Original Article
A Simplified Method for Assessing Cytotechnologist Workload Louis J. Vaickus, MD, PhD and Rosemary Tambouret, MD
BACKGROUND: Examining cytotechnologist workflow and how it relates to job performance and patient safety is important in determining guidelines governing allowable workloads. This report discusses the development of a software tool that significantly simplifies the process of analyzing cytotechnologist workload while simultaneously increasing the quantity and resolution of the data collected. METHODS: The program runs in Microsoft Excel and minimizes manual data entry and data transcription by automating as many tasks as is feasible. RESULTS: Data show the cytotechnologists tested were remarkably consistent in the amount of time it took them to screen a cervical cytology (Gyn) or a nongynecologic cytology (Non-Gyn) case and that this amount of time was directly proportional to the number of slides per case. Namely, the time spent per slide did not differ significantly in Gyn versus Non-Gyn cases (216 6 3.4 seconds and 235 6 24.6 seconds, respectively; P 5.16). There was no significant difference in the amount of time needed to complete a Gyn case between the morning and the evening (314 6 4.7 seconds and 312 6 7.1 seconds; P 5.39), but a significantly increased time spent screening Non-Gyn cases (slide-adjusted) in the afternoon hours (323 6 20.1 seconds and 454 6 67.6 seconds; P 5.027), which was largely the result of significantly increased time spent on prescreening activities such as checking the electronic medical record (62 6 6.9 seconds and 145 6 36 seconds; P 5.006). CONCLUSIONS: This Excel-based data collection tool generates highly detailed data in an unobtrusive manner and is highly customizable to the individual working C 2013 American Cancer Society. environment and clinical climate . Cancer (Cancer Cytopathol) 2014;122:15-22. V
KEY WORDS: workload; efficiency; computer; cytotechnologist; automation.
INTRODUCTION In the current clinical climate, demand for quick turnaround time and accuracy must be balanced with efficient resource allocation both in terms of equipment and supplies and in terms of personnel. Cytotechnologists form the backbone of most cytopathology services, and increasing demands on their time and mental resources must be carefully evaluated and re-evaluated to prevent fatigue, and reduce error rates while maintaining a reasonable quality of life and job satisfaction. Clinical Laboratory Improvement Amendments of 1988 (CLIA 88) legislation placed a limit of 100 on the number of slides that may be screened by a cytotechnologist in an 8-hour day. With the advent of liquidbased cytology preparations and computer-assisted screening, that number was increased to 200 (for ThinPrep slides with only the imager designated Field of View areas being evaluated). However, recent studies have found a decreased sensitivity for significant cytologic abnormalities at the currently permissible levels for cervicovaginal cytology slides. Indeed, the American Society of Cytopathology has issued recommendations that reduce the number of hours the cytotechnologist may screen to 7 per 24-hour period and that limit the number of slides screened per day to no more than 70.1-6
Corresponding author: Louis J. Vaickus, MD, PhD, Department of Pathology, Massachusetts General Hospital, 55 Fruit Street, Warren Building 2, Boston, MA 02114; Fax: (781) 223-5657; [email protected] Department of Pathology, Massachusetts General Hospital, Boston, Massachusetts Received: August 15, 2013; Revised: September 12, 2013; Accepted: September 16, 2013 We acknowledge Andrew Renshaw, MD, for many helpful discussions and advice. Published online November 6, 2013 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/cncy.21364, wileyonlinelibrary.com
Cancer Cytopathology
January 2014
15
Original Article
The studies on which these recommendations were based evaluated only gynecologic (Gyn) cytology slides. Nongynecologic (Non-Gyn) slide screening and the impact of other laboratory activities on the daily workload of the cytotechnologist have not been evaluated, nor has the actual time spent performing the various caserelated tasks. Previous studies of workload have relied on triedand-true methods such as keeping work logs by hand or manually in a computerized spreadsheet to finally be tabulated and analyzed computationally. This procedure adds complication to the technologist’s day and makes data analysis more difficult. In addition, manual recording of data limits the amount of information that can reasonably be tabulated. Reducing the demand on the cytotechnologist’s time during data collection and employing computerized “time stamping” can not only improve compliance but also result in more meaningful data, because a less obtrusive collection system minimizes distraction and workflow disruption while vastly increasing the amount of data that can feasibly be collected. With the advent of new technologies, new methods and new guidelines for cytopathology, the roles of cytotechnologists have gradually changed. Computerization of records and requisitions, although not yet universal, is becoming ever more prevalent as is the utilization of automated systems for generating reports and preventing human error. Furthermore, these changes are occurring at an accelerating rate due to emergence of a technological/ biomedical Moore’s law. Because of this, cytotechnologists’ workload should be reevaluated periodically to reflect these changes and to take advantage of modern data-keeping techniques. Therefore, given the need for reevaluation of cytotechnologists’ workloads (especially where Non-Gyn cases are concerned) and the desire to minimize distraction and manual datakeeping, we created a simple-to-use program for tracking a cytotechnologist’s workflow in a computational fashion with minimal manual data entry and unlimited potential for postanalysis in the versatile Excel environment.
