Content uploaded by Petr Jarolim
Author content
All content in this area was uploaded by Petr Jarolim on Feb 28, 2014
Content may be subject to copyright.
Feature
How Laboratory Automation Can Help
Laboratories, Clinicians, and Patients
Stacy E.F. Melanson, MD, PhD, Neal I. Lindeman, MD, Petr Jarolim, MD, PhD
(Department of Pathology, Division of Clinical Laboratories, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA)
DOI: 10.1309/YK4WXG5T1P8UCNG4
labmedicine.com March 2008
Volume 39 Number 3
LABMEDICINE 137
Feature
138
LABMEDICINE
Volume 39 Number 3
March 2008 labmedicine.com
aboratories of varying sizes are considering partial
or total laboratory automation due to labor short-
ages and demands for process improvements and
cost savings. The cost of automation, previously
a significant deterrent to widespread implementa-
tion, has decreased to current levels that enable
automation to be a viable option for many labora-
tories that would never have considered it a few years ago. Bene-
fits of automation can be realized in the preanalytical, analytical,
and post-analytical phases of processing and include a safer
workplace for employees, decreased errors, more consistent turn-
around times, increased productivity, and enhanced flexibility
for resource reallocation. Using a series of vignettes, this review
will demonstrate several ways by which laboratory automation
can benefit laboratories, clinicians, and patients.
Automation in the Laboratory
The laboratory and its interaction with clinicians during
specimen processing, analysis, and storage are critical compo-
nents of effective patient care in all health care settings. Improve-
ments in the clinical laboratory workflow and the timeliness
and accuracy of results
reporting will not only
benefit the laboratory but
can also translate into
increased clinician and
patient satisfaction with
laboratory services. Labo-
ratory automation has
become a popular option
to improve internal and
external laboratory
services for both the em-
ployees and their clients.
Automation designs
and systems are available
for laboratories of various
sizes and needs, from pre-
analytical modules to total
laboratory automation.
Updates and improve-
ments are constantly
being made and released
to the market, and while
specific details on each
vendor’s products will
not be discussed here, a
comprehensive tabular
overview is published
yearly in CAP Today.
1
Once a laboratory consid-
ers automation, selecting
the appropriate products
is a complex and time-
consuming process. In a
recent report,
2
we discussed essential features of automation,
presented our automation and instrument selection process,
emphasized the importance of site visits, and suggested site
visit questions and evaluation criteria.
Automation offers a variety of benefits that range from
decreasing the number of manual, error-prone steps to increasing
operational capacity and additional revenue for the laboratory.
Although several articles provided metrics to assess the benefits
of automation,
3-8
in the course of our selection process we found
that concrete examples were helpful, especially for laboratory
personnel who would be operating the equipment on a daily
basis. Here, several scenarios will be used to illustrate ways that
automation can improve the laboratory, as both an element of
patient care and a workplace. Although each of these examples
may not apply to every laboratory, as a whole they represent a
breadth of scenarios that may occur in each individual laboratory,
depending upon its size, specific needs, and information system.
Each laboratory must make an individualized assessment of the
areas in which these benefits can be realized.
Vignette One
M.L. is an experienced laboratory technician who has been
receiving and processing specimens for 15 years. Over the past few
months she has noticed numbness and tingling in her hand, which
has caused considerable discomfort and decreased her productivity.
After meeting with a nurse from the occupational health services and
her primary care phy-
sician, M.L.
is told she has carpal
tunnel syndrome, most
likely as a result of the
repetitive motion in
her job. Her physi-
cian suggests several
remedies, including
physical therapy, an
immobilizing brace,
hand surgery, or a
transfer to a new
position that doesn’t
require extensive
repetitive motion.
