Skip to main content
. Author manuscript; available in PMC: 2009 Jul 1.
Published in final edited form as: Am Heart J. 2008 Mar 6;156(1):106–111. doi: 10.1016/j.ahj.2007.12.035

Figure 2.

Figure 2

Mean and 95% confidence limits on CRP are shown stratified by previous cardiovascular symptoms and after multivariate adjustment. Estimates were made in log-space, and results exponentiated for graphic presentation. CRP level are reported in mg/L. P-value for effect modification of aspirin and statins for those without evidence of CVD is 0.25, and for those with a history of CVD is 0.16; however, the three-way interaction was not significant, suggesting the pooled interaction between ASA and statin use of 0.014 is appropriate for effect modification between these factors.