Net clinical benefit of antiplatelet therapy was affected by patient preferences: A personalized benefit-risk assessment.

OBJECTIVE To assess the effect of patient preferences on the net clinical benefit (NCB) of an antiplatelet therapy for the secondary prevention of cardiovascular complications. STUDY DESIGN AND SETTING Risk equations were developed to estimate the individual predicted risk of key outcomes of antiplatelet treatment in patients with a prior myocardial infarction using the Clinical Practice Research Datalink linked to the Hospital Episode Statistics and UK Office of National Statistics databases. Patient preferences for outcomes of antiplatelet therapies were elicited in a separate discrete choice experiment survey. Trial hazard ratios, relative to placebo, were used to calculate the per-patient NCB using equal or preference weighting of outcomes. RESULTS Risk equations were estimated using 31,941 adults in the Clinical Practice Research Datalink population, of which 22,125 were included in the benefit-risk assessment. The mean NCB was lower in the preference-weighted than in the equal-weighted analysis (0.040 vs 0.057; p < 0.0001), but the direction of effect was unchanged by the weighting. In analyses stratified by the presence of bleeding risk factors, including preference weighting altered the ranking of subgroups by NCB. CONCLUSION Patient preference weighting may have a significant effect on NCB and should be included in personalized benefit-risk assessments.