Repetitive
motion injuries and
other occupational
hazards are common
in laboratories that
process specimens
manually. Similar to
M.L., many techni-
cians suffer from
work-related repeti-
tive motion injuries
such as carpal tunnel
syndrome. Large
hospital laboratories
may process hun-
dreds of specimens
per hour during peak
times and thousands
of specimens per day. These specimens need to be accessioned
into the laboratory information system, labeled, centrifuged,
decapped, aliquoted, and recapped, all of which involve repet-
itive hand motion. Specimens also need to be transported to
various sections of the laboratory. A workflow analysis performed
L
M.L. is told she has carpal tunnel
syndrome, most likely as a result of
the repetitive motion in her job. Her
physician suggests several remedies,
including physical therapy, an
immobilizing brace, hand surgery,
or a transfer to a new position.
Feature
labmedicine.com March 2008
Volume 39 Number 3
LABMEDICINE 139
in our laboratory found that a technologist traveled an average of
1,025 feet over a 30-minute period, which may potentially lead
to work-related injuries. In addition, employees are at risk for
biohazard exposure from handling body fluids during specimen
processing. In one study, Hawker and colleagues reported that
each tube was handled by at least 10 individuals, magnifying the
biohazard risk.
3
Automation of manual and potentially dangerous steps can
decrease repetitive motions, biohazard exposure, and walking
traffic in the laboratory. With an adequate information tech-
nology infrastructure, specimens can be placed directly on the
automation track, eliminating manual accessioning and relabel-
ing. Some automation systems accommodate a variety of tube
sizes, further reducing manual aliquoting, and obviating the
need for the standardization of collection tubes. The specimens
can be automatically centrifuged prior to analysis, using one or
more centrifuges depending on both the type and volume of
testing. Most automation lines have a decapper and a recapper
or resealer, while others offer closed-tube sampling. Both options
result in fewer repetitive motions for technical staff. Our work-
flow analysis and visits to automated laboratories suggested that
automation can decrease the number of manual steps by at least
50% and decrease the
number of manipula-
tions by at least 75%.
One study reported
that the number of
biohazard exposure
events decreased from
2,658 per month to
6 per month after
implementation of
automation.
6
The principal
drawback to auto-
mated prenanalytical
specimen processing
and transportation is
the inability of tech-
nologists to visually
inspect samples for
potential interferences
(ie, lipemia, hemo-
lysis, icterus) prior
to analysis. Most
modern analytical
systems detect these
interferences prior to
analysis, but, in our
opinion, the quality
of these on-board
interference checks
is variable, and this
feature is more critical
in automated than in
nonautomated
systems.
Vignette Two
R.D. was recently hired as a laboratory technician for specimen
processing in a laboratory that prepares hundreds of manual aliquots
per day. Her primary responsibility is to make and label aliquots
for hepatitis and electrophoresis testing. One week after R.D. was
hired, her supervisor received several questions regarding unexpected
positive results for hepatitis. When she reviewed the incident reports
for the last week, she found several additional complaints from
clinicians and other laboratory staff, all apparently related to
specimens processed by R.D.
The laboratory for which R.D. works prepares a significant
number of aliquots per day. Preparation of aliquots is associated
with an increased risk for errors in specimen sorting, routing,
pouring, and labeling. R.D. made several of these errors during
her first week of employment. During decapping of specimens,
a splash of blood from a patient with active hepatitis B infection
contaminated tubes from patients without hepatitis, causing
false-positive hepatitis B surface antigen results to be reported.
This not only presented an occupational health risk to R.D.,
it also caused significant stress for several patients and inconve-
nience for their physicians, as well as an unnecessary administra-
tive burden for the state public health department. R.D. also
mislabeled several
aliquot tubes, causing
results to be associated
with the wrong patients,
and she lost several
aliquots, necessitating
that patients return
to phlebotomy, some
from out of state, to
be redrawn. Although
employees are trained
on proper techniques,
manual processes are,
by their nature, prone
to errors such as these.
