Unwarranted Certainty: Psychiatry's Common Enemy

Across The Great Divide:

The raging debate on diagnosis in mental health is at fever pitch right now, and I find myself drawn into discussions on Twitter that are sometimes fun, sometimes interesting and sometimes feel like trying to suck Marmite through a straw. Although the two sides of this great divide are extremely heterogeneous, and though there are more bridges than is sometimes apparent, there is often a sense that some differences are too great to be reconciled. When intellectual rifts emerge it is all too easy to find yourself positioned (sometimes by your own lazy thinking; sometimes by that of other people) on one or other side of the divide. Theoretical positions become ossified and it gets harder to be heard across the gulf that has opened up. There are good and reasonable arguments coming from both directions.

On the one hand, classifying people's problems is a prerequisite for understanding their nature in a scientific way. You can't say with confidence that a person's mental distress is, say, a response to traumatic experiences unless you are able to draw on knowledge gleaned from other similar cases that confirms this to be a possibility. A person's problems may seem like they constitute such a response, but how can you know if you aren't able to confidently say that you have seen cases of this nature before and can rule out alternative conceptions?

On the other hand, reasoning about individual cases from general information is a probabilistic business. Such statistical information as we have about mental health problems is well equipped to inform us about the relative likelihood of particular causes and of therapies/medications being effective. It is less help in providing a straightforwardly biological understanding akin to that we have about physical health problems. This is why extra sources of case information are so useful in this field and why approaches like psychological formulation are a worthwhile technique for conceptualising the contextual and cognitive contributions to any given person's situation. In fact the DSM-5 itself recognises this fact in an opening section on clinical case formulation:

"The case formulation for any given patient must involve a careful clinical history and concise summary of the social, psychological, and biological factors that may have contributed to developing a given mental disorder." (DSM-5, 2013)

Fighting The Common Enemy:

This brings me to the real common enemy that should be uniting the psy-professions; unwarranted certainty. There is much mud slung between the anti and pro diagnosis positions about the extent to which either side is "scientific", but to the extent that they have any common ground in this regard, it is over the problematic tendency of human clinicians to be more certain about their conclusions than they have reason to be. Unwarranted uncertainty is a highly unscientific way of thinking--although some commentators, noting the tendency of overly-certain people to claim a scientific position, get this back to front. This is what seems to have happened in this David Brooks Op-Ed in the New York Times, in which he calls psychiatry a semi science and psychiatrists "heroes of uncertainty". Psychiatry may well be a semi-science, but that has more to do with the fact that it involves ethical and political reasoning which science can guide but not determine, than with the fact its practitioners are "heroes of uncertainty". It is also probably this confusion which accounts for the moment in this interview in which Tom Burns implies that being "ultra-scientific" in psychiatry is a bad thing. In fact being ultra-scientific would entail precisely the kind of skepticism and careful humane thought that Burns appears to be advocating.

The essence of the strongest argument against diagnosis is it can blinker clinicians and limit their capacity to see beyond the boundaries they define. Many of the evils identified by Peter Kinderman here and elsewhere are the consequences not of diagnosis per se, but of mental health staff naively taking a diagnosis to mean a particular sort of thing. It is not classification that leads to over-medicating of people with psychosis; it is the belief that a classification like "Schizophrenia" picks out a disease that cannot be treated any other way (incidentally, doesn't the quote from DSM-5 above suggest that the authors don't intend for clinicians to think in those terms?). The enterprise of "critical psychology" often lays claims to undermining this kind of unwarranted semantic certainty, but really it is the business of all psychology to bear uncertainty about the nature of its constructs and to explore them further. Mary Boyle's book on Schizophrenia undermines the "disease construct" not by drawing not on evidence from some sub-discipline called "critical psychology" but from...psychology.

Of course, the reification of formulation and psychological narrative is no less of a problem than the assumption that diagnosis is the same thing as disease. Information about how a person's problems have been affected by their life story and consequent beliefs is essential for the practice of psychotherapy, but if we drift into unwarranted certainty about that information's importance we aren't doing scientific clinical work anymore, we're just seeking confirmation of our biases.

I often find myself agreeing in debates that DSM is taken too seriously and in the wrong way; that its use can represent a confident assertion of knowledge that is unwarranted by the reality of what that a diagnosis picks out in nature. Noticing this is an important step for the well-being of service users, but it is not sufficient to conclude from it that diagnosis is conceptually impossible. Although it is frequently claimed that the diagnosis debate is not a turf war, there can be no doubt it is becoming increasingly war-like. A clarification of what is at stake helps us to direct our energies against the real common enemy.