To evaluate the effectiveness of the AutoPap System in detecting abnormal and normal cervical smears when used in a primary screening/quality control mode, as compared with currently established laboratory practices. Slides were obtained prospectively and were initially processed in the routine fashion with cytotechnologist screening followed by 10% random quality control rescreening. Slides were then processed on the AutoPap System and allocated into the following groups: (1) approximately 25% of the lowest-ranking slides were placed in the laboratory's archives as within normal limits; (2) the remaining approximately 75% of slides were subjected to manual screening. Approximately 15% of the highest-ranking slides in this group underwent quality control rescreening. For each slide needing manual screening, the cytotechnologist was supplied with a report giving the ranking score of that slide. All discrepant slides for either adequacy or diagnosis were subjected to a truth-determination process. The results obtained from the two arms of the protocol were then compared. The AutoPap System-assisted arm of the study was superior to the current practice arm for the identification of abnormal slides at the level of atypical squamous cells of undetermined significance and above (ASCUS+), low grade squamous intraepithelial lesion (LSIL) and higher LSIL+. AutoPap System-assisted practice was equivalent to current practice for the identification of unsatisfactory and satisfactory but limited by slides. All results showed statistical significance. In addition, AutoPap System-assisted practice in the study indicated improved specificity of diagnosis. AutoPap System-assisted practice shows superior sensitivity and specificity when compared to current practice. Its clinical use as a primary screening device should improve the overall practice of cervical cytology as well as provide potential enhancement in overall laboratory productivity.