Recent studies have demonstrated the utility of measuring subsets of CD8+ T cells as prognostic markers in epidemiology cohort studies of human immunodeficiency virus (HIV)-infected patients. Most of these studies evaluating the value of CD8+ T-cell subsets have been performed at single centers, and few data are available on variability in the measurement of the CD8+ cell populations in multicenter trials. In the current study, we addressed this question by evaluating interlaboratory variability from the five laboratories enrolled in the Women and Infants Transmission Study sponsored by the National Institutes of Health. This study evaluated 35 HIV-positive and 28 HIV-negative proficiency testing samples sent to the laboratories for evaluation. The study focused on the robust coefficient of variation (RCV) for CD38 (11%), HLA-DR (21%), and CD57 (15%) expression on the CD8+ population. Data from the current study indicated that the variability in these measurements is greater than that for CD3+ CD4+ (RCV, 5%) and CD3+ CD8+ (RCV, 5%) cells. Knowledge of the variability of the CD8+ subset measurements should guide investigators in the design and analysis of clinical trials and epidemiology studies. Ability to obtain improved interlaboratory agreement on CD8+ subset measurements will facilitate further evaluation of these markers in HIV studies.