Gynecologic Consensus Conference Topic Group 3 GYN Workload

Gynecologic Consensus Conference Topic Group 3: GYN Workload Limits June 4, 2011 cap. org v. 1

Background • Although the issue of workload was not included in the original survey questions, this workgroup recognizes that workload is an area of cytology laboratory quality that warrants further discussion. • Evaluation of individual workload assessments, as well as laboratory daily screening limits, have been questioned. © 2011 College of American Pathologists. All rights reserved. 2

Background • Workload Limits mandated by CLIA were established to prevent excessive workloads, which could decrease sensitivity and result in false negative results. • Current literature suggests that workload limits should be reevaluated, based on daily screening sensitivity and laboratory abnormal rates. © 2011 College of American Pathologists. All rights reserved. 3

Background • The American Society of Cytopathology (ASC) Productivity and Quality Assurance in the Era of Automated Screening Task Force has compiled preliminary recommendations for imaged guided screening workloads • This committee is presenting for discussion: o A summary of the ASC preliminary recommendations for automated screening o Items for consideration for manual (non-imaged guided) screening © 2011 College of American Pathologists. All rights reserved. 4

Discussion: Preliminary ASC Recommendation #1 • Cytotechnologists’ screening time should not exceed 7 hours in an 8 -hour work shift, provided there are no additional duties or distractions, and should not exceed 7 hours in a 24 hour day. o Based on the ARTISTIC Trial and other literature indicating sensitivity decreases below 95% after 5 hours of screening in a work day. o In comparison, US studies show a sensitivity of 80 -85% when screening for an entire work day. Kitchener HC, et al. Lancet Oncol 2009 Jul; 10(7): 672 -82. Ellis K, et al. Diagn Cytopathol 2010 Nov 2. Wilbur DC, et al. Am J Clin Pathol 2009 Nov; 132(5): 767 -75. Biscotti CV, et al. Am J Clin Pathol 2005; 123: 281 -7. © 2011 College of American Pathologists. All rights reserved. 5

Voting Recommendation #1: Cytotechnologists’ screening time should not exceed 7 hours in an 8 -hour work shift, provided there are no additional duties or distractions, and should not exceed 7 hours in a 24 hour day. 61. Do you agree with recommendation #1? A. Yes 86. 67% B. No 11. 67% C. Other 1. 67% © 2011 College of American Pathologists. All rights reserved. 6

Discussion: Preliminary ASC Recommendation #2 • Workload limits should not be extrapolated to 8 hrs of screening, rather it should take into account screening time plus necessary mini breaks. o Workload limits set for the FDA TIS trial and BD Focal. Point GS were based on extrapolated data o Literature suggests that these limits are too high and sensitivity decreases after 4 hours for most cytotechnologists Thin. Prep Imaging System. Package Insert Becton Dickinson Co. BD Focal. Point GS Imaging System Package Insert. Biscotti CV, et al. Am J Clin Pathol 2005; 123: 281 -7 Elsheikh T, et al. Cancer Cytopathol 2010; 118: 75 -82 © 2011 College of American Pathologists. All rights reserved. 7

Voting Recommendation #2: Workload limits should not be extrapolated to 8 hrs of screening, rather it should take into account screening time plus necessary mini breaks. 62. Do you agree with recommendation #2? A. Yes 96. 67% B. No 3. 33% © 2011 College of American Pathologists. All rights reserved. 8

Discussion: Preliminary ASC Recommendation #3 • Although the False Negative Rate (FNR) is not an ideal measurement, it should be calculated at the ASCUS (not LSIL) threshold, if utilized by the laboratory. o No studies have used the FNR as a successful primary measure of screening sensitivity. o Rare studies show the results of FNR calculated at the threshold of ASCUS correlate with other measures associated with sensitivity. o Performance of screening for LSIL may not be a good surrogate for screening of other more significant lesions Elsheikh T, et al. Cancer Cytopathol 2010; 118: 75 -82 Renshaw AA, et al. Arch Pathol Lab Med 2006; 130: 332 -6. Renshaw AA, et al. Arch Pathol Lab Med 2005; 129: 733 -5. Renshaw AA, et al. Arch Pathol Lab Med 2006; 130: 456 -9. Renshaw AA, et al. Arch Pathol Lab Med 2005; 129: 23 -5 © 2011 College of American Pathologists. All rights reserved. 9

Voting Recommendation #3: Although the False Negative Rate (FNR) is not an ideal measurement, it should be calculated at the ASCUS (not LSIL) threshold, if utilized by the laboratory. 63. Do you agree with recommendation #3? A. Yes 68. 52% B. No 22. 22% C. Other 9. 26% © 2011 College of American Pathologists. All rights reserved. 10

