Why there are so pitifully little evidence for psychotherapy effectiveness? Part 2


The neuroses are ‘cured’ by Christian Science, osteopathy, chiropractic, nux vomica and bromides, benzedrine sulfate, change of scene, a blow on the head and psychoanalysis which probably means that none of these has yet established its real worth in the matter … moreover since many neuroses are self-limited, anyone who spends two years with a patient gets credit for the operation of nature.[1]

And this is not an isolated view, since

You can take a ward full of patients of whatever diagnosis, age and sex, and you can give them all a new drug, or a new kind of therapy, or simply a change in their routine, and a third of them will get better, a third will stay the same, and a third will get worse, give or take a few each way. Of course, a few weeks or months later, some of those who got better will get worse, and some of those who got worse will get better. But you cannot be sure that when patients say they are better they really are.[2]

These paragraphs introduced us to another difficulty in analyzing evidence for therapy effectiveness and immediately provided us with its explanation. Following their line of thought, we can in one breath add the following items to the list of “efficient” therapies: neuro-linguistic programming, facilitated communication, neurokinesiology, neurotransgression, regression, numerology and tarot. The only variable, which accounts for the efficacy of all those “therapies”, is… time; there is also the natural tendency of an organism to regain a state of balance, a tendency which developmental psychopathologists call resilience. In their view, it is an ability to recover one’s lost or weakened strength as well as resistance to harmful factors. This individual ability to revert to the initial state of balance depends on the proper functioning of the neurohormonal structures, including the pituitary gland, renal cortex, and hypothalamus.[3]

In his study which was devastating blow to psychoanalysis Hans Eysenck  asked a very simple, yet accurate, research question: is the Freudian psychoanalysis any more effective than any other face-to-face interaction, for example a meeting with a preacher who talks to his patients about the Bible? Eysenck compared psychoanalysis to various talk-based “therapies” and discovered that 44% of patients with symptoms of depression or anxiety declared improvement. Patient’s treated with other therapies managed to improve in 64% of cases over 2 years of treatment. Patients who were left untreated for 2 years reported improvement of their symptoms in… 60% of cases. [4]

It is because of this tendency that 60% of those examined in Eysenck’s research showed an improvement without any therapy. Whenever research is carried out without reliable comparison with control groups not subjected to any therapy, obtained results might be due to the influence of the variable of time (sometimes described as the regression to the mean). Sooner or later the organism recovers its balance, and if someone “assisted” in this natural process, he can easily claim the laurels of being a healer. But is he really one?

This credit certainly cannot go to one psychoanalyst who on terminating years-long psychotherapy was told by his patient that through all those years he had told him nothing that he couldn’t learn from his friends in a pub.[4] A certain woman drew similar conclusions, but before she completed her therapy. Having noted that that the professional psychotherapy she had gone into did not come up to her expectations, she hit upon an excellent idea. She began to regularly invite her friend out to a restaurant. In return for her advice and the possibility of confiding in her, she agreed to pay for food and drinks. Those meetings were probably more profitable to her than visits to the psychotherapist and certainly more pleasant because accompanied by eating.[5]

There are many similar cases and anecdotes. This is how Raj Persaud describes this:

The tendency for large numbers of people to get better without help, but with the passage of time, is so well known that some psychiatrists have suggested that the waiting list to see a psychiatrist could be seen as form of treatment in itself. Sometimes if you managed to see clinician too soon after an upset, when you were in a terrible state, he or she might prescribe treatment which in the end could prove unnecessary because recovery was possible without it – given enough time.[6]

Perhaps it is the time, but perhaps some other factors, that may cause an improvement in patients in similar situations. Without going too deep into this phenomenon, scientists have referred to it as placebo therapy, sometimes also calling it “talk therapy”. To make this notion clearer, let us once more give the floor to Eysenck:

“‘Placebo treatment’ is a pseudotreatment” which has no rationale or meaning, and is not intended to benefit the patient; it is simply instituted to make him believe that he is being treated, while in actual fact he is receiving no kind of effective treatment whatsoever. A placebo treatment is a control for non-specific effects, sych as a patient’s going to see a therapist, believing that something is being done for him, and possibly talking to the psychiatrist or psychologist. It should therefore be a control.[7]

Therefore, if a comparison of a given therapy with a placebo therapy does not point to statistically relevant differences, we can confidently assume that it is worth as much as a chat with friends in a bar or a restaurant[8]. The converse is also true: if such differences do appear, they may be a rationale for recommending such a therapy to a person who is looking for it. Do psychotherapies emerge victorious from such comparisons?

