In abstract P150, “Differences in Real-World Cardiovascular Event Rates Between Atorvastatin and Simvastatin Among New Users When LDL-Lowering Is Controlled,” by Willke et al, which published with the Abstracts From the 47th Annual Conference on Cardiovascular Disease Epidemiology and Prevention in the February 27, 2007, issue of the journal (Circulation. 2007;115:e254), an error was made.
In subsequent re-analysis of the data while responding to referee comments, the authors found that the primary result presented in the abstract was incorrect because of the inadvertent use of observation periods of inappropriate length for some patients. This issue created a bias in favor of atorvastatin use. When corrected, the primary result no longer shows a statistically significant cardiovascular event rate reduction for atorvastatin beyond that seen with simvastatin. The authors regret this error.
The secondary analysis end point, which is now included, serves 2 purposes: It is a correction of an erroneous result featured in the poster that was presented at the conference, and it highlights that the time frame of the analysis made a significant difference in the results.
This error has been corrected in the current online version of the abstract.