Cost-effectiveness of expanded human immunodeficiency virus-testing protocols for donated blood.

Abstract

This study was designed to estimate the cost-effectiveness of expanding the human immunodeficiency virus (HIV)-testing protocol for donated blood beyond screening for HIV antibodies to further reduce the risk of HIV transmission through transfusion. A Markov decision analysis model was developed to estimate the cost-effectiveness of HIV antibody testing (at a cost of $5/unit) and of adding to that protocol a second HIV test, either plasma p24 antigen detection or RNA polymerase chain reaction (PCR) (at costs of $5/unit and $8/unit, respectively). Test efficacy was projected from anticipated window-period reductions (6 days for p24 antigen, 11 days for RNA PCR), and donor seroconversion rates were derived from the Retrovirus Epidemiology Donor Study. On the basis of current estimates of HIV prevalence rates in blood donors (1/10,000) and 16 million annual transfusions in the United States HIV antibody testing prevents 1568 cases of transfusion-acquired HIV infection each year at a cost of $3600 per quality-adjusted year of life saved. The addition of p24 antigen testing would prevent 8 more cases at a net additional cost of $60 million annually ($2.3 million/quality-adjusted life year); RNA PCR testing would prevent 16 more cases at a net additional cost of $96 million annually ($2.0 million/ quality-adjusted life year). Although expanding the donor HIV screening protocol with p24 antigen or RNA PCR testing will prevent rare cases of transfusion-associated HIV, the cost-effectiveness of such an addition is predicted to be far below that of most medical interventions. Thus, HIV test protocol additions are unlikely to provide cost-effective improvements to blood safety in the United States.