Clinical Chemistry / IMPROVING PREANALYTIC PROCESSES
Improving Preanalytic Processes Using the Principles
of Lean Production (Toyota Production System)
Thomas J. Persoon, MS,1,2 Sue Zaleski, MA,1 and Janice Frerichs1
Key Words: Preanalytic; Process improvement; Lean production; Turnaround time
DOI: 10.1309/865V7UMFPUKGCF8D
Abstract
The basic technologies used in preanalytic
processes for chemistry tests have been mature for a
long time, and improvements in preanalytic processes
have lagged behind improvements in analytic and
postanalytic processes. We describe our successful
efforts to improve chemistry test turnaround time from a
central laboratory by improving preanalytic processes,
using existing resources and the principles of lean
production. Our goal is to report 80% of chemistry tests
in less than 1 hour and to no longer recognize a
distinction between expedited and routine testing. We
used principles of lean production (the Toyota
Production System) to redesign preanalytic processes.
The redesigned preanalytic process has fewer steps and
uses 1-piece flow to move blood samples through the
accessioning, centrifugation, and aliquoting processes.
Median preanalytic processing time was reduced from
29 to 19 minutes, and the laboratory met the goal of
reporting 80% of chemistry results in less than 1 hour
for 11 consecutive months.
16 Am J Clin Pathol 2006;125:16-25
16 DOI: 10.1309/865V7UMFPUKGCF8D
Downloaded from https://academic.oup.com/ajcp/article/125/1/16/1759625 by guest on 06 May 2021
Laboratories are under constant pressure to improve their
services. The mantra of “faster, cheaper, better” is heard in the
laboratory world as loudly as in other sectors of society.1 Clinical
chemistry laboratories have responded to these demands by
acquiring analyzers with improved throughput to speed up the
analytic process. The postanalytic process of reviewing results
and distributing them to the appropriate providers has been
largely automated by adoption of laboratory information sys-
tems (LISs) that use autoverification rules to reduce the number
of results manually reviewed by laboratory scientists. The use of
autoverification makes results available on-screen to clinicians
much sooner. Less attention has been given to preanalytic
processes. In this article, we describe our successful efforts to
improve chemistry test turnaround time from a central (core)
laboratory by improving preanalytic processes, using existing
resources and the principles of lean production.
The basic technologies used in preanalytic processes for
chemistry tests (centrifugation and aliquoting) have been
mature for a long time. There is little financial or technologi-
cal incentive to replace preanalytic equipment (centrifuges
and pipetters) unless they fail and become uneconomical to
repair. Operating the centrifuges, preparing sample aliquots,
and distributing samples to analysis stations traditionally have
been manual processes. Within the last decade, systems have
been developed to automate these processes, but they require
significant capital expenditures and long payback periods.
Thus, improvements in preanalytic processes have lagged
behind improvements in analytic and postanalytic processes.
Our laboratory supports a 763-bed acute care hospital, an
emergency treatment and level I trauma center, and numerous
general and specialty clinics located in the same physical
structure. The clinical chemistry laboratory handles 1,000 to
© American Society for Clinical Pathology