MATERIALS AND METHODS The cytotechnologists at the Massachusetts General Hospital cytology laboratory have a varied daily workload that includes combinations of screening Gyn and Non-Gyn cases, assisting on rapid on-site evaluation (ROSE) of 16
cytology slides, performance of testing for high-risk human papillomavirus (hrHPV) by the COBAS (Roche) system and teaching of residents. All Gyn cases are prescreened by the FocalPoint GS imager that provides the cytotechnologist with 11 fields of view (FOV) for screening. The Non-Gyn cases contained a wide variety of specimens including urines, cerebrospinal fluids, lung lavages, lung nodule lymph node fine-needle aspirates (FNAs), pancreatic ultrasound-guided FNAs of nodules and cysts, breast FNAs, thyroid smears, thyroid ThinPreps, FNAs of subcutaneous nodules and cysts of all varieties, and other less common specimens. Slide number averaged between 2 and 5, but ranged as high as 18. The cytotechnologists chosen for the initial study had 8 and 16 years of experience, respectively. They recorded the data presented within this article during roughly 1 month of their typical workload. Namely, they carried out their jobs as usual but used the program to record the times spent on the various tasks each case entails. During this time, they worked full days and received an unaltered case load. The program (loosely referred to as TimeStamper) is written in the Visual Basic Architecture 6.0 and employs Microsoft Excel (a ubiquitous and powerful data collection and analysis suite). It uses a series of “if-then” loops and other contingencies to record information concerning time, date, time of day, day of the week, case number, number of slides, case type, work-related interruptions, and non–work related interruptions into an Excel spreadsheet. Following data collection, the user indicates that he or she would like to save and archive his or her data, and the program creates date-stamped Excel files preprogrammed to analyze the data in whatever fashion the investigator deems necessary (by default, average times, standard deviations, slides adjusted averages and standard deviations and analyses dependent on time of day and day of week) (Fig. 1A). The buttons and interfaces for user input are as follows: GYN: The first click prompts an input box for case number, which may be typed in or entered with a bar code scanner. Next, a time stamp is placed in the first cell of the case appropriate row indicating the time the cytotechnologist began doing the pre-work for the case such as checking the paperwork and consulting the electronic medical record (longitudinal medical record, or LMR). Cancer Cytopathology
January 2014
Simplified Cytotechnologist Workload/Vaickus and Tambouret
FIGURE 1. Sample screens are shown of TimeStamper user interface and input boxes.
The second click indicates that the user has finished prework and has begun screening the slide and places a corresponding time stamp. The third click indicates that screening is completed and the user has begun postscreening work including signing out the case and brings up a prompt for the cytotechnologist to enter the type of screening (eg, field of view [FOV], full screening [Full], or both FOV and Full [Both]) in addition to placing another time stamp. The fourth and final click indicates that the case is complete and has been moved from the “in” pile to the “out” pile and places the final time stamp (Fig. 1B,C). Cancer Cytopathology
January 2014
Non-Gyn: The first click prompts an input box for case number as in the above case. Next, an input box for number of slides is generated and a time stamp is placed. From this point on, the second, third, and fourth clicks perform analogous functions to those described above (Fig. 1D). Other: This button is intended to record work related interruptions to Gyn and Non-Gyn slide screening such as ROSE or hrHPV testing. The first click records a starting time and the second click records an end time to the event. 17
Original Article
FIGURE 1. (Continued).