Reduction in the num-
ber and extent of errors
is extremely important
for patient safety and
adequate health care
delivery. Recently, the
Centers for Medicare
and Medicaid Services
(CMS) emphasized the
importance of error re-
duction by announcing
that they will no longer
reimburse for hospital
mistakes.
9,10
Automation can
minimize preanalytical
processing errors,
resulting in improved
accuracy of reported
results and fewer patient redraws. In addition, automation
systems can locate and track specimens throughout the
preanalytical, analytical, and post-analytical processes, reducing
the chances of losing a sample. A study investigating the
implementation of automation at a large reference laboratory
One week after R.D. was hired,
her supervisor received several questions
regarding unexpected positive results
for hepatitis…apparently related to
specimens processed by R.D.
Feature
140
LABMEDICINE
Volume 39 Number 3
March 2008 labmedicine.com
illustrated a 58% reduction in the number of lost specimens.
4
Most manufacturers provide an aliquoting module that can
make several aliquots of predetermined volumes. Prior to
aliquoting, automatic clot detection and liquid level sensing are
used to optimize specimen integrity and ensure that the correct
volume is dispensed. In addition, aliquots can be labeled and
sorted for easier distribution to other areas of the laboratory or
to a reference laboratory. Some systems can dispense aliquots
into nonstandard tube types that may be required by certain
analyzers. Implementation of preanalytical automation at
Hershey Medical Center decreased specimen sorting and rout-
ing errors by 95% and decreased specimen pour-off and labeling
errors by more than 98%.
6
A potential drawback to relying upon an automated sys-
tem to identify, aliquot, and route samples is that it requires a
capable information system. Information systems that assign the
same identification number to all samples collected at the same
time from the same patient cannot distinguish among the
different sample types (eg, heparinized versus EDTA blood)
and cannot, therefore, aliquot or route the samples properly.
To truly realize this benefit of automation, a laboratory
information system that assigns a unique specimen identifier
to each individual tube of
blood is essential.
In addition, auto-
mated sample handling
requires clear, uniform,
legible bar codes to be
applied to the samples.
Where a technician can
recognize—and handle—
a label that is applied
backwards, or is applied
too low on the tube, or
is wrinkled, or stained,
some automation systems
may fail to recognize
these samples properly.
Ideally, these samples
would be routed to an
“error” area for manual
intervention.
Vignette Three
M.N. is the
supervisor for the high-
throughput clinical
chemistry analyzers.
At the weekly staff
meetings, she frequently
receives complaints about
how time consuming it
is to retrieve specimens
for add-on testing. M.N.
investigates the complaints
further and determines that her laboratory performs hundreds of
add-on tests per week and estimates that this testing requires 1 to 2
full-time equivalents (FTEs).
Add-on tests are tests that are requested by clinicians to be
performed on specimens that have already been tested for other
analytes. Add-on testing is an inefficient process that requires
a technologist to search for individual specimens in racks of
hundreds or thousands of other specimens. As a result, add-on
testing consumes a disproportionate number of FTEs when
compared with routine testing.
11
A study of 2 large academic
centers found that both laboratories performed a considerable
number (approximately 300 and 500 per week) of add-on tests
and confirmed the inefficiency of add-on testing.
12
Among the
authors’ solutions to decrease add-on testing were automation,
improvements in information technology and systematic
implementation of guidelines.
Refrigerated sample storage is available from several
automation vendors. Once specimens are analyzed, they are
transported and archived in the storage unit. When an add-on
test is requested, the system can automatically retrieve the
specimen and send it for analysis, without a technologist’s
intervention. Automatic repeats and dilutions can also be
performed in this manner. Evaporation of the specimen is
prevented by either resealing the tube or using a cap-piercing
technology that does not remove the cap during specimen sam-
pling. One vendor offers laser technology that can record the
volume remaining in the specimen prior to storage, which en-
ables the system
to notify the tech-
nologist whether or
not enough speci-
men is available for
the additional test to
be performed. Auto-
mation systems allow
add-on tests to be
managed from the
central workstation.