1,500 samples per day and performs between 7,000 and
10,000 analyses per day. Our interest in improving the prean-
alytic chemistry test process was prompted by complaints
from our customers about slow turnaround times and com-
plaints from laboratorians about workplace stress. The word
most commonly used by laboratorians to describe the environ-
ment in the preanalytic process work area was chaos.
Interviews with laboratorians and direct observation of the
workplace confirmed that the processes in place were no
longer meeting the needs of the laboratory or its customers.
The laboratory’s traditional approach to resolving cus-
tomer complaints about turnaround time was to establish an
expedited process for that customer’s samples. Examination of
workload data revealed that approximately 47% of chemistry
tests were being ordered with the expectation of expedited pro-
cessing—the meaning of the term stat. Some customer groups,
particularly the emergency treatment center, were requesting
that service be expedited even beyond the stat classification.
We believed that providing additional categories of expe-
dited testing would result in additional chaos by adding even
more processes and pathways to an already complicated envi-
ronment. Additional processes and process complexity also
contribute to cyclic propagation of error-generating behavior.2
Conversations with key customer groups revealed that all cus-
tomer groups desired faster turnaround time for chemistry
tests. Therefore, we chose to take a different approach to
redesigning the processes. We believed that we could improve
service for all customers by using the tools of lean production
to create preanalytic processes that eliminated wasteful activ-
ities and delays. Our goal was to be able to report 80% of all
chemistry tests (including therapeutic drug monitoring and
thyroid hormones) in less than an hour, and we would no
longer recognize requests for expedited (stat) testing.
The concepts of lean production were first developed by
the Toyota Motor Corporation during the last half of the 20th
century and since have become known as the Toyota Production
System.3 The major goal of lean production is to reduce waste
in 7 categories: overproduction, inventory, transportation,
inspection, correction, motion, and waiting. Waste is defined as
the elements of production that do not add value to the product
or service being produced. Lean production has been adopted in
many industries but has been applied to health care only recent-
ly, and there are few reports in the peer-reviewed literature.4-8
Materials and Methods
Preanalytic Process
Blood samples were received in evacuated glass tubes of
various sizes, containing no additives, lithium heparin anticoag-
ulant, polyester separation gel, or a combination of anticoagu-
lant and separation gel (Vacutainer, Becton Dickinson, Franklin
© American Society for Clinical Pathology
Clinical Chemistry / ORIGINAL ARTICLE
Lakes, NJ). Tests were accessioned into an LIS (Cerner Classic,
Cerner, Kansas City, MO) that assigned a unique accession num-
ber to each sample and printed barcode labels that were affixed
to the parent sample tube and any child (aliquot) tubes.
For samples requiring serum or plasma for analysis, cen-
trifugation of the blood collection tubes was performed at
3,000 rpm for 6 minutes (approximately 1,600g) with a Jouan
C-412 centrifuge (Jouan, Winchester, VA) equipped with
swinging bucket heads.
Analytic Process
Chemistry tests were performed on a Modular analyzer
Downloaded from https://academic.oup.com/ajcp/article/125/1/16/1759625 by guest on 06 May 2021
(Roche Diagnostics, Indianapolis, IN) consisting of 1 ISE1800
module and 3 P800 modules as the primary analyzer. The sec-
ondary analyzer initially consisted of a Hitachi 737 (Roche
Diagnostics). The 737 was replaced with a Modular consisting
of an ISE and 1 P module in April 2002. Immunochemical
assays were performed on an Elecsys 2010 (Roche
Diagnostics). No significant changes were made to the opera-
tion of the chemistry analyzers during the study period, other
than replacing the secondary analyzer.
Analyzers were linked to the LIS via a computerized
interface (Dawning Technologies, Fairport, NY). The LIS
uses a rule-based system to verify results from the analyzers
that meet defined criteria and transmit them to the hospital’s
computerized medical record system. Results failing to meet
the acceptability criteria are held until reviewed by a clinical
laboratory scientist.
Baseline Cycle Time Study
To determine a baseline for benchmarking process
improvement toward our goal, we conducted a chemistry test
cycle time study before implementing changes. Adhesive labels
were prepared with a time code that masked the arrival time to
minimize the occurrence of the Hawthorne effect. A sample
identification station was established approximately 50 ft from
the laboratory sample drop-off window. As incoming chemistry
samples were brought to the sample identification station, we
affixed a label to each that identified the time the sample passed
the station. The samples then were delivered and processed in
the usual manner. As the samples reached the input rack of the
main chemistry analyzer, we recorded the assigned time code as
well as the LIS-assigned accession numbers. The time the
accession number was assigned was obtained retrospectively
from the LIS. From these data, we calculated the time delay
before accessioning (time from sample delivery to accession-
ing) and the preanalytic cycle time (time from accessioning to
delivery at the chemistry analyzer input rack).
Performance Index
Because significant resources are required to conduct a
complete cycle time study, we sought to develop a simpler,
Am J Clin Pathol 2006;125:16-25 17
17 DOI: 10.1309/865V7UMFPUKGCF8D 17
Persoon et al / IMPROVING PREANALYTIC PROCESSES

