Sunday, 7 September 2008

Clinical judgement and individual cases

Advances in Psychiatric Treatment sent me a manuscript by Robin Downie (pictured) and Jane Macnaughton to review a few weeks ago. Although very much in agreement with the overall aim of the paper, I didn’t really get on with the way it worked. I suspect that, precisely because I agreed with its conclusions, I wished that it had set about arguing for them in a slightly different way.

Perhaps in response to my slight qualms, the editor who had accepted the paper asked me, this week, to write a commentary on it. Thus I’m presented with the problem of critically engaging with the paper whilst still making it plain that I think that its authors are very much on the side of the angels. I'm not sure, therefore, that this draft (knocked up as the rain rained down yesterday) has the right tone, yet. (PS: a year later the commentary came out thus.)

Clinical judgement and individual cases

Like Robin Downie and Jane Macnaughton, I think that judgement lies at the heart of good clinical practice in psychiatry [Downie and Macnaughton 2009]. I fully agree with the central thrust of their paper. But in this commentary, I wish to sound a note of caution about their likely success in defending judgement against those who criticise or neglect it without some further augmentation of their strategy. My assumption is that their paper is intended to be programmatic. Thus I do not wish to criticise it as incomplete (they have written much more elsewhere, eg.). Rather, my concern is that the route to a defence of judgement that it suggests is not the best route. Of course, my own brisk criticism and positive outline is even more programmatic.

In their paper, Downie and Macnaughton suggest that two factors disguise the central role of judgement in good clinical practice. One is the misapplication of numerical codification to judgement based on qualitative research (‘qualitative judgement’ in what follows) and the other is the rise of a consumer model of healthcare. In this short note, I can consider only the former.

Downie and Macnaughton on qualitative judgement

Downie and Macnaughton argue that the important connection between numerically codified analysis, generalisation and objectivity in quantitative research is mistakenly carried over into the domain of qualitative research and hence into qualitative judgement. They blame reductionism: ‘the process of seeing human beings and their interactions in terms of a number of discrete features’. And they object:

But to try to understand patients in this way, in terms of a finite number of discrete features, is to abstract from the complexity and totality of a human interaction. Blood pressure can helpfully be abstracted in this way and measured, but not a human response in its complex totality. There is something not only patronising but clinically misleading in the suggestion that the complexity of human relationships can be reduced to a few factors and ‘measured’ with an ‘assessment tool’. [Downie and Macnaughton 2009: **]

The problem with this as a defence of the role of clinical judgement, however, is that without some argument as to why reductionism is false, it remains merely a dismissal of the reductionist dismissal of judgement. Furthermore, whilst, like Downie and Macnaughton, I believe reductionism is false, those who oppose clinical judgement are likely to be those who believe it to be true – who believe that there is no limit to the application of the method of breaking down complex interrelations into discrete features - and thus simply asserting its falsity is unlikely to achieve the end of defending judgement against such critics.

In their corresponding positive characterisation of qualitative judgement, Downie and Macnaughton make a number of claims about it. They say that:

[It] is more akin to the understanding gained from literature and art than that gained from a numerical science…
It requires the active participation of the reader to identify with the situation and relate the findings to his/her own situation…
The route to understanding is through our identification with the situation. Through that identification we reach general features of human emotions…
Through identification with the particular situation the researcher or clinician can recognise the general elements in human emotion...
[E]ven if there is no universality in human emotions and reactions there is a broad similarity, and that may be all that is needed as a basis for individualised judgement.
[Downie and Macnaughton 2009: **]

These comments suggest that the sort of judgement Downie and Macnaughton have in mind is akin to narrative understanding (this is, however, merely my gloss on their phrase ‘literature and art’; I will return to it at the end); that it turns on general features of human emotion; that it requires that the clinician achieves understanding by identification with a subject; and that it is a particular kind of individualised judgement.

Especially within mental health care, narrative understanding looks to be a genuinely useful addition to criteriological understanding and I agree with the broad thrust of this account [IDGA Workgroup 2003; Phillips 2005; Thornton 2008, forthcoming b]. But I have some specific qualms about the proposed defence of clinical judgement generally based upon it.

Firstly, it would be a mistake to base a defence of the general role of clinical judgement on the need to understand individuals’ mental states in the same meaning-laden terms as are found in literary or narrative forms. Secondly and relatedly, a restriction of judgement to an understanding of human emotion (however relevant generalities are to be construed) seems misplaced. Both of these leave open the response by a reductionist critic that judgement may have a role in the broader surroundings or context of clinical care – in mere bedside manner, perhaps - but not in the core application of medical science itself. In other words, Downie and Macnaughton do not go far enough in their defence.

