Click on image to view larger version.

Figure 2 The effects of ramipril on the primary study outcome in key subgroups of women. The data show the consistency of the benefit of ramipril. The dotted line represents the average treatment effect in women. The size of each box is proportional to the number of patients in each individual analysis. The p value is for the subgroup–ramipril interaction term in the Cox proportional hazards model. Similar results were obtained using the test for homogeneity of odds ratios in subgroups. CAD = coronary artery disease; CVD = cardiovascular disease; DM = diabetes mellitus; HRT = hormone replacement therapy; MI = myocardial infarction; PAD = peripheral arterial disease; RR = relative risk; TIA = transient ischemic attack.