Drug or No Drug: Placebos may be more than appeasing

Antidepressant drugs such as Prozac generally fall short of providing significantly more relief to depressed patients than placebo pills do, according to a new analysis of multiple clinical trials obtained from the Food and Drug Administration (FDA).

Antidepressants substantially outperform placebos only among extremely depressed individuals, says a team led by psychologist Irving Kirsch of the University of Hull, England. In these cases, relatively weak placebo responses, rather than any heightened reactions to antidepressants, explain the medications’ superiority, the researchers hold.

“There is little evidence to support the prescription of antidepressant medication to any but the most severely depressed patients, unless alternative treatments with fewer side effects have failed to provide benefit,” Kirsch says.

Alternative depression treatments with demonstrated effectiveness include physical exercise, several forms of psychotherapy, and even certain self-help books, in his view.

Kirsch and his colleagues obtained data from 47 clinical trials submitted to the FDA as late as May 2007. They then weeded out 12, focusing on 35 clinical trials that typically lasted 6 weeks and also showed substantial completion rates and consistent monitoring of depression symptoms. Those trials compared randomly assigned placebo treatment to treatment with any one of four antidepressants—fluoxetine (Prozac), venlafaxine (Effexor), nefazodone (Serzone), and paroxetine (Paxil).

A total of 5,133 depressed patients participated in the clinical trials.

The researchers statistically combined data from qualifying trials and calculated the extent to which antidepressants and placebos alleviated depression. Their findings, based on both published and unpublished data, appear in the February PLoS Medicine.

The placebo response “was exceptionally large,” Kirsch says. Statistically, it accounted for more than 80 percent of the symptom alleviation observed in antidepressant-treated patients. Final depression scores for patients taking antidepressants generally did not indicate any greater effects than those observed for placebo patients, Kirsch holds.

For as yet unclear reasons, the placebo effect, while it remained substantial, was weaker in the most severely depressed patients.

All of the antidepressants, which belong to the newest generation of these medications, alleviated depression comparably well.

Still, the FDA data are from patients with a narrow range of scores on a standard depression-rating scale, Kirsch notes. These clinical trials focused primarily on patients who, at the start of treatment, scored as having “very severe” depression. Only a small number of patients started out with moderate to severe depression.

Future studies with data for patients with a wide array of depression scores might alter the results, Kirsch notes.

The placebo response identified in the new analysis parallels that reported in a previous assessment of FDA data directed by psychiatrist Arif Khan of the Northwest Clinical Research Center in Bellevue, Wash. The team analyzed 52 clinical trials of depression conducted between 1985 and 2000. Antidepressants outperformed placebos in only 25 of those trials.

Clinical trials recruit a select group of patients who, typically, are seriously depressed but not suicidal and not suffering from other psychiatric ailments, as often happens with depressed patients seen in clinical practice, Khan asserts. Thus, clinical trials don’t negate the usefulness of antidepressant medication for depressed patients seen in physicians’ offices.

Moreover, depression symptoms fluctuate over time, so it’s hard to know whether changes observed in clinical trials reflect patients’ spontaneous progress or deterioration.

Despite the drawbacks of clinical trials, “at this point, they’re all we can rely on to assess the effects of antidepressants,” Kirsch responds.

Bruce Bower

Bruce Bower has written about the behavioral sciences for Science News since 1984. He writes about psychology, anthropology, archaeology and mental health issues.

From the Nature Index

Paid Content