Article Text

Download PDFPDF

Effect of Lean method implementation in the histopathology section of an anatomical pathology laboratory
  1. S S Raab1,2,
  2. D M Grzybicki3,
  3. J L Condel1,2,
  4. W R Stewart1,
  5. B D Turcsanyi1,
  6. L K Mahood1,
  7. M J Becich3
  1. 1Department of Pathology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, USA
  2. 2Center for Quality Improvement and Innovation, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, USA
  3. 3Department of Biomedical Informatics, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, USA
  1. Stephen S Raab, Department of Pathology, University of Colorado Denver, 12605 East 16th Avenue, Anschutz Inpatient Pavilion, Room 3022, Aurora, CO 80045, USA; stephen.raab{at}


Background: In the USA, the lack of processes standardisation in histopathology laboratories leads to less than optimal quality, errors, inefficiency and increased costs. The effectiveness of large-scale quality improvement initiatives has been evaluated rarely.

Aim: To measure the effect of implementation of a Lean quality improvement process on the efficiency and quality of a histopathology laboratory section.

Methods: A non-concurrent interventional cohort study from 1 January 2003 to 31 December 2006 was performed, and the Lean process was implemented on 1 January 2004. Also compared was the productivity of the Lean histopathology section to a sister histopathology section that did not implement Lean processes. Pre- and post-Lean specimen turnaround time and productivity ratios (work units/full time equivalents) were measured. For 200 Lean interventions, a 5-part Likert scale was used to assess the impact on error, success and complexity.

Results: In the Lean laboratory, the mean monthly productivity ratio increased from 3439 to 4074 work units/full time equivalents (p<0.001) as the mean daily histopathology section specimen turnaround time decreased from 9.7 to 9.0 h (p = 0.01). The Lean histopathology section had a higher productivity ratio compared with a sister histopathology section (1598 work units/full time equivalents, p<0.001) that did not implement Lean processes. The mean impact, success and complexity of interventions were 2.4, 2.7 and 2.5, respectively. The mean number of specific error causes affected by individual interventions was 2.6.

Conclusion: It is concluded that Lean process implementation improved efficiency and quality in the histopathology section.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

The lack of standardisation in American anatomical pathology laboratories results in less than optimal quality, inefficiencies and increased healthcare costs.1 With the rapid growth of the international patient safety movement, laboratories have increased their focus on quality and decreasing errors,17 but they generally lack the quality improvement tools and training to redesign large-scale processes and to measure effectiveness of process change.

The aviation and automobile industries have demonstrated success in improving safety, quality and efficiency by using a systems-based approach.811 Lean quality improvement initiatives simultaneously address quality, cost and efficiency, and currently are being adapted to healthcare settings.1217 As laboratories exhibit similarities to industrial production lines, Lean methods seemingly could have major effects in improving diagnostic testing patient care outcomes and reducing healthcare costs.1 5

In January 2004, we implemented Lean processes in the histopathology section of an anatomical pathology laboratory. In a qualitative study, Condel et al reported preliminary findings of the Lean implementation.1 This study extends the qualitative work of Condel et al1 and examines the long-term outcomes of Lean implementation on quality and efficiency.



The study was performed at the University of Pittsburgh Medical Center (UPMC) Shadyside Hospital, a 486 bed tertiary hospital located in urban Pittsburgh, USA.

Lean, Toyota Production System, and Perfecting Patient Care theory

Lean theory has been well described in the business and, more recently, in the healthcare literature.817 The Toyota Production System and Perfecting Patient Care (PPC) are types of Lean quality improvement that focus on immediate problem solving and reorganisation of processes and workflow, with the ultimate goal being the improvement of quality, reduction of inefficiencies, and the decrease of waste or unnecessary costs.817 Work can be broken down into a number of elements (ie, activities, connections and pathways) that may be altered to create a one-by-one, continuous flow process.8 Problems are solved to root cause and interventions are designed to alter the elements of work to reduce the probability of those problems occurring in the future.

