Covered by
SCI-Expanded
|
|
[ Summary ]
[ PDF ]
[ Mail to Author ]
[ Mail to Editor ]
|
| Radiological report: expectations of clinicians |
| Nurullah Doğan1, Zeynep Nigar Varlıbaş3, Özge Petek Erpolat2 |
1From the Departments of Radiology, Kütahya Evliya Çelebi Government Hospital, Kütahya, Turkey
2From the Departments of Radiation Oncology, Kütahya Evliya Çelebi Government Hospital, Kütahya, Turkey
3Department of Radiology, Uludağ University School of Medicine, Bursa, Turkey |
| Keywords: • radiology • communication • standardization |
| DOI: 10.4261/1305-3825.DIR.2820-09.1 |
| Summary |
PURPOSE
Although there have been many publications on composing
an accurate radiological report, they usually do not include an
assessment of the clinicians' expectations from a radiological
report. In this study, we aimed to assess the clinicians' expectations
and preferences in terms of radiology report style and
content.
MATERIALS AND METHODS
A multiple-choice questionnaire, containing 19 questions, was
formed. Two-hundreds clinicians, working either in a university
hospital or a public hospital, were allocated into 4 groups
which included equal number of clinicians from surgery and
internal medicine departments. Questionnaire was applied to
participants by face-to-face interview. Results were analyzed
for each group using Pearson chi-square test.
RESULTS
No statistically significant difference was found among four
groups except for the 16th question which was about the
image format pertaining to the report (CD/DVD or negative
film). It has been determined that clinicians preferred detailed,
standardized radiological reports with complete sections (i.e.,
clinical information, technique, findings, conclusion, recommendations).
CONCLUSION
This study provided essential data for radiologists to write
more effective reports. |
Top
Summary
Introduction
Methods
Results
Disscussion
References
|
| Introduction |
The radiological report is the most significant vehicle of communication
between a radiologist and a clinician, but it is naturally a
one-sided communication. For the most part, radiologists do not
know how their reports are evaluated by clinicians. Furthermore, radiologists
have individual and idiosyncratic ideas about composing reports
that vary significantly. Ultimately, there is no consensus on the part of
clinicians or radiologists about radiological reporting.
The aim of this study was to examine how clinicians evaluate radiological
reports and determine what they expect from a radiologist. The
ultimate goal was to contribute to the standardization of radiological
reporting. |
Top
Introduction
Methods
Results
Disscussion
References
|
| Materials and Methods |
Study group selection
This questionnaire-based study included specialists from surgery and
internal medicine departments. Medical practitioners, residents, fellows,
and basic scientists were excluded.
The study was conducted with two hundred doctors; 100 were randomly
selected from public hospitals located in Bursa, Turkey (Çekirge
Public Hospital, İnegöl Public Hospital and Ali Osman Sönmez Oncology
Hospital), and the other 100 were selected from the Uludağ University
School of Medicine Research Hospital. Equal numbers of specialists were
selected from the surgery and internal medicine departments of public
or university hospitals. Thus, four groups were constructed, each including
50 specialists from either university hospitals or public hospitals.
The features of the participants, including gender, age, affiliations, and
academic degrees are shown in Table 1, and the distribution of those
features by medical department is shown in Table 2.
Click to Enlarge |
Table 2: Distribution of the participants by medical department |
Questionnaire
A 19-question multiple-choice questionnaire was prepared for the
study. The questions were administered during a face-to-face interview
conducted by radiology residents. The questions obtained were examined
after classifying the answers according to their goals.
The answers to the first 17 questions are shown in Table 3. Answers to
the last two questions were determined by giving sample reports to the
participants and asking them to choose the most appropriate one (Samples
1 and 2 are shown in Tables 4 and 5, respectively). The results were
assessed with the sample reports that the participants found to be sufficient
and, comparing those results to the answers of other questions.
Click to Enlarge |
Table 3: The first seventeen questions of the questionnaire and their answer choices |
Statistical analyses
Analyses were performed with SPSS software, version 11 (SPSS Inc.,
Chicago, USA). Results from the four groups were analyzed separately and compared using the Pearson chisquared
test. P values less than or equal
to 0.05 were considered statistically
significant.
