Background: Because most strokes cause neurological impairment rather than death, stroke prophylaxis may improve quality of life more than length of life. Thus, an understanding of how stroke and stroke prophylaxis affect quality of life is central to clinical decision making for many patients. Methods: We elicited quality-of-life estimates, known as utilities, for 3 degrees of severity of anticipated stroke-mild, moderate, and major-and for stroke prophylaxis with either warfarin sodium or aspirin therapy. We used the time tradeoff and standard gamble methods to elicit these utilities from 83 patients who had atrial fibrillation. Results: Seventy patients completed the interview successfully. Their utilities for stroke ranged from worse than death (<0) to as good as current health (1.0). The median utilities for mild, moderate, and major stroke were 0.94, 0.07, and 0.0, respectively. Although the median utilities decreased with increasing severity of stroke (P<.001), there was high interpatient variability within each degree of stroke severity. For example, 7 subjects (10%) rated a major stroke above 0.5, while 58 subjects (83%) rated it as equal to or worse than death. In contrast to the stroke utilities, the median utilities for warfarin and aspirin therapy were high-0.997 and 1.0, respectively. However, the interpatient variability for warfarin therapy was also important: 11 patients (16%) with atrial fibrillation rated the utility of warfarin therapy so low that their quality- adjusted life expectancy would be greater with aspirin. Conclusion: Patients' utilities for stroke prophylaxis and anticipated stroke vary substantially. Many patients view the quality of life with major stroke as tantamount to or worse than death. These findings highlight the relevance of incorporating patient preferences when choosing stroke prophylaxis.