automated measure of chemistry test turnaround time that Process Map

could be obtained on a daily basis. We determined that our LIS
could provide us with a report listing the percentage of chem-
istry tests completed in an hour for hourly and daily intervals.
Because our goal was to report 80% of tests in an hour or less,
we subtracted 80% from each period’s percentage completion
rate to arrive at a performance index (PI). A positive PI is the
percentage of tests completed above the 80% goal, and a neg-
ative PI is the percentage below the 80% goal. We monitored
hourly and daily PIs. The monthly PI was obtained by averag-
ing the weekday PIs for the month.
We observed the existing accessioning and preanalytic
processes, with close attention to the process steps that might
be wasteful, unnecessary, or inappropriate. Laboratorians
involved in the process also were asked to describe the
processes and environment in the preanalytic work area.
These observations and descriptions were used to create the
preintervention map of the process ❚Figure 1❚. Like most clin-
ical laboratories, our laboratory purportedly provided 2 levels
of service, expedited (stat) and routine. However, the only dis-
tinction between the process for an expedited sample and a

Downloaded from https://academic.oup.com/ajcp/article/125/1/16/1759625 by guest on 06 May 2021

Call for
Information

No
Accessioner Labeler

Accumulate Information Accession Print labels

requisitions
Yes tests
complete
Triage

Match
Open Separate
requisition
package paper from
number with
sample
tube number

Place tubes Find tube

in rack in rack

Store
requisitions
in holding
basket

Allow large
Sort aliquot Load Run Empty Prepare
Label tubes batch to
tubes centrifuge centrifuge centrifuge aliquots
accumulate

Centrifuge operator Centrifuge operator Centrifuge operator

Figure Key
Sort tubes Deliver Process activity Decision

and to multiple
aliquots locations

Blood specimen Barcoded blood specimen

Lab requisition Barcoded empty aliquot tube

Barcode label
Barcoded full aliquot tube

❚Figure 1❚ Preintervention process map, preanalytic process.

18 Am J Clin Pathol 2006;125:16-25 © American Society for Clinical Pathology

18 DOI: 10.1309/865V7UMFPUKGCF8D
 Clinical Chemistry / ORIGINAL ARTICLE

routine sample was that the expedited sample would be moved
to the front of whatever queue it was in at the time. With more
than 40% of the samples being marked stat, this distinction
was essentially meaningless.
Intervention
By using the information obtained from our observations
and interviews, we redesigned the preanalytic process using
principles of lean production and the Toyota Production
System.9 In our process redesign, we were constrained to use
the same laboratory space and human resources for preanalytic
functions, although duties could be reassigned. The redesigned
constitute a process are performed on each object under-
going the process before the work is begun on the next
object; there is no batching. For example, when a requi-
sition arrives in the laboratory accompanied by multiple
sample containers, the laboratorian handling each requi-
sition accessions all tests and labels all containers asso-
ciated with that requisition, including containers for
aliquots not yet prepared. Only after accessioning is
complete are the containers sorted based on the next
process or destination.
In the 1-piece flow process, no distinction is made
between stat and routine requests; rather, samples are handled

Downloaded from https://academic.oup.com/ajcp/article/125/1/16/1759625 by guest on 06 May 2021

process was submitted to supervisors and laboratorians for on a “first in, first out” (or FIFO) basis.
comments, and their suggestions were incorporated. The
process map for the redesigned process is shown in ❚Figure 2❚. Toyota Production System
In addition to redesigning the flow of work, we also
One-Piece Flow redesigned how the work would be accomplished, using 4
The redesigned process incorporates the concept of rules of the Toyota Production System.3 The rules and an
1-piece flow. In a 1-piece flow system, all activities that example application are given in ❚Table 1❚.

Store
requisitions
in holding
basket

Accessioner Accessioner Accessioner

Allow small
Open Information Accession batch to
Yes Print labels Label tube
package complete? tests accumulate

No
Send to Resolve
exception
handler
problems
Process activity
Figure Key
Decision

Blood specimen Barcoded blood specimen

Lab requisition Barcoded empty aliquot tube

Barcode label
Barcoded full aliquot tube
Exception handler

Centrifuge operator

Load Deliver
centrifuge Run Empty Aliquot Prepare aliquots to
with small centrifuge centrifuge
Sort tubes
needed?
Yes aliquots single
batch location

No

Deliver tubes
without
aliquots

Sorter-
circulator

❚Figure 2❚ Process map for the redesigned preanalytic process.