Discussion: Preliminary ASC Recommendation #4 • The average laboratory cytotechnologist productivity should not exceed: 80 slides/day using the new FDA/CLIA/CAP recommendations for calculating workload (imaged slide only = 0. 5 slide, full manual review = 1. 0 slide, imaged + manual review = 1. 5 slide) Or 100 slides/day, where all examined slides (imaged or not) are counted as 1. 0 slides o Studies that document the move from manual to automated screening show a consistent increase in ECA sensitivity when workload is restricted to these values, and a consistent decrease in ECA sensitivity when workload exceeds these limits. o Studies examining ECA rate when cytotechnologists increase their workload over a relatively short time period show a decrease in ECA rate above these limits. Renshaw AA, et al. Diagn Cytopathol 2010 Oct 13 Elsheikh T, et al. Cancer Cytopathol 2010; 118: 75 -82. Levi A, et al. Cancer Cytopathol 2010; 118: 307. © 2011 College of American Pathologists. All rights reserved. 11

Voting Recommendation #4: Do you: A. Agree with statement as is B. Agree with FDA only C. Agree with 100/day only D. Disagree with the whole statement © 2011 College of American Pathologists. All rights reserved. 32. 61% 26. 09% 8. 70% 12

Discussion: Preliminary ASC Recommendation #5 • The percentage of imaged slides that undergo full manual review should be at least either 15%, or twice (2 x) the epithelial cell abnormality (ECA) rate, whichever is greater. o Recommendation based on survey data presented at the 2009 ASC Annual Scientific meeting. The majority of laboratories manually review at least 15% of cases. Miller FN. What is a Reasonable and Realistic Workload Limit in the Age of Image-Assisted Pap Testing? ASC Annual Scientific Meeting, Nov 16, 2009 © 2011 College of American Pathologists. All rights reserved. 13

Voting Recommendation #5: The percentage of imaged slides that undergo full manual review should be at least either 15%, or twice (2 x) the epithelial cell abnormality (ECA) rate, whichever is greater. 65. Do you agree with recommendation #5? A. Yes 72. 41 B. No 22. 41% C. Other 5. 17% © 2011 College of American Pathologists. All rights reserved. 14

Discussion: Preliminary ASC Recommendation #6 • The Epithelial Cell Abnormality (ECA)-adjusted workload measure is a promising new method for calculating and monitoring cytotechnologist workload, and can be correlated to desired laboratory screening sensitivity. This method may be especially useful in the setting of higher ECA rates. Further studies of this method are encouraged before full endorsement of this method, by this task force, is issued. o ECA-adjusted workload is the only measure that has been directly correlated with screening sensitivity. o ECA-adjusted workload has been shown to directly correlate with screening sensitivity at the level of ASCUS, LSIL, and HSIL in the FDA TIS trial and when comparing manual and automated screening to assess the workload cut off in Recommendation #4 Renshaw AA, Elsheikh TM. Diagn Cytopathol 2010 Nov 22. Renshaw AA, et al. Diagn Cytopathol 2010 Oct 13 Elsheikh T, et al. Cancer Cytopathol 2010; 118: 75 -82. Levi A, et al. Cancer Cytopathol 2010; 118: 307. © 2011 College of American Pathologists. All rights reserved. 15

Voting 66. Should the ECA adjusted workload measure be further investigated A. Yes B. No 87. 27% 12. 73% © 2011 College of American Pathologists. All rights reserved. 16

Discussion: Preliminary ASC Recommendation #7 • Cytotechnologist productivity and workload limits are just one aspect of a good quality assurance program in a cytology laboratory. Other quality indicators to assess cytotechnologist performance (screening not interpretation) should be implemented. • The other quality indicators, many of which of will be discussed and voted on today, are also important components in determining screening limits. © 2011 College of American Pathologists. All rights reserved. 17

Discussion: Preliminary ASC Recommendation #7 • Although there are studies documenting the utility of the recommended measures for general QA, few correlate them with screening sensitivity • No current studies correlate these QA measures with workload. Renshaw AA, et al. Diagn Cytopathol 2010(38): 180 -3. Renshaw AA, et al. Cytojournal 2009; 6: 19 © 2011 College of American Pathologists. All rights reserved. 18

Voting 67. Should QA and workload correlation be further investigated? A. Yes B. No 94. 55% 5. 45% © 2011 College of American Pathologists. All rights reserved. 19

Discussion: Additional recommendations for manual workload • Anecdotal evidence suggests that most cytotechnologists cannot achieve adequate sensitivity at workloads set at the upper limit of the CLIA workload limit. • Many of the suggested recommendations for imaged -guided workload are also appropriate for manual screening: o Recommendation #1 o Recommendation #2 o Recommendation #3 o Recommendation #6 o Recommendation #7 © 2011 College of American Pathologists. All rights reserved. 20