If you're in crisis, this page is the wrong place. Samaritans, Ireland and the UK: 116 123, free, 24 hours.
"I'm a bit OCD about my desk" is one of the more casually misused phrases in English. The condition it names is not a preference for order. It is one of the more distressing things a person can live with, people wait years to get help for it, and the treatment that works is specific enough that asking for the wrong thing wastes that time twice.
What it actually is
OCD has two parts that feed each other.
Obsessions are intrusive, unwanted thoughts, images or urges that cause significant distress. They are not enjoyed and not chosen. Crucially, they are usually repugnant to the person having them — which is exactly why they cause such distress, and why people so often keep them secret for years.
Compulsions are repetitive behaviours or mental acts performed to reduce that distress or to prevent some feared outcome. Checking, washing, counting, seeking reassurance, mentally reviewing, silently repeating phrases.
For a diagnosis, this has to be time-consuming — conventionally more than an hour a day — or causing significant distress or interference with work, study, relationships or daily life. Occasional intrusive thoughts are close to universal; almost everyone has had the thought of swerving the car or shouting in a quiet room. Having them is not the condition. Being tormented by them, and building rituals around them, is.
The cycle is the whole condition
The compulsion works — briefly. The relief is real and short-lived, and performing the ritual teaches the brain two things at once: that the threat was genuine, and that the ritual is what prevented it. Neither gets tested, so neither gets disproved, and the cycle reinforces itself. Over time the rituals tend to grow — longer, more elaborate, more rigidly specified — because the relief they buy gets shorter.
Understanding that mechanism explains why the treatment looks the way it does, and why the intuitive responses — reasoning with the thought, seeking certainty, avoiding the trigger — all make it worse.
And most people know it's irrational
This is widely misunderstood. Insight is usually intact: people with OCD generally know the fear is disproportionate and the ritual illogical. That knowledge doesn't help, and it is part of the torment — doing something for two hours that you know makes no sense is its own distinct misery, and it is a common reason people don't disclose it.
What it looks like, beyond the stereotype
Contamination fears and handwashing are the version everyone knows. The others are common and far less recognised:
- Checking — locks, appliances, that you haven't harmed someone, that an email didn't say something terrible
- Intrusive harm thoughts — unwanted images of hurting people you love, including children. These are among the most distressing and most secret, and having them is not a sign you want to act on them. The distress is the evidence of that, not evidence against it
- Sexual and religious themes — unwanted taboo sexual thoughts, or relentless doubt about faith (sometimes called scrupulosity). Both are heavily under-reported for obvious reasons
- Relationship OCD — unremitting doubt about whether you love your partner or they love you, driven by a search for certainty that no answer satisfies
- Symmetry, ordering and "just right" feelings — the closest to the stereotype, and still nothing like a preference for tidiness: the driver is an intolerable sense of incompleteness, not a liking for neatness
- "Pure O" — where the compulsions are entirely mental and invisible from outside. Someone can spend hours a day on internal review and appear, to everyone around them, to be sitting quietly. This is a major reason it goes undetected for years
Two things classified nearby but distinct: hoarding disorder is now a separate diagnosis rather than a form of OCD, and body dysmorphic disorder and the body-focused repetitive behaviours such as hair-pulling and skin-picking sit in the same family of conditions with related but not identical treatments. Tics and Tourette's also co-occur more often than chance.
The reassurance trap
Asking for reassurance is itself a compulsion. Giving it — which is what everyone who loves the person instinctively does — relieves the distress briefly and strengthens the cycle.
This has a name in the clinical literature: family accommodation. Answering the question again, checking the door on their behalf, taking over the task they can't face, adjusting household routines around the rituals. All of it is kind, all of it is understandable, and all of it is fuel.
It is genuinely hard for families, because withholding reassurance feels cruel and is often the more useful thing. The workable version is not cold refusal but a shared agreement made in advance, ideally with a therapist involved — which is one of the reasons this needs proper treatment rather than good intentions.
The treatment is specific — and this is the practical point
The evidence-based psychological treatment is exposure and response prevention (ERP) — a form of CBT that involves deliberately encountering the trigger while not performing the compulsion, so the brain gets to learn that the anxiety subsides on its own and the feared outcome doesn't arrive.
It is done gradually, by agreement, working up a hierarchy rather than being thrown in at the deep end. It is demanding, and it works.
It is not the same as general talking therapy, and general counselling can be actively unhelpful here — particularly if it involves exploring and analysing the content of the intrusive thoughts. Searching for the meaning of an unwanted thought is a compulsion. A therapist who does that with you, sincerely and at length, is running the ritual with you.
So the single most useful sentence in this article: when you seek help, ask specifically whether the therapist does ERP for OCD. That one question filters a great deal.
Medication
SSRIs also have good evidence, alone or alongside ERP, with two differences from their use in depression that matter: the doses used are often higher, and the trial needs to be longer — commonly ten to twelve weeks before judging whether it is working. People stop too early because they are applying a depression timeline. Clomipramine is an older option with good evidence, and in treatment-resistant cases an antipsychotic is sometimes added. All of that is a psychiatrist conversation. If you already take supplements, the interactions worth raising with mental-health medication are set out separately.
Why the delay matters
OCD is not rare — it affects on the order of one to two percent of people — and the gap between symptoms starting and getting effective treatment is frequently measured in years, sometimes more than a decade.
The reasons are structural rather than mysterious. The content of the obsessions is often shameful enough to prevent disclosure. The stereotype tells people OCD means being tidy, so those with harm or sexual themes don't recognise themselves in it. And when they do present, they may be treated for generalised anxiety or depression — both of which commonly co-occur — without the OCD being identified.
Worth adding: OCD is not classified as an anxiety disorder in current diagnostic systems, which is not a technicality. It sits in its own group, and it needs its own treatment rather than generic anxiety management. The anxiety disorders, and how they differ from each other, are here, and the treatments ranked by evidence are here — with ERP listed for exactly this reason.
If you're reading this having recognised yourself in something after years of not having a name for it, that experience is more common than it feels. We've written separately about late diagnosis in adults, which covers a lot of the same ground emotionally.
Where we sit
Nowhere, honestly. There is no supplement with evidence for OCD, ours included, and we're not going to construct an angle where one belongs.
We've written this because it's badly understood, because people wait years to get help, and because "ask whether they do ERP" is genuinely useful information that costs nothing to pass on and can save someone a year of the wrong therapy.
This article is not medical advice and cannot diagnose OCD. If you recognise yourself here, a GP is the place to start — and it is worth saying the word OCD out loud at that appointment, even if the content of the thoughts feels impossible to describe.