© American Society for Clinical Pathology Am J Clin Pathol 2006;125:16-25 19

19 DOI: 10.1309/865V7UMFPUKGCF8D 19
Persoon et al / IMPROVING PREANALYTIC PROCESSES

Redesigned Process to ring at this workstation. The accessioning workstations

In the redesigned process, 5 new workstations were cre-
ated, with specific descriptive titles. ❚Table 2❚ describes the
principal duties of each workstation. Laboratorians are
assigned to workstations for 2-hour periods, alternating
between sitting–cognitive task workstations (accessioner,
exception handler, customer service/phones) and
standing–physical task workstations (centrifuge operator,
sorter-circulator) as much as possible. This rotation ensures
that laboratorians are cross-trained on all tasks in the area,
maintain proficiency in these tasks, and experience task vari-
ety. In cases of staff shortage, this organizational structure can
were configured to include a computer, label printer, rack con-
taining clean aliquot tubes, and baskets for incoming samples
and samples ready to be distributed to the next workstation.
The centrifuge workstation included open bench space to load
and unload centrifuge buckets, 2 centrifuges, biohazard waste
containers, and above-counter bins for transfer pipettes and
other consumables. The design of the centrifugation process
called for a 6-minute centrifugation cycle at 1,500g followed
by a 3-minute deceleration. The procedure for centrifuge oper-
ation called for starting a centrifuge batch every 6 minutes;
thus, the centrifuge operator has 3 minutes from the time the

Downloaded from https://academic.oup.com/ajcp/article/125/1/16/1759625 by guest on 06 May 2021

be collapsed to require fewer laboratorians by combining
some duties.
To accommodate the new workstations, we made minor
modifications to the laboratory work space. A customer ser-
vice workstation was established separate from the accession-
ing stations, and the primary telephone lines were configured
deceleration is complete to empty the centrifuge and load the
next batch. Aliquot preparation was performed when the oper-
ator was not loading or unloading.
Following a training period, the redesigned lean process
was introduced. We conducted additional observations to
monitor performance. Supervisors conducted daily evaluation

❚Table 1❚
Toyota Production System Rules3

Rule
No. Rule Example

1 All work shall be highly specified as to
content, sequence, timing, and outcome.
2 Every customer-supplier connection must
be direct, and there must be an
unambiguous yes-or-no way to send
requests and receive responses.
3 The pathway for every product and service The sorter-circulator delivers accessioned samples directly from the accessioners to the
must be simple and direct.
4 Any improvement must be made in
accordance with the scientific method,
under the guidance of a teacher, at the
lowest possible level in the organization.
Samples needing centrifugation are loaded into waiting centrifuge bucket inserts in a
specified pattern (tubes not requiring aliquots in the center, tubes requiring aliquots on
the perimeter). A centrifuge batch is started every 6 minutes, regardless of the number
of samples to be centrifuged or in the accessioning queue. When the centrifugation is
complete, the tubes not requiring aliquots are removed and placed in a transport rack.
The tubes requiring aliquots are removed from the buckets one at a time and the
aliquots dispensed into prelabeled tubes. When the bucket inserts are empty, they are
returned to the loading point.
If a requisition is received without complete information, the accessioner does not
accession, but completes an “exception ticket” citing the missing information, attaches
it to the requisition, and places it in a basket. The sorter-circulator picks up exceptions
and delivers them to the exception handler. The exception handler resolves the missing
information issue, signs off on the exception slip, and returns the sample to an
accessioner to restart the accessioning process.
centrifuge workstation and delivers centrifuged samples directly to the chemistry
analyzer loading workstation.
Laboratorians wanting to change a “best practice” submit their ideas to a supervisor or
process engineer. An experiment is designed to test the effects of the change and
conducted with appropriate measurements. If desired effects are achieved, the new
information is incorporated into a best practice.

❚Table 2❚
Principal Duties for Each Workstation

Position Principal Duties

Accessioner (1 to 3 workstations) Unpack incoming samples; check for completeness of information; accession tests into LIS; affix labels
to parent and aliquot containers
Sorter-circulator Unload pneumatic tube carriers; distribute incoming workload to accession workstations; deliver
accessioned samples to next workstation; deliver centrifuged samples to next workstation; restock
supplies at accession workstations as needed
Centrifuge operator Load and unload centrifuges (2); prepare aliquots as needed
Exception handler Resolve problems with incoming workload (eg, lack of information, inappropriate samples); measure and
aliquot 24-hour urine specimens; back up accessioners during busy periods
Customer service Answer incoming telephone calls; accession requests for “add-on” tests

LIS, laboratory information system.

