Printer Friendly

The role of monitoring interpretive rates, concordance between cytotechnologist and pathologist interpretations before sign-out, and turnaround time in gynecologic cytology quality assurance: findings from the College of American Pathologists Gynecologic Cytopathology Quality Consensus Conference Working Group 1.

Monitoring interpretive rate categories for Papanicolaou tests (Pap tests), concordance of cytotechnologist and pathologist interpretations before sign-out, and turnaround time are 3 categories of monitors that may be useful for quality assurance in gynecologic cytology. These quality assurance measures were evaluated as part of a larger project by the College of American Pathologists (CAP) that surveyed a wide range of quality assurance monitors in gynecologic cytology. The monitoring of interpretive rates is mandated by the Clinical Laboratory Improvement Amendments of 1988 (CLIA '88) and by accrediting agencies, including the CAP. Most frequently, such rates are monitored for the entire laboratory and for cytotechnologists, and less frequently for pathologists. Determining what rates should be monitored, who should be monitored, how frequently, and how such monitoring should occur within the laboratory were issues addressed in this study.

Cytotechnologists and pathologists work in tandem to identify potentially precancerous and cancerous cells on Pap tests. Disagreements may arise between the cytotechnologist and pathologist as to the presence of abnormal cells, or the degree of abnormality identified on the Pap test. The manner in which these disagreements are handled will potentially impact patient care. Tracking such disagreements between cytotechnologists and pathologists may be used as a quality metric. Objectives for quality assurance guidelines are to determine if and how concordance/ discordance of cytotechnologist and pathologist interpretations should be used as a metric. Other questions are how and when to adjudicate discrepancies between cytotechnologists and pathologists, and what types of cases should be shown to a third party before sign-out of the Pap test by the pathologist.

Turnaround time (TAT) for Pap tests is a readily quantified measure in the cytology laboratory. However, the effectiveness of TAT as a quality metric is debatable. Turnaround time may relate more to customer satisfaction or to laboratory staffing than to laboratory quality. Monitoring TAT may have a negative impact if its use causes undue pressure on cytotechnologists, resulting in rushing Pap test screening. If and how monitoring TAT should be part of a quality assurance plan in the cytology laboratory is at issue.

METHODS

The CAP, with support from the Centers for Disease Control and Prevention (CDC), conducted a national survey of quality assurance practices in gynecologic cytology. This Web-based survey was developed by the 3 senior authors of the CAP project with input from national organizations (CAP Cytopathology Resource Committee, the American Society of Cytopathology, the American Society for Cytotechnology, and the American Society of Clinical Pathology), and CDC colleagues. The survey was distributed to all laboratories that participate in gynecologic cytology proficiency testing in the United States, and more than 540 survey responses were received. For details of the complete process of this study, including the development of the survey, enhanced Web-based input, and culmination in a consensus conference, see the introductory article. (1) In short, expert cytopathologists and cytotechnologists were recruited to become part of 5 working groups that studied the survey data on different aspects of quality assurance. These working groups added follow-up questions to the survey, which were available online and elicited additional opinions. Evaluating the data and follow-up questions, together with a review of the literature, the working groups developed a series of preliminary statements on good laboratory practices in cytology quality assurance and presented these at a consensus conference in Rosemont, Illinois, on June 4, 2011. Participants in the conference included working group members, representatives from national cytopathology and cytotechnology organizations, Centers for Medicare and Medicaid Services (CMS), the CDC, and individuals who accepted invitations after completing the written survey. Representatives from the working groups presented their draft statements to the audience participants who voted electronically on the issues. Some statements received a clear consensus from the audience, some had clearly no consensus, and for others, consensus was questionable. The stratified consensus responses were categorized as agreement": 70% to 79%; moderately strong agreement": 80% to 89%; strong agreement": 90% to 98%; and nearly complete agreement": 99% to 100%. The voting resulted in a number of consensus good laboratory practice statements. Limitations of the process include the small number of participants in the consensus conference, and the basis largely on expert opinion, rather than on evidenced-based practices.

RESULTS

Monitoring Interpretive Rates

The monitoring of interpretive rates is common practice among the laboratories surveyed, although which rates are monitored and the frequency of monitoring varies (Tables 1 and 2). Some interpretive rates are either mandated by CLIA regulations and/or required as a component of laboratory certification by deemed organizations, especially the CAP. Theoretically, some interpretive rates, such as the percentage of cancer, are population-specific indices that are not expected to vary over time unless there is a shift in the demographics of the patient population served. And, as different laboratories serve populations of women with different demographics, interpretative rates would be expected to vary among laboratories. For example, CAP benchmarking data from 2006 showed that the rate of low-grade squamous intraepithelial lesion (LSIL) on liquid-based Pap tests varied among laboratories from a low of 1.1% at the fifth percentile to more than 7% at the 95th percentile. Rates for high-grade squamous intraepithelial lesion (HSIL) varied more than 20-fold from 0.1% to 2%. (2)

That being said, cervicovaginal cytology does have a subjective component to interpretations, and thus interpretive rates could vary as a reflection of thresholds for specific interpretations, such as between atypical squamous cells of undetermined significance (ASCUS) and LSIL, for example. In this regard, the monitoring of interpretive rates might be useful for intralaboratory monitoring of consistency, which cytotechnologists and pathologists use in applying interpretative criteria. Published scientific data concerning the usefulness of the application of the monitoring of specific diagnostic rates to an effective quality assurance program, however, is quite scant, and thus this area is wide open for future research exploration. That which might seem logical or intuitive on the surface may not always bear out after subjection to scientific scrutiny. For example, the limited published data concerning the correlation of overall abnormal pickup rates and false-negative fraction (3) showed no correlation. Findings such as this are important to consider when making choices as to which metrics to use and for which purposes.

Survey Results.--Another example would be that of regular monitoring of categories of abnormalities subject to poor interobserver and intraobserver variability, such as ASCUS. Such monitoring makes sense in regard to an individual laboratory's reproducibility of results, and the subsequent trust that clinicians may put in such reports. Indeed, the monitoring of ASCUS was reported by 83.9% of 528 laboratories (Table 1), the highest rate for any of the diagnostic categories, indicating a great degree of consensus that ASCUS is an important diagnostic rate to monitor.

ASCUS is also one of the categories that is most subject to diagnostic drift" and variability among both pathologists and cytotechnologists. Because the ASCUS to squamous intraepithelial lesion ratio (ASCUS:SIL ratio) has much less variability than the ASCUS rate, this monitor is more useful for interlaboratory comparisons, such as in benchmarking. (4) Indeed, in our written survey, the ASCUS:SIL ratio was the only interpretive rate opined to be very useful by more than half of respondents (53.5 %) (Table 3). A study correlating cytotechnologists' ASCUS:SIL ratios with sensitivity (5) showed that the mean screening sensitivity for cytotechnologists with ASCUS:SIL ratios less than 1.5 was significantly less than that for cytotechnologists whose ASCUS:SIL ratio was more than 3.0. The authors also suggested that an ASCUS:SIL ratio less than 1.5 for a cytotechnologist might be useful as a surrogate marker for inadequate screening sensitivity. Note that this study compared ASCUS:SIL ratios of individual cytotechnologists before final interpretation of cases by pathologists, and thus higher ratios are expected. Indeed, the 2006 CAP benchmarking data (2) showed that the average laboratory ASCUS: SIL ratio was about 1.5. However, too much emphasis on a low ASCUS:SIL ratio in a laboratory could potentially have a negative impact on laboratory sensitivity. A recent clinical trial showed that several laboratories with ASCUS:SIL ratios of 1.5 or less had less than 50% adjusted screening sensitivity for cervical intraepithelial neoplasia grade 2+. (6) Therefore, benchmarking data without data correlating it to quality outcomes can be deceptively reassuring.

