninefold difference, and the likeliest explanation is not that one design is wrong. It is that the rate you find depends on whether you ask.
Trials that record adverse events only when a participant volunteers them find few. Studies that go looking, and that follow people who practise seriously for a long time, find many. Which means the honest summary is that nobody knows the true rate, and the low end of the range is probably an artefact of not looking.
The commonest difficulties reported were anxiety, then depression, then cognitive changes, with anxiety appearing in about a third of the studies that reported any adverse events at all. [farias-2020-adverse]
A separate mixed-methods study of experienced Western meditators documented the range in more detail: fear, anxiety, changes in the sense of self, and re-emerging traumatic material, across traditions and levels of experience. [lindahl-2017-varieties-worse]
All three prevalence figures are taken directly from the systematic review cited. They are real reported values rather than an illustration.
The right-hand bar is the one worth remembering, because observational studies are closer to how most people actually meditate: unsupervised, for a long time, without anybody checking.
Why it happens
The popular framing treats meditation as inert, something that either helps or does nothing. It is not inert, and four ordinary features of it explain most of the difficulty.
It removes distraction. That is the point, and it is helpful when the problem is that you never stop. It is unhelpful when distraction was doing necessary work, because what it was holding off now arrives with nothing in the way.
It points attention at the body. Scanning bodily sensation is a standard instruction and a poor fit for health anxiety and panic, where the existing problem is already too much attention on the heartbeat and the breath. Our guide to reassurance seeking covers why body-checking sustains that particular loop.
It has no brakes built in. A therapist paces exposure to difficult material. An app does not know what is in your history and will run a forty-minute silent session on request.
The instruction to sit with discomfort makes it worse when it is wrong. Meant for restlessness and boredom, it is routinely applied to panic, dissociation and distressing memories, which is precisely the material that should end the session rather than be endured.
The problem with the research
There is a reason the risks are not better known, and it is not a conspiracy.
A widely cited critical evaluation of the field argued that methodological problems and misinformation in mindfulness research risk leaving the public harmed, misled and disappointed, and set out how difficult the basic construct is to define and measure in the first place. [vandam-2018-hype]
Two consequences follow for a reader. Effect sizes in this literature are frequently smaller than the coverage implies, and adverse events have historically gone unrecorded, because a trial only reports harms if it collects them and many did not.
None of that means mindfulness does not work. It works well enough for specific problems to be recommended in clinical guidance. It means the gap between the evidence and the enthusiasm is unusually wide here, which is worth knowing before you conclude that a bad experience was your fault.
Is the practice doing something it should not?
Tick anything you have noticed during or after meditating. This is a reflection prompt rather than a test, and it produces no diagnosis.
0 of 7 ticked
That is the pattern the research describes, and none of it means you are meditating incorrectly. The most useful immediate change is format: shorter, eyes open, an external anchor such as sound or the feet on the floor. If it persists after that, it is worth raising with a professional rather than with an app.
One of these on its own can be ordinary restlessness, which is a normal part of learning to sit still. The distinguishing question is duration: discomfort that fades within minutes of finishing is different from a mood that stays low for the rest of the day.
Nothing here matched. If you are looking for the right practice for a specific difficulty rather than troubleshooting one, our mindfulness guide covers which format suits which problem.
How to practise without this happening
Short, and eyes open. The highest-risk format is the one most associated with seriousness: long, silent, eyes closed, attention inward. Starting at the other end costs nothing.
Use an external anchor. Sound in the room, the feet on the floor, contact with the chair. All of them keep some attention outside your own body, which is the specific adjustment that helps when interoception is the problem.
Stop when it is the wrong kind of difficult. Boredom, restlessness and an itchy leg are what “sit with it” was written for. Rising panic, unreality and distressing memories are not, and continuing does not build tolerance for them.
Get a person if you have a trauma history. Not an app. This is the single clearest recommendation available, and it is the one most often skipped because guided sessions are free and a professional is not.
Do not conclude you are bad at it. The most common response to an adverse experience is to assume the practice was done wrong, which leads directly to doing more of it.
Our guide to mindfulness for intrusive thoughts covers a case where the standard instruction needs adjusting, and mindfulness for emotional regulation covers what it does well. The mindfulness guide covers which practice suits which problem.
When to seek help
See a GP or a mental health professional if meditation has brought up distressing memories or dissociative experiences, if your mood has been persistently lower since starting a practice, or if you have been enduring difficult sessions on the basis that pushing through is the point.
Do the same if the anxiety a practice was meant to help has increased instead, particularly if the sessions have become something you dread.
Go urgently if you have thoughts of harming yourself. In the UK, Samaritans is free on 116 123 at any hour.
How MyFreud can help
The reason a practice that is making things worse can continue for months is that nobody compares the days with sessions to the days without them. MyFreud gives you daily mood tracking that takes seconds, which is enough to see whether a practice is helping or whether you have been assuming it must be.
Download MyFreud and start today: App Store or Google Play.