20 Am J Clin Pathol 2006;125:16-25 © American Society for Clinical Pathology

20 DOI: 10.1309/865V7UMFPUKGCF8D
 Clinical Chemistry / ORIGINAL ARTICLE

meetings with laboratorians to find additional process prob- Increased PI

lems and reinforce the use of the lean process. As laboratori- ❚Figure 5❚ and ❚Figure 6❚ show the daily PIs for November
ans became more familiar with the lean process, their com- 2001 (before implementation of the lean production system)
ments were used to develop best practices for operating each and November 2002 (6 months after implementation of the
workstation. These best practices were disseminated at the lean production system). The monthly PIs for the 18 months
daily evaluation meetings and incorporated into the procedure following the beginning of the project are shown in ❚Figure 7❚.
manual and training for new staff. Following complete implementation of the lean production
During the study period, we experienced the usual system in June 2002, the monthly PIs became positive and
amount of employee turnover but neither added nor reduced remained positive for 9 consecutive months. The monthly
the number of budgeted full-time equivalent positions index became negative in March, April, and May 2003. Review
assigned to the preanalytic process. of the daily PIs for these 3 months revealed that each of these
months had 3 “bad days” on which the daily PIs were less than

Downloaded from https://academic.oup.com/ajcp/article/125/1/16/1759625 by guest on 06 May 2021

–10. Further investigation revealed that performance on these 3
days was suboptimal owing to failures of analytic instrumenta-
Results
tion. When the data for the analytic instrumentation failure
days were removed, the monthly PI became positive for March
Decreased Cycle Time and April and less negative for May. The 12-month period of
The redesign of the preanalytic process in our laboratory sustained improvement supports our conclusion that the per-
significantly improved chemistry test turnaround time. The formance improvements are the result of the improved preana-
results of the baseline and postimprovement cycle time stud- lytic lean process and not a manifestation of the Hawthorne
ies are shown in ❚Figure 3❚ and ❚Figure 4❚. The redesigned pre- effect or the result of seasonal or periodic changes in workload.
analytic process resulted in a 30% improvement in the 50th, Workload increased by an average of 0.54% per month during
80th, and 90th percentiles of the cycle time distribution. the study period, or 6.5% per year.

35 100

90
30
80
50th percentile 29 min

25 90th percentile 43 min

70

Cumulative Percent
60
20 80th percentile 35 min
No. of Samples

50

15
40

Samples
10 30
Cumulative %

20
5
10

0 0
0:00
0:02
0:04
0:06
0:08
0:10
0:12
0:14
0:16
0:18
0:20
0:22
0:24
0:26
0:28
0:30
0:32
0:34
0:36
0:38
0:40
0:42
0:44
0:46
0:48
0:50
0:52
0:54
0:56
0:58
1:00

Cycle Time (Minutes After Accessioning)

❚Figure 3❚ Preanalytic process cycle time (before lean production).

© American Society for Clinical Pathology Am J Clin Pathol 2006;125:16-25 21

21 DOI: 10.1309/865V7UMFPUKGCF8D 21
Persoon et al / IMPROVING PREANALYTIC PROCESSES

35 100

50th percentile 19 min

90
30
80

25 70

Cumulative Percent
60
No. of Samples

80th percentile 24 min

20

50

15

Downloaded from https://academic.oup.com/ajcp/article/125/1/16/1759625 by guest on 06 May 2021

40
90th percentile 29 min

10 30

Samples
Cumulative % 20
5
10

0 0
0:00
0:02
0:04
0:06
0:08
0:10
0:12
0:14
0:16
0:18
0:20
0:22
0:24
0:26
0:28
0:30
0:32
0:34
0:36
0:38
0:40
0:42
0:44
0:46
0:48
0:50
0:52
0:54
0:56
0:58
1:00
Cycle Time (Minutes After Accessioning)

❚Figure 4❚ Preanalytic process cycle time (lean production system).

20

15

10
Performance Index

–5

–10

–15

–20
13

14

15

16

19

20

21

26

27

28

29

30
7

9
6

ov

ov

ov

ov

ov

ov

ov

ov

ov

ov

ov

ov

ov
v

ov
No
No

Date

❚Figure 5❚ Performance index, November 2001 (before lean production).