Interestingly, 41.9% of written survey respondents reported that other undesignated diagnostic rates, not listed on the survey, were useful to them in monitoring quality (Table 3). This obviously requires more study.

The most common interval for monitoring of diagnostic rates was monthly (Table 2). As diagnostic rates for entities other than ASCUS tend to be rather stable over time unless there is a shift in demographics, the monthly frequency of monitoring probably reflects a logistic convenience, as many laboratories collect the raw data for calculating error rates on a monthly basis. For rates with expected fluctuations, such as ASCUS or ASCUS:SIL, monthly monitoring might make more sense. However, such frequent monitoring may pose a problem for small laboratories with extremely small volumes, such as 500 or less cases per year, as their rates may fluctuate widely just by statistical chance. Also, comparing their rates to published benchmarks may also be misleading for the same reason. Some laboratories use a rolling 12 months of data collection, which can be analyzed monthly, quarterly, or at some other interval, based upon the suitability to individual laboratories. This is an area that needs more research.

Follow-up Questions Posted on Internet Site.--The detailed comments from the follow-up online questions were interesting. Only 56% of 87 respondents reported ever seeing a shift in an actively monitored diagnostic rate laboratory-wide. A similar percentage reported ever seeing a shift in an actively monitored individual diagnostic rate, but the reasons attributed to the changes were different. The most common factors (mentioned in 25% of the comments each) attributed to the shift were a change in technique/ methodology, such as adding imaging or switching to a different liquid-based method and personnel receiving feedback about recently missed cases, review of diagnostic criteria, and others. For shifts in individual diagnostic rates, the common factor attributed was the individual receiving feedback or participating in open discussion, supporting the intuitive importance of providing feedback to individuals. Common but less frequently cited reasons were a change in personnel (pathologists or cytotechnologists) and implementation of human papillomavirus (HPV) testing or review of HPV results.

When asked why participants found that monitoring ASCUS was helpful, most alluded to the fact that ASCUS is the clinical decision point, and an indicator of the threshold of abnormality. Participants felt that monitoring ASCUS rates prevents overcalling and undercalling and helps to keep interpretations within the laboratory uniform. However, the follow-up online questions revealed that of 87 respondents, 71% and 68% reported that ASCUS rates varied greatly both among cytotechnologists and pathologists, respectively. This seems to challenge the impression that merely monitoring the ASCUS rates would necessarily produce uniformity. One respondent mentioned that cytotechnologists who miss ASCUS also tend to miss HSIL, an observation that is supported by the literature. (3)

When asked which interpretive rates were felt to not be useful, most comments included relatively rare but significant interpretations such as atypical glandular cells, atypical squamous cells, cannot exclude high-grade squamous intraepithelial lesion (ASC-H), and cancer. This issue was discussed in detail at the conference, and the suggestion was made to combine these categories into a monitor that could be followed even in low-volume laboratories. However, 43.5% of attendees at the consensus conference felt that this was not useful. Only future research will clarify this newly identified controversy.

Another category felt to be problematic, and one documented now to lead to many unnecessary procedures, was normal endometrial cells in women older than 40 years. This was also a topic of discussion at the conference. Scientific data show that the only women with normal-appearing endometrial cells who may require endometrial sampling or ultrasound assessment are those who are postmenopausal or having clinically abnormal bleeding. (7)

From the follow-up online questions, most reporting laboratories (95%) monitor the total abnormal rate. Eighty-two percent monitor both laboratory and individual rates. More than half of the respondents (58%) felt that cytotechnologists and pathologists should have access to their diagnostic rates. Such feedback was felt to be helpful because it provided "peer pressure" to outliers, making counseling easier when needed; helped people know how they were doing in relation to others in the laboratory and to CAP benchmarks; improved education and sharing of cases; helped to refine skills and cytologic criteria; and helped with accuracy and managing of diagnostic thresholds.

Fifty-nine percent of online respondents felt that each person's individual interpretive rates should be shared confidentially with him or her. Several commented that it was helpful to openly publish/display interpretive rates within the laboratory, but that people should not be individually identifiable. A few commented that the sharing of interpretive rates could actually be a negative, depending upon how they were used, for example, as a basis for merit raises. While there are correlations between some rates and performance, such as the correlation between ASCUS:SIL ratios and sensitivity cited above, such correlations are not absolute; therefore, it is important to use more than 1 metric. This once again brings up the importance of tying benchmarks to clinically significant metrics such as error rates or sensitivities for significant lesions.

Consensus Good Laboratory Practice Statements and Comments.--Table 4 lists the good laboratory practice statements generated from the survey, the Internet follow-up questions, and opinions of the authors for monitoring interpretive rates in gynecologic cytology. These statements were voted on at the consensus conference, and in cases where there was no clear initial consensus, the statements were revised and voted on again.

Concordance of Cytotechnologist and Pathologist Interpretations

Tracking discrepancies between cytotechnologists and pathologists is incorporated into the evaluation of individual performance as mandated by CLIA. However, the methods for monitoring and analyzing discrepancies are not specified. The CAP checklist requires documentation of each individual's diagnostic discrepancies and corrective actions taken (CYP.07660). The CAP checklist also requires comparison of individual cytotechnologist interpretation to the final diagnosis in gynecologic specimens signed out as abnormal, as part of the 6-month workload assessment (CYP 0.08575). Most laboratories incorporate some evaluation of discrepancy analysis as part of the every-6-month workload review, and this has been advocated as a possible quality metric. (8-10) Establishing a baseline level of discrepancies is important so that trends can be analyzed. Individual discrepancy rates can be compared to the laboratory rate, and individual rates can be tracked over time. Laboratories also need to determine what level of discrepancy is important to track and whether upgraded or downgraded interpretations, or both, need to be monitored. Laboratories should also consider formalizing procedures and policies to adjudicate discrepant interpretations of Pap tests between cytotechnologists and pathologists before sign-out of the Pap test. Adjudication of discrepancies may be challenging in laboratories with 1 pathologist and cytotechnologist or in laboratories with only a pathologist.

