Figure 2From: Subgroup effects despite homogeneous heterogeneity test results Modified forest plot of a hypothetical meta-analysis of five studies, ordered by the proportion of women in each study. The treatment effects and corresponding 95% confidence intervals of five studies included in a hypothetical meta-analysis are shown. The dashed line (- - - -) indicates no treatment effect. The studies are ordered by the proportion of women that was included in each study. The studies differ on the proportion of included women (0%, 25%, 50%, 75%, and 100%). The hypothetical treatment is effective in women (RR = 0.7), but not in men (RR = 1.0).Back to article page