Reader Comments
Post a new comment on this article
Post Your Discussion Comment
Please follow our guidelines for comments and review our competing interests policy. Comments that do not conform to our guidelines will be promptly removed and the user account disabled. The following must be avoided:
- Remarks that could be interpreted as allegations of misconduct
- Unsupported assertions or statements
- Inflammatory or insulting language
Thank You!
Thank you for taking the time to flag this posting; we review flagged postings on a regular basis.
closeTwo Sources of Bias
Posted by plosmedicine on 31 Mar 2009 at 00:24 GMT
Author: Robert Waldmann
Position: professor
Institution: Universita di Roma "Tor Vergata"
E-mail: robert.waldmann@gmail.com
Submitted Date: March 20, 2008
Published Date: March 25, 2008
This comment was originally posted as a “Reader Response” on the publication date indicated above. All Reader Responses are now available as comments.
Kirsch et al estimate the average benefit of new-generation antidepressants (NDAs) using a simple standard meta-analysis procedure “We conducted two types of data analysis …and another using each study's drug and placebo groups' arithmetic mean (weighted for the inverse of the variance) as the meta-analytic “effect size” They first calculated an estimate of the overall average change of the Hamilton Rating Scale of Depression (HRSD) for patients who received a NDA as the precision weighted average of the changes reported for the treated subsample in each trial. Then they calculated the average change for patients who received the placebo in the same way and calculated the difference. This approach involves two methodological choices. First the use of precision weights (weights inverse to the reported variance of the estimate of the mean change in each study). Second the decision to calculate overall mean changes for treated and control groups and then take the difference rather than first calculating the difference in the HRSD then calculating a weighted average across trials of those differences. In each case, Kirsch chose a method which, under strong assumptions, gives an efficient and unbiased estimate of the true overall average benefit. In each case there are alternative approaches which are less efficient under those assumptions but which are unbiased not only when the Kirsch et al estimates are unbiased, but also for many cases in which the Kirsch et al estimates are biased. That is they are less efficient under the null but more robust. In each case the null hypothesis that the Kirsch et al estimator is unbiased has been tested and overwhelmingly rejected. The available unbiased estimate of the overall average benefit of NDA’s is equal to 2.65 HRSD units, which is considerably higher than Kirsch et al’s biased estimate.
The proof of these claims would put me over the character limit. It is available here
http://rjwaldmann.blogspo...