Friday, 31 May 2013

Clinical Psychology- Anti or Ante-Science?

Watch out, you might get what you're after
Cool babies, strange but not a stranger
I'm an ordinary guy
Burning down the house
Hold tight, wait 'till the party's over
Hold tight, we're in for nasty weather
Burning Down the House (Talking Heads)

Have some clinical psychologists developed a bad case of ...anti-science?

Burning down the House, Pull up the Roots and I Get Wild

Although scientists thought this disabling disorder had been eradicated in the previous century, we are seeing increasing numbers of clinical psychologists presenting with a variety of anti-science symptoms. I start here with the symptom of formulation (recently also accompanied by paradigma shiftitis ) - other symptoms will follow in later posts

I understand the reservations that psychologists have routinely and historically expressed about psychiatric diagnoses. Indeed, questions can be always be raised about the reliability and validity of any diagnosis -psychiatric or otherwise. Often these questions about diagnosis are framed in a low evidence, high hyperbole manner - for example saying they are "...hardly more meaningful than star signs". One thing is sure, much research has attempted to assess the reliability and validity of diagnoses like schizophrenia - whether people decide the evidence is sufficiently impressive is then at least a matter of empirical - rather than simply ideological - debate.

Making Flippy Floppy and Slippery People

Given the recent 'position statement' by the British Psychological Society's (BPS) Division of Clinical Psychology (DCP) - Time for a Paradigm Shift in Psychiatric Diagnosis (link to full document at foot of that page) - it is worth taking a closer look at the alternative to diagnosis proposed by the DCP -so-called Formulation

Pere Ubu (Non-Alignment Pact: 1977)

In their Good Practice Guidelines on the use of Psychological Formulation, the DCP states "there is no universally agreed definition of formulation", but do rather nebulously state that:

"Psychological formulation is the summation and integration of the knowledge that is acquired by this assessment process that may involve psychological, biological and systemic factors and procedures"

In the same document, they reference Clinical psychologist Gillian Butler (1998) who says

"A formulation is the tool used by clinicians to relate theory to practice… It is the lynchpin that holds theory and practice together… Formulations can best be understood as hypotheses to be tested.”
and later Kuyken (2006) is quoted as saying

'...formulation is ‘a balanced synthesis of the intuitive and rational cognitive systems’

So, Formulation is a hypothesis that links (any specific?)theory and (any specific?) practice that balances intuitive and rational cognitive systems?

and then later still, what formulation is not?
"A formulation is not an expert pronouncement, like a medical diagnosis, but a ‘plausible account’ (Butler, 1998, p.1), and as such best assessed in terms of usefulness than ‘truth’  (Johnstone, 2006)"
Plausible to whom? How do we assess usefulness as opposed to truth? It seems from the way that some clinical psychologists speak that formulation is viewed as orthogonal to veracity - indeed, it is implicit that multiple formulations of the same case are not only possible but possibly desirable(?)

This Heat: 24 Track loop (1978)

Girlfriend is Better

In this context, it is worth unpacking this very recent post - So... What happens next? by the clinical psychologist Peter Kinderman in the light of the DCP paradigm shift document:
Of course, traditional psychiatrists, and many members of the public, say that they find a diagnosis helpful and even comforting. But the truth is that this comfort comes from knowing that your problems are recognised (in both senses of the word), understood, validated, explained (and explicable) and that the person you’re speaking to has a decent plan to help you. A problems list and a formulation can do that. Paradoxically, better than a diagnosis – since, for example, two people with a diagnosis of ‘schizophrenia’ or ‘personality disorder’ might have absolutely nothing in common, not even the same ‘symptoms’, any comfort from a diagnosis is likely to be illusory.

Doesn't Peter Kinderman seem to speak about diagnosis as meeting just those criteria set out by Lucy Johnstone for assessing formulation? As Kinderman says, psychiatrists and patients often view diagnosis as a ‘plausible account’ and presumably diagnosis may also be assessed in terms of usefulness rather than ‘truth’?  It does sometimes seem as though clinical psychologists want to use different criteria for assessing diagnoses and formulations (Science and anti-science).  Further, don't these claims sound somewhat modest and mundane for a paradigm shift? On the basis of saying that no two people with a schizophrenia diagnosis have anything in common, he leaps to the conclusion that "any comfort from a diagnosis is likely to be illusory" - presumably we do return to 'truth' as opposed to ill-usion. Indeed, it would be interesting to hear how Professor Kinderman delineates illusory comfort from real comfort in his patients - that would take some expertise!