Consequently, less
time is spent
searching for the
specimens. For all
these reasons,
analytical and
post-analytical
automation
eliminates most
inefficiencies
associated with
add-on testing and
markedly reduces
labor requirements
associated with
add-on testing
in most hospital
laboratories.
Vignette Four
J.K. is an
emergency
department (ED) clinician who directs the cardiac triage unit.
He needs to provide his administrators with data on the manage-
ment of patients with suspected acute coronary syndrome (ACS) and
his strategy for reducing their length of stay. Troponin results are a
critical component for the diagnosis of suspected ACS in the cardiac
M.N.’s laboratory performs
hundreds of add-on tests per week
and estimates that this testing requires
1 to 2 full time equivalents.
Feature
labmedicine.com March 2008
Volume 39 Number 3
LABMEDICINE 141
unit, and J.K. believes that timely results can reduce the length
of stay. J.K. frequently calls the laboratory supervisor, in
frustration, when a troponin result takes over an hour to be
reported, and he blames the technologist for delaying the diagnosis
of ACS and increasing the length of stay. The supervisor reminds
him that the average turnaround time for troponin is 55 minutes,
but J.K. tells him that he sees significant delays in result reporting
every week.
Turnaround time (TAT) is critical for many laboratory
tests, as the results are necessary for rapid and accurate patient
diagnosis and management. The need for a TAT of less than 60
minutes for troponin testing, to aid in the diagnosis of ACS, is a
standard of care for cardiology, emergency medicine, and clinical
laboratory practice.
13
Although the laboratory described above
has an acceptable average TAT of 55 minutes, it
frequently has outlier samples with TATs greater than this
mean value, which are the source of complaints from physicians.
Problems that increase TATs include delays in specimen
processing, delays in transport to the analyzer, and technical
delays with the analyzer. Furthermore, the workflow in many
clinical laboratories promotes batch processing, which leads to
frequent periods when specimens are sitting idle. An internal
workflow evaluation
performed by our
laboratory showed
that from arrival in
the laboratory to
reporting of results,
specimens were
stationary, without
any processes
being carried out,
approximately
40% of the time.
Most laborato-
ries that have
implemented
automation have
reported an
improvement in
TATs,
4,14,15
and
all have seen more
consistent TATs
with considerably
fewer outliers.
5
This translates into
increased patient and
physician satisfaction
with laboratory
services and fewer
calls to the laboratory
regarding outliers.
Holland and
colleagues found that
laboratory automa-
tion provided more
consistent TATs and eliminated the laboratory as a factor in
the ED length of stay, which would be helpful for clinicians
like J.K.
5
Several features of automation systems contribute to
improved or more consistent TATs. Automation optimizes
specimen processing by supporting a continuous flow of
specimens and reducing the time specimens spend waiting to be
placed into the centrifuge or aliquoted. This optimization allows
both routine and stat tests to be reported in a timely manner
and decreases the need for priority stat testing. The analytical
phase is also improved, as the detection modules and software
will alert the technologist to any interferences without delaying
the processing of other samples and will automatically divert
specimens to another analyzer when one analyzer has technical
problems. Finally, several vendors offer a post-analytical storage
unit or tray, enabling quick and efficient retrieval for automatic
dilutions, repeat analysis, or add-on testing.
Vignette Five
G.K. is an experienced medical technologist with board
certification and 25 years of experience who has become frustrated
that the majority of her time is spent performing simple tasks rather
than the more interesting work she did earlier in her career as a
medical technologist. She remembers labor-intensive assays and
intellectually challenging procedures that required extensive training
and education, and she misses this component of clinical chemistry.
Many technologists
who trained in clinical
chemistry began their
careers performing
manual procedures, such
as radioimmunoassay and
chromatography, and
enjoyed the intellectual
and labor requirements
of these procedures.