PPC implementation

We elected to implement the PPC system18 in a histopathology laboratory section, because we believed that the work processes in this section most resembled an assembly line on an industrial shop floor.1 The timeline of major events for PPC implementation is shown in fig 1.

Figure 1 Timeline of Perfecting Patient Care (PPC) implementation. The timeline depicts major events in the history of the learning line and includes the introduction of technology and training. The team leader (JLC), who had no prior PPC experience, was hired in August 2003, and received PPC training at the Pittsburgh Regional Healthcare Initiative. Organisational commitment was obtained from the University of Pittsburgh Medical Center (UPMC) Chief Executive Officer, Chief Operating Officer, Department of Pathology Chairman, UPMC Shadyside Hospital Chief of Pathology, and Director of Anatomic Pathology.

Condel et al1 previously reported several of the steps of the implementation process conducted in project years 2003 and 2004. Briefly, the early steps consisted of education, determining the current condition (fig 2), designing and implementing an ideal state (fig 3), and implementing and sustaining the PPC learning line in March 2004.

Figure 2 Current condition. The current condition was established by determining the specimen mix, volume and timing; assessment of the physical space; and evaluation of existing laboratory personnel, consisting of histotechnologists and aides. The initial physical layout and specimen flow in the University of Pittsburgh Medical Center Shadyside laboratory is depicted. Specimens were first accessioned and grossed (1) in the gross room that was located in a physical location separate from the histopathology section. Tissue was then placed in tissue processors (2) that were located in the gross room. The following morning, all specimens were batched and embedded at one of three embedding stations (3), placed in the refrigerator (4), and then cut at one of four cutting stations (5), located in different parts of the laboratory. Slides were then placed in an oven (6), stained (7), coverslipped (8), labelled (9) and sent to the pathologists (10). PA, pathologists' assistant; Proc, processor; Tech, technologist.
Figure 3 Ideal state. The ideal condition was created following detailed observations that led to the design of a one-by-one, continuous flow work process in which a specimen moves from one work station to the next without waiting or adding waste (non-value added work) to the system. The physical layout and specimen flow in the University of Pittsburgh Medical Center Shadyside laboratory following introduction of the learning line is depicted. Following grossing, the tissue processors were moved to the histopathology section (1). A single, continuous flow line was created in which tissue was embedded (2) and cut (3). In the current condition, multiple histotechnologists embedded and then cut; in the ideal state only one histotechnologist embedded and one histotechnologist cut, unless volumes required additional staff being added to the line. Slides were then placed in an oven (4), stained (5), coverslipped (6), labelled (7), checked for quality (8) and sent to the pathologists (9). PA, pathologists' assistant; Proc, processor; Tech, technologist.


Interventions varied in complexity, impact and success. We classified interventions into two categories: (1) real time, and (2) A3 documented. Real-time interventions were performed on the fly, generally without documentation, and were of relatively low complexity. An example of a real-time intervention was a worker who rearranged part of her workspace to make a specimen hand-off more efficient. Although we recognised that real-time interventions were an integral part of culture change and affected efficiency and quality, we did not design detailed processes to measure the direct success or failure of these interventions.

A3 interventions were planned using an A3 diagram that documented the problem, causes, solutions, and proposed interventions. In the PPC learning line, the team leader worked with front-line personnel to devise and test these interventions.


We compared the pre- and post-PPC implementation efficiency metrics and qualitatively assessed the impact, success and complexity of 200 A3 interventions.


We measured the efficiency metrics of: (1) turnaround time (TAT) and (2) productivity. In the USA, the overall TAT of anatomical pathology specimens is an important metric for laboratory accreditation,19 20 and TAT may be measured for any anatomical pathology section. For the histopathology section, we defined specimen TAT as the time when the gross examination was complete to the time when the slides from a case was verified and sent to a pathologist. We calculated TAT using business hours (08:00–21:00). The laboratory information system (LIS) (CoPath Plus Anatomic Pathology, Cerner Corporation, Kansas City, Missouri, USA) was not efficiently designed to record the histopathology-verified TAT for all cases and we excluded cases that were processed over several days (eg, >36 h) because of grossing issues (eg, delays due to decalcification) and cases with a negative TAT secondary to data entry verification problems.

