Human immunodeficiency virus test evaluation, performance, and use. Proposals to make good tests better

Abstract

Human immunodeficiency virus (HIV) tests are essential for detecting asymptomatic infection and are helpful in confirming the diagnoses of acquired immunodeficiency syndrome-related complex and acquired immunodeficiency syndrome. Nonetheless, many aspects of their use remain controversial, partly because of concerns about test accuracy. This article reviews the scientific basis for the evaluation, performance, and use of the most commonly employed HIV assays. Current test performance could be improved by better standardization of test procedures and institution of mandatory proficiency testing and licensure of clinical laboratories that perform HIV testing. Test utility could be enhanced by sequencing tests more appropriately and by interpreting test results in conjunction with the clinical purpose for which the test is being used and the characteristics of the population under study. Finally, HIV tests should be evaluated in a manner that minimizes spectrum and referral bias and inadequate reference standard confirmation, problems that have affected the evaluation of current tests.

KIE: The authors review the literature on the accuracy of currently employed tests for human immunodeficiency virus (HIV) infection and urge improvements in standardization of procedures, mandatory proficiency testing, and licensing of clinical laboratories doing such tests. They discuss test use in several voluntary and mandatory situations; attendant problems of anonymity, confidentiality, and informed consent; and the need to consider ethical principles in defining, quantifying, and balancing the risks and benefits to the parties involved.