A cost-benefit analysis of voluntary routine HIV-antibody testing for hospital patients

The objectives of this paper are threefold: first, to develop a taxonomy of potential benefits from voluntary, routine HIV-antibody testing of hospital patients; second, to inform attending healthcare workers, hospital patients and policy makers of the potential benefits from such testing; and third, to make inferences about whether such testing is warranted by a cost-benefit criterion. Benefits stemming from information about a patient's HIV serostatus accrue to: (1) healthcare workers if extra precautions reduce their HIV-exposure rate; (2) the patient if knowledge of HIV serostatus allows life-extending prophylactic treatment to be initiated; and (3) the patient's sex partners if the patient is less likely to transmit HIV after undergoing HIV testing. Using recent estimates on the value of life, hospital-specific HIV-prevalence rates, the effectiveness of prophylactic treatment, rates of HIV exposure and conversion by healthcare workers, and reduction in high-risk sexual behaviors by seropositive patients, we estimate the benefits of testing as the value of statistical life saved. The opportunity cost of HIV testing is calculated as the reported cost of a standard HIV-test protocol with pre- and post-test counseling. Information about a patient's HIV serostatus provides small expected benefits to healthcare workers ($3.34) because the risk of HIV transmission is small; benefits to attending healthcare workers are insufficient to warrant routine HIV testing on a stand-alone basis even in high HIV-prevalence hospitals. However, an HIV-seropositive test result provides large expected benefits to the patient ($11,202) and to the patient's sex partners ($5271). Adding these nonrivalrous benefits, we find that routine, voluntary HIV-testing of hospital patients passes a cost-benefit test even in low HIV-prevalence hospitals. Four major qualifications of the cost-benefit analysis should be considered. (1) The benefits to some parties can only be achieved if the patient's serostatus is disclosed to them. (2) The net benefits may be negative if HIV-testing induces riskier behavior. (3) The analysis does not incorporate the significant potential for third-parties (employers, insurers, healthcare workers) to use the test to impose costs on HIV-seropositive patients. (4) The sample of inpatients choosing voluntary HIV testing may not be representative of the overall hospital population. These unmeasured factors suggest that policymakers should exercise caution in implementing a voluntary HIV-testing program.