Cognitive therapy to treat moderate to severe depression works just as well as antidepressants

Apr 5 2005

Cognitive therapy to treat moderate to severe depression works just as well as antidepressants, according to an authoritative report appearing today in the Archives of General Psychiatry. The study, conducted by researchers at the University of Pennsylvania and Vanderbilt University, challenges the American Psychiatric Association guidelines that antidepressant medications are the only effective treatment for moderately to severely depressed patients.

Either form of treatment worked significantly better than a placebo, but the researchers demonstrated that cognitive therapy was more effective than medication at preventing relapses after the end of treatment.

"We believe that cognitive therapy might have more lasting effects because it equips patients with the tools they need to learn how to manage their problems and emotions," said Robert DeRubeis, professor and chair of Penn's Department of Psychology. "Pharmaceuticals, while effective, offer no long term cure for the symptoms of depression. For many people, cognitive therapy might prove to be the preferred form of treatment."

The study, which follows years of debate on the relative merits of cognitive therapy versus medication for more severe forms of depression, is the largest trial yet undertaken on the topic; it involved 240 depressed patients. The patients were randomly placed into groups that received cognitive therapy, antidepressant medication or a placebo. Patients in the antidepressant group, which was twice as large as the other two, were treated with paroxetine (Paxil). Lithium or desipramine was also given, as necessary.

After 16 weeks of treatment, patients in both the medication and cognitive therapy groups showed improvement at about the same rate; however, cognitive therapy patients were less likely to relapse in the two years following the end of treatment. According to the researchers, the return of symptoms might demonstrate that the medication may have blunted the appearance of depression but did not affect underlying disease processes.

"Medication is often an appropriate treatment, but drugs have drawbacks, such as side effects or a diminished efficacy over time," DeRubeis said. "Patients with depression are often overwhelmed by other factors in their life that pills simply cannot solve. In many cases, cognitive therapy succeeds because it teaches the skills that help people cope."

The researchers also noted slight differences in the response to treatment between the two testing locations, with cognitive therapy performing better at Penn and medications performing better at Vanderbilt. Researchers surmise that the medication worked so well at the Vanderbilt clinic because more of the patients there were markedly anxious, in addition to being depressed, and the medications used in the research have anti-anxiety properties.

The researchers further believe that cognitive therapy patients might have done better at Penn due to the experience level of the therapists involved. Just as the experience of therapists may be important in cognitive therapy, so, too, can the expertise of prescribing physicians play a role in the success of antidepressant medication treatment. Studies have shown that antidepressant medication dosages are still largely a matter of physiciansdiscretion.

"Clearly, cognitive therapy is not for everyone, and its success could depend on variables such as the expertise of the therapist and the patient's willingness or ability to take the therapy to heart," DeRubeis said. "The key to establishing any form of treatment is rating its effectiveness in comparison to treatments currently in use, and this study has shown cognitive therapy to be a viable alternative."

Clinical researchers at the Penn School of Medicine's Department of Psychiatry involved in the study were Jay D. Amsterdam, Paula R. Young, John P. O'Reardon and Madeline M. Gladis. Vanderbilt researchers include Steven D. Hollon of the Department of Psychology and Richard C. Shelton, Ronald M. Salomon, Margaret L. Lovett, and Laurel L. Brown of the Department of Psychiatry. Contributing author Robert Gallop is with West Chester University's Department of Mathematics and Applied Statistics.

The work was supported by a grant from the National Institutes of Health. GlaxoSmithKline provided medication and placebos.