You are on page 1of 8

bs_bs_banner

Journal of Evaluation in Clinical Practice ISSN 1365-2753

Delays and errors in abnormal chest radiograph follow-up:


a systems approach to promoting patient safety in radiology
Basrull Najmi Bhaludin MA MBBChir MRCP FRCR,1 Susan Cheng Shelmerdine BSc MBBS MRCS
FRCR,1 Sonal Arora PhD MRCS MBBS BSc,2 Taiwo Senbanjo MBBS MRCP FRCR3 and Arum Parthipun
MBBS MRCP FRCR3
1
Specialty Registrar in Radiology, Department of Radiology, St. Georges Healthcare NHS Trust, London, UK
2
Clinical Lecturer, Clinical Safety Research Unit, Department of Bio-Surgery and Surgical Technology, Imperial College London, London, UK
3
Consultant Radiologist, Department of Radiology, Epsom and St. Helier University Hospitals NHS Trust, Carshalton, UK

Keywords Abstract
clinical audit, clinical governance, clinical
safety Rationale, aims and objectives This study aimed to apply the systems approach to
patient safety in order to identify causes for delays and errors in lung cancer diagnoses
Correspondence following an abnormal chest radiograph.
Dr Basrull Najmi Bhaludin Methods In the first part of this study, the systems approach to patient safety was com-
Department of Radiology prehensively reviewed by three radiologists and seven patient safety experts. In the second
St. Georges Healthcare NHS Trust part of this study, a retrospective review was performed of all patients referred to the lung
Blackshaw Road cancer multidisciplinary team (MDT) meeting over a 1-year period. All abnormal chest
London SW17 0QT radiograph reports were examined and a rootcause analysis performed of cases where
UK errors and delays in diagnoses were deemed to have occurred.
E-mail: bnb21@cantab.net Results A total of 124 cases were reviewed, of which 36 (29%) patients had an abnormal
preceding chest radiograph prior to MDT referral. In six cases, serious errors from delay
Accepted for publication: 17 April 2014 and lack of follow-up were identified. These are analysed and discussed in detail in this
article. Application of the systems approach to each case identified poor communication
doi:10.1111/jep.12178
and lack of clinical action as prime causes.
Conclusions Both reporting radiologists and referring clinicians have a responsibility to
ensure appropriate action following an abnormal chest radiograph. The main error lies in
communication between the referring clinicians and the radiologists. Direct electronic
communication is potentially a more robust method to overcome this.

culture, a systems approach to analysing error is imperative [4].


Introduction This method of conducting a rootcause analysis hypothesizes
Iatrogenic errors remain a significant source of harm with approxi- that errors in the care process are rarely due to the fault of one
mately 1 in 10 patients suffering an adverse event during their single individual but rather a series of inter-related factors within
hospital stay [1]. Improving quality and safety is consequently an organization or team. Although well recognized as a concept
a high priority for patients, the public and the profession alike. in identifying shortcomings relating to surgical and medical
Radiology is not exempt from this problem with errors in the care, the systems approach in radiology has never been previously
order of 1020% when reporting plain radiographs. Up to 30% of published [57].
medical malpractice claims in the United States are due to diag- Most current literature pertaining to radiological error focuses
nostic errors in radiology [2]. High levels of inter-observer vari- primarily on specific inadequacies of a single radiologist such as
ability among radiologists have been cited a significant source of lack of experience or knowledge in interpretation [8] yet little is
such error [2,3]. written relating to contributory external factors, nor the subsequent
However, as in many high-risk industries, the aetiology of error delays or impact on patient care. In this paper we aim to expand
is multifactorial and not simply due to poor performance of a our assessment of radiological error using the systems approach.
single radiologist. Other high-risk, high-reliability industries such Specifically, we aim to use this validated methodology to investi-
as oil and aviation have identified that in order to promote a safety gate the extent of delays in diagnosis of patients with lung cancer

Journal of Evaluation in Clinical Practice 20 (2014) 453459 2014 John Wiley & Sons, Ltd. 453
Patient safety in radiology B.N. Bhaludin et al.

