Psychosis is a symptom rather than a diagnosis, and treating it as a single frightening condition is the source of most of what people get wrong about it. It appears in schizophrenia, in bipolar disorder, in severe depression, after enough sleep loss, with some drugs and in physical illness, and what is producing it changes what helps. [nice-2014-psychosis] This guide covers what it actually involves, what the evidence says about causes and outcomes, and why the delay before treatment is the thing most worth acting on.
What it involves
Three features, usually in combination.
Hallucinations are perceptions with no external source. Hearing voices is much the commonest form; visual, tactile and olfactory hallucinations occur and point toward particular causes.
Delusions are beliefs held with conviction despite strong evidence against them. Persecutory delusions, being watched, followed or targeted, are the most frequent. The belief is not a failure of intelligence and cannot be argued away, which is the single most useful thing for a family member to know.
Disorganised thinking shows up in speech that becomes progressively harder to follow, with connections between ideas that are clear to the speaker and not to anyone else.
Beneath these sit what clinicians call negative symptoms: reduced motivation, flattened emotional expression, and social withdrawal. These attract far less attention than voices and delusions, respond less well to medication, and often affect daily life more.
More common than the reputation suggests
Around three people in a hundred experience a psychotic episode at some point. Psychotic-like experiences without illness are considerably more common: McGrath and colleagues, across eighteen countries, found that a meaningful proportion of the general population reported at least one such experience, most without ever developing a disorder. [mcgrath-2015-psychotic]
That finding matters because it undercuts the idea of a clean line between people who have these experiences and people who do not. Hearing a voice once is not the same as having a psychotic illness, and knowing that reduces the terror of a single episode.
Causes, and the stress-vulnerability picture
No single cause explains psychosis. The model that fits best is an interaction: a degree of underlying vulnerability, partly genetic, meeting enough stress or exposure.
Childhood adversity is one of the better-established environmental contributions. Varese and colleagues pooled the evidence and found childhood adversities substantially increased the odds of psychosis. [varese-2012-childhood] Cannabis use, particularly high-potency and particularly in adolescence, is another. Migration and urban upbringing both show consistent associations whose mechanisms are debated.
None of these is deterministic, and most people exposed to any of them never develop psychosis. Our guide to bipolar disorder covers one of the conditions psychosis occurs within.
The direction of the association reported by Penttilä and colleagues (2014), where longer untreated psychosis predicted poorer long-term outcome. Values illustrate the gradient rather than reporting measured scores.
The one thing most worth acting on
Duration of untreated psychosis is the most modifiable predictor of outcome available. Penttilä and colleagues pooled the evidence and found longer untreated psychosis predicted worse long-term outcomes across multiple measures. [penttila-2014-dup]
That single finding is the entire rationale for early intervention services, and it is the reason this article does not counsel watchful waiting. Where psychosis is suspected, the useful action is assessment now rather than observation for a few more months.
Treatment is generally a combination. Antipsychotic medication is the mainstay for acute symptoms. Cognitive behavioural therapy for psychosis is recommended alongside it, and family intervention has some of the strongest evidence in the field for reducing relapse. [nice-2014-psychosis] Supported employment and education matter more to most people’s actual lives than any of the above.
Signs worth getting assessed
For yourself or someone you know. This is not a test and cannot diagnose anything. Early assessment is the point, since the delay before treatment is what most affects outcome.
0 of 6 ticked
This is the pattern where the delay before treatment is the thing that most affects how well it goes. Ask specifically about early intervention services if they exist where you are.
Enough here to be worth describing to a doctor rather than waiting. Assessment also rules things out, including physical causes and medication effects.
Little of this is present. If something else is difficult, it is worth raising on its own terms.
No psychosis screener is published on this site, and self-assessment is a poor tool here since reduced insight is part of the picture. What another person has noticed carries real weight.
What to do if it is someone else
Do not argue with the content of a delusion, and do not pretend to agree with it. Arguing entrenches the belief and recruits you into it; agreeing removes you as a link to the ordinary world.
Respond to the feeling instead. Being convinced you are being followed is terrifying, and saying that it sounds terrifying is both true and non-committal about whether it is happening. That keeps the conversation open, which is the thing that matters, because you are trying to get to an assessment rather than to win a point.
Practical help is worth more than persuasion: offering to go with them, making the appointment, being there. If anyone is at immediate risk, treat that as an emergency.
When to seek help
Seek assessment promptly rather than waiting to see whether it settles. That is the single strongest evidence-based recommendation in this whole area. Ask specifically about early intervention services, which exist in many systems and are designed for exactly this.
A first episode should be medically investigated as well as psychiatrically assessed, because physical causes, drug effects and medication reactions all produce psychosis and all are treatable.
If you or someone else is at immediate risk, contact your local emergency services or a crisis helpline.
How MyFreud can help
Sleep collapse and withdrawal often precede an episode by weeks, and both are much easier to see in a record than from inside. Daily tracking gives an assessment something concrete about when things started, which is exactly the question that decides how urgently a service responds.