Patient Safety in the Clinical Laboratory

A Longitudinal Analysis of Specimen Identification Errors

Elizabeth A. Wagar, MD; Lorraine Tamashiro, BS; Bushra Yasin, PhD; Lee Hilborne, MD; David A. Bruckner, ScD

● Context.—Patient safety is an increasingly visible and im- of an automated processing system (14 months) for a 24-
portant mission for clinical laboratories. Attention to im- month period, using trend analysis and Student t test sta-
proving processes related to patient identification and tistics.
specimen labeling is being paid by accreditation and reg- Results.—Of 16 632 total specimen errors, mislabeled
ulatory organizations because errors in these areas that specimens, requisition mismatches, and unlabeled speci-
jeopardize patient safety are common and avoidable mens represented 1.0%, 6.3%, and 4.6% of errors, re-
through improvement in the total testing process. spectively. Student t test showed a significant decrease in
Objective.—To assess patient identification and speci- the most serious error, mislabeled specimens (P ⬍ .001)
men labeling improvement after multiple implementation when compared to before implementation of the 3 patient
projects using longitudinal statistical tools. safety projects. Trend analysis demonstrated decreases in
Design.—Specimen errors were categorized by a multi-
all 3 error types for 26 months.
disciplinary health care team. Patient identification errors
were grouped into 3 categories: (1) specimen/requisition Conclusions. —Applying performance-improvement
mismatch, (2) unlabeled specimens, and (3) mislabeled strategies that focus longitudinally on specimen labeling
specimens. Specimens with these types of identification er- errors can significantly reduce errors, therefore improving
rors were compared preimplementation and postimple- patient safety. This is an important area in which laboratory
mentation for 3 patient safety projects: (1) reorganization professionals, working in interdisciplinary teams, can im-
of phlebotomy (4 months); (2) introduction of an electron- prove safety and outcomes of care.
ic event reporting system (10 months); and (3) activation (Arch Pathol Lab Med. 2006;130:1662–1668)

I mproving the safe care of patients has recently evoked

national attention. Although recognized for centuries,
patient safety first made major headlines in 2000 with the
Accurate specimen identification is a challenge in all hos-
pitals, and a mislabeled specimen can lead to devastating
consequences for patients. In an effort to decrease the risk
publication of ‘‘To Err Is Human: Building a Safer Health of potential harm caused by labeling errors, some hospi-
System.’’ 1 This extensive review recognized earlier re- tals have implemented a zero tolerance laboratory speci-
search showing that preventable adverse events leading to men labeling process.5
death occur as frequently as 44 000 to 98 000 times per Error rates in laboratory practices are routinely collected
year, according to 2 studies in New York and Colorado for a number of performance measures in clinical pathol-
and Utah.2–4 Patient safety is affected by the frequency and ogy laboratories in the United States; however, a critical
seriousness of errors that occur in the health care system. performance measures list has not yet been recommended.
