Question: How much placebo effect is seen with antidepressants

  • Patient or Population: Patients who are treated with antidepressants
  • Intervention or Indicator: Treatment with antidepressants for depression
  • Comparison or Control: Comparing antidepressants with placebos in treating depression
  • Outcome: Controversial and debatable conclusions

1: Am J Health Syst Pharm. 2004 May 15;61(10):1059-63.

Full Text

Library Article

Role of the placebo effect in evaluating antidepressant efficacy.

Carls KA.

Pharmacy Department, Fairview University Medical Center, 2450 Riverside Avenue South, Minneapolis, MN 55454, USA.

Publication Types:

  • Clinical Trial
  • Randomized Controlled Trial

PMID: 15160783 [PubMed - indexed for MEDLINE]

2: Harv Ment Health Lett. 2003 Mar;19(9):1-5.

Full Text

Are antidepressants placebos? A persistent controversy could affect the choice of psychiatric symptoms.

[Author: Miller, MC]

In mid-2002, Irving Kirsch and his colleagues published a meta-analysis of placebo-controlled clinical trials of antidepressants reported to the FDA. The report covered six drugs: fluoxetine (Prozac, others), paroxetine (Paxil), sertraline (Zoloft), venlafaxine (Effexor), nefazodone (Serzone), and citalopram (Celexa). The controversial and widely reported result was that as much as 80% of the positive response to these medications may be a placebo effect. The authors concluded that the popular drugs "may have no meaningful pharmacological effect at all."

Response to an antidepressant is often measured by the Hamilton Rating Scale for Depression (HAM-D), a 17- or 21-item scale that is completed by a mental health professional. The items cover mood, thoughts, and other symptoms of depressive illness, such as sleep and appetite disturbance. The Kirsch study showed an average drug-placebo difference of less than 2 points on the HAM-D.

The authors grant that this difference, though small, was too great to be attributed to chance alone, but they label its clinical (practical) significance as "dubious." This assertion has provoked fierce rebuttals from mainstream pharmacology researchers.

PMID: 12654576 [PubMed - indexed for MEDLINE]