De-implementing treatments from the healthcare system: trauma-focused versus non-trauma-focused therapy of post-traumatic stress disorder as an example

When “evidence-based” starts sounding like “only one kind of therapy counts”

I have been reading a recent BMJ Mental Health piece about de-implementing therapies, and I think there is something bigger sitting underneath it. Not just the usual argument about what works. Something more political than that. Something about who gets to decide what counts, what stays, and what quietly gets pushed out.

The first thing worth saying is this. The report itself is not calling for therapy choices to be stripped away. Quite the opposite. It is pushing back against that kind of narrowing. It looks at PTSD, compares trauma-focused and non-trauma-focused approaches, and argues that the case for removing non-trauma-focused therapies has not been properly made. The authors say those therapies cannot simply be dismissed as an “implementation failure”, and they even raise the issue of researcher allegiance, which is another way of saying that bias can creep in when people are heavily invested in one model.

That matters because once a system starts talking about “de-implementation”, it can sound clean and sensible. Remove what does not work. Fine. Nobody sensible wants harmful or useless therapy kept in place. But the problem begins when that language gets used too quickly, or too selectively, or with a very narrow idea of what counts as proper evidence in the first place. Then it stops being about protecting patients and starts becoming a gatekeeping tool.

And this is where the wider picture matters. In the UK, the current NICE anxiety guideline is still heavily restricted. For generalised anxiety disorder, the recommended high-intensity psychological treatments are basically CBT or applied relaxation. For panic disorder, CBT is the named therapy. That is not a broad field of choice. That is a narrow funnel.

That narrowness is now being openly challenged. UKCP, BACP, Mind, Rethink Mental Illness and many others have backed calls for NICE to urgently review the anxiety guideline. Their argument is simple enough. The guideline is outdated, inconsistent with newer depression guidance, and too restrictive in the therapies it recommends. Their published position is that evidence has not been evaluated consistently, and that this has created a bias in favour of the treatments already sitting at the top of the list, namely CBT and applied relaxation.

That point should not be brushed aside. Because once a model becomes the model, the whole system starts bending around it. Training pathways. Commissioning. Audit tools. targets. Workforce design. What gets measured. What gets funded. What younger therapists are taught to respect. What patients are told is “recommended”. After a while it can stop feeling like one approach among many, and start feeling like the only respectable language left in the room.

Now, to be fair, there is a reason CBT keeps getting centre stage. It is manualised. It lends itself well to trials. It fits neatly into systems that like measurable outcomes, short treatment windows and standardised delivery. There is nothing automatically wrong with that. Good CBT can help a lot of people. The problem is what happens when “well studied” quietly turns into “superior”, and then “superior” quietly turns into “everything else becomes suspect”. That is not science. That is drift.

And once that drift sets in, patients can lose out in a very ordinary way. Not always through obvious harm. Sometimes through mismatch. Through being offered the wrong kind of help because it is the approved kind. Through not being given a real choice. Through being told, politely and professionally, that the menu is wider than it actually is. The 2022 NICE depression guideline moved more towards patient choice and recognised that all treatments in its first-line table can be used as first-line options. The anxiety guideline has not caught up.

So do I think there is bias in the system? Yes, I think there is good reason to say that. Do I think some of that bias favours CBT-shaped evidence, CBT-shaped services and CBT-shaped commissioning? Yes, I think that is hard to deny. Do I think we can prove that the people involved are consciously trying to wipe out anything that threatens their viewpoint, without caring who gets harmed? No, I do not think the evidence takes us that far, and I would be wary of pretending it does. What we can say is serious enough. A narrow evidence culture can still do damage, even when the people inside it think they are being rigorous.

This is why patient choice is not some soft extra. It is not a luxury item to be added later once the proper therapies have been selected. It is part of good therapy itself. People are different. Their histories are different. Their defences are different. The kind of relationship they can work in is different. A system that forgets that will keep mistaking standardisation for wisdom.

The deeper question here is not whether CBT should exist. Of course it should. The real question is whether one model, helped by the way institutions define evidence, has become too powerful in deciding what counts as legitimate therapy in the first place. That is the question under this report. And it is a question worth asking before more doors quietly close.

Source material

Recently, a proposal was made to de-implement ineffective or harmful psychotherapeutic treatments from the healthcare system. Given the potential wide-ranging implications of this proposal, a critical review is warranted.

Source: BMJ Mental Health current issue

Read the original article