Magnitude of benefit according to inherent stroke risk in non-valvar atrial fibrillation
| Primary prevention | Secondary prevention | ||||
|---|---|---|---|---|---|
| Low | Moderate | High | |||
| Intrinsic rate (no treatment) | 1%/year | 3.5%/year | 6%/year | 12%/year | |
| Number of strokes prevented/year/1000 treated | |||||
| Adjusted dose warfarinv none | 6 | 21 | 36 | 84 | |
| Aspirinv none | 2 | 7 | 12 | 12 | |
| Warfarinv aspirin | 4 | 14 | 24 | 72 | |
General estimates to illustrate the effect of risk stratification on the magnitude of treatment effects. Relative risk reductions are 60% for warfarin v none; 20% for aspirin v none; and 40% for warfarin v aspirin for primary prevention with corresponding reductions of 70%, 10%, and 60% for secondary prevention.1