UTILITY OF QUANTITATIVE ENZYME-IMMUNOASSAY REACTIVITY FOR PREDICTING HUMAN-IMMUNODEFICIENCY-VIRUS SEROPOSITIVITY IN LOW-PREVALENCE AND HIGH-PREVALENCE POPULATIONS

To assess the utility of quantitative enzyme immunoassay (EIA) reactiv ity for predicting human immunodeficiency vims seropositivity, we eval uated 22,823 serum samples from homo- and bisexual men, heterosexual i ntravenous drug users, and other heterosexuals with initial screening by EIA, retesting of reactive samples in duplicate, and confirmatory W estern blot (immunoblot) testing. Quantitative EIA reactivity was dete rmined by a mean of the optical density ratio of the three assays perf ormed for each reactive specimen. A total of 1,773 samples (7.8%) were repeatedly reactive, and 1,747 (7.7%) were confirmed Western blot pos itive. All 26 EIA-reactive-Western blot-negative samples had low-level EIA reactivity (ratio <2.2), while most (86%) of the Western blot-pos itive samples had high-level reactivity (ratio, >3.0). The positive pr edictive value for samples with moderate-to-high-level EIA reactivity (ratio, >2.2) was 100% for all risk groups. These results support the value of quantitative EIA reactivity in predicting human immunodeficie ncy vims seropositivity and suggest that confirmatory testing of speci mens with high-level reactivity is not necessary in all situations.