Survey Results.--From the survey, greater than 73% of laboratories activelymonitor the rates at which a pathologist upgrades a cytotechnologist's diagnosis at the time of initial sign-out, and greater than 62% of laboratories actively monitor the rates at which a pathologist downgrades a cytotechnologist's diagnosis at the time of initial sign-out (Table 5). The most critical upgrades monitored are from negative for intraepithelial lesion or malignancy (NILM) to either LSIL or to HSIL or greater, monitored respectively by 97.9% and 96.6% of the 381 laboratories that responded to this survey question (Table 6). Upgrades from NILM to ASCUS and to ASC-H are also frequently monitored by 74.8% and 81.9% of laboratories, respectively, as are upgrades from ASCUS to HSIL, monitored by 83.2% of laboratories. Monitoring upgrades from LSIL to HSIL and ASC-H to HSIL was not as frequent, with only 65.9% and 57.0% of laboratories, respectively, following these rates.

The most frequent and most important downgrades monitored by laboratories are from HSIL to NILM and LSIL to NILM, each monitored by more than 94% of the 320 laboratories that responded to this question in the survey (Table 7). Monitoring of downgrades from either ASC-H or ASCUS to NILM is followed respectively by 80.6% and 72.5% of laboratories, and monitoring downgrades of HSIL to either LSIL or to ASC-H is followed respectively by 64.4% and 60.6% of laboratories.

In adjudicating discrepancies between cytotechnologists' and pathologists' diagnoses, only 49.5% of laboratories have a written policy specifying the process to resolve 2-grade discrepancies, and only 25.2% have such a written policy in cases of 1-grade discrepancies (Table 8). For 2-grade discrepancies, the Pap test in question is frequently shown by the pathologist to a second person before sign-out: 32.2% to the original cytotechnologist, 24.7% to another pathologist, and 4.5% to another cytotechnologist. In only 33.7% of laboratories that responded did the pathologist diagnosis stand without further action. By contrast, in the case of 1-grade discrepancies, such as from LSIL to ASCUS, most laboratories (68.5%) responded that the pathologist's diagnosis stands and only 27.6% responded that the case was shown to a second person.

Follow-up Questions Posted on Internet Site.--Seven additional questions were posted on a Web site in an attempt to supplement the written survey questions. The number of responses, ranging from 62 to 87, was low when compared to the number of responders to the written survey. From the responses, most pathologists (60%) seek additional review of particular cases before downgrading a cytotechnologist diagnosis (Table 9). The most frequent cases that elicit second review are HSIL (78%), atypical glandular cells (54%), and ASC-H (41%). Similarly, 37 of 61 cytotechnologists indicated that they seek additional review before forwarding a case to a pathologist (Table 10). The most frequent case shown by a cytotechnologist is a Pap test interpreted as atypical glandular cells.

Forty-nine percent of pathologists indicated that they did not routinely confirm an abnormal Pap test result by showing it to another individual before sign-out (Table 11). Of those pathologists that did confirm a diagnosis, this was most frequently done at the discretion of the pathologist and not for any specific interpretive category. Frequently, Pap tests not interpreted as abnormal or reactive are routinely reviewed by pathologists before sign-out (Table 12). The most frequent Pap test in this category is one containing herpes, shown by 81% of responders. Unsatisfactory Pap tests and those with benign endometrial cells follow those with herpes at 59% and 44%, respectively. Written responses were solicited in the "other" category, chosen by 29%. Many of these comments indicated that endometrial cells identified in women either older than 40 years or older than 50 years, or the presence of Actinomyces, were Pap tests that were frequently shown: (45% and 35%, respectively).

Consensus Good Laboratory Practice Statements and Comments.--Table 13 lists the good laboratory practices generated by data from the written survey, the Internet discussion site, and opinion of the authors for monitoring concordance of cytotechnologist and pathologist interpretations. These were voted on at the consensus conference and in cases where there was no clear consensus, the statements were reworded and resubmitted for voting.

As shown in Table 13, statement 1a, while there was strong support to actively monitor upgrades of cytotechnologist interpretations, particularly of NILM to SIL+, by pathologists, there was not clear consensus on which other cytotechnologist interpretations should also be monitored. The statement was revised and resubmitted.

Cytotechnologists are responsible for screening slides for potential abnormalities. The rate at which interpretations are upgraded was deemed to be an important quality monitor by 79.7% of conference participants (Table 13, statement 1b). However, there was no consensus as to the definition of a significant discrepancy. Some laboratories may choose to monitor any case upgraded from negative to abnormal (ASCUS+) plus upgrades from ASCUS or LSIL to HSIL. Other laboratories may prefer to define upgraded discrepancies more narrowly with normal to SIL+ counted. At the very least, upgrades of NILM to SIL+ should be considered as a monitor.

As shown in Table 13, statement 2, there was less consensus as to whether downgraded interpretations should be monitored. Some laboratories only monitored significant downgrades, while other laboratories had broader definitions. Pathologists are responsible for making the final determination on abnormal cases, based not only on cellular features but also clinical history and management implications. There is an expectation that many specimens interpreted as potentially atypical by cytotechnologists will be downgraded to negative by pathologists, and similarly, there will be some cases in which a minor downgrade is made because there is uncertainty as to the nature of an abnormality. An example is a case downgraded from HSIL to ASC-H when there is uncertainty, in order to prevent potential overtreatment. In the written survey sent to laboratories, many participants agreed that only significant downgrades should be monitored. For example, 96.3% of laboratories monitor downgrades of HSIL to negative, whereas only 60.6% monitor downgrades from HSIL to ASC-H.

Laboratories may address significant discrepancies in interpretations between cytotechnologists and pathologists in a variety of ways. Often the pathologist uses individual discretion in determining which types of cases should be adjudicated by a third person. In the written survey relating to 2-step discrepancies, 33.7% of respondents stated that the pathologist's diagnosis stands without further action, while most either reviewed the case with the cytotechnologist (32.2%) or consulted a second pathologist (24.7%). At the conference, 62.5% strongly agreed that discrepancies of 2 degrees or more should be showed to a third person when possible, and 29.2% agreed with reservations. Since clear consensus was not obtained on which diagnostic categories may benefit from review by a third person (Table 13, statement 3a), the question was restated and resubmitted for a vote (statement 3b). Examples of cases meeting these criteria would be upgrades from NILM to HSIL or downgrades of cases from HSIL+ to negative. Small laboratories with only a single cytotechnologist and pathologist have challenges in adjudicating discrepancies. Some mechanisms for addressing discrepancies in small laboratories include reviewing the case at a multiheaded microscope, correlating the interpretation with later biopsies, and sending select cases out of the laboratory for outside consultation. Obtaining HPV testing in certain situations may be a possibility, but this has to be tempered by the fact that there is a small false-negative rate of HPV testing in HSIL and cancer cases. (11,12) Furthermore, HPV testing should be requested in consultation with the clinician after discussion of possible pitfalls. Regardless of how individual laboratories may handle discrepancies, 73.7% of consensus conference participants agreed that laboratories should have policies about which categories of discrepancies should be reviewed by a third individual before sign-out. These policies will clarify expectations for both cytotechnologists and pathologists and provide more uniform handling of specimens, which may impact significantly on patient care. Policies dealing with 1-grade discrepancies, such as from HSIL to ASC-H, are not as critical as policies handling a 2-grade discrepancy such as from HSIL to NILM. In the former case, there is no or minimal change in patient management, while in the latter, patient management will be different.