22 Am J Clin Pathol 2006;125:16-25 © American Society for Clinical Pathology

22 DOI: 10.1309/865V7UMFPUKGCF8D
 Clinical Chemistry / ORIGINAL ARTICLE

20

15

10
Performance Index

Downloaded from https://academic.oup.com/ajcp/article/125/1/16/1759625 by guest on 06 May 2021

–5

–10

–15

–20
11

12

13

14

15

16

19

20

25

26

27
1

7
v

v
No

No

No

No

No

No

No

No

No

No

No

No

No

No

No

No
Date

❚Figure 6❚ Performance index, November 2002 (lean production system).

4
Performance Index

–2

–4

–6

–8
02

02
2

3
02

02
02

02

02

03

03
01

02

02

02

03

03
01

02
0

0
l

ct
b

b
ov

ay

ay
n

ov

n
ec

ec
ar

pr

ar

pr
Ju
Fe

Fe
Ja

Ja
Ju

Au

Se

O
A

A
M

M
D

M
M
N

Month

❚Figure 7❚ Monthly performance index average.

© American Society for Clinical Pathology Am J Clin Pathol 2006;125:16-25 23

23 DOI: 10.1309/865V7UMFPUKGCF8D 23
Persoon et al / IMPROVING PREANALYTIC PROCESSES
❚Table 3❚ lists examples of the kinds of waste eliminated
with the lean process. Reduction of mislabeled and missing
tube errors was an additional benefit of the lean process.
Under the old process, the batching and unbatching of tubes
occasionally led to missing tubes, and laboratorians would
spend considerable time locating the missing tubes. With the
new lean process and 1-piece flow, there are few opportuni-
ties for a tube to be lost or misplaced, and these errors no
longer occur.
The redesign of the preanalytic process also revealed
some hidden wastes in the system. In 1 clinic, the phle-
botomists had developed an unwritten and unapproved prac-
tice of obtaining an extra tube of blood from most patients in
case the physician decided to order additional tests after the
patient had left the clinic. Investigation revealed that only
10.8% of these tubes actually were used, and in nearly half of
those cases, the additional tests could be done on specimens
already in the laboratory. Changing the practice to obtain extra
tubes only when specifically requested by the physician result-
ed in more than 1,000 fewer specimens being processed per
month, with a concomitant decrease in the number of collec-
tion tubes used and amount of biohazardous waste discarded.
Discussion
We demonstrated that redesigning the preanalytic
processes to incorporate the elements of lean production and
the Toyota Production System can result in significant
improvement in chemistry test turnaround time without the
addition of automation or other resources. Laboratories that
desire to improve their preanalytic processes but are not able
to acquire preanalytic automation systems may benefit from
this approach.
❚Table 3❚
Types of Waste Eliminated by Using the Lean Process
Original Process
Before accessioning, specimens were separated from
requisitions and sorted into racks by priority. Tests
were accessioned in batches from the sorted
requisitions. Barcode labels were printed for an
accession batch, and the tube, requisition, and
barcode label were rematched before the barcode
label was affixed to the tube. Tubes occasionally were
misplaced, requiring additional search time.
Processing of specimens was delayed when specimens
were received accompanied by a requisition with
incomplete or incorrect information. The accessioning
process would be halted while the accessioner called
the originating location to obtain the missing information.
Specimens that had been processed were stored in racks
or baskets and were transported to the analytic work-
stations when someone had the time or when the
laboratorian operating the workstation came to get them.
24 Am J Clin Pathol 2006;125:16-25
24 DOI: 10.1309/865V7UMFPUKGCF8D
A key element in our success is the use of the Toyota
Production System rules in the implementation of the
redesigned process. All 4 rules impose a discipline on the
process that results in consistency of output and quality.
Convincing laboratorians that they needed to maintain this
discipline was a major part of the change effort involved in the
implementation of the lean process.
In addition to the process changes, the lean production
process required a paradigm shift by laboratorians. Incoming
workload would no longer be classified as stat or routine,
because 1-piece flow and the lean process eliminated any dis-
tinction in the handling of these samples. It is interesting to
Downloaded from https://academic.oup.com/ajcp/article/125/1/16/1759625 by guest on 06 May 2021
note that the stat vs routine paradigm is still widely accepted
in the clinical laboratory industry. Mohammad et al10
described the use of simulation modeling to determine the
impact of increasing numbers of routine samples on routine
and stat turnaround times for a chemistry analyzer. Their study
presumed that there is a constant demand for stat testing and
examined the effect of increasing routine workload on stat
turnaround time. In our experience, requests for rapid turn-
around time are increasing, requiring laboratories to improve
their processes for these samples. Carried to its logical conclu-
sion, this trend will result in no “routine” testing. Developing
a process that improves turnaround time for all samples is a
better long-term strategy for delighting laboratory customers
than categorizing requests as stat and routine and trying to
improve service for only 1 category.
Review of laboratory processes with the goal of improv-
ing them inevitably reveals wasteful practices, as we discov-
ered when we looked at the use of extra tubes of blood
obtained in 1 clinic. The informal initiation of this practice by
front-line workers was well-intentioned—they wanted to
reduce the number of patient call-backs and additional phle-
botomies on a population of very sick patients—but it had the
Revised Process Wastes Reduced
Tests are accessioned for each specimen/requisition set, Motion; waiting;
and the barcode label is printed and affixed to the inspection;
specimens. There is no sorting or batching during the correction
accessioning process.
Specimens arriving with incomplete or incorrect Motion; waiting;
information are routed to an exception handler, inspection;
reducing the delay for specimens that have complete correction
and correct information.
A laboratorian is designated as the sorter-circulator Waiting;
whose job includes delivering specimens from the motion
specimen processing area to the analytic workstations
on a regular basis.
© American Society for Clinical Pathology

