Sensitivity analyses
| Pooled estimate of the relative risk, based on all studies (95% CI) (number of effect estimates) | Without studies at greater risk of information bias (exposure) | Without studies at greater risk of information bias (outcome) | Without studies at greater risk of confounding | Without studies at greater risk of bias in at least one domain | |
|---|---|---|---|---|---|
| Coronary heart disease | 1.29 (1.04 to 1.59) (n=11) | 1.34 (1.03 to 1.74) (n=9) | 1.28 (1.01 to 1.63) (n=10) | 1.34 (1.03 to 1.76) (n=7) | 1.42 (1.00 to 2.01) (n=7) |
| Stroke | 1.32 (1.04 to 1.68) (n=8) | 1.42 (1.09 to 1.85) (n=7) | 1.30 (0.98 to 1.71) (n=4) | 1.34 (1.05 to 1.73) (n=6) | 1.30 (0.98 to 1.71) (n=4) |