Certain types of high-risk specimens may especially benefit from review by a third person; examples are atypical glandular cells and those results, such as HSIL, that will impact on the patient's receiving colposcopy and biopsy. Glandular lesions are problematic and are not infrequently the cause of litigation. The voting at the consensus conference reflects the awareness of difficulties in detecting glandular lesions. Only 27.7% thought it unnecessary to show a premalignant or malignant glandular lesion to a third person, whereas 39.4% thought this was not needed for similarly severe squamous lesions (Table 13, statements 4 and 5).

Laboratories should have policies as to which cases benefit from review by a second person (cytotechnologist or pathologist), even if not required by CLIA. These may include unsatisfactory, endometrial cells in women older than 40 years, glandular cells in women post hysterectomy, and herpes. CLIA requires that all reactive and potentially abnormal cases be reviewed and confirmed by a pathologist. However there is no requirement that certain Bethesda categories be confirmed by a second individual. Some of these types of specimens have important clinical management implications. Examples include unsatisfactory specimens, endometrial cells, and certain types of organisms including herpes simplex virus (Table 13, statement 6a). Greater than 90% of consensus conference participants agreed that laboratories should have policies as to which cases benefit from such review (Table 13, statement 6b). Specimens designated as "unsatisfactory" generally require early repeat according to American Society for Colposcopy and Cervical Pathology management guidelines. (13) Additional review helps to promote intralaboratory reproducibility in application of adequacy criteria. Furthermore, patients who have received chemotherapy or radiation therapy may have lower-cellularity specimens, and there are no data to suggest a minimum numeric threshold; thus, the pathologist may evaluate clinical history and exercise clinical judgment in certain cases designated by the cytotechnologist as unsatisfactory. (13) Endometrial sampling is recommended for postmenopausal women with benign-appearing endometrial cells, while women age 40 years or older who are still having menstrual cycles and are asymptomatic can return to routine screening. (14) In addition, differentiating atypical glandular cells from shed endometrial cells is challenging, and such cases benefit from additional review.

Turnaround Time

Turnaround time has been historically associated with causing pressure on cytotechnologists to increase productivity at the expense of quality. (15) With the implementation of CLIA '88, however, strict regulations have limited the number of slides cytotechnologists are allowed to screen per day and individual workload limits are assessed biannually. Given these quality measures, during the last decade the concept of TAT has evolved to become a readily quantifiable measure in the cytology laboratory and to have an impact on quality. (16) A timely reporting of results translates into timeliness of patient care. However, it may be argued that gynecologic cytology is a screening test, which decreases the sense of urgency for the results.

Turnaround time is generally defined as the time a specimen is accessioned in the laboratory to the time the report is signed out or finalized. However, there is variability among laboratories in defining when the TAT cycle starts and when it ends.

Turnaround time can be considered a reliable indicator for evaluation of laboratory staffing by identifying bottleneck areas in any part of the test cycle. Turnaround time is monitored to examine the functionality of the overall service, including slide preparation, cytotechnologist screening, and pathologist sign-out. Turnaround time also relates to customer satisfaction and may influence decisions about where to refer gynecologic cytology work, especially considering the competitive environment in the current health care arena. (17) In addition, as patients become more knowledgeable about laboratory testing through targeted marketing and other means, the demand for prompt reporting increases, thus influencing the necessity to monitor TAT.

Survey Results.--From the survey, a laboratory's expectations on Pap test TAT varied widely, from 1 to 7 or more days, with a median Pap test TAT of 3 business days (Table 14). Laboratories reported that actual TAT was less than the expected TAT, as shown in the percentile distributions in Table 14. How laboratories define and measure TAT varied; 57.2% of respondents defined the starting point as the date/ time of accessioning the Pap test, while 24.1% used date/ time of specimen receipt, and 14.6% used date/time of specimen collection. There was more universal agreement on the definition of the ending point for TAT measurement, as date/time report finalized (89.1%). Other definitions of ending points included date/time results delivered to physician (4.2%) and date/time case reviewed by technologist or pathologist (4.0%). The most common frequency of TAT monitoring was monthly (46.6%), followed by daily (20.4%) and quarterly (11.6%). The most common metric used to measure TAT variance was percentile distribution within a certain TAT (54.2%), followed by mean TAT (32.4%) and percent of cases deviating from TAT expectation (15.5%) (Table 15).

Follow-up Questions Posted on Internet Site.--Five additional questions were posted on the Web site to attempt to supplement the written survey questions. The number of responses on the Web site (76) was much lower than the responses to the written survey. From the responses, 59% of laboratories agreed that monitoring TAT is an effective quality metric. Monitoring TAT was reported to be useful for monitoring staffing needs in the laboratory, and as a metric for customer service. Thirty-three percent of laboratories felt that monitoring TAT is not an effective quality metric, with several opinions stating that a faster TAT does not equate to quality work. Eight percent of respondents were "unsure" if monitoring TAT is an effective quality metric. There was evidence of some misinformation for the requirement of TAT monitoring, with 61% of respondents believing that TAT monitoring was mandated, primarily by clinical guidelines. Thirty-two percent of respondents report that monitoring TAT negatively affects quality, including causing undue pressure on cytotechnologists to meet screening quotas. Of the 21% who report that monitoring TAT positively affects quality, timely patient care and use as a staffing monitor were given as examples. Forty-seven percent of respondents were "unsure" if TAT constraints negatively or positively affect quality, with many stating TAT did not affect quality, or that TAT could have both positive and negative effects.

Table 16 lists the good laboratory practice statements generated from the written survey, Internet questions and comments, and opinions of the authors, on monitoring Pap test TAT. These statements were voted on at the consensus conference. Most consensus conference participants agreed that TAT should be monitored in gynecologic cytology (Table 16, statement 1). Awareness of TAT should be a consideration in the overall laboratory quality performance, including addressing individual capabilities and limits. However, the use of TAT monitoring should never compromise the quality of the Pap test evaluation in any phase of the cycle. Laboratory directors and managers should be aware of the potential negative implications of monitoring TAT. Thirty-two percent of respondents report that monitoring TAT negatively affects quality, including causing undue pressure on cytotechnologists to meet screening quotas, possibly leading to increase in false-negative rates and screening errors, and possibly leading to increased ASCUS rates.

There was strong agreement among consensus conference participants that we should not attempt to establish a universally acceptable TAT in gynecologic cytology (Table 16, statement 2). A specific TAT for Pap tests is not required as certification criteria for laboratory inspections (18) (CAP, American Society of Cytopathology, CMS). There is no evidence in the literature to support the establishment of a TAT limit for Pap tests. The survey results showed a wide range of laboratory TAT for Pap tests, ranging from 1 to 7 or more days, with a median TAT of 3 business days. Turnaround time is best used as an internal measure to assess the workflow in gynecologic cytology rather than as a benchmark for interlaboratory comparison.