referred under the 2-week rule where a report on an abnormal


chest radiograph had been issued. The 2-week rule is a UK
Results
government initiative whereby patients suspected of a new diag- Based on the systems approach, Table 1 highlights a list of factors
nosis of cancer must be seen by a specialist within 2 weeks of that may impact upon the provision of safe radiological practice.
referral from their general practitioner. These include patient factors, individual factors, team factors,
Reasons for delays in follow-up imaging and clinical manage- equipment factors, environmental factors and organizational/
ment are scrutinized and suggested improvements proposed by governance factors. They have been adapted from Vincent et al.s
which similar future events may be avoided. Framework of factors influencing clinical practice to reflect
radiology [9].
Regarding the results from the second part of this study, a
total of 144 patients were diagnosed with lung cancer during the
Methods 1-year period of this study. Twenty patients were subsequently
In the first part of this study, the systems approach to patient safety excluded from the study as 15 patients did not have any imaging
was comprehensively reviewed by three radiologists and seven reports or images available for review on our system and five had
patient safety experts (defined as >5 years of employment and >10 been referred with an established diagnosis of mesothelioma. The
publications in patient safety) from the UK Centre of Patient patients with mesothelioma were excluded because the diagnostic
Safety and Service Quality. This group worked together in a focus pathway is often complicated and the diagnosis is rarely made on
group style format using the systems approach to develop a frame- the basis of a single abnormal chest radiograph report.
work of factors in radiological practice, which may contribute to Of the remaining 124 patients, 115 (92.7%) had previous chest
delays and errors in care. radiographs that had been appropriately reported and acted upon.
The study was carried out at Epsom and St. Helier University Thirty-six out of the 124 patients (29%) had findings suspicious
Hospitals NHS Trust over a 1-year period (1 January to 31 Decem- for a pulmonary malignancy in their preceding radiograph. In 9 of
ber 2009). Ethical approval was not necessary for this study as this the 36 abnormal cases (25%), inadequate follow-up was demon-
was deemed to form part of a service evaluation. strated. Seven of these were referrals from within our trust (either
A retrospective review was performed of all patients referred via the outpatient or emergency department) and two referrals
under the 2-week rule to the lung cancer multidisciplinary team were from general practice.
(MDT) meeting. In 3 of the 9 cases, the final diagnosis of the abnormal chest
The patient details were obtained via the local lung cancer MDT radiograph was found not to be due to cancer. These have been
meeting database. For each case referred over this 1-year period, termed near misses in our study flow chart (2.4% of total referrals).
the chest radiograph images and reports were reviewed using the In the remaining six cases (16.6%), a diagnosis of lung cancer was
radiology information system (RIS) and patient archiving and subsequently made after a delay in diagnosis from a failure to
communications system (PACS). Both RIS and PACS packages follow up a chest radiograph report (Fig. 2).
are software used commonly in UK radiology departments to store Reasons for such delays were multifactorial (Table 2) and
patient information and images. PACS software is the viewing included contribution due to the lack of awareness for or sugges-
platform by which clinicians can view radiological images; RIS tion of further imaging in radiology reports (three cases), lack of
software allows radiological departments to place reports relating awareness of need for clinical action by clinician based on abnor-
to the images acquired and book appropriate imaging appoint- mal chest radiographs (four cases), administrative errors in faxing
ments for patients. reports and checking medical documentation had been received
The review team consisted of two experienced consultant radi- (two cases), delays in reporting chest radiographs (one case),
ologists who are part of the lung cancer MDT. Patients with a failure of clinicians to read radiology reports (two cases) or to
previously reported abnormality on a chest radiograph were review imaging in outpatient clinic appointments (one case). Our
identified by both reviewers from the database. Using a detailed results demonstrate the prevalence of poor communication links
questionnaire (Fig. 1), each reviewer screened the clinical notes of between the radiology and referring clinician. There is also a lack
these patients, noting down whether a primary lung carcinoma was of action or realization for follow-up by clinicians even once
reported on the previous chest radiographs and decided whether abnormalities in a report had been mentioned.
the chest radiographs had been followed up appropriately. The
reviewers identified whether any adverse events had occurred for
the cases that were not followed up appropriately. An adverse
Discussion
event was defined as a significant delay in the patients diagnosis Out of the 124 patients identified in our retrospective analysis,
of lung cancer, which affected the patients life expectancy and there were 36 (29%) patients with a previous abnormal chest
was categorized as a serious error. Incidents that did not cause radiograph showing an abnormality suspicious of malignancy. Of
any adverse events to the patient at the time but may have had the these, nine (25%) patients demonstrated previous abnormal chest
potential to cause harm were categorized as near miss. The radiograph findings that were not acted on appropriately. In six of
reviewers also identified the underlying causes for the errors using these cases there was a delay in cancer diagnosis and treatment.
the systems approach to patient safety. Our findings echo the results from similar studies both abroad
All cases where there was a previously reported abnormality on and locally. The largest and most frequently quoted study on the
the chest radiograph were discussed between the two reviewers in subject of delayed lung cancer diagnosis has been the Lung Cancer
a formalized meeting, and a joint consensus was reached for each Claims Study performed by the Physician Insurers Association
case whether the patients had been appropriately followed up. of America and the American College of Radiologists [10]. This

454 2014 John Wiley & Sons, Ltd.


