eTable 1.Ovid search terms modified for MEDLINE, EMBASEand CENTRAL (up to September 10, 2015)

Search steps / Search terms
1 / atrial fibrillation.mp. or heart atrium fibrillation/
2 / atrial flutter.mp. or heart atrium flutter/
3 / 1 or 2
4 / (dabigatran or BIBR1048 or BIBR-1048 or "BIBR 1048").mp.
5 / (rivaroxaban or "BAY 59 7939" or "BAY 59-7939" or "BAY 597939" or BAY59-7939 or BAY597939).mp.
6 / (apixaban or BMS-562247 or BMS562247 or "BMS 562247").mp.
7 / (edoxaban or DU-176b or DU176b or "DU 176b").mp.
8 / 4 or 5 or 6 or 7
9 / ("vitamin K antagonis$" or "antivitamin K").tw.
10 / warfarin.mp. or Warfarin/
11 / 9 or 10
12 / 3 and 8 and 11
13 / (propensity adj5 score).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword]
14 / propensity.mp. or exp Propensity Score/
15 / (random$ or placebo$ or single blind$ or double blind$ or triple blind$).ti,ab.
16 / 13 or 14 or 15
17 / 12 and 16

eFigure1. Funnel plot for stroke or systemic embolism in the PS studies

eFigure2. Funnel plot for major bleeding in the PS studies

eFigure3. Funnel plot for stroke or systemic embolism in the RCTs

eFigure4. Funnel plot for major bleeding in the RCTs

eTable2. Summary of findings for NOACs compared with warfarin for stroke or systemic embolism and major bleeding from the PS studies and RCTs in patients with atrial fibrillation

Patient or population:Patients with nonvalvular atrial fibrillation (AF)
Settings:Multicenter, multinational
Intervention: NOACs
Comparison: Warfarin
Outcomes / Illustrative comparative risks* (95% CI) / Relative effect
(95% CI) / No of Participants
(studies) / Quality of the evidence
(GRADE) / Comments
Assumed risk
Warfarin / Corresponding risk
NOACs
Results of the PS studies
Stroke or systemic embolism
Follow-up: median 7 months / High1 / High1 / 0.882
(0.83- 0.94) / 484,452
(8 studies) / ⊕⊕⊕⊕
High / -
Major bleeding
Follow-up: median 7 months / Not estimable / Not estimable / 0.912
(0.79 - 1.05) / 493,856
(9 studies) / ⊕⊕⊕⊝3
Moderate / -
Results of the RCTs
Stroke or systemic embolism
Follow-up: median 1.8 years / High1 / High1 / 0.792
(0.72 - 0.87) / 59,912
(5 studies) / ⊕⊕⊕⊕
High / -
Major bleeding
Follow-up: median 1.8 years / Not estimable / Not estimable / 0.852
(0.73- 1.00) / 59,912
(5 studies) / ⊕⊕⊕⊝3
Moderate / -
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; PS: propensity score; RCT: randomized controlled trial; NOACs: new oral anticoagulants;
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1 Risk of stroke or systemic embolism is relatively high in patients with atrial fibrillation;

2 Hazard ratio (HR) was used for treatment effect estimate;

3 Unexplained heterogeneity

1