Thirdly, the claim that even within the context of a narratively structured understanding of another subject’s emotional states, judgement depends on an identification by a clinician with a subject is contentious. Of course, Jaspers held that such identification was a central aspect of empathy which was itself at the heart of psychiatric understanding [Jaspers [1913] 1974]. But, again, as a defence of clinical judgement against a reductionist critic, it ignores the widely influential approach to interpersonal understanding that claims that it is mediated by implicit knowledge of a ‘theory of mind’: the ‘theory theory’ approach. This approach likens an understanding of another person’s mental states to inference to the best explanation and thus, if it were true, would undermine the contrast Downie and Macnaughton rely on to distinguish qualitative clinical judgement from scientific research.

Perhaps the most telling argument against theory theory turns on the normativity of mental content and the impossibility, in general, of codifying those norms. Refuting theory theory in such a way would not, however, justify Downie and Macnaughton’s position without some further argument as to why direct awareness of another’s mental state was also rejected in favour of the indirect route they outline via identification. (Why would one need to identify with how things are for another person to understand how they are for them? Why would one need to imagine, for example, being in pain oneself to grasp that another is in pain? Might one not simply see in what they say and do, in what they express, that they are in pain) Thus characterising qualitative judgement in these terms seems needlessly contentious as a defence of clinical judgement in general.

Fourthly, as I have argued elsewhere, it is a grave mistake to think that judgement of individual cases requires a form of ‘individualised judgement’ [Thornton 2008; Thornton forthcoming a]. Such judgement, at best, falls prey to Sellars’ criticism of the Myth of the Given [Sellars 1997]. Downie and Macnaughton may merely mean a potentially general judgement about a particular situation but, as a defence of clinical judgement, the phrase is best avoided.

Thus whilst I agree with Downie and Macnaughton’s aims, I suspect a successful defence of judgement in clinical practice needs to be both broader and deeper than the approach outlined in this paper.

Towards a defence of clinical judgement

Clinical judgement lies at the heart of good clinical practice: in the core application of medical science as well as in the broader context of understanding service users and patients. That, at least, is the claim that needs defence. Here is one way to start to defend it.

Consider the way criteriological diagnosis is codified in DSM and ICD manuals. Syndromes are described and characterised in terms of disjunctions and conjunctions of symptoms. The symptoms, in recent years, have tended to be described in ways influenced by operationalism and with as little aetiological theory as possible. (That they are neither strictly operationally defined nor strictly aetiologically theory free is not relevant here.) Thus one can think of such a manual as providing guidance for or a justification of a diagnosis offered by saying that a subject is suffering from a specific syndrome. Thus, presented with an individual, the diagnosis of a specific syndrome is justified because he or she has enough of the relevant symptoms.

The following further thought is tempting. Whilst the overall syndrome is quite general and is characterised in a way that abstracts it away from individuals, the specification of why it applies to someone is more specific in two respects. Firstly, because of the way both ICD and DSM base syndromes on a combination of conjunction and disjunction of symptoms, it is possible that a syndrome so defined may apply to two individuals with little, or even no, overlap of symptoms. The specification of symptoms is thus more tailored to individuals than the overall syndrome. Secondly, and independently of that, the heritage of operationalism suggests that individual symptoms are more closely tied, than syndromes, through a kind of measuring operation to individuals. Symptoms seem to tie more abstract syndromes to particular individuals.

There remains, however, a gap between the description or articulation of a symptom and an individual. The concepts of specific symptoms are, despite their specificity, general concepts that can be instantiated in an unlimited number of actual or potential cases. So how can one judge that a general concept applies to a specific individual case or individual person? One can attempt to bridge this gap. Textbooks of psychiatry can describe, rather than merely list, symptoms. But whatever descriptive account they give of symptoms, there will always be a gap between their general descriptions and concepts (which potentially apply to any number of individuals) and any particular individual. Bridging this gap calls for expertise. It calls for a skilled recognitional clinical judgement. In a nutshell, clinical judgement involves skilled coping with individual cases, both people and their situations, and this requires a kind of non-deductive expertise.