Most consensus conference participants agreed that individual laboratories should determine how to measure or define TAT (Table 16, statement 3). The most common measurement of TAT starts with date/time of accessioning and ends with report sign-out. A minority of laboratories use different definitions of TAT, and this should not pose a problem, since the measurement is largely used for internal assessments. There was strong agreement that individual laboratories should determine the frequency of TAT monitoring and the metric to determine TAT variance (Table 16, statements 4 and 5). The needs for TAT monitoring may vary by differences in individual laboratories, especially given differences in laboratory information systems.

COMMENT

The good laboratory practice statements presented herein have a range of consensus agreement or even disagreement, reflecting differences in opinion and practice patterns of consensus conference attendees. While there are many strengths of this process, there are also shortcomings to the methodology of this process. This process was not a prospective study, but a survey of practice patterns and of opinions from working group members and consensus conference attendees. However, it would be difficult to construct a prospective study of the numerous good practice statements set forth in this survey, and many of the quality monitors in this survey are mandated by regulations from CLIA or accreditation criteria by CAP and other agencies. While literature was reviewed and cited when possible, the literature was not formally evaluated for strength of evidence. Not infrequently, however, there was a dearth of literature on certain quality topics, such as monitoring of concordance between cytotechnologist and pathologist interpretations of Pap tests before sign-out.

The working group authors and consensus conference attendees were sensitive to the weaknesses and strengths of this process. The good laboratory practice statements were often crafted to not be proscriptive given the limitations of this survey process and the vast differences among many cytopathology laboratories. Indeed, the revisions to some of the good laboratory practice statements noted above reflect this struggle. The objective of these laboratory practice statements are not to proscribe a regulated quality assurance program, but rather to frame both areas of agreement and disagreement so that cytopathology laboratories may choose quality metrics, in addition to those mandated by regulations, that may be suited to their particular practice. In cases of metrics proscribed by CLIA regulations, the survey process can highlight methods that may make the use of these metrics more meaningful to the daily operation of the laboratory.

This report was supported in part from a contract (GS-10F-0261K) funded by the Centers for Disease Control and Prevention/ Agency for Toxic Substances and Disease Registry. The authors would like to acknowledge Barbara Blond, MBA, MT(ASCP) for her many contributions to this project.

References

(1.) Tworek JA, Henry MR, Jones BA. College of American Pathologist Consensus Conference on Good Laboratory Practices in Gynecologic Cytology: background, rationale, and organization. Arch Pathol Lab Med. 2013;137(2): 158-163.

(2.) Eversole GM, Moriarty AT, Schwartz MR, et al. Practices of participants in the College of American Pathologists Interlaboratory Comparison Program in Cervicovaginal Cytology 2006. Arch Pathol Lab Med. 2010;134(3):331-335.

(3.) Krieger PK, Naryshkin S. Random rescreening of cytologic smears: a practical and effective component of quality assurance programs in both large and small cytology laboratories. Acta Cytologica. 1994;38(3):291-298.

(4.) Davey DD, Naryshkin S, Nielsen ML, Kline TS. Atypical squamous cells of undetermined significance: interlaboratory comparison and quality assurance monitors. Diag Cytopathol. 1994;11(4):390-396.

(5.) Renshaw AA, Deschenes M, Auger M. ASC/SIL ratio for cytotechnologists: a surrogate marker of screening sensitivity. Am J Clin Pathol. 2009;131(6):776-781.

(6.) Castle PE, Stoler MH, Wright TC Jr, Sharma A, Wright TL, Behrens CM. Performance of carcinogenic human papillomavirus (HPV) testing and HPV16 or HPV18 genotyping for cervical cancer screening of women aged 25 years and older: a subanalysis of the ATHENA study. Lancet Oncol. 2011;12(9):880-890.

(7.) Kapali M, Agaram NP, Dabbs DJ, Kanbour A, White S, Austin RM. Routine endometrial sampling of asymptomatic premenopausal women shedding normal endometrial cells in Pap tests is not cost effective. Cancer Cytopathol. 2007; 111(1):26-33.

(8.) Cibas ES, Dean B, Maffeo N, Allred EN. Quality assurance in gynecologic cytology: the value of cytotechnologist-cytopathologist discrepancy logs. Am J Clin Pathol. 2001;115(4):512-516.

(9.) Voytek TM, Mody DR, Davey DD. Quality assessment and improvement in cytopathology. In: Nakhleh RE, Fitzgibbons PL, eds. Quality Improvement Manual in Anatomic Pathology. Northfield, IL: College of American Pathologists; 2002:115-116.

(10.) Bongiovanni M, De Saussure B, Kumar N, Pache JC, Cibas ES. A quality control study on cytotechnologist-cytopathologist concordance and its relationship to the number of dots on the slide. Acta Cytol. 2009;53(6):653-658.

(11.) Katki HA, Kinney WK, Fetterman B, et al. Cervical cancer risk for women undergoing concurrent testing for human papillomavirus and cervical cytology: a population-based study in routine clinical practice. Lancet Oncol. 2011;12(7): 663-672.

(12.) Kang WD, Kim CH, Cho MK, et al. Comparison of the hybrid capture II assay with the human papillomavirus DNA chip test for the detection of high-grade cervical lesions. Int J Gynecol Cancer. 2009;19(5):924-928.

(13.) Davey DD, Cox JT, Austin RM, et al. Cervical cytology specimen adequacy: patient management guidelines and optimizing specimen collection. J Lower Genital Tract Dis. 2008;12(2):71-81.

(14.) Wright TC, Massad LS, Dunton CJ, et al. 2006 consensus guidelines for the management of women with abnormal cervical cancer screening tests. Am J Obstet Gynecol. 2007;197(4):346-355.

(15.) Bogdanich W. The Pap test misses much cervical cancer through lab's efforts. Wall Street Journal. November 2, 1987:1.

(16.) Jones BA, Valenstein PN, Steindel SJ. Gynecologic cytology turnaround time: a College of American Pathologists Q-Probes Study of 371 laboratories. Arch Pathol Lab Med. 1999;123(8):682-686.

(17.) Bewtra C. Cytology turnaround time: are we being too fast? Diag Cytopathol. 2003:29(5):241-242.

(18.) Yu GH, Gupta PK. The pathologic obsession with turnaround time in gynecologic cytology: is it warranted? Diag Cytopathol. 1998:19(5):321-322.

Karen M. Clary, MD; Diane D. Davey, MD; Sonya Naryshkin, MD; R. Marshall Austin, MD, PhD; Nicole Thomas, CT(ASCP); Beth Anne Chmara, CT(ASCP); Chiara Sugrue, MBA, MS, SCT(ASCP); Joseph Tworek, MD

Accepted for publication May 15, 2012.

From the Department of Pathology, Rochester General Hospital, Rochester, New York (Dr Clary);the Department of Clinical Sciences, University of Central Florida, and Orlando Veterans Administration Medical Center, Orlando, Florida (Dr Davey); the Department of Pathology, Mercy Health System, Janesville, Wisconsin (Dr Naryshkin); the Department of Pathology, Magee-Womens Hospital, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania (Dr Austin); CAP Pap Program (Ms Thomas) and the Department of Cytologic Surveys (Ms Chmara), College of American Pathologists, Northfield, Illinois; the Division of Cytopathology, Hofstra Northshore Long Island Jewish School of Medicine, New Hyde Park, New York (Ms Sugrue); and the Department of Pathology, St Joseph Mercy Hospital, Ann Arbor, Michigan (Dr Tworek).

