In an effort to assess sampling techniques, diagnostic accuracy, and variations in nomenclature used for cervical lesions, a comprehensive quality assurance system for gynecologic cytology has been instituted at the University of Utah Medical Center. A formal review is performed on all cases with poor correlation, ie, a difference of ±2 grades of severity in the cytohistologic diagnosis, discrepancy between negative and mild dysplasia, and discrepancy between carcinoma in situ and invasive carcinoma. This system allows one to identify problems such as sampling errors due to inadequate endocervical components, and cytologic and histologic misinterpretations, especially those involving condylomatous lesions and invasive carcinoma. It also provides a mechanism for continual clinico-pathologic review and serves as an additional educational device.