The field has evolved
to include many high-
throughput automated
analyzers. Medical
technologists are still
essential for result
interpretation and
verification; however,
most of the labor-
intensive procedures
have been replaced with
analyzer maintenance,
specimen sorting,
transport, loading,
and unloading. While
this may improve TATs
and reduce errors,
technologists may lose
their professional skills
as they spend more
time performing mun-
dane tasks.
8
These tasks
can be intellectually
unsatisfying, decreasing
morale and productivity. Workflow analysis in our laboratory
illustrated that technologists working on the large-volume chem-
istry analyzers spend 42% of their time loading and unloading
specimens and 15% of their time waiting for specimens to arrive
for analysis.
G.K. remembers labor-intensive assays
and intellectually challenging procedures
that required extensive training and
education, and she misses this
component of clinical chemistry.
Feature
142
LABMEDICINE
Volume 39 Number 3
March 2008 labmedicine.com
Laboratory automation, by automating manual processes,
reduces the need for technologists to perform simple steps. This
allows the laboratory to maximize scarce resources and better
match job assignments to skills. In one study, automation of a
medium-sized laboratory created an opportunity to reallocate
resources such that routine work could now be done by 1 FTE
assigned to load samples, freeing 4 FTEs to verify results and
deal with complex issues, and a sixth FTE to manage the other
technologists.
13
Many employees may fear that automation will replace
their function in the laboratory; however, the technologists’
time saved by automation is precious and can be used at more
manual workstations, or to bring new tests or technologies into
the laboratory. Productivity, as measured by the number of tests
per employee, is also increased by automation systems.
7
From
one central location, one technologist can manage the system
and be alerted to any interferences, clots, low sample volumes, or
technical issues. During the majority of troubleshooting, results
continue to be reported, further increasing the number of tests
per employee.
Vignette Six
T.D. is in charge
of the clinical laboratory
budget and finances. Each
fiscal year he is asked to
supply, with the help of
the laboratory directors,
detailed plans for cost
savings and increased
revenue. This year the
request for savings is
significantly more
burdensome than in
previous years. The usual
cost-savings solutions are
not sufficient and the
directors of the laboratory
are struggling to find more
substantial savings.
When solicited for
ideas to reduce the budget
or increase the revenue,
many laboratory directors
begin with contracts that
can be renegotiated or
tests that can be brought
in house at a lower cost;
however, executing
these two scenarios is
not straightforward. To
maximize cost savings, no
additional FTEs should
be requested when in-
sourcing tests. In addition,
the new assays should
ideally be FDA-approved
and implementable on
automated analyzers for
maximum efficiency.
Automation offers several potential opportunities for
cost savings and increased revenue.
4,7,15
Cost savings can
be generated both within the laboratory and in other depart-
ments. Because fewer laboratory FTEs are required to run the
automation, FTEs can be transferred to different areas of the
laboratory. This will not only alleviate the stress placed on the
laboratory director to find medical technologists but will also
give technologists an opportunity to use new technologies
and perform new tests. The reduction in errors and improved
efficiency of add-on testing with automation will also save
time and the cost of employing technologists to troubleshoot
errors and find specimens. More consistent TATs can translate
into decreased ED lengths of stay and cost savings for the
hospital. Automation can expand the throughput capability
of the laboratory, enabling expansion of outreach programs,
thereby generating additional revenue.
These potential cost savings must be balanced against
the cost of the automation systems themselves. While total
laboratory automation has historically been a costly endeavor,
we found that the price has decreased considerably, and we
received favorable proposals for automation systems from several
vendors. The overall financial impact of automation projects,
however, will be dif-
ferent for each labo-
ratory and the extent
to which it becomes
automated.
Conclusion
Automation can
offer a variety
of benefits, includ-
ing improvements
in preanalytical
processing efficiency
and errors, reduc-
tion in occupational
risks and biohazard
exposures, answers to
labor shortages, and
potential revenue
generation. The
improvements in
laboratory operations
can translate into
better laboratory
services, improved
patient care, and
increased physician
satisfaction.