Skip: The skip button allows the user to invalidate a row of data if they are interrupted before completing a case by a work or nonwork related interruption. Prescreen: The prescreen button is intended to record the time, date, case number, and diagnosis for rapid prescreening cases. This data is later compared to routine screening and the final diagnoses. The first click prompts an input box for case number and then a second input box for diagnosis. The data is automatically saved to a dated Excel file (a new one is generated for each new day); throughout the day, additional data is added onto to the end of the current table. 18
Save: The “save” button takes all current data and generates a dated Excel file containing only the completed cases (invalidated cases are removed) and performs a series of calculations to render a table listing the times to complete the various tasks (prescreening, screening, postscreening, and total case time) in seconds (although this can be changed to any measure of time the user desires). The table also contains the case number, the number of slides per Non-Gyn case, the time of day, and the day of the week. Additional data recorded on the same day will be added to the table every time “save” is clicked. In addition, the “save” button adds the current cleaned and analyzed Cancer Cytopathology
January 2014
Simplified Cytotechnologist Workload/Vaickus and Tambouret
FIGURE 2. (A) Sample screen demonstrating the format of data recording in the user interface. (B) Daily logs are kept in separate dated files which are automatically generated for each day.
data to the file “DataDump” which contains all data recorded by the particular user over the entire timeframe that the program has been used. In addition to containing the date, time, time of day, day of the week, and time spent per task, this file can also be customized to perform any number of analyses on the data thus collected in a dynamic fashion. Namely, once this file has been set up and customized, all analyses are performed automatically no matter how much data is added to the file and without any additional investigator input (row number is limited by the version of Excel being used; Excel 2003 has a row limit of 65,536, whereas later versions have row limits of 4 3 106 and greater) (Fig. 2A). Reset: The “reset” button is used at the end of the data-collecting day to return TimeStamper to the default setting, ready to begin another data collection session. Cancer Cytopathology
January 2014
The cytotechnologists were each provided with a “Netbook”-style laptop computer with an independent mouse. The operating system was Windows XP or Windows 7 and the version of Excel used was 2003. The computers had 512 megabytes of RAM or higher and at least an Intel N450 Atom processor. The preliminary data was collected by 2 cytotechnologists over a period of 1 month covering a variety of case loads and service types. Graphing and statistical analysis (Student t test, unpaired, 2-tailed and 1-way analyses of variance with Bonferroni post-test) were performed in Prism software GraphPad Pro, version 4.0.
RESULTS Our preliminary results were assembled over a period of 1 month using 2 cytotechnologists employing our Excel 19
Original Article
FIGURE 3. Time spent per task on Gyn and Non-Gyn cases. G is Gyn cases, NG is Non-Gyn cases, sum is the total time spent on each case, Pre is time spent prescreening the case (checking electronic medical record, verifying patient information), Screen is time spent screening the slide (adjusted for slide number), and Post is the time spent doing postscreening work and signing out the case. ***P
A Simplified Method for Assessing Cytotechnologist Workload Louis J. Vaickus, MD, PhD and Rosemary Tambouret, MD
BACKGROUND: Examining cytotechnologist workflow and how it relates to job performance and patient safety is important in determining guidelines governing allowable workloads. This report discusses the development of a software tool that significantly simplifies the process of analyzing cytotechnologist workload while simultaneously increasing the quantity and resolution of the data collected. METHODS: The program runs in Microsoft Excel and minimizes manual data entry and data transcription by automating as many tasks as is feasible. RESULTS: Data show the cytotechnologists tested were remarkably consistent in the amount of time it took them to screen a cervical cytology (Gyn) or a nongynecologic cytology (Non-Gyn) case and that this amount of time was directly proportional to the number of slides per case. Namely, the time spent per slide did not differ significantly in Gyn versus Non-Gyn cases (216 6 3.4 seconds and 235 6 24.6 seconds, respectively; P 5.16). There was no significant difference in the amount of time needed to complete a Gyn case between the morning and the evening (314 6 4.7 seconds and 312 6 7.1 seconds; P 5.39), but a significantly increased time spent screening Non-Gyn cases (slide-adjusted) in the afternoon hours (323 6 20.1 seconds and 454 6 67.6 seconds; P 5.027), which was largely the result of significantly increased time spent on prescreening activities such as checking the electronic medical record (62 6 6.9 seconds and 145 6 36 seconds; P 5.006). CONCLUSIONS: This Excel-based data collection tool generates highly detailed data in an unobtrusive manner and is highly customizable to the individual working C 2013 American Cancer Society. environment and clinical climate . Cancer (Cancer Cytopathol) 2014;122:15-22. V
KEY WORDS: workload; efficiency; computer; cytotechnologist; automation.
