Objective: We sought to explore the patient-specific cost-effectiveness in a community-based sample for a therapy for which the average cost-effectiveness in a clinical trial has been well-described.
Study design and setting: Based on a validated multivariate model, we generated predictions of the effectiveness and cost-effectiveness of t-PA compared to streptokinase on 921 consecutive patients who received thrombolytic therapy for acute myocardial infarction.
Results: The average cost-effectiveness of t-PA was US dollar 40,140 per life-year saved. For the quartile of patients most likely to benefit, the incremental cost-effectiveness of t-PA was US dollar 15,396. However, only 44% of patients who received thrombolytic therapy had an estimated cost-effectiveness ratio below US dollar 50,000 per year of life saved; the ratio was greater than US dollar 100,000 in 37% of treated patients. Patients in the lowest quartile of expected benefit are, overall, more likely to be harmed than to benefit from t-PA.
Conclusion: Compared to the pattern of thrombolytic agent choice observed, targeting t-PA to the half of patients most likely to benefit could save 247 lives and US dollar 174 million nationally per year.