Research ethics compliance
This study was approved by the
Uludağ University Medical Research
Committee (decision number, 2008-
12/30, June 10, 2008) and was conducted
from July 1, 2008, to August 31,
2008. Informed consent was obtained
from all participants. |
Top
Introduction
Methods
Results
Disscussion
References
|
| Results |
We first asked the clinicians questions
about the sufficiency of the reports,
and 60%, 29%, 4.5%, 3.5%, and
3% of the clinicians stated that 75%,
50%, <25%, 25%, and 100% of the radiologic
reports were sufficient, respectively.
When we examined the origins
of the insufficient reports, those originating
from universities were found
to be rated sufficient more frequently
(only 3.5% were rated insufficient).
The results from all institutions (37%), public hospitals (33.5%), and private
imaging centers (26%) were similar,
and no statistically significant difference
was found.
When clinicians were asked their
opinions about the sufficiency of the
clinical information they provided,
53.5% rated their clinical information
as sufficient, while 41.5% reported that
they noted some clinical information,
though it was short. Only 5% of clinicians
admitted that they did not give
any clinical information due to patient
load. No clinicians indicated that providing
information was unnecessary.
When asked about their reaction to
receiving a long report, 46% of clinicians
stated that they read the results,
but only read the rest if they needed
more information; 39% stated that
they read the whole report. However,
in response to the 18th question which
was about radiology reports with different
levels of detail (Table 4), 72% of the
participants preferred detailed reports,
22.5% preferred reports without much
detail, and 5.5% preferred a very short
report. In response to the 19th question
which was also about radiology reports
with different levels of detail (Table
5), 64.5% of the participants preferred
a standardized, detailed report, and
35.5% preferred a summarized report.
Questions five through twelve addressed
the types of content that would
be expected in a sufficient report. Most
of the clinicians (92.5%) requested a
detailed description of the features of
the lesion. If there was more than one
lesion, they (65%) requested a detailed
description of each lesion. In addition,
56% of the clinicians requested
lesion descriptions without radiological
terminology (e.g., Westmark sign,
hypointense on T1-weighted images,
hyperintense on T2-weighted images).
Instead, clinicians preferred that the
features be indicated with terms that
they were more familiar with (e.g., calcification,
necrosis, hemorrhage). Only
a subgroup of participants (30%) considered
the use of radiological terms
as necessary. Sixty-five percent of the
participants thought that the lesions
should be described in detail for examinations
containing many similar
lesions. Ninety-two percent of the
clinicians believed that the anatomic
localization of the lesion should be
indicated in detail. Almost half of the
clinicians (46.5%) requested reporting
of pathologic lesions in an order that started with the most important lesion.
Interestingly, 53.5% thought that
there should be a printed report format,
and the lesion should be defined
there (in italics or bold) when describing
the lesion structure and pathology.
Although 37.5% of clinicians evaluated
the description of basic anatomic
structures as sufficient, 36% percent of
clinicians asked for a description of all
of the examined anatomical structures.
Other clinicians (26.5%) evaluated the description of normal anatomic structures
as unnecessary, and 35% of the
participants considered reporting of
the normal results as unnecessary. Interestingly,
32% of the participants
asked for the reporting of the anatomic
structure, even if it was normal because
they wanted the information to
assist in the assessment of the clinical
condition (e.g., sizes of the liver for a
patient who is under follow up due to
the diagnosis of hepatitis). In addition, 19.5% of the clinicians asked for the reporting
of measures of basic anatomic
structures (e.g., size of the spleen), and
13.5% of the subjects requested measurement
results (e.g., size, diameter) of
all of the anatomical structures, even if
they were in the normal range.
In response to the 11th question
which was about the certainty with
which the radiologist reports an abnormal
finding (Table 3), 56% of clinicians
commented that they were sometimes uncertain of the radiological diagnosis
based on the clinical data they provided.
The majority of clinicians (70.5%)
thought that a recommendations section
at the end of the report would be
helpful. However, a recommendations
section was not believed to be helpful
by 29.5% of the clinicians: 9.5% expressed
that “patients reading those
recommendations put the clinicians in
a tight spot”, 8.5% stated that “the clinician
will decide which examination
he will request”, and 11.5% suggested
both of the causes.
