Example 2 Table. Sensitivity of the Rate Ratio for Cardiovascular Outcome to an Unmeasured Confounder
Prevalence of Unmeasured Binary Confounder in the Exposed Group, % | Prevalence of Unmeasured Binary Confounder in the Comparator Group, % | Unmeasured Binary Confounder Rate Ratio | High Exposure Rate Ratio (95% Cl)* |
90 | 10 | 1.5 | 1.20 (1.01-1.42) |
90 | 50 | 1.5 | 1.43 (1.22-1.67) |
50 | 10 | 1.5 | 1.39 (1.18-1.63) |
90 | 10 | 2 | 0.96 (0.81-1.13) |
90 | 50 | 2 | 1.27 (1.11-1.45) |
50 | 10 | 2 | 1.21 (1.03-1.42) |
90 | 50 | 3 | 1.18 (1.01-1.38) |
50 | 10 | 3 | 0.99 (0.85-1.16) |
90 | 50 | 5 | 1.08 (0.85-1.26) |
Cl, confidence interval.
Adjusted for age, sex, cardiovascular drug use, and unmeasured binary confounder.
Table adapted from Wei et al. [184].