WHEN Dr. Robert DeRubeis began writing up the results of a recent study about the effectiveness of cognitive therapy for severely depressed patients, he and the two graduate students assisting him found themselves approaching the task with even more than the usual professional caution.
“It was an intense writing experience, because we knew that everything we said would be scrutinized carefully,” recalls DeRubeis, associate professor and director of clinical training in psychology. “So I’ve never been as careful in writing anything in my life.”
DeRubeis and his grad students–Lois A. Gelfand C’88 G’95 and Tony Z. Tang –knew that their findings would challenge what has become an accepted wisdom in the psychiatric community: For severely depressed patients, antidepressant medication works; cognitive therapy doesn’t. And what made that accepted wisdom all the harder to buck was that it appeared to have a solid scientific underpinning in the form of a 1989 study by the National Institute of Mental Health’s Treatment of Depression Collaborative Research Program.
“That study, for a variety of reasons, became the single most important study in the area,” says DeRubeis, “and when people went to make inferences about treatment effectiveness for depression, they tended to cite that study and that study only.” (The Agency for Health Care Policy and Research’s guidelines, for example, state: “For severe and psychotic depressions, there is strong evidence for the efficacy of medication and little or none for the efficacy of psychotherapy alone.” The American Psychiatric Association has taken a similar position in its guidelines.)
As a result, says DeRubeis, both patients and professionals “came to have greater and greater confidence that medications are potent treatments for severe depression, and cognitive therapy is not. This was at odds with what I understood from other studies that were published or about to be published. So I wanted to figure out what was going on.”
To that end, he and Gelfand and Tang spent “hundreds and hundreds of hours” examining the raw data that had informed the NIMH study and three other studies. (Dr. Anne Simons of the University of Oregon, who had been involved in one of the three additional studies, provided access to that data and is listed as a co-author.) What they found was that not all the conclusions from the NIMH study were warranted once all the relevant data were examined–and that across the four studies of severely depressed patients, cognitive therapy actually “has fared as well as antidepressant medication.” The results of their study were published in the July issue of The American Journal of Psychiatry –which, as it happens, is published by the American Psychiatric Association. (The article can be found online at: http://ajp.psychiatryonline.org/cgi/content/full/156/7/1007.)
“I used original data, so that each patient counted as one data point, as opposed to each study counting as one data point,” DeRubeis explains. “That allowed me then to compare the effects that medicine had in the 1989 study versus the effects it had in the others.” It also allowed him to compare the NIMH study’s cognitive-therapy results with those in the other studies.
“The effects of the medications were very consistent across those four studies, suggesting that they were properly implemented” in all four, he says. “On the other hand, the effects of cognitive therapy were quite variable across the four studies. And the cognitive-therapy patients fared worst in the 1989 study, which can account for the reason cognitive therapy did poorly relative to the medicines in that study.”
While the variable results of cognitive therapy in that study may, to some extent, mirror the results for patients nationwide, that says more about the relatively uneven quality of psychotherapy in the nation–and in research–than it does about the effectiveness of good psychotherapy. DeRubeis also points out that the NIMH study only examined 53 severely depressed patients, of whom 26 had received cognitive therapy.
But DeRubeis says he is “absolutely not” trying to downplay the benefits of antidepressant medications.
“Medicines can be life-saving,” he acknowledges. “And as my reading of the literature tells me, they are as potent as the best of psychotherapies. The advantages of medicines are that they are easier to engage in; they are somewhat cheaper –though less so than many imagine; they’re readily available; and one knows what one is getting when one takes Prozac or Paxil, whereas this isn’t so true with psychotherapies.
“The reasons to consider psychotherapy are that, as far as we know, there are very few side effects, and their benefit appears to last longer than do the benefits of medicines.”
Not everyone is overjoyed with his findings, acknowledges DeRubeis, who has received some skeptical comments from psychiatrists. Some psychiatrists, he suggests, “have seen many patients who have benefited greatly” from medications but might not have been “exposed to patients who have benefited from psychotherapy.”
A more serious concern, he says, is that there is a “great imbalance in the funding of research on the two different sets of methods,” since pharmaceutical companies “fund depression-treatment research at a vastly higher scale than psychotherapy research is funded.” As a result, “there are many more studies, and more kinds of studies, on the effects of medications.”
DeRubeis and his colleagues are currently conducting a large-scale study of treatments of severe depression, in conjunction with a group at Vanderbilt University. “It will be larger than all of the others combined,” he says. “We’re comparing cognitive therapy versus Paxil versus placebo in the treatment of 240 severely depressed patients.” It should be completed in “about a year,” he says, though it may be another year after that before it gets published.
For those coping with depression now, the message DeRubeis offers is: “If they can find someone with a track record in successful treatment of depression, either with medications or with an active, focused psychotherapy for depression, then they can proceed with some confidence. But we don’t know how available are the kinds of treatments that we study in our research.”
