Therapy v. Medication

Psychologist Robert DeRubeis searches for a better weapon in the battle against depression.
January 1, 2009

A groundbreaking study on the effects of cognitive therapy by Robert DeRubeis, Professor of Psychology and Associate Dean for the Social Sciences, and Steven Hollon at Vanderbilt University has continued to generate new findings since its initial publication in 2005.

Published in the Archives of General Psychiatry, it showed that cognitive therapy worked just as well as the antidepressant paroxetine (also known as Paxil) in treating depression, challenging the American Psychiatric Association’s guidelines that antidepressant medications are the only effective treatment for moderately to severely depressed patients. The study also found that patients who received cognitive therapy were less likely to relapse after discontinuation of treatment than those who received medication.

DeRubeis’ study, which involved 240 patients with moderate to severe depression, was the largest trial to date on the subject. “Because it was such a big study,” DeRubeis says, “we have lots of very interesting data on a range of variables.” As a result, several papers have subsequently been published on different aspects of these data.

Penn psychology doctoral student Yan Leykin, along with DeRubeis and a team of researchers, explored whether patients with zero, little or extensive experience with antidepressant medication responded differently to cognitive therapy compared to medication. Their findings, published in 2007 in the Journal of Consulting and Clinical Psychology, revealed that although the two treatments are equally effective in patients who have taken medications for little or no time, cognitive therapy is significantly more effective in patients who have tried medications extensively.

"We ask ourselves as clinical scientists, what clues can these patterns provide about the mechanisms that these treatments engage—in other words, how these treatments work." - Robert DeRubeis

In a paper published last year in the British Journal of Psychiatry, Jay Fournier, also a Penn psychology doctoral student, and DeRubeis found that medication was more effective for depressed patients with personality disorders, while cognitive therapy was more effective for those with depression alone.

According to DeRubeis, “The general expectation—based on highly indirect evidence—was that cognitive therapy would work well for patients who have personality disorders comorbid with depression, and medications would work best for patients suffering from depression alone.” He explains that the field traditionally thinks of depression as being more firmly rooted in biology, whereas a personality disorder indicates trouble negotiating one’s social and personal world—issues that cognitive therapy can help train a patient to handle. “For the first time we had data to test this assumption, and we found the reverse of the expected pattern,” he says.

DeRubeis suggests that paroxetine and other SSRIs (a class of antidepressants) could inhibit hostility, aggression and impulsivity. “This may allow people with personality disorders to experience life in a different way, and allow the people in their social networks to experience them in a different way—thereby reducing tension in their lives and allowing them to calm down,” he says. “We don’t normally think of depression as not being calm, but in fact depression can have in it hostility and disgruntlement. If a person could inhibit these reactions, with or without the help of medications, life could become more manageable.”

Many patients with personality disorders dropped out of cognitive therapy after just a few sessions, and DeRubeis conjectures that the treatment sessions may have made them too uncomfortable. However, of the patients in the study who persisted with the therapy and did well on it, a much smaller proportion relapsed after treatment than did their counterparts who got better on medication and then discontinued it. DeRubeis is now conducting research to test the effects of a much longer period of cognitive therapy than the 16-week treatment used in the 2005 study.

“We’re getting hints that individuals with personality disorders can benefit from cognitive therapy,” he says, “but it can often take six months or a year, rather than the three to four months it often takes to help depressed people who do not also have these troubling personality patterns.”

DeRubeis thinks that therapy’s significant success for depressed patients without a personality disorder might be due to the treatment’s “activating effect,” and his paper forthcoming in the Journal of Consulting and Clinical Psychology further supports this idea. It presents analysis of data from the 2005 study that shows that cognitive therapy was far more effective than antidepressants in improving depression for those who began treatment unemployed. DeRubeis also has preliminary findings that show that unemployment itself decreased among those who received the therapy.

“These findings certainly have practical implications in terms of treatment,” DeRubeis says. “But we also ask ourselves as clinical scientists, what clues can these patterns provide about the mechanisms that these treatments engage—in other words, how these treatments work.”