INTRODUCTION In the current clinical climate, demand for quick turnaround time and accuracy must be balanced with efficient resource allocation both in terms of equipment and supplies and in terms of personnel. Cytotechnologists form the backbone of most cytopathology services, and increasing demands on their time and mental resources must be carefully evaluated and re-evaluated to prevent fatigue, and reduce error rates while maintaining a reasonable quality of life and job satisfaction. Clinical Laboratory Improvement Amendments of 1988 (CLIA 88) legislation placed a limit of 100 on the number of slides that may be screened by a cytotechnologist in an 8-hour day. With the advent of liquidbased cytology preparations and computer-assisted screening, that number was increased to 200 (for ThinPrep slides with only the imager designated Field of View areas being evaluated). However, recent studies have found a decreased sensitivity for significant cytologic abnormalities at the currently permissible levels for cervicovaginal cytology slides. Indeed, the American Society of Cytopathology has issued recommendations that reduce the number of hours the cytotechnologist may screen to 7 per 24-hour period and that limit the number of slides screened per day to no more than 70.1-6
Corresponding author: Louis J. Vaickus, MD, PhD, Department of Pathology, Massachusetts General Hospital, 55 Fruit Street, Warren Building 2, Boston, MA 02114; Fax: (781) 223-5657; [email protected] Department of Pathology, Massachusetts General Hospital, Boston, Massachusetts Received: August 15, 2013; Revised: September 12, 2013; Accepted: September 16, 2013 We acknowledge Andrew Renshaw, MD, for many helpful discussions and advice. Published online November 6, 2013 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/cncy.21364, wileyonlinelibrary.com
Cancer Cytopathology
January 2014
15
Original Article
The studies on which these recommendations were based evaluated only gynecologic (Gyn) cytology slides. Nongynecologic (Non-Gyn) slide screening and the impact of other laboratory activities on the daily workload of the cytotechnologist have not been evaluated, nor has the actual time spent performing the various caserelated tasks. Previous studies of workload have relied on triedand-true methods such as keeping work logs by hand or manually in a computerized spreadsheet to finally be tabulated and analyzed computationally. This procedure adds complication to the technologist’s day and makes data analysis more difficult. In addition, manual recording of data limits the amount of information that can reasonably be tabulated. Reducing the demand on the cytotechnologist’s time during data collection and employing computerized “time stamping” can not only improve compliance but also result in more meaningful data, because a less obtrusive collection system minimizes distraction and workflow disruption while vastly increasing the amount of data that can feasibly be collected. With the advent of new technologies, new methods and new guidelines for cytopathology, the roles of cytotechnologists have gradually changed. Computerization of records and requisitions, although not yet universal, is becoming ever more prevalent as is the utilization of automated systems for generating reports and preventing human error. Furthermore, these changes are occurring at an accelerating rate due to emergence of a technological/ biomedical Moore’s law. Because of this, cytotechnologists’ workload should be reevaluated periodically to reflect these changes and to take advantage of modern data-keeping techniques. Therefore, given the need for reevaluation of cytotechnologists’ workloads (especially where Non-Gyn cases are concerned) and the desire to minimize distraction and manual datakeeping, we created a simple-to-use program for tracking a cytotechnologist’s workflow in a computational fashion with minimal manual data entry and unlimited potential for postanalysis in the versatile Excel environment.
MATERIALS AND METHODS The cytotechnologists at the Massachusetts General Hospital cytology laboratory have a varied daily workload that includes combinations of screening Gyn and Non-Gyn cases, assisting on rapid on-site evaluation (ROSE) of 16
cytology slides, performance of testing for high-risk human papillomavirus (hrHPV) by the COBAS (Roche) system and teaching of residents. All Gyn cases are prescreened by the FocalPoint GS imager that provides the cytotechnologist with 11 fields of view (FOV) for screening. The Non-Gyn cases contained a wide variety of specimens including urines, cerebrospinal fluids, lung lavages, lung nodule lymph node fine-needle aspirates (FNAs), pancreatic ultrasound-guided FNAs of nodules and cysts, breast FNAs, thyroid smears, thyroid ThinPreps, FNAs of subcutaneous nodules and cysts of all varieties, and other less common specimens. Slide number averaged between 2 and 5, but ranged as high as 18. The cytotechnologists chosen for the initial study had 8 and 16 years of experience, respectively. They recorded the data presented within this article during roughly 1 month of their typical workload. Namely, they carried out their jobs as usual but used the program to record the times spent on the various tasks each case entails. During this time, they worked full days and received an unaltered case load. The program (loosely referred to as TimeStamper) is written in the Visual Basic Architecture 6.0 and employs Microsoft Excel (a ubiquitous and powerful data collection and analysis suite). It uses a series of “if-then” loops and other contingencies to record information concerning time, date, time of day, day of the week, case number, number of slides, case type, work-related interruptions, and non–work related interruptions into an Excel spreadsheet. Following data collection, the user indicates that he or she would like to save and archive his or her data, and the program creates date-stamped Excel files preprogrammed to analyze the data in whatever fashion the investigator deems necessary (by default, average times, standard deviations, slides adjusted averages and standard deviations and analyses dependent on time of day and day of week) (Fig. 1A). The buttons and interfaces for user input are as follows: GYN: The first click prompts an input box for case number, which may be typed in or entered with a bar code scanner. Next, a time stamp is placed in the first cell of the case appropriate row indicating the time the cytotechnologist began doing the pre-work for the case such as checking the paperwork and consulting the electronic medical record (longitudinal medical record, or LMR). Cancer Cytopathology
January 2014
Simplified Cytotechnologist Workload/Vaickus and Tambouret
FIGURE 1. Sample screens are shown of TimeStamper user interface and input boxes.
