Story of Psychology - Morton Hunt [397]
But Does It Really Work?
In his autobiography, the late H. J. Eysenck proudly termed himself “rebel with a cause.” Indeed, many causes. After leaving Germany for England in his youth, he enthusiastically laid about him in sundry educational, political, and scientific battles, even while making solid contributions in several areas of psychology. Long a professor and researcher at the Institute of Psychiatry, University of London, and with an impressive list of published and widely cited contributions on intelligence, testing, and personality, he, like Ellis (but on a serious plane), was always a resolutely cheerful bad boy of psychology.
None of his imbroglios was more heated than the one brought about by his historic assault on psychotherapy in 1952. Eysenck had always been contemptuous of psychotherapy, which he felt was unsupported by any scientific evidence. To prove the point he reviewed the data of nineteen studies reporting the results of psychotherapy and came to some shocking conclusions. The different studies claimed “improvement” in as few as 39 percent and as many as 77 percent of the cases, a range so broad as to justify suspicion, he said, that something was amiss. Far worse, Eysenck added up the findings and calculated that, on average, 66 percent of the patients had improved—and then cited other studies reporting that of neurotic patients who had custodial care but no psychotherapy, 66 to 72 percent had improved. His conclusion: There was no evidence that psychotherapy was responsible for its supposed effects. His radical corollary to that conclusion: All training in psychotherapy should be abandoned forthwith.111
“The sky fell in,” he later commented. “I immediately made enemies of Freudians, of psychotherapists, and of the great majority of clinical psychologists and their students.”112 As was to be expected, many of his newly made enemies—including prestigious names in British and American psychology—wrote angry replies. Anger aside, they had good grounds for discrediting his findings, and published rebuttals in a number of leading British and American psychology journals. Their most telling criticisms were that Eysenck had lumped together data derived from different forms of therapy, different kinds of patients, and different definitions of improvement; moreover, the untreated group was not truly comparable to the treated groups.113 Still, he had thrown down the gauntlet; it was now up to those who believed in psychotherapy to prove that it was effective, a task they had never seriously undertaken.
Ever since, there has been a steady flow of psychotherapy outcome studies—many hundreds, in fact—differing greatly in scientific quality, in the size of the samples studied, in the criteria of improvement, and in the use or lack of use of control groups. Their findings, accordingly, have shown great variation.
But meta-analyses that rate the studies by scientific quality, adjust for differences in method, and only then sum up the results, have repeatedly found that the weight of evidence is clearly in favor of psychotherapy. In 1975, a painstaking meta-analysis of nearly a hundred controlled studies, by Lester Luborsky of the University of Pennsylvania, concluded that most of them found a high proportion of patients benefiting from psychotherapy. And, contrary to Eysenck’s claim, two thirds of the studies showed that significantly more treated than untreated patients improved.114 (If studies involving minimal treatment had been excluded from the Luborsky review, the superiority of therapy over no therapy would have appeared still greater.)
A comprehensive review of outcome studies made in 1978 by a team at the National Institute of Mental Health came to a similar conclusion.115 In 1980 a still more comprehensive meta-analysis by another team of psychologists reviewed and