We compared the pre- and post-PPC histopathology section productivity ratio, which we defined as the number of work units divided by the number of personnel full time equivalents (FTEs). We considered that the histopathology section produced two types of work: tissue blocks and slides. Based on our previous assessment of time to produce product, we considered the work unit of production of a block as 1.0 and a slide as 1.0. We recognised that block and slide production depended on a number of factors, such as tissue type and specimen size, and that individual blocks and slides would take longer or shorter times to produce. As a main goal was to compare the pre- and post-PPC productivity, we viewed this estimate as reasonable, considering that the work type remained constant.

Productivity depends on system capacity, and productivity may increase in systems that are operating at levels below capacity when work is added and this productivity gain is not secondary to improved efficiency. Histopathology section productivity is poorly quantified in the medical literature.21 Thus, we compared the 2006 Lean histopathology section productivity ratio to the 2006 productivity ratio of a sister histopathology section that did not implement Lean. The number of histopathology section FTEs was averaged per month.


Two of the authors (SSR and JLC) separately qualitatively classified each intervention in terms of success, complexity and impact. Success was assessed in terms of sustainability, ranging from not sustainable beyond the initial implementation phase to fully sustainable and adopted as a new means to perform work. Complexity ranged from interventions affecting single activities to affecting areas outside the department and requiring more than 1 month to implement. For each intervention, success and complexity were graded on a 5-part Likert scale (Box 1). Impact was graded on the sum of 4 additional metrics: (1) effect on workflow change, (2) effect on process change, (3) effect on overall PPC objectives, and (4) staff perception of effect of change. Each of these metrics was also scored on a 5-part Likert scale and these scores were added and the total was divided by four. Each intervention was classified as primarily affecting quality, cost or efficiency, recognising that interventions ultimately could be linked to all three metrics. In addition, we assessed if the specific interventions were primarily focused on the overall system or individual pathways, connections or activities.

Box 1: Scales for grading interventions

Success scale
  1. No sustainable implementation

  2. Minimal sustainable implementation

  3. Moderate sustainable implementation

  4. Significant sustainable implementation

  5. Highly sustainable implementation

Complexity scale
  1. No complexity for implementation (simple, straightforward and isolated to activity)

  2. Minimal complexity for implementation (activity to connect areas and greater than 3 days to fully implement)

  3. Moderate complexity for implementation (overlapping areas and departments affected and greater than 1 week to implement)

  4. Significant complexity for implementation (requires obtaining resources and supplies, multiple areas affected, and greater than 1 month to implement)

  5. High complexity for implementation (requires alterations in staffing, multiple departments affected, and greater than 1 month to implement)

Impact on outcome scale
  • 4–5, no significant impact

  • 6–10, minimal impact

  • 11–15, moderate impact

  • 16–20, significant impact

Impact on outcome was determined by separately measuring workflow changes, process changes, advancement of overall staff objectives, and staff perception (each measured on a 5-part Likert scale). Thus, the overall impact ranged from 4 (a score of 1 on each metric) to 20 (score of 5 on each metric).

The UPMC Shadyside histopathology section tracked 30 domain specific errors (Box 2) prior to and after PPC implementation, and we classified each intervention in terms of the specific errors affected and theoretically decreased.

