Reader Comments

Post a new comment on this article

These findings depend on the use of the standardised mean difference

Posted by plosmedicine on 31 Mar 2009 at 00:22 GMT

Author: PJ Leonard
Submitted Date: March 03, 2008
Published Date: March 4, 2008
This comment was originally posted as a “Reader Response” on the publication date indicated above. All Reader Responses are now available as comments.

Although it is admirable that Kirsch et al have been able to obtain the complete pre-licencing data for these antidepressants, their results, and thus conclusions, are dependent on the use of a specific statistical measure, the standardised mean difference (SMD).

If the weighted mean difference is used (an equally, or more valid approach given that all studies utilised the same outcome measure, namely the HRSD) effect sizes expressed in HRSD scores are larger than reported in this study (2.8 vs 1.8), and paroxetine and venlafaxine reach the NICE criteria for 'clinical significance' (HRSD change > 3):


Similarly, regression on baseline HRSD scores against HRSD change scores reaches 'clinical significance' at lower scores than found by Kirsch et al (26 vs 28) and minimal decrease in placebo responsiveness with increasing baseline severity:


It is unfortunate that this study has received so much uncritical media coverage when the results are more nuanced than they first appear.

Competing interests declared: I declare that I have no competing interests