Most of the clinicians who participated
in our questionnaire (73%) opposed
a report written using Turkish
terms. The clinicians requested the use
of universal medical terminology for
the following reasons: 15% suggested
that “patients read the report”, 28.5%
suggested that “everybody knows this
universal medical terminology”, and
29.5% put forward both of the reasons.
The 14th question addressed the
marking of lesions, and 73% of the
clinicians requested marking of the lesion.
In answer to the question, 14% requested
this as a cross-sectional image
number of the lesion, 16.5% requested
this as marking on the film, and 42.5%
requested both.
When we asked the clinicians how
they preferred to receive the report,
we found that sending it with the patient
or their relatives was sufficient
with 85% of the clinicians. Only 27%
requested the report in a closed envelope,
2% of the clinicians preferred receiving
the report by courier, and 13%
of the clinicians preferred receiving the
report electronically (e.g., e-mail, hospital
operating system).
With the questions so far, no statistically
significant difference was found
between clinicians at public or university
hospitals. Interestingly, there was a
significant difference between these two
groups of clinicians (P = 0.005) when
they were asked about the examination
imagery format attached to the report
(16th question). The answers to the
16th question are shown in Table 6. Although
37% of clinicians at public hospitals
preferred the examination to be
printed on negative film, we found that
clinicians working at a university hospital
preferred a CD or DVD. Interestingly,
only 18% of the clinicians working in
public hospitals preferred a CD or DVD.
When the answers to other options of
both the groups were evaluated together we found that 13.5% of the clinicians
marked the option “presentation of the
image is not important if the patient
has a sufficient report”, and 10% of the
clinicians marked the option “only the
pathological lesions should be printed
on a film”. Furthermore, we did not
find any differences between groups according
to the classification of clinicians
into their surgery or internal medicine
department (P > 0.05).
Click to Enlarge |
Table 6: The answers of the participants to the 16th question, according to the type of institution |
Finally, we attempted to understand
how much assistance the clinicians
wanted to get from consultations with
radiologists, and 75.5% percent of the
clinicians indicated at least some level
of assistance: 43.5% of the subjects
answered “occasionally”, and 32% of
the subjects expressed that they would
“regularly” consult radiologists. Conversely,
16.5% of the clinicians did not
feel that they needed help from the
radiologists and stated, “If I need to,
I will call and ask my radiologist colleagues,
so there is no need for an extra
consultant”. Another 4.5% of clinicians
stated, “It is not required, I rarely
need to discuss cases and I can always
contact my other specialist colleagues
when I do (not a radiologist)”.
The answers to all of the questions
according to the choices given are
shown in Table 7.
Click to Enlarge |
Table 7: The percentages of answers to all questions according to the choices given |
|
Top
Introduction
Methods
Results
Disscussion
References
|
| Discussion |
The only communication vehicle
between a radiologist and a clinician
is a report, which is composed by the
radiologist. A clinician only knows
the radiologist by his/her reports, and
the radiologist usually does not know
the clinician who is receiving his/her
report. Most radiologists do not know
how their not reports are evaluated or
what is expected by clinicians in the
radiological reports.
Although there are many reports
about radiological examinations and
the quality of reports[1–6], few studies
have examined the clinicians' expectations
of radiologists. Clinger et al.[7]
were the first researchers to examine
clinicians' expectations for radiological
reports. In a study by McLoughlin
et al.[4], they found that radiologists
did not pay sufficient attention to the
requests of the clinician who referred
the patient.
In our study, we initially asked the
clinicians some questions about the
sufficiency of radiological reports.
The reports of university origin were deemed to be more sufficient (96.5%).
The latter percentage was higher than
the reports from public hospitals or
private imaging centers. These findings
may be the result of the high quality of
examinations in the university hospital,
which was reported in the study of
Ozsunar et al.[2]. In the Ozsunar et al.
study, the examinations were categorized
into three groups: university hospital,
state hospital and private health
center. These groups were compared
for overall quality of examinations.
