Meditation is presented in popular coverage as beneficial or, at worst, ineffective. The possibility of harm is rarely mentioned.
Research examining this has found that difficult experiences occur in a meaningful minority of practitioners, and that they're underreported in trials.
What the research has found
Studies specifically investigating adverse experiences have reported that a substantial minority of meditators experience unpleasant effects at some point, with a smaller proportion experiencing effects severe enough to cause functional impairment.
Reported figures vary considerably depending on how the question is asked, the population studied and the intensity of practice.
Notably, reviews of meditation trials have found that a large proportion do not monitor or report adverse events at all — which is unusual for clinical research and means the trial literature systematically underrepresents this.
What the difficulties look like
Qualitative research with practitioners has described a range of experiences.
Anxiety and fear. Including panic during or after practice, sometimes in people with no prior anxiety history.
Emotional material surfacing. Reduced distraction can bring difficult memories or feelings into awareness. For some this is therapeutic; for others it's overwhelming, particularly with trauma history.
Depersonalisation and derealisation. A sense of detachment from one's self or surroundings. Some contemplative traditions describe related states as part of the path; experienced without context or support they can be frightening and can persist.
Changes in perception. Altered sense of time, body sensation, or perceptual clarity.
Increased emotional sensitivity. Some practitioners report becoming more reactive rather than less, at least for a period.
Social withdrawal and loss of motivation. Reported particularly after intensive retreat practice.
Who appears more at risk
The evidence here is limited and some patterns emerge.
Intensity matters. Extended retreats with many hours of daily practice appear more likely to produce difficult experiences than brief daily practice.
History of trauma is frequently identified, which is why trauma-sensitive approaches to teaching have developed. Standard instructions to attend to bodily sensation can be destabilising for somebody whose body carries traumatic memory.
Existing psychiatric conditions, particularly those involving psychosis or dissociation, warrant caution and specialist advice.
Practising without a teacher is frequently mentioned by those reporting difficulties, since there's no one to normalise, contextualise or adjust the practice.
The tradition's own account
Worth noting that contemplative traditions themselves describe difficult stages.
Various traditions have terminology for challenging phases of practice — periods characterised by fear, disgust, restlessness or a sense of dissolution. These are described as expected stages rather than as failures, with teachers available to guide practitioners through them.
That framing is largely absent from secular presentation, which tends to describe meditation exclusively in terms of calm and clarity.
Which creates a specific problem: somebody encountering a difficult phase, with no framework for it and no teacher, reasonably concludes something is wrong with them.
Why it isn't discussed
Several reasons.
Commercial incentive. An industry selling meditation has no reason to highlight risk.
Research culture. Trials in this field have historically not monitored adverse events, partly because the intervention was assumed benign.
Attribution. Practitioners experiencing difficulty frequently attribute it to themselves rather than to the practice, and don't report it.
And a genuine reluctance to discourage something that helps most people. That's understandable and it isn't a good reason to withhold information.
Practical guidance
None of this argues against practising. It argues for practising with information.
Start modestly. Short daily sessions rather than intensive retreat, particularly initially.
Have a teacher or a group. Someone to ask when something unexpected happens. This is the most consistently useful protection.
Know that difficulty can be part of it. Restlessness, emotional material and boredom are normal. That doesn't mean severe distress should be endured.
Stop if it's causing significant distress. Persisting through severe difficulty is not required and the instruction to sit with discomfort has limits.
Be cautious with trauma history. Trauma-informed teachers exist, and practices with an external anchor rather than internal body focus may be more appropriate. Worth discussing with a therapist if you have one.
Be cautious with certain conditions. If you have a psychotic disorder, dissociative disorder, or bipolar disorder, discuss meditation practice with your clinician before beginning intensive work.
The reasonable framing
Meditation is an intervention with effects. Things with effects can have adverse effects — that's not a criticism, it's what having effects means.
Presenting it as risk-free is both inaccurate and unhelpful to the minority who encounter difficulty, who are left without an explanation for what's happening to them.
The honest version — beneficial for most, occasionally difficult, worth doing with support — is more useful and no less encouraging.
General information only. If meditation practice is causing significant distress, stop and seek advice from a qualified healthcare professional.
What good teaching looks like
Some markers worth looking for, given that having a teacher is the main protection available.
Someone who asks about mental health history before you begin. Someone who describes difficult experiences as possible rather than presenting practice as uniformly pleasant. Someone who offers alternatives when an instruction is uncomfortable rather than encouraging you to persist through it. And someone who will say when something is outside their competence and suggest clinical support.
Trauma-informed teacher training exists and is worth asking about directly. A teacher who has never considered the question is not necessarily unsafe, and one who has thought about it is a better bet.