The study of this kind, conducted by Eysenck, was described above. Its findings were shocking because they demonstrated differences, which spoke against psychoanalysis. It turned out that more patients had made an improvement when they were left to themselves than those who were subjected to psychoanalysis[9]. The publishing of the results provoked a stormy debate and Eysenck found himself on the receiving end of criticism. Granted, this contributed to identifying certain flaws in his methodology, but his research was later replicated many times and in various ways. Many years afterward, as late as 1985, Eysenck had already realized that psychoanalysis did not necessarily produce negative results as compared with placebo therapy, but he also knew that it was not worth much more. Here is how he summed up the state of knowledge on the subject:

Even now, thirty years after the article in which I pointed out the lack of evidence for therapeutic effectiveness, and some five hundred extensive investigations later, the conclusion must still be that there is no substantial evidence that psychoanalysis or psychotherapy have any positive effect on the course of neurotic disorders, over and above what is contributed by meaningless placebo treatment. Treatment or no treatment, we get rid of our colds, and treatment or no treatment, we tend to get rid of our neuroses, although much less quickly and much less surely.[10]

As a well-known and intransigent opponent of psychoanalysis, Eysenck was always suspected of one-sidedness. But he was not the only one to prove that there are no differences between psychotherapy and placebo treatment, and not only with regard to psychoanalysis. Gavin Andrews demonstrated that good psychiatric care has better effects, poses fewer risks, is cheaper than dynamic psychotherapy and is only slightly less effective than cognitive-behavioral therapy.[11] Eysenck and Andrews’s views are not isolated. As Richard Stuart claims in his analysis of therapeutic failures based on the review of 21 empirical studies:

In view of this evidence, it can be said that persons who enter psychotherapy do so with a Modest Chance of marked improvement, a much greater chance of experiencing little or no change, and a Modest chance of experiencing a deterioration in their functioning[12].

Similarly in reviewing the literature on therapist effectiveness Robyn Dawes concluded that: “there is no positive evidence supporting the efficacy of professional psychology. There are anecdotes, there is plausibility, there are common beliefs, yes – but there is no good evidence.”[13]   These findings are concurred with by William Epstein, who in his book Psychotherapy as religion contends that: “Psychotherapeutic intervention has not demonstrated any benefit to any patient group under any circumstances.”[14]

A similar view is presented by Tana Dineen: “Outcome evaluations and cost-benefit studies suggested, at best, that psychotherapy was somewhat effective with some of the clients some of the time.”[15]

Parloff, in reviewing nearly 500 rigorously controlled studies, concluded: “the research evidence… has not met the needs of the policy makers and does not greatly enhance the credibility of the field of psychotherapy.”[16]

What remains for us then, after analyzing all these studies and their summaries? Shall we switch the light off and leave? Is there nothing left?

To be continued.

[1] A. Myerson, The attitude of neurologists, psychiatrists, and psychologists toward psychoanalysis. „American Journal of Psychiatry”, 96, s. 623-641, 1939 p. 641.

[2] Rowe, Introduction.

[3] Curtis, W. J., & Cicchetti, D. (2003). Moving research on resilience into the 21st century: Theoretical and methodological considerations in examining the biological contributors to resilience. Development and Psychopathology, 3, 773-810.

[4] Shorter, E. (1996). A history of psychiatry. New York: Wiley, p. 312.

[4] Symington, The analytic experience…

[5] Feltham, C. (1995). What is counselling? The promise and problem of the talking therapies. London: Sage Publications.

[6] Persaud, R. (1977). Stay sane. How to make your mind work for you. London: Metro Books, pp. 24-25.

[7] Eysenck, H. J. (2004). Decline and fall of the Freudian empire. Transaction Publishers, p.78.

[8] Authors of this book would like to point out that in “classic” situation, when you meet with friends in a pub, you could actually split the bill!

[9] Scientists call a therapy or treatment that inhibits (or slows down) patient’s ability to regain physical or emotional balance, a counter-effective therapy.

[10] Eysenck, Decline and fall…p 80.

[11] Andrews, G. (1993). The essential psychotherapies. British Journal of Psychiatry, 162, 447-451.

[12] Stuart, R. B. (1970). Trick or treatment: How and why psychotherapy fails. Champaign: Research Press, p. 50.

[13] Dawes. House of Cards. p. 58.

[14] Epstein, Psychotherapy as religion. p. xi.

[15]  Dineen, Manufacturing victims. p. 122.

[16] Parloff, M. B. (1982). Psychotherapy research evidence and reimbursement decisions: Bambi meets Godzilla. American Journal of Psychiatry, 139(6), 721.


One response to “Why there are so pitifully little evidence for psychotherapy effectiveness? Part 2

  1. Reblogged this on wrsurya and commented:
    is this perhaps pointing to the seemingly inherent volatility of the human mind that it continues to elude us despite decades of attempts to box and define it…? 🙂

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s