There was no difference between the
state hospitals and the private health
centers. However, there were significant
differences between the university
and state hospitals (P = 0.03) and the
university and private health centers (P
= 0.04). They stated that quality control
and standardization was becoming
more important in radiological services.
According to the Ozsunar et al.
study, we believe that the high quality
of the examinations was related to the
sufficient reports.
The greatest request of radiologists
to clinicians is that they include some
clinical information; however, when
this happens, the information is usually
insufficient or it is not legible. The
conditions that contain written clinical
information on the request form,
which is required for the diagnosis of
the patient, are not common. According
to our study, 53.5% of clinicians
write sufficient clinical information,
and 41.5% note only short amounts of
clinical information. The description
of clinical information is different for
radiologists and clinicians.
One of the issues amongst radiologists
pertains to what should be contained
in the results section of the report.
Some radiologists compose short
reports by focusing on the pathological
state while others draw up very detailed
reports about every structure
observed (sometimes about structures
not observed as well). When we asked
the clinicians how they react to a long
report, 39% stated that they read the
whole report. We decided to evaluate
the answer to this question in combination
with the answers to the last
two (18th and 19th) questions of the
questionnaire. In the 18th question,
very short, not detailed and detailed
types of reports were presented, and
the participants were asked which one
they preferred. Most of the participants
(72%) preferred the detailed report. In the 19th question, two report examples
for the same patient were presented;
one was in short form, and the other
was a standardized, detailed report. Almost
65% of the clinicians preferred
the standardized, detailed report. In
both of these questions, pathology
was indicated in the same manner.
The only difference was the addition
of the normal structures to the report
or a change in the presentation of the
report, which included sections for
clinical information, methods, findings
and conclusion. When the report
was evaluated with the results of the
three questions, we found that clinicians
preferred standardized, detailed
reports regardless of whether they read
the whole report. In a study by Naik
et al.[8] that examined 25 radiologists
and 95 clinicians, six different report
samples with three clinical scenarios
were formed similarly to our 18th and
19th questions, and participants were
asked which report they preferred.
This study also found that most of the
participants preferred standardized, detailed
reports.
A series of questions was designed
to understand what qualified as a sufficient
report. Most of the clinicians requested a detailed description about
all of the features of each lesion regardless
of the number of lesions. In addition,
most of the clinicians requested
lesion descriptions that indicated the
features that could be commented on
by clinicians (e.g., calcification, necrosis,
hemorrhage) rather than the use of
radiologic terminology. However, this
request was for a diagnosis that is more
appropriately made by a pathologist,
though it was still expected from the
radiologist.
Furthermore, there was no consensus
amongst the clinicians about the
format of the report. Similarly, there
was no consensus about reporting the
results of normal measures. In a study
by Naik et al.[8], the authors were not
sure about the examination of the organ
if it was not noted in the report.
However, in the same study, most of
the participants preferred standardized,
printed reports, which did not agree
with the present findings. In another
study, Gagliardi[3] emphasized the
significance of standardized reports.
Similarly, in an evaluation performed
on 104 clinicians, Lafortune et al.[9]
concluded that “radiological reports
should be explicit, should give direct answers to clinical questions requested
and should contain findings and conclusion
sections”.
When the clinicians were asked how
they interpreted an uncertain finding
in the radiologist's report, 28% stated
that they considered all uncertain sentences
as positive for pathology. This
finding is very interesting and must be
taken seriously. Interpreting all uncertain
sentences as positive for pathology
will affect the treatment of the
patient, and mistakes that are made
in the comments of suspected lesions
may have serious consequences. Only
15.5% of the subjects stated that they
might request additional examinations
due to uncertain expressions.
Interestingly, in situations where the
conclusion is not definitive but there
is suspicion of a positive pathology, a
recommendation of additional examinations
would be an integral part of
the report. Similar to the findings in
the Lafortune et al. study[9], which
found that the radiological report
should contain conclusion and recommendation
sections, the present
study found that a recommendations
section was requested by the majority
of the clinicians.