The authors have no relevant financial interest in the products or companies described in this article.

The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention/Agency for Toxic Substances and Disease Registry and are not intended to take the place of applicable laws or regulations.

Reprints: Karen M. Clary, MD, Department of Pathology, Rochester General Hospital, 1425 Portland Ave, Rochester, NY 14621 (e-mail: karen.clary@rochestergeneral.org).
Table 1. Actively Monitored Diagnostic Rates (N = 528)

Diagnostic Rate           Laboratory            Cytotechnologist

                     Frequency   Percentage   Frequency   Percentage

None                      9          1.7          17          3.2
NILM                    407         77.1         374         70.8
Unsat                   443         83.9         380         72.0
LSIL                    437         82.8         398         75.4
HSIL                    435         82.4         399         75.6
SCC                     383         72.5         337         63.8
Other malignancies      355         67.2         311         58.9
ASCUS                   443         83.9         404         76.5
ASC-H                   398         75.4         359         68.0
AGC                     399         75.6         361         68.4
ASCUS:SIL ratio         424         80.3         324         61.4
NILM:SIL ratio           96         18.2          82         15.5
Other                    90         17.0          75         14.2

Diagnostic Rate            Pathologist

                      Frequency    Percentage

None                      56          10.6
NILM                     169          32.0
Unsat                    182          34.5
LSIL                     208          39.4
HSIL                     207          39.2
SCC                      180          34.1
Other malignancies       161          30.5
ASCUS                    215          40.7
ASC-H                    194          36.7
AGC                      192          36.4
ASCUS:SIL ratio          198          37.5
NILM:SIL ratio            46           8.7
Other                     46           8.7

Abbreviations: AGC, atypical glandular cells; ASC-H, atypical
squamous cells, cannot exclude high-grade squamous intraepithelial
lesion; ASCUS, atypical squamous cells of undetermined
significance; ASCUS:SIL ratio, the ratio of cases with an
interpretation of atypical squamous cells of undetermined
significance to cases with squamous intraepithelial lesions; HSIL,
high-grade squamous intraepithelial lesion; LSIL, low-grade
squamous intraepithelial lesion; NILM, negative for intraepithelial
lesion or malignancy; NILM:SIL ratio, the ratio of cases with an
interpretation of negative for intraepithelial lesion or malignancy
to cases with squamous intraepithelial lesions; SCC, squamous cell
carcinoma; Unsat, unsatisfactory.

Table 2. Monitoring Frequency

How Frequently Are the Diagnostic Rates
Monitored?

                Frequency   Percentage

Daily                9          1.8
Weekly               3          0.6
Monthly            333         64.8
Bimonthly            4          0.8
Quarterly           62         12.1
Semiannually        65         12.6
Annually            32          6.2
Other                6          1.2

Table 3. Usefulness of Diagnostic Rates in Monitoring Quality

Diagnostic      N    Average    Very    4, %   3, %   2, %    Not
Rate                  Rank     Useful                        Useful
                                5, %                          1, %

NILM           439     3.6      33.3    21.6   24.1   11.8     9.1
Unsat          492     4.0      44.3    23.8   20.3    7.1     4.5
LSIL           490     4.0      41.8    26.9   22.9    5.7     2.7
HSIL           490     4.1      46.5    24.1   22.2    4.5     2.7
SCC            432     3.7      36.6    21.8   23.4   10.9     7.4
Other          409     3.6      35.7    20.3   24.9   10.3     8.8
  malignancy
ASCUS          492     4.2      51.0    23.8   18.1    5.3     1.8
ASC-H          459     4.0      44.2    24.4   21.4    6.5     3.5
AGC            451     3.8      33.3    28.8   23.9    8.9     5.1
ASCUS:SIL      465     4.2      53.5    23.0   16.1    5.2     2.2
  ratio
NILM-SIL       132     3.5      33.3    18.9   23.5   11.4    12.9
  ratio
Other           86     3.5      41.9    11.6   17.4   12.8    16.3

Abbreviations: AGC, atypical glandular cells; ASC-H, atypical
squamous cells, cannot exclude high-grade squamous intraepithelial
lesion; ASCUS, atypical squamous cells of undetermined significance;
ASCUS:SIL ratio, the ratio of cases with an interpretation of
atypical squamous cells of undetermined significance to cases with
squamous intraepithelial lesions; HSIL, high-grade squamous
intraepithelial lesion; LSIL, low-grade squamous intraepithelial
lesion; NILM, negative for intraepithelial lesion or malignancy;
NILM:SIL ratio, the ratio of cases with an interpretation of
negative for intraepithelial lesion or malignancy to cases with
squamous intraepithelial lesions; SCC, squamous cell carcinoma;
Unsat, unsatisfactory. Rankings based on a scale of 1 to 5: not
useful to very useful.

Table 4. Consensus Good Laboratory Practice
Statements: Monitoring Interpretive Rates

1. Monitoring of interpretive rates for all Bethesda System
categories is potentially useful, as each Bethesda System
category is clinically relevant.

Do you agree with the consensus statement?

Yes: 94.3%

No: 5.7%

Should standard categories of interpretive rates be
monitored in all laboratories?

Yes: 85.1%

No: 6.3%

Should each individual laboratory choose which
interpretive rates to monitor?

Yes: 20.7%

No: 79.3%

2. It is most useful to monitor interpretive rates for
cytotechnologists individually and in comparison for
the entire laboratory.

Do you agree with the consensus statement?

Yes: 100%

3. It is currently unclear whether or not monitoring
interpretive rates for individual pathologists beyond
laboratory rates as a whole is useful.

Is monitoring interpretive rates of individual pathologists
useful to you?

Yes: 85.7%

No: 12.9%

Other: 1.4%

Is this an area that should be explored?

Yes: 90.5%

No: 3.2%

Other: 6.4%

4. Consider monitoring combined interpretive rates of
"dangerous abnormals," defined as cancer, suggestive
of cancer, HSIL, AGC, and ASC-H.

Do you think that using the combined category
"dangerous abnormals" could be useful?

Yes, in low-volume/low-prevalence laboratories
only: 15.9%

Yes, in any laboratory: 34.8%

No: 43.5%

Don't know: 5.8%

5. Monthly monitoring of interpretive rates may be
useful, if possible.

Is monthly monitoring:

Too frequent: 43.1%

Not frequent enough: 1.7%

Just right: 55.2%

5a. Revised statement: Regular monitoring of interpretive
rates may be useful and the individual laboratory should
determine the frequency of monitoring.
Do you agree?

Yes: 98.2%

No: 1.9%

6. Providing feedback of interpretive rates is important.
Should individual interpretive statistics be provided to
cytotechnologists and pathologists as feedback?