Box 2: Errors attributable to the histopathology section of the laboratory

  • Loss of tissue due to excessive microtomy

  • Blocks over processed (too dry), delay in slides for soaking

  • Histopathology has no slide trays causing delay in slides to staff

  • Tissue specimen is lost during embedding

  • Contaminant/floater picked up off water bath

  • Wrong block pulled and ordered stains performed

  • Wrong case number written on glass slide or impossible to read

  • Delay in turnaround time due to shortage in staff

  • Block not properly faced

  • Tissue improperly embedded

  • Poor tissue processing

  • Tissue processor failure

  • Slide broke during staining

  • Inadequate amount of stain

  • Stainer failure

  • Wrong pre-treatment used

  • Wrong control used

  • Coverslipper equipment failure

  • Stains completed but not verified

  • Inappropriate staining of slide (wrong stain performed)

  • Delayed receipt of slides

  • Contaminant/floater on slide other than waterbath

  • Poor tissue microtomy

  • Ordered stain not performed

  • Failure to notify pathologist of missing/incomplete case materials

  • Poor quality staining

  • Tissue fell off slide

  • Paper-labelled slides mislabelled

  • Slides received in poor condition

Data retrieval

Specimen TAT and daily work volume were obtained from the LIS. Personnel data were obtained by communication with the laboratory manager and histopathology section personnel.

Data analysis

We used statistical process control charts to track mean daily specimen TAT.2226 Differences in mean daily specimen TAT and mean monthly productivity ratio for the pre- and post-PPC implementation time frames (years 2003 and 2006, respectively) were examined using a Student t test. Statistical significance was assumed with a p⩽0.05.


The mean daily pre- and post-PPC implementation specimen TAT is shown in fig 4. The mean daily specimen TAT for project years 2003 and 2006 improved from 9.7 to 9.0 h, respectively (p = 0.01).

Figure 4 Statistical process control chart depicting histopathology section turnaround time (TAT) in business hours. The mean specimen TAT is based on 2003 (pre Perfecting Patient Care) monthly data. Upper and lower control limits were established at 3 SD above and below the mean, respectively. Upper and lower warning limits were established at 2 SD above and below the mean, respectively. Process control charts are used to differentiate common cause variation from special cause variation. A special cause was indicated when: (1) a single point fell outside the control limit, (2) two out of three successive values were on the same side of the centreline and more than 2 SD from the centreline, (3) eight successive values were on the same side of the centreline, or (4) there was a trend of six or more values in a row steadily increasing or decreasing. In July 2005, the mean daily specimen TAT decreased to below the mean daily 2003 specimen TAT (pre Perfecting Patient Care ) and in 2006, only two data points exceeded the mean daily 2003 specimen TAT.

The number of Lean histopathology section FTEs/day in 2003 and 2006 were 4.5 and 5.1, respectively, and the productivity ratios were 3439 and 4074 work units/FTE, respectively (p<0.001). In 2006, the sister histopathology section produced 23 972 mean monthly work units with 15.0 FTEs/day and had a productivity ratio of 1598 work units/FTE, which was statistically significantly lower compared with the Lean histopathology section (p<0.001).

Examples of A3 problems and interventions and the summary of the success, complexity and impact of the 200 A3 quality interventions are shown in table 1.

Table 1 Examples of specific problems and solutions and summary of A3 intervention quality metrics

Box 3: Summary quality data

Mean intervention impact score: 2.4; one (0.5%) intervention had a score above 4

Mean intervention success score: 2.7; 51 (25.5%) interventions had a score above 4

Mean intervention complexity score: 2.5; 39 (19.5%) interventions had a score above 4

Mean number of specific quality metric failures affected by individual interventions: 2.6; 17 interventions affected five or more specific errors

Interventions primarily affecting:

  • Quality, 102 (51%)

  • Efficiency, 91 (45.5%)

  • Cost, 7 (3.5%)

  • Overall system pathway, 41 (20.5%)

  • Individual pathway, 25 (12.5%)

  • Connections, 71 (35.5%)

  • Activities, 63 (31.5%)


Our findings indicate that the implementation of Lean processes in a histopathology section decreased specimen TAT and increased productivity. Although not directly quantified, improved efficiency correlated with the histopathology section operating at a decreased level of expenditure (ie, higher production per FTE). Lean processes also affected histopathology section quality, although the culture in our laboratory only allowed us to measure this qualitatively.