B.N. Bhaludin et al. Patient safety in radiology

Figure 1 Detailed questionnaire used by the


case reviewers.

study closely evaluated data from 184 paid claims, closed between less, there has been no substantial research to date to identify the
1985 and 2004, arising from failure to diagnose lung cancer. The prevalence of incidents specifically involving failure to read or
study cites communication issues as the most prevalent cause for a follow-up radiology reports.
delay in diagnosis, reported in 52.6% of cases. Failure to respond Standards of care and guidelines for recommending further
to an abnormal chest radiograph occurred in 45.7% of cases. imaging do exist and have been published within the Royal
The largest study from the United Kingdom performed by College of Radiologists framework for risk management [12].
Turkington et al. [11] in 2002 found that of a total of 58 patients They state that, where there are unexpected findings likely to
with a proven diagnosis of lung cancer, 14 (24%) had previously affect the patients management, or if the severity of the condition
reported abnormal chest radiographs. Of these, an accurate report is greater than expected, it is the responsibility of the radiologist
was issued by the radiologist but not seen or acted upon by the to inform a member of the clinical team by direct discussion or
physician in 5%. In 11% of cases follow-up was suggested in the other means, however this is not a legal duty. An updated stand-
radiologists report but not carried out by the physician. Neverthe- ards document on the topic of communication of critical, urgent

2014 John Wiley & Sons, Ltd. 455


Patient safety in radiology B.N. Bhaludin et al.

Table 1 Systems approach to radiological practice: principle and influencing factors contributing to safe patient care

Factor types Influencing factors

Patient factors Limitations in patient positioning/preparation for radiological procedures


Poor communication of clinical information by patient to clinicians/radiologists
Poor attendance to radiology procedures
Individual factors Experience with radiological procedures/reporting
Self-awareness of radiologists knowledge and limitations
Diagnostic challenges due to abnormal presentation of pathology
Fatigue, stress, illness, competing commitments
Team factors Effective communication between clinicians and radiologists; for requesting images, adding information, clarification
of reports
Adequacy of information provided on request forms by clinicians
Appropriate radiological follow-up recommended by radiologists and acted on by clinicians in a timely manner
Situational awareness: ability to rapidly relay information to clinicians where change of management/urgent
treatment/unexpected result
Appropriate supervision during image acquisition and reporting
Inter-observer variability
Equipment factors Dated machinery with poor image production
Inadequate settings/protocol for type of scan
Poor software design resulting in end-user confusion when reporting/reformatting images
Poor computation of reporting workflow resulting in confusion over images to be reported/accidentally authorized
Environment factors Access to viewing monitors and appropriate lighting conditions
Access to scanned patient/online order comms request forms
Up to date PACS software on all reporting ports
Quiet reporting room
Availability of patient records online
Accuracy of voice recognition software to produce accurate reports
Distractions and interruptions due to phone calls/MDT meetings/emergency requests/teaching
Organizational/ Workforce management availability of personnel to carry out procedures/perform reporting (cross cover minimized)
governance factors Quality assurance checks in place and not out of date
Quality of image reporting outsourcing
Timely access of patients to radiological studies and clinicians to reports
Infrastructure: adequate email/fax/telephone facilities to facilitate communication between radiological and hospital
departments
Processes for feedback on radiology procedures in place (both from clinicians and patients)
Adequate review of radiology discrepancies at clinical governance/discrepancy meetings

MDT, multidisciplinary team; PACS, patient archiving and communications system.

and unexpected significant radiological findings was published in


2008 [13], reiterating the responsibility in reading and acting upon
abnormal radiology reports to lie with the referring clinician but
also a need for appropriate information technology support and
resource to achieve ease of access to such reports.
Following 22 reports received by the National Patient Safety
Agency (NPSA) in 2007 related to patient safety incidents result-
ing from lack of imaging follow-up, NPSA has issued Safer
Practice Notice 16 on 5 February 2007. In this notice, guidelines
have been outlined to introduce changes to ensure that radiological
imaging results are communicated and acted on appropriately [14].
The notice also includes clear recommendations to all radiology
departments to ensure that there is a robust system in place to
ensure imaging reports are accurately communicated to the refer-
ring clinicians in a timely manner. The guidelines published by the
National Institute of Clinical Excellence first published in 2005
also recommend copying the report of any findings of suspected
malignancy to the MDT coordinator and the respiratory physician
[15]. Figure 2 A summary of the total patients included in this study and their
In accordance to these guidelines, our current trust policy for outcomes. These six individual cases are described in further detail in
alerting clinicians to all serious or unsuspected diagnoses is to Table 2.

456 2014 John Wiley & Sons, Ltd.