Immanuel Kant was aware of this gap. In his third major work, the Critique of Judgement, he draws an important distinction between what he calls ‘determinate’ and ‘reflective’ judgement. He describes these in this way:

If the universal (the rule, principle, law) is given, then judgment, which subsumes the particular under it, is determinate... But if only the particular is given and judgment has to find the universal for it, then this power is merely reflective. [Kant 1987: 18]

The model at work here is of judgement as having two elements: a general concept and a particular subject. Judgement subsumes a particular under a general concept. The contrast between determinate and reflective judgement is then between an essentially general judgement, when the concept is already given, and a particular or singular judgement, which starts only with a particular. The former, determinate judgement, appears to be relatively mechanical and thus unproblematic. The idea that if a general principle is already given then judgements which deploy it are relatively unproblematic can be illustrated through the related case of logical deduction where a general principle is already given. If, for example, one believes that
1: All men are mortal; and
2: Socrates is a man.
Then it is rational to infer that:
3: Socrates is mortal.

One reason this can seem unproblematic is the following thought. If one has accepted premises 1 and 2 then one has, ipso facto, already accepted premiss 3. To accept that all men are mortal is to accept that Tom, Dick, Harry and Socrates are mortal. So given 1 and 2, then 3 is no step at all [though see Carroll 1895 and Fulford, Thornton and Graham 2006: 98-105]. Furthermore, some central forms of deductive judgement, at least, can be codified using Frege’s logical notation. Given the codification, one can inspect the form of a deductive inference to determine whether true premises could ever lead to a false conclusion. (In fact, neither of these reasons for taking deduction, and thus determinate judgement, is quite so straight forward. Here, however, the perceived relative straight forward nature of determinate judgement is what matters.)

By contrast, for reflective judgement, there is a principled problem in how to get from the level of individuals to the level of generalities, or how to get from people and things to the general concepts that apply to them. That is not a matter of deduction because the choice of a general concept is precisely what is in question. To move from the particular to the general that applies to it is somehow to gain information not to deploy it. Reflective judgement thus cannot be a matter of mechanical derivation. Kant himself suggests that there is a connection between reflective judgement and aesthetic understanding. It may be this connection to which Downie and Macnaughton are referring when they talk of qualitative judgement as being connected to judgements of literature and art (to which I promised to return) However, there is reason to think that art cannot provide a substantial clue to further unpack the nature of the expertise involved in judgement [Thornton 2007].

But what is important about Kant’s account and the illustration of it in the case of psychiatric syndromes and symptoms is that it demonstrates how such judgement is always involved in the application of general knowledge to individuals. Whatever general claims can be gained from quantitative research – which lies at the heart of Evidence Based Medicine – their application to individuals necessarily depends on a kind of skilled expertise in judgement. This does not merely apply to understanding the psychological aspects of service users or patients (hugely important though that is). Even in judgements that are seen as paradigmatically empirical and scientific, skilled and uncodified expertise, or clinical judgement, lies at the heart of seeing that a general concept applies in an individual case or to an individual person.


Carroll, L. (1895) ‘What The Tortoise Said To Achilles’ Mind 4: 278-280
Downie, R. and Macnaughton, J (2009) **
Fulford, K.W.M, Thornton, T and Graham, G. (2006) The Oxford Textbook of Philosophy and Psychiatry Oxford: Oxford University Press
IDGA Workgroup, WPA (2003) ‘IGDA 8: Idiographic (personalised) diagnostic formulation’ British Journal of Psychiatry, 18 (suppl 45): 55-7
Jaspers, K. ([1913] 1974) ‘Causal and “Meaningful” Connections between Life History and Psychosis’, (trans. J.Hoenig) in Hirsch, S.R., and Shepherd, M. (eds.) Themes and Variations in European Psychiatry, Bristol: Wright: 80-93
Kant, I. (1987) Critique of judgment Indianapolis: Hackett
Phillips, J. (2005)
‘Idiographic Formulations, Symbols, Narratives, Context and Meaning’ Psychopathology 38: 180-184
Sellars, W. (1997) Empiricism and the Philosophy of Mind, Cambridge, Mass.: Harvard University Press
Thornton, T. (2007) ‘An aesthetic grounding for the role of concepts in experience in Kant, Wittgenstein and McDowell?’ Forum Philosophicum 12: 227-45
Thornton, T. (2008) ‘Should comprehensive diagnosis include idiographic understanding?’ ’ Medicine, Healthcare and Philosophy 11: 293-302
Thornton, T. (forthcoming a) ‘Does understanding individuals require idiographic judgement?’ European Archives of Psychiatry and Clinical Neuroscience
Thornton, T. (forthcoming b) ‘Idiographic versus narrative approaches to assessment’ Psychopathology