Take-home messages

  • Lean methods may be used simultaneously to improve quality and efficiency metrics in laboratory settings.

  • Lean improvement focuses on work process standardisation, and currently many technical laboratory work processes are not standardised, even within laboratories.

  • Successful Lean implementation necessitates top-level organisational commitment, without which true improvement is not sustainable.

  • Lean methods involve a cultural shift to allow front-line workers to problem solve and change their own work processes.

Our data corroborate the findings of other authors who implemented Lean processes in clinical laboratory sections.1315 2730 Only several studies have presented data on the effect of Lean implementation in anatomic pathology laboratories, and there has been an absence of anatomical pathology studies measuring the effectiveness of Lean implementation in terms of improved efficiency and quality and cost reduction.1 5 16 Anatomical pathology specimen TAT depends on the completion of a number of sequential, yet separate steps and the optimal use of Lean methods would affect all steps. A limitation in our system was the lack of pathologist acceptance of PPC principles; the main reason why PPC methods were first implemented only in the histopathology section was to limit direct effect on the workflow of individual pathologists. Systems that most effectively adopt Lean methods have acceptance at all organisational levels6 11 17; unfortunately, some of the UPMC leadership viewed PPC implementation only as experimental and labelled the implementation team members as troublemakers.

The decrease in the histopathology section TAT was secondary to improved efficiency that resulted in the introduction of a second run of specimens that normally would have been processed the following day. Efficiency was improved by altering work flow through interventions affecting histotechnologist activities and connections.

A challenge in measuring histopathology section productivity was the lack of published baseline and capacity data. A component of the initial PPC implementation phase was the measure of cycle times1 of individual processes to standardise processes and estimate maximum capacity. The impact of PPC implementation was more profound when the Lean histopathology section productivity ratio was compared to the productivity ratio of a sister histopathology section. These data indicated that the pre-PPC histopathology section productivity ratio was not initially low and that PPC implementation improved the productivity ratio in a histopathology section that already functioned at or near capacity.

Zarbo and D’Angelo used Lean methods to change anatomic pathology laboratory work culture by creating a blame-free environment to increase the reporting of specimen quality defects.5 Benchmarking the error or defect frequency is a prerequisite for measuring the affect of Lean implementation on quality.5 28 Leape and Fromson reported that a major limitation in error reduction has been disruptive physician behaviour that interferes with the process of delivering care.31 Such behaviours extend beyond the fear of error reporting, derail quality improvement initiatives, and are tacitly accepted by non-physician administrators.3134 The main challenge in the UPMC Shadyside laboratory was overcoming the resistance to culture change and disruptive physician behaviours.

The majority of PPC A3 interventions affected the metrics of quality and efficiency and only indirectly affected cost by improving efficiency. Histopathology section employees designed these interventions, and their main goal was to improve the quality and efficiency of work, thereby eliminating the anxiety of potentially making errors in a high volume and high stress environment. A goal of PPC implementation was to link histopathology section work to individul patient well-being in order to drive quality improvement processes.

As PPC implementation improved care delivery in a single section of the anatomical pathology testing pathway, problems in adjacent sections became more apparent. Our strategy has been to implement PPC techniques in these adjacent sections as workers in these sections recognise the benefits of PPC process improvement. Laboratories are complex systems and strategies of implementation must be tailored for particular laboratory cultures and designs.


The authors would like to acknowledge the UPMC Shadyside histopathology section employees who worked during and after the implementation of PPC methods.



  • Funding: This study was supported by a grant from the Jewish Healthcare Foundation, Pittsburgh, Pennsylvania, USA, and grant HS13321-01 from the Agency for Healthcare Research and Quality, Rockville, Maryland, USA. The funding agencies had no role in the study design, data collection, data analysis, interpretation of data, writing of the manuscript, and the decision to submit the manuscript.

  • Competing interests: None.