The most extensive databases describing error rates in pa-
thology were developed and are maintained by the Col-
Accepted for publication April 6, 2006. lege of American Pathologists. These databases include the
From the University of California, Los Angeles, Clinical Laboratories, College of American Pathologists’ Q-Probes and Q-Tracks
Department of Pathology and Laboratory Medicine (Dr Wagar); Sutter programs, which provide information on error rates from
Medical Center of Sacramento, Clinical Lab, Sacramento, Calif (Ms
Tamashiro); and University of California, Los Angeles, Pathology and more than 130 interlaboratory studies.6
Laboratory Medicine (Drs Hilborne and Bruchner). A recent article identified 17 types of errors associated
The authors have no relevant financial interest in the products or with invasive procedures, 10 of which were directly relat-
companies described in this article. ed to failure to verify the patient’s identity.7 The Joint
A summation of this data was presented as a speaker presentation Commission for the Accreditation of Healthcare Organi-
at the Institute for Quality in Laboratory Medicine Conference, Atlanta,
Ga, April 30, 2005. zations (JCAHO) standard for ‘‘improving the accuracy of
Presented at the Pittsburgh Patient Safety Meeting, Pittsburgh, Pa, patient identification’’ was accepted as 1 of 6 National Pa-
May 18, 2006 and American Association for Clinical Chemistry, Lab- tient Safety Goals for 2003, shortly thereafter. Patient iden-
oratory Automation 2006, St Louis, Mo, November 9, 2006. tification continues as a major priority for the JCAHO lab-
Reprints: Elizabeth A. Wagar, MD, University of California, Los An- oratory patient safety goals for 2006.8
geles, Clinical Laboratories, Department of Pathology and Laboratory
Medicine, David Geffen School of Medicine at UCLA, Box 951732, A key clinical care delivery concept is the emphasis on
AL-206 CHS 10833 Le Conte Ave, Los Angeles, CA 90095-1732 ‘‘patient-centeredness.’’ 9 Clinical laboratories are, in many
(e-mail: ewagar@mednet.ucla.edu). ways, models of ‘‘patient-centered’’ services. Laboratory
1662 Arch Pathol Lab Med—Vol 130, November 2006 Patient Safety and Specimen Identification—Wagar et al
results provide diagnostic clues, assist in therapeutic de- Specimen Error Information
cision-making, and determine clinical progress for multi-
Specimen Error
ple disease entities. The clinical laboratory is also an area
Category Definition
where errors in patient identification can potentially result
in significant harm. In a typical hospital clinical labora- Clotted specimen A specimen received that otherwise
tory, thousands of specimens are received and analyzed should have been received in an un-
clotted state
daily. These specimens require correct identification of the Container leaking A specimen received that is leaking or
patient from phlebotomy through reporting of results to otherwise has breached the integrity
the requesting health care provider. of the container
Recognizing the importance of clinical laboratories to Contaminated A specimen that otherwise should be
patient safety, the Institute for Quality in Laboratory Med- received using sterile collection tech-
icine was organized in 2003 and incorporated in 2005, niques that has obvious contamina-
tion through handling or appearance
with the assistance of the Centers for Disease Control and Duplicate order A specimen received after a first speci-
Prevention. The College of American Pathologists recog- men has been received within a time
nized the importance of specimen and patient identifica- frame established by the laboratory
tion by initiating a series of quality measurement tools for when repeat specimens are not typi-
clinical laboratories; for example, a 2-year Q-Track study cally required for patient evaluation
of continuous wristband monitoring that demonstrated a Hemolyzed speci- A specimen received that has evident
men hemolysis of red blood cells as evi-
decrease, 7.4% to 3.05%, in wristband errors from volun- denced by pink/red discoloration of
tarily participating institutions. Increased attention to and the serum or plasma
awareness of these errors likely contributed to the decline Improperly collected A specimen received that has been col-
during the 8 quarters of the study.10 specimen lected in the wrong container or tube
Despite this heightened attention and interest in patient required for the type of testing or-
safety in the clinical laboratory, many institutions are un- dered
Improperly handled A specimen received that may not have
sure of the best approach for reducing laboratory-related specimen been transported or handled in a
patient errors. Also, each institution faces unique longi- proper manner, for example, trans-
tudinal effects on patient safety projects related to shifting port at an inappropriate temperature
priorities at the local, regional, and national levels. Mislabeled speci- A specimen that is not labeled with ap-
men propriate patient identifiers
MATERIALS AND METHODS Quantity not suffi- A specimen received in a quantity not
cient (QNS) sufficient for proper testing
Beginning in November 2002, University of California, Los An- Requisition mis- A specimen received with a requisition
geles (UCLA) Clinical Laboratories initiated a study to determine match that does not match the request or
a baseline frequency for blood specimen errors. From this base- patient identified on the tube or con-
line data collection grew the full development of a specimen/ tainer
patient identification project examining 3 blood specimen iden- Specimen not re- Recording of an order, either electroni-
tification errors: (1) unlabeled specimens, (2) specimen/requisi- ceived cally or manually, not matched with
tion mismatch, and (3) mislabeled specimens. Error trends and an actual received specimen at the
statistical analyses were followed through the course of 3 patient time or subsequent to the order re-
safety interventions: (1) reorganization of phlebotomy services, ceipt
(2) implementation of an electronic event reporting system, and Specimen not suit- Some specimen types or tubes/contain-
(3) installation of an automated specimen processing system. able for test ers are not suitable for a given test
To assess patient identification and specimen labeling improve- Tube overfilled Blood collected in a tube that has been
ment, the faculty, nurses, and staff at UCLA established guide- overfilled, excluding some testing
lines and definitions to create 15 specimen error categories (Ta- Tube underfilled Blood collected in a tube that has been
ble). The UCLA Clinical Laboratories began collecting baseline underfilled, excluding some testing
specimen error information for each of these error categories from Unlabeled specimen A specimen received in the clinical
laboratory with no label or without
November 2002 to March 2003. Implementation data for total
2 identifiers on a label
specimen errors were subsequently followed from September 1,
2003 through August 31, 2005 (24 months). As the study pro-
gressed, it became apparent that the intensive care units (ICUs)
were an important variable because phlebotomy was performed around the clock (4 months); nursing in-service education was
both by nursing and laboratory personnel, thus increasing the performed for ICU nursing phlebotomy. Prior to the implemen-
chance of draw errors. When this difference was recognized (Jan- tation of 24-hour phlebotomy service, 13 to 16 phlebotomists per-
uary 1, 2004), data for ICU specimen errors were collected and formed early morning blood draws. There were no phlebotomists
tabulated alongside the total specimen error data. on the evening or night shifts. Twelve additional phlebotomists
Three specimen errors were identified as critical because of were hired to provide the 24-hour phlebotomy service. During
their direct relationship with patient identification. Specimen er- the early morning rounds (4:00 AM to 7:00 AM), 13 phlebotomists
rors in these 3 critical patient identification categories were spe- perform the blood draws. Four phlebotomists serve the day shift
cifically tabulated throughout the project implementation phase, hourly and timed draws, 4 work the evening shift, and 3 operate
during which more than 4.29 million specimens and 2.31 million on the night shift. Prior to implementation of the patient safety
phlebotomy requests were received. The 3 categories of critical projects, the nursing staff ordered all tests for the timed and
patient identification errors were (1) specimen/requisition mis- hourly draws just before they drew blood from the patient and
match, (2) mislabeled specimens (‘‘wrong blood in tube’’), and ensured that all orders were accounted for at the time. Under the
(3) unlabeled specimens (including specimens labeled with only revised process, nursing clerical staff enter orders into the hos-
1 of 2 required identifiers). pital information system for phlebotomy to draw blood from the
Three patient safety activities were introduced at 4, 10, and 14 patients.
months:
2. In 2003, at 10 months, the University of California complet-
1. The laboratory began providing phlebotomy services ed an effort to implement an online electronic event reporting
Arch Pathol Lab Med—Vol 130, November 2006 Patient Safety and Specimen Identification—Wagar et al 1663
Figure 1. Total blood draw errors by error
type (September 1, 2003–August 31, 2005).
Data for all specimen error types that result
in unacceptable specimens were tabulated
for the entire study period (total errors ⫽
16 632). Critical identification errors are not-
ed by red bars. QNS indicates quantity not
sufficient.

system at all 5 medical center campuses, including UCLA. The difference existed between the two under different conditions.
system was built from scratch, and initially developed as an early The paired t test is used to compare preintervention and postin-
risk management notification system at University of California- tervention measurements in a single group for a time period. It
Davis. The system was redesigned to incorporate important pa- is appropriate when the variable of interest is continuous and
tient safety concepts by a consortium of University of California approximately normally distributed. The paired t test is suitable
quality management, risk management, and performance im- for our study because it is a statistical test commonly used in
provement staff. This electronic event reporting system now al- situations in which subjects are tested in a pre-post situation, that
lows nurses, physicians, and other health care professionals to is, across time, with some intervention occurring such as the 3
easily report adverse events and near misses from any computer patient safety activities (phlebotomy services around the clock,
within their organization. Errors are entered into the Event Re- electronic event reporting system, and automated processing sys-
porting System for unlabeled, mislabeled, and specimen/requi- tem, introduced at 4, 10, and 14 months, respectively).
sition mismatch. Error information is taken from the Cancel Com- The second statistical tool applied to this study was linear
ment reports printed daily for issues that occurred the preceding trend analysis. Linear trend analysis (indicating a statistical
day. trend) was conducted on the collected data to allow us to plot
A unique feature compared to other event reporting systems aggregated response data related to time and determine if the
is the ability for each organization to customize the event re- interventions resulted in a statistical trend. This is especially
porting system to meet their specific performance improvement valuable when conducting a long-running survey and measuring
goals. Migration to an electronic event reporting system has re- differences in perception and responses for a period of time. If
sulted in a 2-fold to 3-fold increase in reported events compared an experiment contains a quantitative independent variable, then
with those that were reported using the previous paper system. the shape of the function relating the levels of this quantitative
independent variable to the dependent variable is of interest. To
3. Prior to the installation of the Beckman Coulter Power Pro-
apply this analysis to our study, we considered the effect of the
cessor (Los Angeles, Calif), specimens were received in the lab-
3 patient safety activities (independent variables) on the incidence
oratory, ordered in the laboratory information system, labeled,
of critical errors (dependent variable). Trend analysis can be used
sorted, centrifuged, aliquoted, and sorted again for different test-
to test different aspects of the shape of the function relating the
ing areas. The receiving to final sort and delivery of specimens
independent variable (patient safety initiatives) and the depen-
required a minimum of 6 to 7 full time equivalents. On January
dent variable (incidence of critical errors). Trend analysis consists
23, 2004, the UCLA began a laboratory automation project. As
of testing 1 or more components of trend. These components are
part of this project, in May 2004, at 14 months, an automated
tested using specific comparisons. The linear component of trend
processing system was activated in Specimen Receiving in the
is used to test whether there is an overall increase (or decrease)
UCLA Clinical Laboratories. At present, specimens are still re-
in the dependent variable as the independent variable increases.
ceived, ordered, and labeled manually, but they are sorted, cen-
trifuged, aliquoted, and sorted again by the automated processor.
Instead of the minimum 6 to 7 full time equivalents that were RESULTS
required prior to automation, we are now able to function with The total number of specimens in the 3 critical identi-
4 to 5 full time equivalents for the basic processing operation per
fication error categories (mislabeled, requisition/specimen
shift (with the exception of night shift when fewer specimens are
processed). The Beckman Coulter Power Processor implementa- mismatch, unlabeled specimens) represented 11.9% of all
tion has consolidated centrifugation, aliquoting, and the sorting specimen errors from September 1, 2003, through August
steps of specimen processing. 31, 2005 (1978 critical errors, N ⫽ 16 632 total errors) (Fig-
ure 1). More than 4.29 million specimens and 2.31 million
Statistical analyses were performed on the data collected by 2
methods. The paired Student t test was conducted to measure
phlebotomy requests were received in the same time pe-
and compare the numbers of unlabeled and mislabeled speci- riod. Critical identification errors occurred in fewer than
mens before (November 1, 2002 through July 31, 2003) and after 1 in 1000 of all procedures or specimens received.
(June 1, 2004 through August 31, 2005) interventions by the 3 Major external factors during the entire study period
patient safety activities to assess whether a statistically significant (November 1, 2002, through August 31, 2005) included
1664 Arch Pathol Lab Med—Vol 130, November 2006 Patient Safety and Specimen Identification—Wagar et al
 Figure 2. A, Total blood draw errors for all
error types (September 1, 2003–August 31,
2005). Total blood draw errors are demon-
strated by month for the evaluation interval.
B, Total blood draw errors for selected error
types (September 1, 2003–August 31, 2005).
Requisition mismatch errors represent the
major identification error in cumulative error
data. Mislabeled specimens, the most dan-
gerous identification errors, represent 8.4% of
the identification error total.

outside hospital management consultant oversight begin- 16 632) (September 1, 2003 through August 31, 2005), mis-
ning fall 2002; an institution-wide JCAHO accreditation labeled specimens, requisition mismatches, and unlabeled
survey in April 2004; departure of the outside consultant, specimens represented 1.0%, 6.3%, and 4.6%, respectively
June 2004; and a chief executive officer leadership change, (Figure 2, B).
July 2004. There was no change of leadership within the Intensive care units had a disproportionate frequency,
laboratory during this period. No significant differences 31.2% (range, 21.6%–37.7%) of specimen errors even
were distinguishably associated with any of the external though they accounted for only 16.1% of total beds (Figure
factors (Figure 2, A). 3, A). Within the ICUs, requisition mismatches were the
Initially after the implementation of in-service educa- most common category of misidentification (55.1%) fol-
tional activities and full-time around-the-clock phleboto- lowed by unlabeled specimens (38.2%) (Figure 3, B).
my, there was an increase of 33.7% in reported errors (data The trend analysis data for 26 months (July 1, 2003–
not shown), because of process changes and higher re- August 31, 2005) are shown in Figure 4. The incidence of
porting rates related to the educational activities. The total critical errors decreased as the patient safety initiatives
number of blood draw errors decreased after implemen- were introduced. Linear trend analysis shows a general
tation of the electronic error reporting (March 2004) and decline during this time in errors for mislabeled speci-
installation of an automated specimen processor (Septem- mens, requisition mismatch, and unlabeled specimens.
ber 2004) (Figure 2, A). Of the total specimen errors (N ⫽ The numbers in all 3 categories decreased after interven-
Arch Pathol Lab Med—Vol 130, November 2006 Patient Safety and Specimen Identification—Wagar et al 1665
Figure 3. A, Total blood draw errors for all
error types (intensive care units [ICUs] only)
(January 1, 2004–August 31, 2005). ICU
blood draw errors mirror total blood draw er-
rors irrespective of location. B, ICU blood
draw errors for selected error types (January
1, 2004–August 31, 2005). Data from the
ICUs indicates similar proportions for the 3
critical identification error types, with misla-
beled specimens representing 6.7% of the
identification error total.

tions by the 3 patient safety activities (June 1, 2004–August ysis of specimen errors, including an in-depth study of
31, 2005) compared to performance before any of the pa- patient identification errors during the course of 3 patient
tient safety projects were implemented. Statistical analysis safety implementation projects. Although these data rep-
using paired Student t test showed P ⬍ .001, for misla- resent the experience of a single academic institution, the
beled specimens, and P ⫽ .15 for unlabeled specimens results share commonalities with many other clinical lab-
when compared to November 1, 2002, through July 31, oratories.9
2003. Of the 3 categories of critical patient identification er-
rors, the most frequently documented occurrence was the
COMMENT receipt of a mismatched requisition with a specimen la-
Correct patient identification is an important patient beled with another patient’s name. This probably repre-
safety initiative. Its prominence is validated by the rec- sents the ease with which this type of error is recognized.
ognition of patient identification by JCAHO as a primary As soon as a specimen is received, the mismatch is readily
patient safety goal. Analysis of the frequency and risk fac- apparent in the clinical laboratory. The second most fre-
tors for patient identification errors, however, is infre- quently occurring specimen identification error was the
quently reported in the literature. We present here an anal- receipt of an unlabeled specimen, also easily recognized
1666 Arch Pathol Lab Med—Vol 130, November 2006 Patient Safety and Specimen Identification—Wagar et al
Figure 4. Trend analysis: critical errors by type/month (July 1, 2003–August 31, 2005). Longitudinal data for the 3 critical identification errors by
month demonstrates a decrease in errors by trend analysis. The 3 patient safety initiative implementation dates are noted by dotted lines.

by laboratory personnel. For purposes of this study, a par- study underestimates the actual frequency of mislabeled
tially labeled specimen missing 1 of 2 identifiers was also specimens.12
considered unlabeled, to emphasize the importance of the An important observation of this study was that patient
2-identifier initiative by JCAHO in our institution. identification errors occur more frequently on specimens
The most infrequently reported specimen identification received from ICUs (Figure 3, A and B). This is somewhat
error was the notorious ‘‘mislabeled’’ specimen. Some- anticipated, given that many critical care specimens are
times referred to as ‘‘wrong blood in tube,’’ this specimen drawn from lines and by nurses, in more complex care
represents blood from 1 patient accompanied by both a environments. When the percentage of specimens for all
requisition and label that indicates another patient. In the 3 identification error types was compared to the percent-
current study the reduction of this type of error post pa- age of ICU beds, 21.6% to 37.7% of errors were noted in
tient safety interventions was greater than the reductions ICU locations (ICU beds represent 16.1% of total inpatient
of the other 2 error types (Figure 4). The other 2 error beds at UCLA). The increase in ICU errors was also ob-
types are more apparent and hence more easily recog-
served when patient days were analyzed (ICU patient
nized, documented, and controlled. The detectability of
days vs total patient days, data not shown). Higher acuity
such errors as mislabeled specimens is very low, because
care frequently requires nursing performance of phlebot-
the errors generally are not discovered until a clinician
questions a result that is highly atypical for a given patient omy from peripheral and central lines. These processes
(eg, a highly elevated glucose level in a nondiabetic pa- are inherently more complicated with more process points
tient). Hence with the introduction of the patient safety at which errors can occur (failure mode and effects anal-
interventions, the most significant decrease was seen with ysis data not shown), possibly contributing to increased
this error type. These errors are probably the most severe error rates in ICUs.
of patient errors, because there is no reason to question a When viewed from the perspective of all types of spec-
clinical action taken in response to the laboratory result. imen errors, critical identification errors occurred in fewer
This type of error would score high for severity and low than 1 in 1000 procedures or specimens. This information
for detectability using assessments such as the failure may seem somewhat reassuring; however, it still repre-
mode and effects analysis.11 It is reasonable to conclude sents approximately 3 to 4 ‘‘recognized’’ errors per day.
that the number of mislabeled specimens observed in this The absolute magnitude of unrecognized mislabeled spec-
Arch Pathol Lab Med—Vol 130, November 2006 Patient Safety and Specimen Identification—Wagar et al 1667
imens remains a cause for concern and an opportunity for than unobserved subjects. First observed in a research
improvement. project at the Hawthorne Plant of the Western Electric
Additionally, of all error types, duplicate specimens, Company from 1927 to 1932, this study was the first to
specimens received with duplicate orders or requisitions, propose that, regardless of the experimental design, work-
represented the bulk of identified specimen errors (Figure ers responded to the process of being observed.13 This ef-
1). As such they represent marked overutilization of phle- fect may well have contributed to decreased patient iden-
botomy services, inappropriate drawing of excess blood tification errors, especially with initiation of an electronic
from already ill patients, and significant waste of reagents event reporting system that automatically forwarded re-
and supplies. Also, the duplicate specimens may not rep- ported labeling errors to respective nursing and clinical
resent the entire waste profile because some duplicate or- services. Awareness is probably a primary component of
ders may go unrecognized if they are not exact duplicates patient safety projects, even though difficult to define and
at the time of order entry. Increasing attention is being objectively capture. Awareness is also a likely key element
paid to waste within the health care system; although not for maintaining successful patient safety projects. This
the focus of this study, duplicate specimens and duplicate may be an interesting area of study as we examine the
orders are opportunities to examine and reduce waste in sustainability of our patient safety efforts.
health care practice. Patient identification can be improved for clinical labo-
We demonstrated through a combination of both trend ratory specimens. Institutions considering such projects
and statistical analyses that the patient safety implemen- for improving their patients’ safety should consider data-
tation projects of a 24/7 phlebotomy service, electronic driven longitudinal models. Also, awareness is a key fac-
event reporting, and automated processing contributed to tor in ensuring successful implementation; awareness may
decreased patient identification errors. During the course additionally be a key component of patient safety sustain-
of the study, other unrelated changes in institutional struc- ability.
ture did not appear to interfere with successful imple- This study was supported in part by a Developmental Center
mentation. It is important for laboratory professionals to Grant HS11512–03 from the Agency for Healthcare Research and
be reassured that a well-designed patient safety imple- Quality. We thank Merian Raz, BS, for her invaluable assistance
mentation study can produce positive results, even when in assembling the data for this study. We also thank Ralph Bow-
unforeseen changes occur in the immediate health care man Sr, MS, and Pamela Bearly, MS, for their useful assistance
environment during the course of an implementation pro- with information technology and electronic event reporting.
ject. References
1. Institute of Medicine. To Err is Human: Building a Safer Health System.
As laboratories and laboratory users consider patient Washington, DC: National Academy Press; 2000.
safety and quality improvement initiatives, it is important 2. Brennan TA, Leape LL, Laird NM, et al. Incidence of adverse events and
to design them so that they easily prove, through data- negligence in hospitalized patients: results of the Harvard medical practice study,
I. N Engl J Med. 1991;324:370–376.
driven studies, their success to hospital administration 3. Leape LL, Brennan TA, Laird NM. The nature of adverse events in hospital-
and other stakeholders. Laboratory professionals, by using ized patients: results of the Harvard medical practice study, II. N Engl J Med.
the 2 described methods, Student t test analysis and trend 1991;324:377–384.
4. Thomas EJ, Studdert DM, Burstin HR, et al. Incidence and types of adverse
analysis can provide administrators with useful patient events and negligent care in Utah and Colorado. Med Care. 2000;38:261–271.
safety follow-up. The Student t test method and trend 5. Dock B. Improving the accuracy of specimen labeling. Clin Lab Sci. 2005;
analysis additionally have the advantages of ease of use 18:210–212.
6. Howanitz PJ. Errors in laboratory medicine: practical lessons to improve
and familiarity for clinical laboratory professionals as patient safety. Arch Pathol Lab Med. 2005;129:1252–1261.
compared to other statistical tools available for longitudi- 7. Chassin M, Becher EC. The wrong patient. Ann Intern Med. 2002;136:826–
nal analyses. 833.
8. Joint Commission on Accreditation of Healthcare Organizations. 2006
The specific elements that contribute to success are National Patient Safety Goals. Available at: http://www.jcipatientsafety.org. Ac-
sometimes difficult to identify in a longitudinal analysis cessed January 12, 2006.
schematic. Cumulative effects may in part be defined by 9. Institute of Medicine. Crossing the Quality Chasm: Improving the 21st Cen-
tury Health Care System. Washington DC: National Academy Press; 2001.
subtractive analysis, but external trends and multiple out- 10. Howanitz PJ, Renner SW, Walsh MK. Continuous wristband monitoring
side variables also affecting an open project as compared over 2 years decreases identification errors: a College of American Pathologists
to a closed controlled study will still not be fully account- Q-Tracks Study. Arch Pathol Lab Med. 2002;126:809–815.
11. Joint Commission on Accreditation of Healthcare Organizations. Failure
ed for. Because laboratory professionals are obliged to cre- Mode and Effects Analysis in Health Care: Proactive Risk Reduction. Oakbrook
ate change in a fluctuating open environment, it is difficult Terrace, Ill: Joint Commission Resources; 2002.
to ascertain all the effectors. 12. Murphy MF, Stearn BE, Dzik WH. Current performance of patient sample
collection in the UK. Transfusion Med. 2004;14:113–121.
An example is the ‘‘Hawthorne effect.’’ This phenome- 13. Frank RH, Kaul JD. The Hawthorne experiments: first statistical interpre-
non implies that subjects under observation perform better tation. Am Sociological Rev. 1978;43:623–643.

1668 Arch Pathol Lab Med—Vol 130, November 2006 Patient Safety and Specimen Identification—Wagar et al