Since 2014, the recommended algorithm for laboratory diagnosis of HIV infection in the United States has consisted of an HIV-1/2 antigen/antibody (Ag/Ab) test followed by an HIV-1/2 antibody (Ab) differentiation test and, if necessary, a diagnostic HIV-1 nucleic acid test (NAT) to resolve discordant or indeterminate results.Using stored specimens from persons seeking HIV testing who had not received a previous diagnosis or treatment, we compared the performance of a three-step alternative algorithm consisting of an Ag/Ab test followed by a quantitative HIV-1 RNA viral load assay and, if viral load is not detected, an Ab differentiation test, to that of the recommended algorithm. We calculated the sensitivity and specificity of five Ag/Ab tests and the proportion of specimens correctly classified by the alternative algorithm compared to the recommended algorithm. Results were examined separately for specimens classified as early infection, established infection, and false-reactive screening RESULTS: Sensitivity and specificity were similar among all Ag/Ab tests. Viral load quantification correctly classified all specimens from early infection, all false-reactive screening specimens, and the majority of specimens from established infection.Although cost, regulatory barriers, test availability, and the ability to differentiate early from established infection must be considered, this alternative algorithm can potentially decrease the total number of tests performed and reduce turnaround time, thereby streamlining HIV diagnosis and initiation of treatment.