Fig 2. Risk of gastrointestinal bleeding with low-dose aspirin.
Data are shown as adjusted OR, adjusted RR (unless otherwise stated), adjusted IRR, or multivariate HR, plus 95% CIs, for the risk of any gastrointestinal bleeding with low-dose aspirin vs no aspirin. A test for heterogeneity (I2 statistic) is provided. aCase-crossover study; badjusted OR; cadjusted RR; dadjusted IRR; emultivariate HR. CI, confidence interval; HR, hazard ratio; IRR, incidence rate ratio; OR, odds ratio; RR, relative risk.