LM
T.D. is in charge of the clinical
laboratory budget.... This year...
the usual cost savings solutions are not
sufficient and the directors of the
laboratory are struggling to find more
substantial savings.
Feature
labmedicine.com March 2008
Volume 39 Number 3
LABMEDICINE 143
1. Laboratory automation systems and workcells. CAP Today. 2007;21:30–48.
2. Melanson SE, Lindeman NI, Jarolim P. Selecting automation for the clinical
chemistry laboratory. Arch Pathol Lab Med. 2007;131:1063–1069.
3. Hawker CD, Garr SB, Hamilton LT, et al. Automated transport and sorting
system in a large reference laboratory: Part 1. Evaluation of needs and
alternatives and development of a plan. Clin Chem. 2002;48:1751–1760.
4. Hawker CD, Roberts WL, Garr SB, et al. Automated transport and sorting
system in a large reference laboratory: Part 2. Implementation of the system
and performance measures over three years. Clin Chem. 2002;48:1761–1767.
5. Holland LL, Smith LL, Blick KE. Total laboratory automation can help
eliminate the laboratory as a factor in emergency department length of stay.
Am J Clin Pathol. 2006;125:765–770.
6. Holman JW, Mifflin TE, Felder RA, et al. Evaluation of an automated
preanalytical robotic workstation at two academic health centers. Clin Chem.
2002;48:540–548.
7. Sarkozi L, Simson E, Ramanathan L. The effects of total laboratory
automation on the management of a clinical chemistry laboratory.
Retrospective analysis of 36 years. Clin Chim Acta. 2003;329:89–94.
8. Tatsumi N, Okuda K, Tsuda I. A new direction in automated laboratory
testing in Japan: Five years of experience with total laboratory automation
system management. Clin Chim Acta. 1999;290:93–108.
9. CMS. CMS-1533-FC. Available at: http://www.cms.hhs.gov/center/hospital.
asp. Accessed on: 9/18/2007.
10. Marlantes L. Medicare won’t cover hospital mistakes. Available at: http://www.
abcnews.go.com/WN/story?id=3499053. Accessed on: 9/5/2007.
11. Melanson SF, Hsieh B, Flood JG, et al. Evaluation of add-on testing in the
clinical chemistry laboratory of a large academic medical center: Operational
considerations. Arch Pathol Lab Med. 2004;128:885–889.
12. Melanson SE, Flood J, Lewandrowski K. Add-on testing in the clinical
laboratory: Observations from two large academic medical centers. Lab Med.
2006;37:675–678.
13. Apple FS, Jesse RL, Newby LK, et al. National Academy of Clinical
Biochemistry and IFCC Committee for Standardization of Markers of
Cardiac Damage Laboratory Medicine Practice Guidelines: Analytical
issues for biochemical markers of acute coronary syndromes. Clin Chem.
2007;53:547–551.
14. Markin RS, Whalen SA. Laboratory automation: Trajectory, technology, and
tactics. Clin Chem. 2000;46:764–771.
15. Seaberg RS, Stallone RO, Statland BE. The role of total laboratory automation
in a consolidated laboratory network. Clin Chem. 2000;46:751–766.
To the right is an image of a field of cells
in a conventional Pap smear
that has been Pap stained
(photomicrograph x400).
What are the artifacts
to which the 4 arrows
are pointing?
Send your guesses by e-mail to
LABMEDICINE at labmed@ascp.org
with “Cytology Teaser” in the subject line.
Winners will be drawn from all correct
answers and announced in the May issue.
We’re looking for reader contributions!
If you have images you’d like to submit for a Teaser,
please send them to labmed@ascp.org.
This month’s Cytology Teaser was generously
submitted by Gary W. Gill, CT(ASCP),
Cytology Consultant, Indianapolis, IN.
Cytology Teaser
