In Reply: Dr Ridker and colleagues discuss differential effects in subgroups. Due to the inherent limitations of study-level meta-analyses, we did not attempt to define the clinical benefit and risk of aspirin therapy in particularly high-risk subgroups such as persons with advanced age or diabetes. We agree that age is a risk factor for cardiovascular events and would expect that more events would be prevented by an effective treatment in a higher risk population. Within any population of human subjects, even those enrolled in a single clinical trial, there is a range of benefit and risk associated with any intervention, as pointed out by the Women's Health Study. The NNT and NNH are only averages of responses of patients across the spectrum of risk and benefit and should always be interpreted as such.
Berger JS, Brown DL. Use of Aspirin as Primary Prevention of Cardiovascular Events—Reply. JAMA. 2006;296(4):391–392. doi:10.1001/jama.296.4.391-c