Writers of NICE guidelines have to take on three tasks. They have to marshal the evidence and integrate new findings into the existing guidance. They have to present their findings in different formats to suit different audiences. And they have to ensure that any guidance can reasonably be taken up, by taking into account such research as there is on barriers to implementation.
In the context of self-harm these tasks are more than usually demanding. Most obviously there just isn’t that much new evidence – a symptom of underfunding as much as it is of researcher indifference or incompetence. And such research as there is may be based upon populations far removed from the typical UK patient seen in ED or primary care. Much practice is therefore consensus or opinion based and new guidance cannot simply take the form – now modify treatment regime A by substitution or replacement to make Regime B.
Second, much self-harm is seen initially by practitioners who have no great depth of knowledge of self-harm theory or research; even when working in mental health services they are often generalists rather than part of a specialist self-harm team. Guidance therefore needs to take the form of clear and specific pointers to action, of a sort that does not rely on implicit knowledge.
And third, existing culture could not be described as conducive to dispassionate evidence-based practice. Influential and unhelpful ideas about personality disorder, non-suicidal self-harm or emotional instability exert a dismaying amount of influence.
It strikes me that these three demands are impossible to reconcile into a single document, as the latest NICE guidelines on the management of self-harm illustrate.
In an effort to stay based upon evidence the review recommends only two psychological treatments, neither with advice about how the recommendations will be read and implemented: CBT-informed therapy the definition of which does not take account of interpersonal factors and which will therefore be interpreted as “CBT-lite” by most people, and DBT for young people with “significant emotional dysregulation” without indicating how that will be determined in routine clinical practice or what happens to all the other young people who aren’t getting a label of emerging borderline personality.
Some of the advice to non-specialists is unhelpfully vague – I don’t think the average GP needs to be advised to refer to mental health services if they are concerned after assessment in primary care, and I don’t know that school staff will know without some specifics how to meet the recommendation that they identify “self-harm behaviours”. On the other hand some key questions – about the role of diagnosis in management decisions, how to get to the meaning of suicidal thinking with a patient, the important differences between one-off presentations in crisis and long-standing repeated self-harm, the origins and response to professional stigmatising – are hardly touched on.
Implementation challenges don’t get much of a look in. To give one example: it’s impossible to discuss how to improve self-harm services if we don’t acknowledge that aftercare is likely to require referral to CMHT who can’t/won’t see people without a “serious mental illness” or clinical psychology who won’t see suicidal people or who will if they’ll sit on a months-long waiting list.
I don’t think the problem here is primarily down to the expertise of the guideline group, which contained some experienced and sensible clinicians. A major problem resides in remit. This is a three-in-one document: a (short!) document giving a synthesis of what’s new in research; an assessment and treatment manual written with specific readers in mind (so actually a series of manuals) and a trouble-shooting guide to tackling current problems in service provision. NICE guidance just isn’t set up to meet such diverse aims and I wouldn’t therefore expect these latest guidelines to make much difference to the grim reality of service provision for people who present after self-harm.
There’s more to the debate than disagreement about research findings: prejudice against psychological diagnoses and treatments is also a factor
NICE’s decision to defer publication of its guidance on ME/CFS is welcome, allowing as it does time for reflection about what is going wrong that prevents consensus on the management of this group of conditions.
Actually, just a read through the draft (consultation) guidance provides some clues. It reads less like a dispassionate review of the evidence and more like a one-sided position statement from the ME Association. There is talk about listening to and believing patients, about energy envelopes, and criticism of trials that use “subjective” outcome measures – like self-reported fatigue and quality of life. What exactly these ideas mean isn’t always made clear. Disbelief isn’t likely to be about somebody being ill (symptomatic with functional impairment) but more about that illness being caused by a yet-to-be-identified physical pathophysiology. The energy envelope is an interesting term that I have not come across in this context (as opposed to engineering or acoustics) except when used by people who dislike the idea of activity pacing, but its physiological (“objective”) basis is as far as I know undefined. Treatments that “only” change symptoms are described as non-curative but then so are stroke rehabilitation, paracetamol and hip replacement.
What is being played out in these guidelines is a long-standing opposition to (sometimes expressed as hostility, anger or aggression) treatments such as CBT and GET. What is that opposition about? The debate is sometimes framed as being about the status of trials’ evidence but I don’t think that explains the emotional tone. I detect in the background a drive to reject these therapies because they suggest there is a large psychological or motivational element to the illness.
We live in a society in which the idea is not widely accepted that physical illness can have predominantly psychological causes or treatments, and of course the personal experience of physical symptoms is so pressing that psychological ideas can just not seem to work as explanations. Beyond that starting point there is something more going on and I think it is an active prejudice against the idea of psychological (and specifically psychosomatic) diagnosis. There are several reasons for such an aversion. The diagnosis may be misunderstood as meaning that an illness is imagined or even deliberately feigned. No doubt this is also part of the “not believed” complaint. On the other hand it is assumed that if the diagnosis is accurate, the person so diagnosed must be really odd. This idea has its roots, I think, in the history of hysteria and its conflation of the hysterical illness with the hysterical personality. And of course there is a threat inherent in the idea that somebody can be so lacking self-awareness that their body generates an illness in a mysterious way that is outside their control.
Is prejudice too strong a word? I don’t think so: it explains the animus in the reaction to psychological research and practice in the field, which clearly reflects more than disagreement about the interpretation of trial results. And it surfaces even in some of the milder critical responses to NICE’s decision, such as use of the word “relegated” to describe people offered psychological diagnosis or input, with its implications of inferiority and failure.
How can professionals with an interest respond? Of course there is a continuing need to engage with constructive critical appraisal of the research literature, including discussion of the use of so-called PROMs (patient-reported outcome measures) in applied health research. In addition there is a need to explore new ways to discuss attitudes to psychological causes and treatment for functional or psychosomatic illnesses, tackling the idea that these are somehow disreputable or demeaning approaches.
This last question is sensitive because of the anger and resistance it can provoke in some quarters. My own sense is that the result of this sensitivity is some professional foot-shuffling. Too much is made of findings that many functional states are not preceded by severely threatening or traumatic life experiences; psychotherapy is discussed at times as if it is synonymous with CBT which in this context functions largely as an adjunct to rehabilitation rather than as a way of exploring more widely into the psychosocial context of illness development and maintenance; a welcome interest in the physiological mediators of symptom production is unnecessarily and incorrectly accompanied by claims that such research heralds the end of dualism (implication, using a psychological vocabulary is so passé…).
Time will tell, but my own prediction is that for these reasons the impasse will not be resolved merely by a renewed critical appraisal of the literature. Maybe a “form of words” will be found, but a more useful conclusion would be an acceptance that psychological treatments are important because psychological disorders are important – real, potentially serious, treatable and not inferior to physical diseases.