This isn’t an attack on talk therapy, which is genuinely effective for a great many things. It’s a statement about one specific condition and one specific mechanism. In OCD, examining the thought is the compulsion. So a therapy built on examining thoughts becomes, without anyone intending it, an hour of supervised compulsion once a week.
How it happens
You bring the obsession. The therapist, doing what good therapists do, takes it seriously and explores it. Where does this come from? What might it mean? What was happening in your childhood? Let us look at the evidence for and against.
You leave feeling better. That’s the problem. The relief is exactly the relief a compulsion provides, and your brain draws exactly the same conclusion: the thought was important, and analysing it’s what made the feeling stop. The loop closes, with a professional inside it.
Then next week you bring it again, because it came back — and it came back because it was reinforced.
The specific things that go wrong
Reassurance. A therapist saying “you would never do that” is doing what any kind person would. It functions identically to asking your partner for the ninth time. Four minutes of relief, and the doubt comes back with more force.
Meaning-making. Treating an intrusive thought as symbolic — as expressing a repressed wish, an unmet need, a buried memory — takes the one premise OCD depends on, that the thought means something, and gives it professional endorsement. For harm and taboo themes this isn’t merely unhelpful; it can be devastating.
Cognitive restructuring, misapplied. Weighing the evidence for and against the obsession is standard CBT and it’s a compulsion in this context. You can’t argue a doubt disorder into certainty, and every attempt teaches the machine that certainty is the goal.
Accommodation. A therapist who agrees to avoid a topic because it’s too distressing, or who lets sessions be organised around managing anxiety, is doing the same thing a family does when it accommodates.
Why so many clinicians get this wrong
Because most graduate programmes teach very little about OCD specifically, and because everything that goes wrong here looks, from inside the room, like therapy going well. The client is engaged. They feel better afterwards. They come back. There’s no obvious signal that anything is amiss — the signal takes a year to appear, and by then it looks like the OCD being stubborn rather than the treatment being wrong.
What to do
If you’ve been in therapy for OCD for six months or more and it isn’t measurably better, that’s the information. Ask your therapist directly whether they’re trained in ERP, and ask the questions that actually distinguish training from enthusiasm. A good clinician won’t be offended; a good clinician will often refer you.
This doesn’t mean your therapist is bad, or that the work you did was wasted. It means OCD is a specialist condition with a specific treatment, and general skill isn’t a substitute for it. Plenty of people keep a therapist they trust for everything else and add an ERP clinician for the OCD.