Yes, regularly: 88%

No, not at all: 1%

Only as part of scheduled employee reviews: 11%

Abbreviations: ACG, atypical glandular cells; ASC-H, atypical squamous
cells, cannot exclude high-grade squamous intraepithelial lesion;
HSIL, high-grade squamous intraepithelial lesion.

Table 5. Monitoring Change in Diagnosis

                     Frequency   Percentage

The rates at which a pathologist upgrades a
  cytotechnologist's diagnosis at the time of
  initial sign-out are actively monitored
  per cytotechnologist:

Yes                     376         73.3
No                      137         26.7

The rates at which a pathologist downgrades
  a cytotechnologist's diagnosis at the time
  of initial sign-out are actively monitored
  per cytotechnologist:

Yes                     312         62.5
No                      187         37.5

Table 6. Upgrade Rates Monitored for Cytotechnologists (N = 381)

Cytotechnologist                   Pathologist Diagnosis
Diagnosis

                            ASCUS                ASC-H           LSIL

                    Frequency   Percentage   Freq   Percentage   Freq

NILM                   285         74.8      312       81.9      373
ASCUS                  ...         ...       179       47.0      216
LSIL                   ...         ...       169       44.4      ...
ASC-H                  ...         ...       ...       ...       ...

Cytotechnologist        Pathologist Diagnosis
Diagnosis

                       LSIL       HSIL or Greater

                    Percentage   Freq    Percentage

NILM                   97.9      368        96.6
ASCUS                  56.7      317        83.2
LSIL                   ...       251        65.9
ASC-H                  ...       217        57.0

Abbreviations: ASC-H, atypical squamous cells, cannot exclude
high-grade squamous intraepithelial lesion; ASCUS, atypical squamous
cells of undetermined significance; Freq, frequency; HSIL, high-grade
squamous intraepithelial lesion; LSIL, low-grade squamous
intraepithelial lesion; NILM, negative for intraepithelial lesion or
malignancy.

Table 7. Downgrade Rates Monitored for Cytotechnologists (N = 320)

Cytotechnologist          Pathologist Diagnosis
Diagnosis

                        NILM                ASCUS

                   Freq   Percentage   Freq   Percentage

ASCUS              232       72.5      ...       ...
ASC-H              258       80.6      158       49.4
LSIL               302       94.4      189       59.1
HSIL or greater    308       96.3      274       85.6

Cytotechnologist         Pathologist Diagnosis
Diagnosis

                       ASC-H               LSIL

                   Freq   Percentage   Freq   Percentage

ASCUS              ...       ...       ...       ...
ASC-H              ...       ...       ...       ...
LSIL               ...       ...       ...       ...
HSIL or greater    194       60.6      206       64.4

Abbreviations: ASC-H, atypical squamous cells, cannot exclude
high-grade squamous intraepithelial lesion; ASCUS, atypical
squamous cells of undetermined significance; Freq, frequency;
HSIL, high-grade squamous intraepithelial lesion; LSIL,
low-grade squamous intraepithelial lesion; NILM, negative for
intraepithelial lesion or malignancy.

Table 8. Additional Monitoring Characteristics

                                                Frequency   Percentage

In cases of a 2-grade or greater discrepant
  diagnosis between a pathologist and a
  cytotechnologist, how is the discrepancy
  most commonly resolved?

  The pathologist's diagnosis stands without       172         33.7
    further action
  By reviewing the Papanicolaou test with          164         32.2
    original cytotechnologist
  Discrepancies are shown to a second              126         24.7
    pathologist
  Other                                             24          4.7
  Discrepancies are shown to a second               23          4.5
    cytotechnologist
  By HPV testing                                     1          0.2

There is a written laboratory policy
  specifying the process for resolution of
  a 2-grade discrepancy:

  Yes                                              252         49.5
  No                                               257         50.5

In cases of a 1-grade or greater discrepant
  diagnosis between a pathologist and a
  cytotechnologist, how is the discrepancy
  most commonly resolved?

  The pathologist's diagnosis stands               350         68.5
    without further action
  By reviewing the Papanicolaou test with           94         18.4
    original cytotechnologist
  Discrepancies are shown to a second               37          7.2
    pathologist
  Other                                             18          3.5
  Discrepancies are shown to a second               10          2.0
    cytotechnologist
  By HPV testing                                     2          0.4

There is a written laboratory policy
  specifying the process for resolution of
  a 1-grade discrepancy:

  Yes                                              129         25.2
  No                                               383         74.8

Abbreviation: HPV, human papillomavirus.

Table 9. For Pathologists, Are There Particular
Cases for Which You Seek Additional Review Before
Downgrading a Cytotechnologist Diagnosis?

                     Frequency   Percentage

Yes                     37           60
No                      18           29
Unsure                   7           11

Total                   62          100

If Yes, for Which of the Following Diagnoses?
  (Check All That Apply)

ASCUS                   10           27
ASC-H                   15           41
LSIL                     7           19
HSIL                    29           78
AGC                     20           54

Abbreviations: AGC, atypical glandular cells of
undetermined significance; ASC-H, atypical
squamous cells, cannot exclude high-grade squamous
intraepithelial lesion; ASCUS, atypical squamous
cells of undetermined significance; HSIL,
high-grade squamous intraepithelial lesion; LSIL,
low-grade squamous intraepithelial lesion.

Table 10. For Cytotechnologists, Are There
Particular Cases for Which You Seek Additional
Review Before Forwarding a Case on to a Pathologist?

                         Frequency   Percentage

Yes                         37           61
No                          22           36
Unsure                       2            3

Total                       61          100

If Yes, for Which of the Following Cases?
  (Check All That Apply)

ASCUS                       16           46
ASC-H                       15           43
LSIL                         2            6
HSIL                         9           26
AGC                         20           57

Abbreviations: AGC, atypical glandular cells of
undetermined significance; ASC-H, atypical
squamous cells, cannot exclude high-grade squamous
intraepithelial lesion; ASCUS, atypical squamous
cells of undetermined significance; HSIL,
high-grade squamous intraepithelial lesion; LSIL,
low-grade squamous intraepithelial lesion.

Table 11. Are Abnormal Interpretations Confirmed by
Showing the Papanicolaou Test Slide to Another
Individual Before Final Sign-Out?

                               Frequency   Percentage

Yes                               33           49
No                                33           49
Unsure                             1           1

Total                             67          100

If Yes, for Which of the Following Diagnoses?
  (Check All That Apply)

ASCUS                              9           22
ASC-H                             11           28
LSIL                               4           10
HSIL                              12           30
AGC                               12           30
Individual cases at the           25           62
  discretion of the
  pathologist only

Abbreviations: AGC, atypical glandular cells of
undetermined significance; ASC-H, atypical squamous
cells, cannot exclude high-grade squamous
intraepithelial lesion; ASCUS, atypical squamous
cells of undetermined significance; HSIL, high-grade
squamous intraepithelial lesion; LSIL, low-grade
squamous intraepithelial lesion.

Table 12. For Which Papanicolaou Test Findings
(Other Than Abnormal and Reactive) Do You Require
Pathologist Review?

