Tarik M. Elsheikh

BACKGROUND: With the current and projected shortage of a cytotechnologist (CT) workforce and the desire to reduce laboratory costs, increased productivity with automated assisted primary screening has become an attractive option for many laboratories. To the best of the authors' knowledge, longitudinal studies examining the effect of increasing workload on the performance of individual CTs have not been performed previously.METHODS: Using the ThinPrep imaging system (TIS), the performance of 3 CTs with variable levels of experience were evaluated. Their productivity was noted to increase from an average of 87 to 118 slides per day. The analysis included comparisons of error rates, screening rates, and screening times, including a review of 22 fields of view (FOV). Poststudy interviews of the CTs were also performed.RESULTS: Increased workload was found to be proportional to the decreased percentage of cases that underwent full manual review (25.2% to 20.1%; P < .001), and decreased actual screening times (7.3 hours/day to 6.7 hours/day, and 5.0 minutes/slideto 3.7 minutes/slide). This resulted in a lower detection of total abnormal findings (10.4% to 8.3%; P < .001), atypical squamous cells (6.7% to 4.9%; P < .001), and high‒grade squamous intraepithelial lesion (0.9 %to 0.7%; P = .37), as well as an increased false‐negative fraction rate (3.8% to 7.0%; P = .08).CONCLUSIONS: The results of the current study indicate that an increased average CT workload >100 slides per day with the TIS appears to have been accomplished mostly through a reduction in the amount of time spent reviewing the 22 FOV and the percentage of cases that underwent full manual review, which resulted in a significantly reduced screening performance. Cancer (Cancer Cytopathol) 2010. © 2010 American Cancer Society.

Major components of good quality assurance (QA) in cytology laboratories are measuring the screening accuracy of cytotechnologists (CTs) and determining appropriate workload limits for them. Currently, the FDA approved workload limits for image assisted Paps, however, are too high. In addition, the CLIA 88 mandated QA measures, by themselves, are insufficient to accurately measure the screening performance of CTs. Although the use of “controls” is fundamental to good practice in the clinical laboratory; this has not been emphasized in gynecologic cytology. In this review, we underscore the importance of using controls, such as monitoring morning and afternoon abnormal rates, in assessing CT performance and assigning reasonable workload limits. Using controls, however, requires significant alteration in the laboratory workflow, and much lower and variable workload limits for CTs, compared to those approved by FDA. Diagn. Cytopathol. 2010;38:772–775.