Moon Rocks

What about the evidence on formulation?
Bieling & Kuyken (2003) state in their paper Is Cognitive Case Formulation Science or Science Fiction?
In terms of the scientific status of the cognitive case formulation process, current evidence for the reliability of the cognitive case formulation method is modest, at best. There is a striking paucity of research examining the validity of cognitive case formulations or the impact of cognitive case formulation on therapy outcome.
One problem, of course, is that all humans are prone to biases and influence of short-cut heuristics that include halo effects, illusory correlations, framing biases, recency effects, confirmatory biases, and failure to consider normative standards. Bieling states that
"Clinicians may make these errors so habitually that in cognitive case formulations of identical cases using identical formulation methods it is not possible to accurately establish consensus."

Of course, some clinical psychologist essentially argue for a science of the individual. In their review of case formulation in mental health, Rainforth & Laurenson 2013 state
… there are difficulties in promoting commonality due to the individual nature of the formulation, based on the service user presentation, traits, personality experiences and needs, and issues relating to practitioner skills and experience...The complex nature of formulation-based approaches to treatment planning contains vulnerability due to judgemental and inferential bias. Benefits for standardizing treatments were noted; however, this also highlights a dilemma in whether to use standardized or individualized approaches to CF.

In other words, it sounds awfully like no two formulations would be the same

Who benefits from formulation?
As noted by Kinderman above, those who use formulation do, of course value it believing it benefits their patients ….but this remains unsubstantiated by any acceptable notions of empirical scientific evidence.

Some evidence suggests that formulation benefits staff rather than the patients or the outcomes for patients
"care planning, staff-patient relationships, staff satisfaction and teamworking, through increasing understanding of patients, bringing together staff with different views and encouraging more creative thinking" Summers 2006.
Kuyken et al (2005) in their paper 'The reliability and quality of cognitive case formulation' say:

Our review suggests that, contrary to the claimed benefits of cognitive case formulation, it is not a panacea, and its evidence base is weak at best. Our review suggests instead that it is a promising but currently limited approach to describing and understanding patients’ presenting problems

They suggest "the quality of formulations ranged from very poor to good, with only 44% rated as being at least good enough." and among mental health practitioners in training this fell to 24.1%. Formulations were distributed across the range from very poor to good (‘‘very poor’’ 22.1%; ‘‘poor’’ 33.6%; ‘‘good enough’’ 34.5%; ‘‘good’’ 9.7%). In other words, only a minority of formulations are rated as "good enough"


Perhaps reliability and validity are irrelevant to the anti-science of formulation?

"Formulations may be reliable and valid but have no impact on treatment outcome. In contrast, they may be unreliable and invalid but lead through some alternative mechanism (e.g., increasing therapist self-confidence or enhanced alliance) to improved outcome." Bieling & Kuyken (2003) - see also p34 Good Practice Document Johnstone et al 2011

What this highlights most is the view that, while evidence for reliability and validity for formulation is lacking, it just doesn’t matter! The implication is that the lack of evidence for formulation is irrelevant, as it may still improve outcome. - Actually, no empirical scientific evidence exists to show that formulation improves outcome. Moreover no evidence at all exists to support the bold claim that formulation is in fact orthogonal to reliability and validity.

Finally, clinical psychologists may see formulation as an art rather than a science. Indeed, the BPS Good Practice Guidelines on Psychological Formulation states
"the subject matter of our discipline [clinical psychology], human beings and human distress, is not best served by the narrow ‘technical-rational’ application of research to practice. Rather, it requires a kind of artistry that also involves intuition, flexibility and critical evaluation of one’s experience. In other words, formulation is ‘a balanced synthesis of the intuitive and rational cognitive systems’ (Kuyken, 2006, p.30)."
Again, it seems little interest in the science rather than the artistry of formulation

Formulation is a treatment in itself?
Interestingly the BPS document on psychological formulation states "It should also be noted that developing a formulation can be a powerful intervention in itself" - this is an interesting notion insofar as it has no typical 'science' oriented evidence-base whatsoever - and if it is an intervention in itself then it ought to be evidence-based

This Must be the Place (naive melody)

So, formulation cannot be defined, it is a hypothesis, a theory-practice link. It has no basis in truth, it is based in usefulness (though possibly not usefulness to the patient it seems) . It may be an intervention in itself, and also not imply an intervention. It is unreliable and lacks validity. It has no evidential link to outcome. It is artistry linked to short, it is anti-science....

An ironic conclusion, that the touted Kuhnian paradigm shift appears to be one going backwards into pre-science or perhaps....formulation its better described as ante-science

Sunday, 12 May 2013

Psychology - Seductive, but is it Science?

Harvey: Tell me. We're alone here. No witnesses.
Art: Tell you?
Harvey: A sort of confirmation.
Art: Tell you what?
 Harvey: About ravishment

I am an unintentional psychologist. As teenager, I was persuaded to add Psychology as my final 'A'-level  "...probably useful to have a science" I mused. Within days, psychology had consumed me with its easy charm - taking me to places that were simultaneously familiar and exotic - and we duly began a relationship that has persisted for 30 years.

Psychology is the perfect partner - being whatever we want it to be. As a psychology undergraduate, one day I was dissecting brains or measuring social behaviour in cockroaches, the next I was mesmerised by William James' poetic words on emotion or a lecturer telling me that I wanted to have sex with my mother.

Can we really refer to this capacious church of psychology as science?

'La Ritournelle' by S├ębastien Tellier
("its awesome" according to my 5 yr old son Vivek)

For most of my academic life as a psychologist, I have endured a dissonance - never doubting that psychology was a science, but believing it is impossible to define a science. Thirty years on, defining science still seems like counting angels on the head of a pin, though now...I believe it's a mistake to refer to psychology as if it were a unitary discipline - rather we have psychologies. And more recently, reservations about the scientific status of some psychologies have begun to feed my obsessive dissonance. Indeed, I would venture that some of these psychologies are...frankly...anti-science (something I will return to in a later post).

Possibly because of a (righteous) historical fear of introspection, psychologists tend to look outwards rather than inwards...which brings me to my point about how psychology currently operates. Psychologists engage in a version of science that is systemically corrupted and blinkered. The evidence on this issue seems unquestionable to me - as indicated by the lack of published replications and null findings, questionable research practices such as selective reporting, hyperbole, evidence denial and even outright fraud, all combined with the shallow pursuit of the curious.

" cannot intellectualise your genes, which make aspects of your life inevitable. You cannot intellectualise yourself out of obsession. You cannot cure yourself of it." Nicolas Roeg

In this context, I am posting a collection of recent pieces where I have discussed problems in the practice of psychology - and the extent to which these issues undermine the scientific status of psychology as it currently practices

It's time for psychologists to put their house in order - My original article in the Guardian where I outline some of the systemic problems inherent in the way that psychology currently operates - especially with regard to publishing and my section editorship at the new journal BMC Psychology, which is addressing some of the issues regarding null findings and replications
Negativland - a home for all findings in psychology - The Open Access paper that I published in BMC Psychology (which I'm delighted to see has over 10,000 downloads in 2 months). This paper reviews many problems and distortions that beset psychology and how these are longstanding - raising questions about the resistance to change amongst psychologists. One upshot of this unwillingness is how it plays out in the minds of the public - whether we believe psychology is a science or not in any technical intellectual sense becomes redundant if the wider perception of psychology is that it has little credibility and masquerades as a perverse charade of science (see Rupert Read's points below).


BBC Radio3 NightWaves audio recording of debate - Is Psychology a Science? - between myself and the philosopher Rupert Read on BBC Radio 3 Night Waves programme (it was linked to my Guardian article). One point argued by Read, is that it is in fact  impossible to replicate experiments in psychology - because of the historical nature of human beings. I am pretty sure no psychologist would agree with this philosophical point - and I explain that we use naive participants. Indeed, Read's argument strikes me as essentially incoherent - at what point does the historical nature of humans kick-in? Presumably, even after one individual has been tested in any experiment - in which case, no experimentation is possible in psychology (nevermind replication)

Why Psychology ain't Science - piece written by Rupert Read following our debate, where he expands on why he thinks psychology is not (and cannot be) a science. This largely seems to consist in his straw-man positioning me as a simplistic Popperian as opposed to his seemingly Kuhnian view of science. "...real science is: roughly, Kuhnian puzzle-solving within a research tradition, in a field that is not one that we construct and inhabit just by virtue (following here Schutz and Garfinkel and Wittgenstein) of being competent social actors"

"Established Psychology is one of those juggernauts that Wittgenstein didn’t like, and rightly so." Rupert Read

Keith Jarrett - The Koln Concert

Storify - this is a collation of the many Tweet discussions that followed our debate on whether Psychology is a Science (compiled by @neurowhoa) - they have been nicely ordered along the line of themes as they emerged in random timeless Twitter space

E=MC2 by Big Audio Dynamite
a paean to Nic Roeg
Even if we psychologists do eventually show the determination to get our house in order - many will still view psychology as a pseudoscience - what is how we psychologists view what we do and how we practice what we do