Gold Coast's Pinniger Clinic


Where cutting-edge neuroscience meets depth psychotherapy.

home . background . practitioners . training . library . events . links

Evidence Based Psychotherapy Outcome Studies, Leon Petchkovsky.

return to the library

download pdf Evidence Based Psychotherapy Outcome Studies. Do they make any sense?

We all have psyches, therefore we all, inevitably, fancy ourselves experts in matters psychological. Furthermore, as Jung put it so pithily, "every psychology is a confession" (Jung M D R). The Freudian paradigm tells us as much about Freud's own psychology as it does about everyman's, as Skinner's does about Skinner, and so on for every model, and every practitioner. To cling too firmly to universal psychological generalisations can be a form of fundamentalism. But we can make ONE generalisation. There will ALWAYS be unique, idiosyncratic aspects of the inner life of every individual which will defy prescriptions; and arguably, it is these unique features which are more significant in the individual's life than the generalisations. Jung's notion of "individuation" as the central thrust of the psychotherapeutic process therefore has much to recommend it, as has Wilfred Bion's advice to the therapist to abandon all preconceived notions.

Nevertheless, if there were no models, no working hypotheses, no generalisations, however provisional; learning the rudiments of this craft would be impossible. And it is very much a craft, halfway between art and science. We yearn for stable verities, and effective and reproducible techniques; hence one of the drives behind Outcome Studies.

But Psychotherapy Outcome Studies have massive methodological problems, as well as their foundational philosophical ones. There are swarms of variables, and some of them may not be obvious; (the age of the therapist, their gender, appearance, social class, political leanings, size of their moustache, etc etc). There is the struggle to keep identified variables constant (moustaches grow, they can be dyed, shaved or trimmed etc etc). Qualitative methods are time and labour intensive, and there are constant trade-offs to be made between a design which captures important phenomenological nuances but eats up massive energy, and a chunkier design which samples thousands to achieve statistical significance, but where the outcome is trivial and of dubious validity ("in a sample of 10,000 patients, the 5,000 who had CBT as opposed to IPT did 0.001 percent better on HAM-D scores; and we can be confident of that at the 0.05 level; isn't that great").

More fundamentally still, human nature being what it is, the goal posts (see Petchkovsky Morris and Rushton 2002) keep changing ("I started off wanting to deal with my social anxiety, but discovered my marriage was actually on the rocks").

Because the CBT paradigm lends itself more readily to operational standardisation (instrumentalism), and because goals are clearly (over-precisely?) defined, a CBT methodology lends itself more readily to "evidence based" paradigms, hence the prominence of CBT in the EB literature; to such an extent, it should be said, that it opens itself to critiques of rhetorical pretence of rigour.

There are some lateral approaches. Rather than comparing therapy models against each other, Michael Lambert and his associates (Lambert in Okiishi et al 2006) have tracked self-reported patient responses over time in the course of psychotherapy with various therapists, and discovered that therapists come in three groups. There is the bulk of therapists (about 70%), who do a fair enough job. There are the "supershrinks" (about 10%), whose patients regularly do better, often dramatically. And there are the "toxic shrinks" (another 10%), who are guaranteed to make their patients worse. And there is an optimal therapeutic input time, for a range of conditions, which is actually longer than that prescribed by various brief intervention paradigms.

The data are so strong that some American Health Insurance Companies insist that patients have MORE sessions rather than fewer, in line with Lambert's paradigms, because the statistical (AND financial) outcome studies support it ! The case to be made for market endorsement cannot be brushed aside lightly. If hard-nosed health insurers recommend certain kinds of psychotherapy, maybe that's as good as it gets in this contentious field.

In the German speaking world, some of my psychiatrist psychotherapist colleagues have been conducting long term effectiveness studies over several DECADES (from 1987 to the present). This project, with centres in Zurich (Prof Mattanza et al) Berlin (Prof Keller et al) and Heidelberg (Prof Rudolf et al), now called the PAL study (Naturalistic Study of Long-Term Analytic Psychotherapies), was initiated by the DGPT (German Society for Psychotherapy). (See Keller 2006, Breyer et al 1997, Grande et al 2006, and Open Door Review 2002). Outcome data for psychodynamic psychotherapy modalities was so strong that the German Government now endorses Jungian Analytic therapy under its health insurance scheme! When it comes to judging art, or fine wines, or heady perfumes, the scientific paradigm has its limitations. Yet connoisseurs can, and do, make very fine gradations, and the markets endorse these judgements accordingly. How much did you pay for that bottle of Grange, that Brett Whitely painting? Perhaps the German Government is onto something.

1. Breyer, F., Heinzel, R. & Klein, Th. (1997). The Konstanz Study - The German Consumer Reports Study; Kosten und Nutzen ambulanter Psychoanalyse in Deutschalnd (Cost and benefit of outpatient analytical psychotherapy in Germany). Gesundheitsökonomie und Qualitätsmanagement, 2, 59-73. 1997

2. Grande T, Dilg R, Jakobsen T, Keller W, Krawietz B, Langer M, Obrebracht C, Stehle S, Stennes M, & Rudolf G. (2006). Differential effects of two forms of psychoanalytic therapy: Results of the Heidelberg-Berlin study. Psychotherapy Research, August 2006; 16 (4): 470_485

3. Jung, C. G. (1983). Memories Dreams and Reflections. Random House 1983, New York .

4. Keller W, Westhoff G, Dilg R, Rohner R, Studt HH. (2006). Effectiveness and utilization of health insurance benefits in long-term analyses: Results of an empirical follow-up study on the effectiveness of Jungian Analysis http://www.uni-saarland.de/fak5/krause/ulm97/keller.htm

5. Open Door Review; Second Revised edition. (2002). Meta-analysis of psychotherapy outcome studies. International Psychoanalytic Association.

6. Okiishi John C., Lambert Michael J., Eggett Dennis, Nielsen Lars, Dayton David D., Vermeersch David A.. (2006). An Analysis of Therapist Treatment Effects:Toward Providing Feedback to Individual Therapistson Their Clients' Psychotherapy Outcome. J Clinical Psychology, Vol. 62(9), 1157-1172 (2006)

7. Petchkovsky Leon, Morris Philip, and Rushton Paul (2002). Choosing a psychodynamic psychotherapy model for an Australian public sector m,ental o 4. Dec 2002. Pp 330-4.

return to the library

home . background . practitioners . training . library . events . links