Reporting in Turkish terms has been
an issue in radiological reports in Turkey
for years. Recently, a number of
reviews[10–13] have recommended
the use of Turkish terms in radiological
reports, which is in opposition to our
study. Indeed, the present study found
that most clinicians do not want the
patients to read the reports, and they
stated that the use of universal medical
terms between medical doctors supported
better communication.
We also determined that opinions
on issues such as marking a lesion on
the film and transporting the report to
the clinician in a way that does not
allow the patient to see the content
of the report (e.g., by courier) were
widely accepted by radiologists but
not agreed upon by clinicians. We
determined that clinicians expected
sections of the reports to be complete
and expected a section describing recommendations.
Additionally, most of
the clinicians stated that they wanted
to communicate with radiologists via
consultation before and after the radiologic
examination.
Limitations of our study included
the general limitations of studies based
on questionnaires. A primary limitation
is that this was a sampling study
conducted only around Bursa, Turkey.
Therefore, we cannot claim that it represents all of the clinicians around the
whole country or world. Additionally,
the experimental groups were heterogeneous
groups containing many specialties.
Different results would likely
be obtained if the questionnaire were
used in a larger, more homogenous
group (e.g., neurologists, surgeons).
The present study also contained some
subjectivity, which is a common feature
of studies based on questionnaires.
The accuracy of the answers is only
possible if the subjects answer honestly.
This accuracy, however, cannot be
tested objectively with a questionnaire
method.
We believe that the present study
provided essential data for radiologists
to write more effective reports. If this
questionnaire was modified and applied
to a larger, more homogenous
group, it would be possible to test our
results and obtain new data.
Acknowledgement
We would like to express our gratitude to
Ercan Tuncel, MD, who gave us the opportunity
to complete our study. The preparation of
the manuscript would not have been possible
without his support. |
Top
Introduction
Methods
Results
Discussion
References
|
| References |
1) Tuncel E. Radyogramların değerlendirilmesi
ve rapor yazma. Tani Girisim Radyol 1996;
2:5–10.
2) Özsunar Y, Çetin M, Taşkın F, et al. The
level of quality of radiology services in
Turkey: a sampling analysis. Diagn Interv
Radiol 2006; 12:166–170.
3) Gagliardi RA. The evolution of the Xray
report. AJR Am J Roentgenol 1995;
164:501–502.
4) McLoughlin RF, So CB, Gray RR, Brandt
R. Radiology reports: how much descriptive
detail is enough? AJR Am J Roentgenol
1995; 165:803–806.
5) Schreiben MH, Leonard JM, Rieniets CY.
Disclosure of imaging findings to patients
directly by radiologists: survey of patients'
preferences. AJR Am J Roentgenol 1995;
165:467–469.
6) Stavema K, Fossa T, Botnmarka O,
Andersend OK, Erikssenb J. Inter-observer
agreement in audit of quality of radiology
requests and reports. Clin Radiol 2004;
59:1018–1024.
7) Clinger NJ, Hunter TB, Hillman BJ.
Radiology reporting: attitudes of referring
physicians. Radiology 1988; 169:825–826.
8) Naik SS, Hanbidge A, Wilson SR. Radiology
reports: examining radiologist and clinician
preferences regarding style and content.
AJR Am J Roentgenol 2001; 176:591–598.
9) Lafortune M, Breton G, Baudouin J. The
radiological report: what is useful for the
referring physician? J Can Assoc Radiol
1988; 39:140–143.
10) Aydıngöz Ü. Radyoloji dili: temel
sorunlarımız. Tani Girisim Radyol 2003;
9:5–9.
11) Akan H. Tıpça üzerine Türkçe düşünceler.
Tani Girisim Radyol 2003; 9:131–134.
12) Berkmen YM. Bilim dilinin Türkçeleşmesi.
Tani Girisim Radyol 2003; 9:275–278.
13) Işık S. Bilim dili Türkçe. Tani Girisim
Radyol 2004;10:93–95. |
Top
Introduction
Methods
Results
Discussion
References
|
|
[ Top ]
[ Summary ]
[ PDF ]
[ Mail to Author ]
[ Mail to Editor ]
|
|
|
Ahead of Print)
)
)
)
)
)
)