                            Frequency   Percentage

Benign endometrial cells       31           44
Unsatisfactory                 41           59
Obscuring factors              10           14
Herpes                         57           81
Individual cases at the        16           23
  discretion of the
  pathologist only
Other glandular                25           36
  processes
Other                          20           29

Table 13. Consensus Good Laboratory Practice Statements: Monitoring
Concordance of Cytotechnologist and Pathologist Interpretations

                                                          Percentage

1a. Actively monitor rates at which a pathologist
  upgrades cytotechnologist interpretations before
  sign-out.

  A. Agree, NILM to SIL+ (negative to SIL or higher)          25.8
  B. Agree, NILM to SIL+, also ASCUS to HSIL                  25.8
  C. Agree, A and B plus NILM to ASCUS                        17.7
  D. Agree, any upgrades to abnormal plus LSIL or             29.0
    ASC-H to HSIL
  E. Do not monitor upgrades                                   1.6

1b. Revised statement: Actively monitor rates at
  which a pathologist upgrades cytotechnologist
  interpretations before sign-out. Definition of
  upgrades should be determined by the laboratory.
  Do you agree?

  A. Yes                                                      79.7
  B. No                                                       15.3
  C. Other                                                     3.4
  D. Other                                                     1.7

2. Actively monitor rates at which a pathologist
  downgrades cytotechnologist interpretations before
  sign- out. Do you:

  A. Agree, HSIL+ or LSIL to NILM only                        65.2
  B. Agree, ASCUS+ (all abnormal) to NILM                     10.1
  C. Agree: all abnormal to NILM, and HSIL to ASC-H           14.5
    or LSIL
  D. Do not monitor downgrades                                10.1

3a. Show discrepancies of 2 degrees or more to a
  third person when possible. Do you:

  A. Strongly agree                                           62.5
  B. Agree with reservations                                  29.2
  C. Disagree                                                  8.3

3b. Revised statement: Laboratories should have
  policies about which categories of discrepancies
  should be reviewed by a third individual before
  sign-out. Do you agree?

  A. Yes                                                      73.7
  B. No                                                       22.8
  C. Other                                                     3.5

4. Some cases benefit from review by a third person
  even if not upgraded/downgraded (squamous). Which
  cases benefit from third-person review (squamous)?

  A. ASC-H and greater                                        19.7
  B. HSIL and greater                                         18.3
  C. Squamous cell carcinoma only                             22.5
  D. Not necessary                                            39.4

5. Some cases benefit from review by a third person
  even if not upgraded/downgraded (glandular). Which
  cases benefit from third-person review (glandular)?

  A. Both atypical glandular cells and adenocarcinoma         55.4
  B. Adenocarcinoma only                                      15.4
  C. Not necessary                                            27.7
  D. Other                                                     1.5

6a. Some cases benefit from routine review by a
  second person even if CLIA does not require
  confirmation by a pathologist. Which cases
  benefit from routine review by second person
  (cytotechnologist or pathologist) even if not
  required by CLIA?

  A. Herpes                                                    4.0
  B. Endometrial cells in women >40 y                            0
  C. Glandular cells in women post hysterectomy                2.7
  D. B and C only (glandular processes)                       16.2
  E. All examples (A, B, C)                                   59.5
  F. Not necessary routinely (only at the discretion          17.6
    of the screener)

6b. Revised statement: Laboratories should have
  policies as to which cases benefit from review by
  a second person (cytotechnologist or pathologist),
  even if not required by CLIA. These may include:

  --Unsatisfactory
  --Endometrial cells in women >40 y
  --Glandular cells in women post hysterectomy
  --Herpes

Do you agree?
                                                              90.7
  A. Yes                                                       9.3
  B. No

Abbreviations: ASC-H, atypical squamous cells, cannot exclude
high-grade squamous intraepithelial lesion; ASCUS, atypical squamous
cells of undetermined significance; CLIA, Clinical Laboratory
Improvement Amendments of 1988; HSIL, high-grade squamous
intraepithelial lesion; LSIL, low-grade squamous intraepithelial
lesion; NILM, negative for intraepithelial lesion or malignancy; SIL,
squamous intraepithelial lesion.

Table 14. Turnaround Time (TAT) Expectations

                       Frequency   Percentage

Laboratory's expected Papanicolaou test TAT
  in business days

  1                        33          7.7
  2                        90         21.0
  3-4                     122         28.4
  5-6                     123         28.7
  >7                       61         14.2

Laboratory's median Papanicolaou test TAT
  in business days

  1                        60         15.2
  2                       109         27.5
  3-4                     163         41.2
  5-6                      40         10.1
  >7                       24          6.1

Table 15. Turnaround Time (TAT) Monitoring

                                           Frequency   Percentage

Which metric is used to measure TAT variance? (N = 491) (a)

  Percentile distribution within a            266         54.2
    certain TAT
  Mean TAT                                    159         32.4
  Percentage of cases deviating from           76         15.5
    TAT expectation/standard
  Median TAT                                   61         12.4
  Average length of deviation from             39          7.9
    TAT expectation/standard
  Other                                        36          7.3

How frequently does the laboratory monitor Papanicolaou
  test TAT?

  Daily                                       104         20.4
  Weekly                                       40          7.9
  Monthly                                     237         46.6
  Bimonthly                                     4           .8
  Quarterly                                    59         11.6
  Semiannually                                 13          2.6
  Annually                                     23          4.5
  Other                                        29          5.7

When does the clock start ticking for the TAT measurement?

  Date/time accessioned                       297         57.2
  Date/time received for processing           125         24.1
  Date/time collected                          76         14.6
  Other                                         8          1.5
  Date/time ordered by provider                 7          1.3
  Date/time received for screening              5          1.0
  Date/time results submitted for               1           .2
    reporting

(a) Multiple responses were allowed.

Table 16. Consensus Good Laboratory Practice
Statements: Turnaround Time (TAT)
in Gynecologic Cytology

1. Turnaround time should be monitored in gynecologic
cytology. Do you agree?

A. Yes: 80.3%
B. No: 19.7%

2. We should not attempt to establish a universally acceptable
TAT in gynecologic cytology. Do you agree?

A. Yes: 90.1%
B. No: 9.9%

3. Individual laboratories should determine how to measure/
define TAT. Do you agree?

A. Yes: 74.3%
B. No: 25.7%

4. Individual laboratories should determine the frequency
of TAT monitoring. Do you agree?

A. Yes: 98.6%
B. No: 1.4%

5. Individual laboratories should determine the metric used to
measure TAT variance. Do you agree?

A. Yes: 91.3%
B. No: 8.7%
COPYRIGHT 2013 College of American Pathologists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Clary, Karen M.; Davey, Diane D.; Naryshkin, Sonya; Austin, R. Marshall; Thomas, Nicole; Chmara, Bet
Publication:Archives of Pathology & Laboratory Medicine
Article Type:Report
Geographic Code:1USA
Date:Feb 1, 2013
Words:9074
Previous Article:General quality practices in gynecologic cytopathology: findings from the College of American Pathologists Gynecologic Cytopathology Quality...
Next Article:Role of napsin A and TTF1 as a diagnostic marker for lung adenocarcinoma.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters