AP Discrepancy Rates - College of American Pathologists

CAP Laboratory Improvement Programs
Patient Safety in Anatomic Pathology
Measuring Discrepancy Frequencies and Causes
Stephen S. Raab, MD; Raouf E. Nakhleh, MD; Stephen G. Ruby, MD
● Context.—Anatomic pathology discrepancy frequencies
have not been rigorously studied.
Objective.—To determine the frequency of anatomic pathology
discrepancies and the causes of these discrepancies.
Design.—Participants in the College of American Pathologists
Q-Probes program self-reported the number of
anatomic pathology discrepancies in their laboratories by
prospectively performing secondary review (post–sign-out)
of 100 surgical pathology or cytology specimens. Reasons
for the secondary review included conferences, external
review, internal quality assurance policy, and physician request.
Participants.—Seventy-four laboratories self-reported
data.
Main Outcome Measures.—Frequency of anatomic pathology
discrepancy; type of discrepancy (ie, change in
margin status, change in diagnosis, change in patient in-
The 1999 Institute of Medicine report increased the national
awareness of medical errors and patient safety. 1
Anatomic pathology errors are reported to occur in 1% to
43% of all anatomic pathology specimens, 2–19 and this exceptionally
wide range depends on the methods of detection
and the definition of what counts as an error. On review
of the literature, Raab2 estimated that the mean anatomic
pathology error frequency ranged from 1% to 5%,
although this frequency was largely based on studies using
single-institution data. No large-scale, multi-institutional
anatomic pathology error studies have been conducted,
and information on the effect of anatomic pathology
error on patient outcome is generally lacking.
Error detection in anatomic pathology most often depends
on some form of secondary case review. 2 Secondary
case review has been built into some pathology quality
assurance practices (eg, review of a set percentage of cases,
intradepartmental ‘‘difficult case’’ conferences, cytolog-
Accepted for publication December 2, 2004.
From the Department of Pathology, University of Pittsburgh, Pittsburgh,
Pa (Dr Raab); the Department of Pathology, St Luke’s Hospital/
Mayo Clinic, Jacksonville, Fla (Dr Nakhleh); and the Department of
Pathology, Palos Community Hospital, Palos Heights, Ill (Dr Ruby).
The authors have no relevant financial interest in the products or
companies described in this article.
Reprints: Stephen S. Raab, MD, Department of Pathology, University
of Pittsburgh, UPMC Shadyside Hospital, 5150 Centre Ave, Pittsburgh,
PA 15232 (e-mail: raabss@msx.upmc.edu).
formation, or typographic error); effect of discrepancy on
patient outcome (ie, no harm, near miss, or harm); and
clarity of report.
Results.—The mean and median laboratory discrepancy
frequencies were 6.7% and 5.1%, respectively. Forty-eight
percent of all discrepancies were due to a change within
the same category of interpretation (eg, 1 tumor type was
changed to another tumor type). Twenty-one percent of all
discrepancies were due to a change across categories of
interpretation (eg, a benign diagnosis was changed to a
malignant diagnosis). Although the majority of discrepancies
had no effect on patient care, 5.3% had a moderate
or marked effect on patient care.
Conclusions.—This study establishes a mean multi-institutional
discrepancy frequency (related to secondary review)
of 6.7%.
(Arch Pathol Lab Med. 2005;129:459–466)
ic-histologic correlation, or review of all malignancies).
Secondary case review also occurs in hospital patient-centered
conferences (eg, tumor board); external consultation
practices; or at the behest of clinicians, who may initiate
communication when the pathology report does not correlate
with the clinical findings. An error detected by 1 of
these processes may be referred to as a discrepancy or a
difference in interpretation or reporting between 2 pathologists.
Error detection frequencies based on the different methods
of secondary review have been variably studied. For
example, the College of American Pathologists (CAP) has
intensively studied gynecologic cytologic-histologic correlation
and reported that paired cervical-vaginal cytology-biopsy
specimens had a sensitivity and specificity of
89.4% and 64.8%, respectively. 20–22 Based on nongynecologic
cytologic-histologic correlation data, Clary et al 6 reported
that 23% of interpretive discrepancies had a major
impact on patient outcome. Other methods of secondary
review have been studied in less detail. As a consequence,
how these methods may be utilized to improve patient
safety and be incorporated into laboratory continuous
quality improvement programs is unknown.
The CAP’s Q-Probes program has measured and defined
a number of key quality indicators in anatomic and
clinical pathology. 21,22 This Q-Probes study is the first multi-institutional
study to measure and to document anatomic
pathology discrepancy frequencies and the effect of
these discrepancies on patient outcome.
Arch Pathol Lab Med—Vol 129, April 2005 Patient Safety in Anatomic Pathology—Raab et al 459

Table 1. Definition of Discrepancy
Discrepancy: A discrepancy has occurred if there is any difference between the original interpretation and the interpretation after the
second review. Discrepancies were further classified by cause into one of the following categories:
Change in margin status: The interpretation of the margin status was changed from benign to malignant or vice versa.
Change in categoric interpretation: An interpretation was changed from one categoric diagnosis, such as benign, to another categoric
diagnosis, such as malignant. For purposes of this study, interpretations were classified within categories that were graded by their
probability of a malignant clinical outcome (eg, a benign diagnosis was assigned a 1; atypical, 2; suspicious, 3; and malignant, 4). We
considered a difference of 2 or more steps between the original and the review interpretation as a discrepancy. For example, if the
original diagnosis was benign, and the review diagnosis was malignant, the difference in steps between these 2 diagnoses was 4 1
3, and this case was considered discrepant. If the step difference between the original and review interpretations was 1, we decided
that a discrepancy had not occurred.
Change within the same category of interpretation: An interpretation was changed from one benign interpretation to another benign
interpretation or from one malignant interpretation to another malignant interpretation. A change from one tumor type to another fell
within this category. For example, if the original interpretation was adenocarcinoma and the review diagnosis was epithelioid sarcoma,
this case was placed within this category of discrepancy.
Change in patient information: There was a change in the organ site, such as the left ovary to the right ovary.
Typographic error
Table 2. Definitions of Effects of Discrepancies
Effect of discrepancy on patient management: Discrepancies were classified into the following categories based on patient outcome:
Harm (significant event): A discrepancy that resulted in patient harm (eg, inappropriate treatment, loss of life or limb, psychological
event). The effect of the significant event on patient outcome was assessed using a 3-point Likert scale (1 severe effect, 2 moderate
effect, 3 mild effect). The pathologists performing the review judged the significance of the event.
Near miss: A discrepancy that was detected before harm occurred, such as a discrepancy that was detected at a clinical pathologic
conference before treatment was initiated.
No harm: A discrepancy that did not result in patient harm, such as a typographic error that had no bearing on patient management.
MATERIALS AND METHODS
Laboratories enrolled in the CAP’s volunteer Q-Probes quality
improvement program participated in this study in 2003. The Q-
Probes program and the format for data collection and handling
have been previously described in detail. 21
Participants prospectively identified 100 consecutive surgical
pathology or cytology specimens that were reviewed by a second
pathologist after a first pathologist had signed out that case. In
order to standardize the data-collection process across all participating
laboratories, pertinent terms were defined (Tables 1 and
2). For each case, the participating laboratory recorded the specimen
type (surgical pathology or cytology), organ or anatomic
site (chosen from a specified list), primary reason for secondary
review (chosen from a specified list), clarity of the report, presence
or absence of a discrepancy (Table 1), effect of discrepancy
on patient outcome (Table 2), and modification of report (if performed).
We subclassified discrepancies into several main groups of
causes that have been previously described in the pathology literature.
2 The detection of particular discrepancy subtypes depends
partly on the method of secondary review, described in
more detail later. For example, change in margin status is more
likely to be detected by a frozen-permanent section review, and
change in categoric interpretation is more likely to be detected
by cytologic-histologic correlation review. We arbitrarily chose a
disagreement of 2 steps as constituting a categoric interpretation
discrepancy, although even a 1-step discrepancy is an error. Raab 2
reported that 2-step discrepancies have a much greater probability
of clinical significance compared with 1-step discrepancies. In
addition, because interobserver variability studies have shown
that 1-step discrepancies are much more common, we did not
want to collect data on a large subset of cases that had little effect
on patient care. 2 For changes in categoric interpretation, the original
and review diagnoses were recorded.
The taxonomy of the effect of a pathology discrepancy on patient
outcome was based on the medical patient safety literature,
23–26 which uses a taxonomy related to the effect of a failure
of a planned action to be completed or the use of a wrong plan
to achieve an aim. 1 Diagnostic error does not fit neatly into the
category of an ‘‘action’’ error, and the resulting outcome of the
patient is often difficult to assess. In particular, distinguishing
between a no-harm and a near-miss event may be problematic.
We defined a near-miss event as an error that was detected before
harm occurred; an example is when a diagnostic error was picked
up at a conference (or another means of secondary review)
before a particular treatment protocol was initiated. In this case,
if the error had not been detected, we assume that some degree
of harm would have occurred. We recognize that harm (eg, psychologic)
may still have occurred in this case, but we classified
this event as a near miss, because the ability of the review pathologist
to identify this type of harm was limited in this study.
We defined a no-harm event as occurring when a diagnostic error
occurred that would not cause patient harm even if the error had
been undetected. An example of a no-harm event was a typographic
error, such as writing the word ‘‘brest’’ instead of
‘‘breast’’ in the diagnostic line.
We defined a significant event as an error resulting in patient
harm. 1 We provided the review pathologist a 3-point Likert scale
to grade the severity of this harm, recognizing that this grading
was subjective. Grzybicki et al 27 showed that even with more
well-defined criteria for patient harm, there is little agreement
among pathologists as to the effect of an error on patient outcome;
thus, although subclassifying harm needs further study,
we wanted to measure the general view of the participant institutions
in assessing harm. We did not require the pathologist to
perform medical record review but only to assess the error severity
based on the information available at the time of detection.
In general, the pathologist had some degree of knowledge of the
clinical outcome. This information varied depending on the method
of detection; for example, more information may be available
if a clinician requests secondary review compared to if a cytologic-histologic
correlation is performed.
Autopsy cases were the only anatomic pathology case type excluded
from this study. Multiple specimens with a secondary
review from the same case or specimens from different cases
associated with the same patient were included in the study. Secondary
review is the main method used to detect anatomic pathology
errors. The recorded reasons why cases were reviewed
were as follows: intradepartmental conference, request by clinician,
interdepartmental conference, specified institutional quality
460 Arch Pathol Lab Med—Vol 129, April 2005 Patient Safety in Anatomic Pathology—Raab et al

assurance policy, and extradepartmental review. These methods
encompass the majority of forms of secondary review, although
participants also could choose the category of ‘‘other’’ if none of
these methods applied. We chose to examine multiple methods,
because single institutions often do not review a sufficient number
of cases using a single method for statistical significance testing
in the time frame provided by this Q-Probes study. We also
wanted to determine if discrepancies were detected at different
frequencies and had different clinical import depending on the
method of detection. Institutional quality assurance policies include
cytologic-histologic correlation, random review, and frozenpermanent
section correlation. We recognize that these methods
of review detect only a fraction of discrepancies (because in most
institutions, the majority of cases do not undergo secondary review)
and exclude those errors detected prior to sign-out. However,
some of these forms of secondary review are the most likely
means to detect clinically significant error, the subcategory of
error that some authors believe is the most important to track
and eradicate.
The ‘‘clarity of the report’’ question referred to how understandable
the original report was perceived to be from the standpoint
of the review pathologist. The review pathologist determined
the report clarity using a 4-point Likert scale (ie, clear,
mildly unclear, moderately unclear, markedly unclear). A limitation
in this assessment was that clinicians, who were the users
of the diagnostic information, did not perform the analysis, and
previous authors have indicated discrepancies between clinician
and pathologist impression of report clarity. 28 We also recognize
that the clarity assessment was subjective, because definitive criteria
of clarity were not provided. The lack of clarity may be an
example of error, particularly for reports assessed to be moderately
or markedly unclear, because this lack of clarity could result
in improper patient management.
Laboratories could choose 3 options regarding computerized
report modification after a discrepancy was detected: (1) the original
report was deleted and replaced with a new report; (2) the
original report was marked with a qualifier, and a new report
was appended; or (3) the original report was not marked or modified,
but a new report was appended.
Institutional demographic and practice variable information
was obtained by completion of a questionnaire. These data included
annual specimen volume, number of institutional practicing
pathologists, type and frequency of institutional review conferences
(eg, breast, chest, or endocrine conference), and institutional
quality assurance methods of secondary review. The discrepancy
frequency (expressed as a percentage) was calculated
for the aggregate data and for each institution. The discrepancy
frequency was calculated as the number of discrepancies divided
by the total number of cases reviewed, multiplied by 100. Discrepancies
and report clarity assessments were subclassified by
organ type (and further broken down by effect on patient outcome).
The correlation of the main indicator variable (discrepancy
frequency) was assessed for each of the predictor variables separately
by using nonparametric Wilcoxon rank sum or Kruskal-
Wallis tests. The 2 goodness-of-fit test was used to test for associations
between the presence of a discrepancy and other caselevel
variables. Statistically significant associations are defined at
the P .05 level. The reason for case review was correlated with
the effect of the discrepancy on patient outcome.
Some participating institutions did not answer all of the questions
on the demographics form or on the input forms. These
institutions were excluded only from the tabulations and the
analyses that required the missing data element.
RESULTS
A total of 74 institutions submitted data for this study.
Most institutions (87.8%) were located in the United
States, with the remainder located in Canada (n 7), Australia
(1), and Saudi Arabia (1). Of the participating institutions,
31% were teaching hospitals, and 23% had a pathology
residency program. The Joint Commission on Ac-
Table 3. Percentile Distribution of Anatomic
Pathology Discrepancy Rate
All Institutional Percentiles
N 10th 25th
50th
(Median) 75th 90th
Discrepancy rate, % 74 21.0 10.0 5.1 1.0 0.0
creditation of Healthcare Organizations inspected the majority
of the institutions (68%) participating in this study,
and the CAP inspected 82% of laboratories contributing
data. The participants’ institutional sizes were as follows:
39.7% had fewer than 150 beds; 39.7%, 151 to 300 beds;
9.5%, 301 to 450 beds; 3.2%, 451 to 600 beds; and 7.9%,
more than 600 beds. Private nonprofit institutions comprised
54.9% of the institutions; 11.1% were private forprofit
institutions; 9.5% were state, county, or city hospitals;
6.3% were university hospitals; 3.2% were federal
governmental institutions; and 15.9% were other. Fiftythree
percent of the institutions were within a city, 28.1%
were suburban, and 15.6% were rural. The annual mean
number (and SD) of surgical pathology, nongynecologic
cytology, and gynecologic specimens per institution were
16 241 (19 567), 1898 (2376), and 21 147 (44 903), respectively.
The mean and median numbers of pathologists at
each institution were 6 and 4, respectively.
The 74 institutions performed secondary review of 6186
anatomic pathology specimens (5268 surgical pathology
specimens and 847 cytology specimens), and each institution
collected data on a range of from 2 to 100 specimens
(median, 99 specimens). In aggregate, 415 discrepancies
were reported, and the overall mean anatomic pathology
discrepancy frequency was 6.7%. The overall
mean surgical pathology discrepancy frequency was 6.8%
(356 discrepancies of 5255 reviewed cases), and the overall
mean cytology discrepancy frequency was 6.5% (55 discrepancies
of 844 reviewed cases); neither of the specimen
types had a higher discrepancy frequency (P .78). The
distribution of the anatomic pathology discrepancy frequencies
is listed in Table 3. Higher percentile ranks indicate
lower discrepancy frequencies.
The number of reviewed specimens, overall discrepancy
frequency, and the discrepancy classification by organ
type are shown in Table 4. The most common organ types
reviewed in this study were female genital tract and
breast. None of the organ types had a higher frequency of
discrepancy than other organ types (P .15). For the organ
types with higher volumes reviewed (200 cases), a
change in categoric interpretation (ie, an error type more
likely to be associated with harm) occurred more frequently
in the female genital tract, male genital tract, and
lymph node. A change in margin status occurred most
frequently in breast specimens.
In Table 5, the effect of the discrepancy on patient outcome
and the report modification in response to a discrepancy
are listed by organ type. Some form of harm was
seen in the majority of organs, although specimen type
(cytology or surgical pathology) or organ type did not
correlate with effect on patient outcome (P .73 and P
.83, respectively). Harm was observed in 20.8% (11 cases)
of breast specimens and in 25.3% (21 cases) of female genital
tract specimens in which a discrepancy was detected.
Neither specimen type nor organ type correlated with the
Arch Pathol Lab Med—Vol 129, April 2005 Patient Safety in Anatomic Pathology—Raab et al 461

Table 4. Number of Specimens and Discrepancy by Organ Type
Change in Change in Change in
Overall Change in Categoric Same Patient Typographic
Specimens, No. Discrepancy, Margin, Interpretation, Category, Information, Error,
Organ
(% of Total) %
%
%
%
%
%
Genital, female
982 (15.9) 7.1 2.3 28.7 36.8 18.4 13.8
Breast
796 (12.9) 8.3 10.9 12.5 42.2 10.9 23.4
Lung
463 (7.5) 5.0 0 13.6 50.0 4.6 31.8
Genital, male
355 (5.7) 7.1 0 41.7 54.1 0
4.2
Soft tissue
345 (5.6) 7.5 0 12.5 66.7 8.3 12.5
Lymph node
288 (4.7) 5.9 5.9 23.5 52.9 0
17.7
Hepatobiliary
240 (3.9) 6.7 0 13.3 53.3 13.3 20.0
Urinary tract
181 (2.9) 7.2 0 30.8 53.9 0
15.4
Pharynx
141 (2.3) 5.0 0 28.6 42.9 0
28.6
Endocrine
125 (2.0) 8.0 0 10.0 50.0 10.0 30.0
Bone marrow
107 (1.7) 6.6 0 14.3 85.7 0
0
Bone
99 (1.6) 2.0 0
0
50.0 0
50.0
Neuropathology
88 (1.4) 6.8 0 16.7 83.3 0
0
Kidney
55 (0.9) 5.5 0 33.3 33.3 0
33.3
Pancreas
31 (0.5) 6.5 0 50.0
0
0
50.0
Salivary gland
29 (0.5) 3.5 0
0 100.0 0
0
Spleen
11 (0.2) 9.1 0
0
0
0 100.0
Gastrointestinal and other
1841 (29.8) 5.6 5.0 19.0 48.0 8.0 20.0
Total 6162 6.7 3.7 21.0 47.7 9.1 18.5
Organ
Genital, female
Breast
Lung
Genital, male
Soft tissue
Lymph node
Hepatobiliary
Urinary tract
Pharynx
Endocrine
Bone marrow
Bone
Neuropathology
Kidney
Pancreas
Salivary gland
Spleen
Gastrointestinal and other
Table 5. Effect of Discrepancy on Patient Management and Pathology Response by Organ Type
Marked
Harm,
%
2.4
1.9
5.3
4.4
8.0
0
0
7.7
0
0
0
0
0
0
0
0
0
0
Moderate
Harm,
%
6.0
7.6
0
0
0
6.2
0
0
0
0
0
0
0
0
0
0
0
2.1
Effect on Patient Outcome
Mild
Harm,
%
16.9
11.3
0
8.7
0
0
7.1
23.1
14.3
0
28.6
0
16.7
0
50.0
0
0
12.6
Near Miss,
%
No Harm,
%
Report Change
Yes, % No, %
462 Arch Pathol Lab Med—Vol 129, April 2005 Patient Safety in Anatomic Pathology—Raab et al
4.8
13.2
21.1
13.0
12.0
18.8
0
7.7
0
11.1
0
0
0
33.3
0
0
0
6.3
69.9
66.0
73.7
73.9
80.0
75.0
92.9
61.5
85.7
88.9
71.4
100.0
83.3
66.7
50.0
100.0
100.0
79.0
58.3
62.3
33.3
56.5
58.3
56.3
42.9
38.5
14.3
33.3
42.9
100.0
50.0
0
100.0
0
100.0
54.1
Total 2.1 3.2 11.3 8.7 7.5 53.3 46.7
report modification in response to a discrepancy (P .15
and P .14, respectively).
In Table 6, the clarity of the report is listed by organ
type. Markedly and moderately unclear reports were seen
infrequently and only in a few specimen types, such as
female genital tract, breast, and lung. Because of the large
number of cells with a value of 0, a 2 goodness-of-fit test
was not performed.
In Table 7, the discrepancy type, original and review
interpretations for categoric discrepancies, and the effect
of the discrepancy on patient outcome are shown. A
change in the same category of diagnosis was the most
common discrepancy detected. For changes in categoric
interpretation, the review diagnosis tended to be shifted
downward to a benign or upward to a malignant diagnosis,
and there were fewer nondefinitive (atypical or sus-
41.7
37.7
66.7
43.5
41.7
43.7
57.1
61.5
85.7
66.7
57.1
0
50.0
100.0
0
100.0
0
45.9
picious) diagnoses compared with the original diagnosis.
When a discrepancy occurred, the most common classification,
based on patient outcome, was a no-harm event.
Anatomic pathology discrepancy specimen-centered
variables, including specimen type and origin, discrepancy
type, primary reason for review, the effect of discrepancy
on patient outcome, and the response to a discrepancy
in the form of a report change, were evaluated
to identify any associations. The statistically significant associations
are shown in Table 8. A request for review directed
by a clinician was much more likely to be associated
with a discrepancy than all other reasons for review. If a
discrepancy occurred, a change in categoric interpretation
was more likely to be seen in cytology specimens compared
with surgical pathology specimens and related to
extradepartmental review compared with all other rea-

Genital, female
Breast
Lung
Genital, male
Soft tissue
Lymph node
Hepatobiliary
Urinary tract
Pharynx
Endocrine
Bone marrow
Bone
Neuropathology
Kidney
Pancreas
Organ Clear, %
Salivary gland
Spleen
Gastrointestinal and other
Table 6. Clarity of Report by Organ Type
97.6
96.5
97.0
98.6
97.7
97.6
96.6
98.3
98.6
99.2
94.4
99.0
95.4
96.2
96.8
100.0
90.9
97.4
Mildly
Unclear, %
Moderately
Unclear, %
Markedly
Unclear, %
Total 97.4 2.3 0.3 0.07
Table 7. Discrepancy Type, the Original and Review
Diagnoses for Categoric Discrepancies, and the Effect
of Discrepancy on Patient Outcome
Discrepancy type
Change within same category
Change in categoric interpretation
Typographic error
Change in patient information
Change in margin status
Original interpretation for categoric
discrepancies
Benign
Atypical
Suspicious
Malignant
Reviewed interpretation for categoric
discrepancies
Benign
Atypical
Suspicious
Malignant
Effect on patient outcome
Harm
Marked
Moderate
Mild
Near miss
No harm
Specimens, No. (%)
415
194
85
75
37
15
25
27
16
16
28
21
7
28
63
8
12
43
33
283
(100)
(47.8)
(20.9)
(18.5)
(9.1)
(3.7)
(29.8)
(32.1)
(16.0)
(16.0)
(33.3)
(25.9)
(8.3)
(33.3)
(16.6)
(2.1)
(3.2)
(11.3)
(8.7)
(74.7)
sons for review. If a near-miss event occurred, the reason
for case review was more likely to be extradepartmental
review compared with all other reasons for review.
Table 9 shows that the reason for case review correlated
with patient outcome in discrepant cases (P .02). Harm
occurred more frequently in discrepant cases that were
reviewed at the request of a clinician (23.5%) and interdepartmental
conference (25.0%). Clinician-directed review
was the most common method that detected a discrepancy
(23.0% of all cases reviewed). The majority of
discrepant cases detected at an intradepartmental conference
were associated with no-harm events.
Arch Pathol Lab Med—Vol 129, April 2005 Patient Safety in Anatomic Pathology—Raab et al 463
1.9
2.1
2.8
1.4
2.1
2.4
3.4
1.7
1.4
0.8
3.8
1.0
4.6
3.8
3.2
0
9.1
2.4
0.3
1.3
0.2
0
0
0
0
0
0
0
1.9
0
0
0
0
0
0
0.2
0.2
0.1
0
0
Of the 74 institutions, the number that had a conference
devoted to breast review was 33; chest, 21; endocrine, 8;
gastrointestinal, 22; general surgical, 29; genitourinary
tract, 18; gynecologic, 26; head and neck, 16; hematopathology,
20; liver, 15; renal, 18; and tumor board, 60. Institutional
quality assurance practices were measured as
well: 52 institutions had an intradepartmental conference
for difficult cases; 31 reviewed a percentage of cases after
sign-out; 22 reviewed all malignancies before sign-out; 19
reviewed a percentage of cases before sign-out; 6 reviewed
all malignancies after sign-out; and 1 reviewed all cases
before sign-out. Of the institutions that made changes to
the reports after an error, 43 issued an amended report
and did not retrieve the original report; 3 retrieved the
original report, stamped it ‘‘in error,’’ and filed the report
in the chart; 3 destroyed the original reports; and 3 handled
the change using other methods.
COMMENT
This is the first study to determine a baseline anatomic
pathology discrepancy frequency across multiple pathology
laboratories. Based on secondary pathologist review,
the mean anatomic pathology discrepancy frequency
(based on cases reviewed through several different methods)
was 6.7%, and the variability across laboratories was
striking, with the 25th and 75th percentiles being 10.0%
and 1.0%, respectively.
Discrepancy represents one form of error, and based on
literature review of a number of error-detection methods,
Raab 2 estimated that the mean laboratory error frequency
ranged from 1% to 5%. The discrepancy frequency established
in this Q-Probes study is based on review of selected
cases (a targeted group of cases studied), a bias that
may overestimate overall laboratory error. Error frequencies
partly depend on the method of case detection, and
the more thoroughly one looks for error, the more frequently
one will find it. 1,29 As expected, some of the secondary
review methods used in this study detected more
error than other methods; for example, clinician-directed
review detected a discrepancy (23.0% of cases) more frequently
than random review (4.3%). In general, methods
0.3
0
0
0
0
0
0
0
0
0
0
0
0
0

Primary reason for review (P .001)
Request by clinician
All other reasons
Discrepancy type: change in categoric interpretation
Specimen type (P .001)
Cytology
Surgical pathology
Primary reason for secondary review (P .03)
Extradepartmental review
All other reasons for review
Discrepancy type: change in same category of diagnosis
Specimen type (P .001)
Cytology
Surgical pathology
Discrepancy type: change in patient information
Primary reason for secondary review (P .001)
Request by clinician
All other reasons for review
Effect on patient outcome: near miss
Primary reason for secondary review (P .001)
Extradepartmental review
All other reasons for review
Effect on patient outcome: no harm
Primary reason for secondary review (P .001)
Intradepartmental review
All other reasons for review
Response to a discrepancy: report change
Primary reason for secondary review (P .001)
Interdepartmental conference
Request by clinician
All other reasons for review
Table 8. Statistically Significant Associations
No. of Specimens No. With Discrepancy
464 Arch Pathol Lab Med—Vol 129, April 2005 Patient Safety in Anatomic Pathology—Raab et al
348
5812
54
349
89
317
349
54
79
327
80
299
44
335
43
77
250
Table 9. Reason for Review Correlated With Effect on Patient Outcome
Errors
Detected, No.
Effect on Patient Outcome
(% of Cases Harm, Near Miss, No Harm,
Reason for Review
Reviewed) No. (%)
No. (%)
No. (%) Total
Intradepartmental review
47 (7.1)
2 (5.3)
2 (5.3)
34 (89.5) 38
Request by clinician
80 (23.0) 12 (23.5) 3 (5.9)
36 (70.6) 51
Interdepartmental review
48 (4.8)
8 (25.0) 2 (6.3)
22 (68.8) 32
Selected by quality assurance review
127 (4.3) 13 (14.6) 8 (9.0)
68 (76.4) 89
Extradepartmental review
92 (8.6)
8 (14.0) 12 (21.1) 37 (64.9) 57
Total 43 27 197 267
of case review that involve clinical input are a better
means to detect error but are harder to perform.
Pathologists have studied error frequency using intralaboratory
quality assurance methods in some detail. The
most commonly studied review method is correlation,
such as frozen-permanent section correlation or cytologichistologic
correlation, a form of review mandated by the
Clinical Laboratory Improvement Amendments of 1988. 20
Using this review process and the total number of cases
as the denominator, Clary et al 6 reported that 2.26% and
0.44% of nongynecologic cytology and histology cases
were discrepant. This Q-Probes study showed that the error
frequency based on extradepartmental conference review
was 7.1%, whereas Raab et al 30 reported an error
frequency of 8.9%, with severe significant events occurring
in 7.0% of all errors. McBroom and Ramsay 11 reported that
80
335
27
56
26
59
179
15
24
13
15
18
40
243
33
53
103
9.0% of cases reviewed at a clinicopathologic conference
had a change in diagnosis. This similarity in error frequency
across studies and across many institutions may
indicate a true benchmark.
Benchmarking and reducing anatomic pathology error
frequency clearly is just beginning, and prior to this study,
the benchmarks were based on single-institution or anecdotal
data. 2 The correlation of error frequencies with particular
secondary review practices and existing laboratory
error reduction programs is largely unknown. Lack of
subspecialty expertise may contribute to higher error frequencies,
although error data detected using extradepartmental
specialty academic institution review may be biased
and may overreport error. 4,7,31 Underreporting of error
because of biased review methods, lack of understandable
error taxonomy, and individual fears invariably exists,

ut this has not been thoroughly studied in pathology laboratories.
Some leading patient safety researchers, such as Resar
et al, 32 have argued that error prevention programs should
target errors that have an effect on patient outcome, rather
than all errors. These Q-Probes study data show that the
majority of anatomic pathology discrepancies do not result
in harm, similar to the data reported in nonpathology
fields. 1 Determining the effect of a pathology error on patient
outcome is challenging because of contact and temporal
barriers between pathology and clinical care. Thorough
medical record review generally is not performed
after a pathology error has occurred, and health care systems
invariably lack personnel trained in the intricacies of
the triad of pathology reporting, patient safety, and assessing
patient outcomes. Grzybicki et al 27 reported that
even with data from medical record review, pathologists
showed poor agreement in determining if harm actually
occurred. We used pathologist self-assessment to determine
error severity, and acknowledging the biases inherent
in this process, we found that 16.6% of all errors resulted
in some form of patient harm. This indicates that
1.1% of all anatomic pathology cases that underwent secondary
review were associated with a harmful significant
event. Because secondary review processes tend to be varied
across institutions, the probability is high that a relatively
large percentage of pathology errors resulting in
harm go undetected.
Statistical significance testing did not show that any
body site was most likely to be associated with an error.
However, some organs (eg, female genital tract and breast)
tended to have higher associations with a discrepancy resulting
in clinical harm compared with other organs. Using
medical chart review to determine clinical follow-up
of errors detected through cytologic-histologic correlation,
Clary et al 6 reported that the most common cytology specimens
associated with harm were pulmonary and breast
specimens. In a study examining errors associated with
malpractice claims, Troxel and Sabella 15 also identified
that diagnostic errors in breast cytology specimens were
associated with harm; in addition, they identified other
organ or specimen types, such as prostate needle biopsies,
Papanicolaou tests, and melanocytic skin lesions, as being
associated with errors resulting in harm. In this Q-Probes
study, discrepancies detected in cytology specimens were
more likely (compared with surgical pathology specimens)
to have a 2-step change in diagnosis, and this
change often meant that the original diagnosis was either
a false-negative or false-positive diagnosis. These types of
errors are more likely to have a clinical effect than other
types of errors. 2 Harm was more likely to be associated
with cases reviewed at the behest of a clinician and cases
sent for extradepartmental review.
Although researchers have studied anatomic pathology
diagnostic variability, the relationship of variability and
error is not sufficiently addressed in the pathology literature.
Strictly speaking, variability is a form of error, and
secondary case review is simply a means to unearth differences
in diagnosis or errors. In practice, some pathologists
would like to maintain that ‘‘true’’ diagnostic errors
are errors that most reasonable pathologists would agree
are errors. However, establishing the ‘‘true’’ diagnosis is a
complex and controversial task (eg, do we use a panel of
practicing pathologists or experts?) that has not been well
studied in anatomic pathology. Our study did not address
methods of establishing the accuracy of the original and
review diagnoses. However, some methods of secondary
review were performed with knowledge of clinical outcome,
which, although biasing the review pathologist
(compared with the original pathologist, who lacked this
bias), provides a window on the actual effect of the diagnosis.
This Q-Probes study recorded errors other than interpretive
errors. Typographic errors and changes in patient
information infrequently result in harm, but when harm
occurs, it may be severe, with far-reaching consequences
(eg, switching of patient specimens). Pathology laboratories
place checks in the system in order to limit these error
types, although these safeguards are generally not formally
shared across laboratories. The frequency of marked
harm as a result of these errors is known only through
small studies or anecdotally, 9,33,34 resulting in a lack of
widespread system learning. Our study was not detailed
enough to drill down into these error types.
The lack of report clarity is an example of error that is
difficult to quantify and falls within the realm of communication
error. Poor communication is an important
source of error in clinical medicine and may result in severe
harm 35 ; Dovey et al 25 reported that 5.8% of family
practice errors were a result of miscommunication. Report
clarity is subjective, and Powsner et al 28 reported that surgeons
misunderstood 30% of pathology reports. As expected,
in this Q-Probes study, pathologists believed that
the majority of reports were clear, and less than 1% were
perceived as markedly unclear. This confirms the conclusion
by Powsner et al 28 that a communication gap exists
between pathologists and clinicians. The fact that clinicians
request that a certain percentage of cases be reviewed
underscores a potential communication problem
but shows as well a functioning means of detecting error.
Conclusions
Using secondary pathologist review, the mean anatomic
pathology diagnostic discrepancy frequency was 6.7%.
More than 1% of all reviewed anatomic pathology cases
may be associated with an error associated with patient
harm.
The College of American Pathologists provided financial support.
The statistical reviewer was Molly Walsh, PhD, College of
American Pathologists.
References
1. Kohn LT, Corrigan JM, Donaldson MS, eds. To Err Is Human: Building a
Safer Health System. Washington, DC: National Academy Press; 1999.
2. Raab SS. Improving patient safety by examining pathology errors. Clin Lab
Med. 2004;24:849–863.
3. Adad SJ, Souza MA, Etchebehere RM, et al. Cyto-histological correlation of
219 patients submitted to surgical treatment due to diagnosis of cervical intraepithelial
neoplasia. Sao Paulo Med J. 1999;117:81–84.
4. Arbiser ZK, Folpe AL, Weiss SW. Consultative (expert) second opinions in
soft tissue pathology: analysis of problem-prone diagnostic situations. Am J Clin
Pathol. 2001;116:473–476.
5. Chan YM, Cheung AN, Cheng DK, et al. Pathology slide review in gynecologic
oncology: routine or selective? Gynecol Oncol. 1999;75:267–271.
6. Clary KM, Silverman JF, Liu Y, et al. Cytohistologic discrepancies: a means
to improve pathology practice and patient outcomes. Am J Clin Pathol. 2002;
117:567–573.
7. Hahm GK, Niemann TH, Lucas JG, et al. The value of second opinion in
gastrointestinal and liver pathology. Arch Pathol Lab Med. 2001;125:736–739.
8. Furness PN, Lauder I. A questionnaire-based survey of errors in diagnostic
histopathology throughout the United Kingdom. J Clin Pathol. 1997;50:457–460.
9. Hocking GR, Niteckis N, Cairns BJ, Hayman JA. Departmental audit in surgical
anatomical pathology. Pathology. 1997;29:418–421.
10. Labbe S, Petitjean A. False negatives and quality assurance in cervicouterine
cytology. Ann Pathol. 1999;19:457–462.
Arch Pathol Lab Med—Vol 129, April 2005 Patient Safety in Anatomic Pathology—Raab et al 465

11. McBroom HM, Ramsay AD. The clinicopathological meeting: a means of
auditing performance. Am J Surg Pathol. 1993;17:75–80.
12. Nakhleh RE, Zarbo RJ. Amended reports in surgical pathology and implications
for diagnostic error detection and avoidance. Arch Pathol Lab Med. 1998;
122:303–309.
13. Ramsay AD, Gallagher PJ. Local audit of surgical pathology: 18 months
experience of peer-review–based quality assessment in an English teaching hospital.
Am J Surg Pathol. 1992;16:476–482.
14. Safrin RE, Bark CJ. Surgical pathology signout: routine review of every case
by a second pathologist. Am J Surg Pathol. 1993;17:1190–1192.
15. Troxel DB, Sabella JD. Problem areas in pathology practice uncovered by
a review of malpractice claims. Am J Surg Pathol. 1994;18:821–831.
16. Whitehead ME, Fitzwater JE, Lindley SK, et al. Quality assurance of histopathologic
diagnoses: a prospective audit of 3 thousand cases. Am J Clin Pathol.
1984;81:487–491.
17. Zuk JA, Kenyon WE, Myskow MW. Audit in histopathology: description of
an internal quality assessment scheme with analysis of preliminary results. J Clin
Pathol. 1991;44:10–15.
18. Lind AC, Bewtra C, Healy JC, et al. Prospective peer review in surgical
pathology. Am J Clin Pathol. 1995;104:560–566.
19. Zardawi IM, Bennett G, Jain S, et al. Internal quality assurance activities
of a surgical pathology department in an Australian teaching hospital. J Clin
Pathol. 1998;51:695–699.
20. Jones BA, Novis DA. Cervical biopsy-cytology correlation: a College of
American Pathologists Q-Probes study of 22439 correlations in 348 laboratories.
Arch Pathol Lab Med. 1996;120:523–531.
21. Zarbo RJ. Monitoring anatomic pathology practice through quality assurance
measures. Clin Lab Med. 1999;19:713–742.
22. Raab SS, Jones BA. Q-TRACKS: Gynecologic Cytologic-Histologic Correlation:
2003 Annual Summary. Northfield, Ill: College of American Pathologists;
2003.
23. Brixey J, Johnson TR, Zhang J. Evaluating a medical error taxonomy. Proc
AMIA Symp. 2002:71–75.
24. McNutt RA, Abrams RI. A model of medical error based on a model of
disease: interactions between adverse events, failures, and their errors. Qual Manag
Health Care. 2002;10:23–28.
25. Dovey SM, Meyers DS, Phillips RL Jr, et al. A preliminary taxonomy of
medical errors in family practice. Qual Saf Health Care. 2002;11:233–238.
26. Tamuz M, Thomas EJ, Franchois KE. Defining and classifying medical error:
lessons for patient safety reporting systems. Qual Saf Health Care. 2004;13:13–
20.
27. Grzybicki DM, Vrbin-Turcsanyi CM, Janosky J, Raab SS. Examining pathology
errors to improve patient safety: pathologists don’t agree on the identification
of errors due to pathologist misinterpretation. Paper presented at:
AcademyHealth Annual Research Meeting; June 8, 2004; San Diego, Calif.
28. Powsner SM, Costa J, Homer RJ. Clinicians are from Mars and pathologists
are from Venus. Arch Pathol Lab Med. 2000;124:1040–1046.
29. Nieva VF, Sorra J. Safety culture assessment: a tool for improving patient
safety in healthcare organizations. Qual Saf Health Care. 2003;12S:ii17–23.
30. Raab SS, Clary KM, Grzybicki DM. Improving pathology practice by review
of cases presented at chest conference [abstract]. Mod Pathol. 2003;67:
312A–313A.
31. Tsung JS. Institutional pathology consultation. Am J Surg Pathol. 2004;28:
399–402.
32. Resar RK, Rozich JD, Classen D. Methodology and rationale for the measurement
of harm with trigger tools. Qual Saf Health Care. 2003;12S:ii39–45.
33. Cree IA, Guthrie W, Anderson JM, et al. Departmental audit in histopathology.
Pathol Res Pract. 1993;189:453–457.
34. Ramsay AD. Errors in histopathology reporting: detection and avoidance.
Histopathology. 1999;34:481–490.
35. Gawande AA, Zinner MJ, Studdert DM, Brennan TA. Analysis of errors
reported by surgeons at 3 teaching hospitals. Surgery. 2003;133:614–621.
466 Arch Pathol Lab Med—Vol 129, April 2005 Patient Safety in Anatomic Pathology—Raab et al