Could CBT be a feasible treatment for Depersonalisation-Derealisation Disorder?

When poor research gets dressed up as progress

I read studies like this with a fair bit of distrust, especially when they come from the usual CBT-shaped world where CBT often gets pushed forward before the evidence is really strong enough to carry that weight. I am not saying CBT never helps. Sometimes it does. But there is a pattern in mental health research where CBT fits the system so well that it gets treated as the sensible answer before the hard questions have really been asked. It is manualised, measurable, easier to package, easier to commission. That can make it look stronger than it is.

This study on depersonalisation and derealisation disorder is a good example. It was a feasibility trial with 30 participants, not a proper trial built to tell us with confidence that CBT works. The authors say themselves that it was underpowered and that inferential analyses could not be done. In plain English, that means this study was not built to prove much about effectiveness in the first place. So any tone of quiet confidence around the results needs pulling back straight away.

Then you get to the delivery, and it starts looking even shakier. Participants in the CBT arm were offered therapy over six months, but only 54% completed what the study had already decided was the minimum dose of six sessions. Twenty-three per cent got only two to five sessions, and another 23% got no sessions at all. Assessment completion also dropped from 87% at one month to 63% at nine months. That is not a small flaw sitting quietly in the corner. That is a major weakness in the study itself. (Springer)

This looks like wasted money dressed up as progress. The whole thing seems to have been set up badly from the start. Limited training, uneven delivery, poor retention, and people in the treatment arm not even getting the minimum number of sessions. That is not solid implementation. It is a weak structure pretending to test a serious clinical question. If you build it badly, of course it struggles. Then the risk is that patients, and the public purse, both pay for research that was half set up to fail.

And the training point really does matter. Clinicians got a half-day workshop, a manual, and fortnightly supervision. Thirty-seven NHS clinicians attended training, but only seven actually took at least one participant for the CBT arm. The paper later admits that future studies should use a longer training period of at least one full day to allow more practice and improve consistency in clinician skills. So even the authors are telling us that what was being delivered here was not yet steady enough.

This is one of the deeper problems with CBT research more generally. CBT often gets spoken about as if it is a clean, standardised intervention, but real delivery depends on who is doing it, how well they understand the condition, how well they are trained, and whether the people in the study even stay in treatment long enough to receive it. If those bits are weak, then the structure is weak. You are no longer testing some solid, well-implemented treatment model. You are testing a patchy version of it and then still trying to draw meaningful conclusions.

The patient response was mixed as well, which gets less attention than it should. In the CBT group, 54% were satisfied with the therapist they worked with. Forty-six per cent rated the treatment outcome positively, and the same percentage said their symptoms were much or a little better. There was also 27% missing data on acceptability. That is not a ringing endorsement. It is a mixed picture with gaps in it.

Even the National Elf review, which was broadly open to the study, pointed to basic design problems. It said the paper does not fully describe how the DDD diagnosis was established, even though DDD can be mistaken for anxiety disorders and other conditions. It also pointed out that the treatment as usual group simply recorded what they were offered, but it is unclear whether they actually engaged with any of it. That makes the comparison arm vague, and once your comparison is vague, the active treatment can start looking better than it really is. (National Elf Service)

So when I read this kind of research, my concern is not just about one study. It is about a wider pattern. CBT research often gets more patience, more institutional goodwill, and more benefit of the doubt because it already speaks the language services like to hear. It comes with manuals, protocols, supervision structures, scoring systems and familiar outcomes. Other approaches do not always get that same generosity. They are often treated as suspect much earlier. CBT, on the other hand, can still be praised as promising even when the setup is plainly weak.

To be fair, this study does not prove that CBT is useless for depersonalisation and derealisation disorder. It may help some people. But that is a very different claim from saying this paper gives us strong evidence. It does not. What it gives us is a small feasibility study with poor retention, uneven implementation, minimal training, unclear diagnostic detail, and a vague comparison group. That is not solid enough to carry the kind of confidence CBT research often seems to borrow from the wider system around it.

The honest conclusion is not complicated. This study does not show us that CBT is the answer for DDD. It shows us how easily weak implementation can be mistaken for meaningful progress, especially when the therapy being studied already has the system behind it. And that is exactly why distrust is sometimes the healthier response. Not cynicism for the sake of it. Just a refusal to call something solid when it plainly is not.

Remeber this is just my opinion if you want to look into the facts do your own reserch don’t just trust what I say.