Outcome No. of participants (studies) | Relative effect (95% CI) | Anticipated absolute effects (95% CI) | Certainty | What happens | ||
VKA | DOACs | Difference | ||||
Thromboembolic events No. of participants: 719 | RR 1.69 (1.09 to 2.62) | 8.1% | 13.8% (8.9 to 21.3) | 5.6% more (0.7 more to 13.2 more) | ⨁◯◯◯ Very low* | DOACs may increase the occurrence of thromboembolic events but the evidence is very uncertain. |
Major bleeding No. of participants: 691 | RR 1.22 (0.72 to 2.07) | 6.6% | 8.0% (4.7 to 13.6) | 1.4% more (1.8 fewer to 7 more) | ⨁◯◯◯ Very low*† | DOACs may increase the occurrence of major bleeding but the evidence is very uncertain. |
All bleeding events No. of participants: 457 | RR 0.79 (0.47 to 1.32) | 32.5% | 25.7% (15.3 to 42.9) | 6.8% fewer (17.2 fewer to 10.4 more) | ⨁◯◯◯ Very low*‡§** | DOACs may decrease the occurrence of all bleeding events but the evidence is very uncertain. |
Mortality No. of participants: 577 | RR 1.17 (0.48 to 2.84) | 2.7% | 3.2% (1.3 to 7.7) | 0.5% more (1.4 fewer to 5 more) | ⨁◯◯◯ Very low*†† | The effect of DOACs on mortality is very uncertain. |
*Three RCTs classified as having some concerns and three cohort studies at serious risk of bias.
†RR 1.22 (95% CI 0.72 to 2.07).
‡I²=66%.
§Mostly no direct comparison.
¶RR 0.79 (95% CI 0.47 to 1.32).
**RR 1.17 (95% CI 0.48 to 2.84).
††The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
DOACs, direct oral anticoagulants; RR, risk ratio; VKA, vitamin K antagonists.