effect of reducing system throughput by adding additional
specimens to the workload that ultimately were never used. As
our efforts to improve laboratory service continued beyond the
time described in this article, we discovered additional oppor-
tunities to reduce waste by changing practices. This also is
consistent with the application of the Toyota Production
System, in which every employee participates in continuous
improvement efforts.
From the Departments of 1Pathology and 2Mechanical and
Industrial Engineering, the University of Iowa, Iowa City.
Address reprint requests to Mr Persoon: Dept of Pathology,
the University of Iowa, 200 Hawkins Dr, Room 6233 RCP, Iowa
City, IA 52242-1009.
10. Mohammad AA, Elefano EC, Leigh D, et al. Use of computer
References
1. Howanitz JH, Howanitz PJ. Timeliness as a quality attribute
and strategy. Am J Clin Pathol. 2001;116:311-315.
2. Kalra J. Medical errors: an introduction to concepts. Clin
Biochem. 2004;37:1043-1051.
3. Ohno T. Toyota Production System: Beyond Large Scale
Production. New York, NY: Productivity Press; 1988.
© American Society for Clinical Pathology
Clinical Chemistry / ORIGINAL ARTICLE
4. Jimmerson C, Weber D, Sobek DK II. Reducing waste and
errors: piloting lean principles at Intermountain Healthcare. Jt
Comm J Qual Patient Saf. 2005;31:249-257.
5. Tragardh B, Lindberg K. Curing a meager health care system
by lean methods: translating “chains of care” in the Swedish
health care sector. Int J Health Plann Manage. 2004;19:383-
398.
6. Condel JL, Sharbaugh DT, Raab SS. Error-free pathology:
applying lean production methods to anatomic pathology. Clin
Lab Med. 2004;24:865-899.
7. Thompson DN, Wolf GA, Spear SJ. Driving improvement in
patient care: lessons from Toyota. J Nurs Adm. 2003;33:585-
595.
8. Bahensky JA, Roe J, Bolton R. Lean Sigma: will it work for
healthcare? J Healthc Inf Manage. 2005;19:39-44.
Downloaded from https://academic.oup.com/ajcp/article/125/1/16/1759625 by guest on 06 May 2021
9. Spear SJ, Bowen HK. Decoding the DNA of the Toyota
Production System. Harv Bus Rev. 1999;77:96-106.
simulation to study impact of increasing routine test
turnaround times of stat samples on ci8200 integrated
chemistry and immunoassay analyzer. Clin Chem.
2004;50:1952-1955.
Am J Clin Pathol 2006;125:16-25 25
25 DOI: 10.1309/865V7UMFPUKGCF8D 25