Reader Comments

Post a new comment on this article

Flawed analysis

Posted by AdamJacobs on 23 Sep 2010 at 17:10 GMT

These 85 manuscripts became a significant portion of all of the articles published on sertraline. A general Medline search of 1998 to 2000 (performed December 2006) found 479 results with the keyword “sertraline,” and 211 with “sertraline” in the title. This suggests that between 18% and 40% of articles on sertraline in this key period were managed by Pfizer through this one MECC, large enough percentages to have substantial effects on the overall shape of the medical literature on this drug.

The analysis in this paragraph is flawed, as the 85 manuscripts written by CMD represented all the manuscripts they had published, whereas the numbers for the denominator (Medline hits) were not obtained in an analagous way: they are only the articles indexed in Medline. Many journals are not indexed in Medline, so the denominator is probably much larger than the 479 articles suggested here. A quick search of Embase for 1998 to 2000 for articles with the keyword "sertraline" gave 1908 hits, so the difference is likely to be significant.

Competing interests declared: My company provides professional medical writing services to pharmaceutical companies.