The impact of expected HIV transmission rates on the effectiveness and cost of ruling out HIV infection in infants.

Abstract

To quantify the costs and effectiveness of different strategies for ruling out HIV infection in infants born to HIV-infected mothers in the United States. The authors assessed 4 different testing strategies that incorporated serial HIV DNA polymerase chain reaction (PCR) testing with or without enzyme-linked immunosorbent assay (ELISA) antibody testing. Testing costs, false reassurance rates, and incremental cost-effectiveness ratios were compared for the 4 strategies. In HIV-exposed infants, HIV DNA PCR testing at birth, 1 month, and 4 months of age results in a false reassurance rate of 21 per million (at a 2% transmission rate). Adding an ELISA test lowers the false reassurance rate to 0.052 per million at a cost of $570,000 per additional case detected; adding another PCR lowers the false reassurance rate to 1.49 per million at a cost of $720,000 per additional case detected compared with the 3-PCR strategy. At a high transmission rate (20%), there would be substantially more erroneously negative results (false reassurance rate is 256 per million with PCR testing at birth, 1 month, and 4 months) and consequently more favorable cost-effectiveness ratios with additional testing: $47,000 per additional case detected by adding 1 ELISA test and $59,000 per additional case detected by adding another PCR test. False-negative HIV results after serial testing in exposed infants are rare, and the incremental cost-effectiveness ratios of additional tests are substantial at low transmission rates. However, the false reassurance rate increases considerably with a 3-PCR strategy and additional testing becomes more cost-effective at greater transmission rates; therefore, additional testing may be warranted in infants at greater risk of infection.