The second click indicates that the user has finished prework and has begun screening the slide and places a corresponding time stamp. The third click indicates that screening is completed and the user has begun postscreening work including signing out the case and brings up a prompt for the cytotechnologist to enter the type of screening (eg, field of view [FOV], full screening [Full], or both FOV and Full [Both]) in addition to placing another time stamp. The fourth and final click indicates that the case is complete and has been moved from the “in” pile to the “out” pile and places the final time stamp (Fig. 1B,C). Cancer Cytopathology
January 2014
Non-Gyn: The first click prompts an input box for case number as in the above case. Next, an input box for number of slides is generated and a time stamp is placed. From this point on, the second, third, and fourth clicks perform analogous functions to those described above (Fig. 1D). Other: This button is intended to record work related interruptions to Gyn and Non-Gyn slide screening such as ROSE or hrHPV testing. The first click records a starting time and the second click records an end time to the event. 17
Original Article
FIGURE 1. (Continued).
Skip: The skip button allows the user to invalidate a row of data if they are interrupted before completing a case by a work or nonwork related interruption. Prescreen: The prescreen button is intended to record the time, date, case number, and diagnosis for rapid prescreening cases. This data is later compared to routine screening and the final diagnoses. The first click prompts an input box for case number and then a second input box for diagnosis. The data is automatically saved to a dated Excel file (a new one is generated for each new day); throughout the day, additional data is added onto to the end of the current table. 18
Save: The “save” button takes all current data and generates a dated Excel file containing only the completed cases (invalidated cases are removed) and performs a series of calculations to render a table listing the times to complete the various tasks (prescreening, screening, postscreening, and total case time) in seconds (although this can be changed to any measure of time the user desires). The table also contains the case number, the number of slides per Non-Gyn case, the time of day, and the day of the week. Additional data recorded on the same day will be added to the table every time “save” is clicked. In addition, the “save” button adds the current cleaned and analyzed Cancer Cytopathology
January 2014
Simplified Cytotechnologist Workload/Vaickus and Tambouret
FIGURE 2. (A) Sample screen demonstrating the format of data recording in the user interface. (B) Daily logs are kept in separate dated files which are automatically generated for each day.
data to the file “DataDump” which contains all data recorded by the particular user over the entire timeframe that the program has been used. In addition to containing the date, time, time of day, day of the week, and time spent per task, this file can also be customized to perform any number of analyses on the data thus collected in a dynamic fashion. Namely, once this file has been set up and customized, all analyses are performed automatically no matter how much data is added to the file and without any additional investigator input (row number is limited by the version of Excel being used; Excel 2003 has a row limit of 65,536, whereas later versions have row limits of 4 3 106 and greater) (Fig. 2A). Reset: The “reset” button is used at the end of the data-collecting day to return TimeStamper to the default setting, ready to begin another data collection session. Cancer Cytopathology
January 2014
The cytotechnologists were each provided with a “Netbook”-style laptop computer with an independent mouse. The operating system was Windows XP or Windows 7 and the version of Excel used was 2003. The computers had 512 megabytes of RAM or higher and at least an Intel N450 Atom processor. The preliminary data was collected by 2 cytotechnologists over a period of 1 month covering a variety of case loads and service types. Graphing and statistical analysis (Student t test, unpaired, 2-tailed and 1-way analyses of variance with Bonferroni post-test) were performed in Prism software GraphPad Pro, version 4.0.
RESULTS Our preliminary results were assembled over a period of 1 month using 2 cytotechnologists employing our Excel 19
Original Article
FIGURE 3. Time spent per task on Gyn and Non-Gyn cases. G is Gyn cases, NG is Non-Gyn cases, sum is the total time spent on each case, Pre is time spent prescreening the case (checking electronic medical record, verifying patient information), Screen is time spent screening the slide (adjusted for slide number), and Post is the time spent doing postscreening work and signing out the case. ***P