B.N. Bhaludin et al. Patient safety in radiology

Table 2 A summary of the six cases where referral to the specialist lung MDT were delayed and recommendations for future improvements

Case Factors leading to


number failure in referral Case description Suggested improvements

1 Communication A chest radiograph demonstrated a small opacity in the right The report should have been sent by fax to the requesting
midzone of the lung with follow-up radiographs clinician, and there should be an acknowledgement system
recommended in 6 weeks time. There was no evidence introduced to ensure that the report has been read, received
that the report was faxed to the referring clinician, and acted upon. There should also be clear emphasis on the
acknowledged by the clinicians or noted within the patient responsibility of the referring clinician to follow up the results
medical records. of the requested imaging.
A repeat chest radiograph performed a year later As a safety net, the patients general practitioner should have
demonstrated a growth in size of the mass but fortunately been sent a copy of the report.
no evidence on subsequent CT imaging of metastatic If faxing is not an effective method for alerting the clinicians to
disease. urgent findings, then an alternative method may be used such
as email or telephone depending on local hospital preferences
and protocols.
2 Personnel, An outpatient chest radiograph demonstrated a large right All radiographs should be reported by a radiologist and a time
communication, upper lobe mass; however, this study was never reported. frame should be implemented to ensure that these
process and A referral from the patients GP 6 months later for a CT radiographs are reported in a timely manner. The IT/PACS
equipment study demonstrated the speculated mass with multiple department of the hospital should be made aware of any
limitations may enlarged mediastinal lymph nodes. irregularities within the reporting software so that further cases
have contributed can be prevented.
The circumstances by which this radiograph was not All clinicians should be aware of their own radiological requests
reported are unclear. It may have been an error in the to the radiology department and ensure that a report has been
radiology reporting software that did not highlight the issued for documentation at patient follow-up and assessment.
study as unreported, or there may not have been
adequate staffing levels at the time for all radiographs to
be reported and in addition, the radiograph was not
reviewed by the requesting clinician as there was no
record of the findings in the patient notes.
3 Communication A chest radiograph referred from the emergency department Appropriate local guidelines should be put in place so the
with right basal consolidation was reported and a follow-up referring clinicians are aware to re-image patients in such
radiograph in 6 weeks time recommended. This was not situations and abnormal radiographs from patients attending
performed. A repeat radiograph 1 year later demonstrated via the emergency department should be fed back to the
multiple pulmonary masses within the right lower lobe. emergency department clinicians to follow up or refer
appropriately.
4 Communication Two consecutive chest radiographs demonstrated right There should be an acknowledgement system in place where
middle zone consolidation without resolution. A referral for both the GP and MDT coordinator can communicate back to
the case to the lung MDT coordinator was made and the the radiologist to say that the report has been successfully
report was faxed to both the GP and the MDT coordinator. received. A record of receipt of the faxed report will need to
be filed or alternatively other methods of communication, such
as email may be utilized.
The case was however never discussed and there are Additionally, the outcome of the MDT meeting should be fed
currently no processes in place to prove whether the fax back to both the GP and the referring radiologist.
was transmitted successfully to the intended recipients.
The patient was admitted to hospital in acute pulmonary
oedema 1 year later and cross-sectional imaging revealed a
right middle lobe mass.
5 Communication, A chest radiograph for an elderly patient demonstrated a left In this case clinicians and radiologists need to be aware of the
process and pleural effusion. Two repeat radiographs taken 3 and fact that rapidly enlarging effusions may be potentially
individual factors 6 months later demonstrated a gradual increase in the malignant. A system should be in place to ensure that any
effusion size. No recommendation for patient referral was abnormal radiographs referred by clinicians are adequately
made in any radiology report and it was only after the third relayed.
radiograph did the GP refer the patient for a respiratory Specialist referral in the radiology report should have been
review. Pleural aspiration revealed malignant cells and the stated.
patient was subsequently diagnosed with an
adenocarcinoma.
6 Communication, Right middle and lower lobe consolidation was identified on It is unclear whether the radiographs were reviewed in the
equipment and an outpatient clinic chest radiograph without any resolution outpatient clinic or whether malignancy was suspected by the
process factors at a repeat radiograph after 3 months. clinician. If this were the case, it would be unlikely that the
An enhanced CT thorax was referred by the GP and clinician would have not seen the large pulmonary mass on the
performed after 6 months from initial presentation radiograph. Facilities for radiograph review are currently not
demonstrating a right middle lobe pulmonary mass. available, but should be made available in clinic for clinician
Follow-up and respiratory referral were not recommended in review and adequate time for patient assessment in clinic
any of the radiographs. should be sought. Follow-up for consolidation should be
explicitly mentioned in all chest radiographs.

CT, computed tomography; GP, general practitioner; IT, information technology; MDT, multidisciplinary team; PACS, patient archiving and communications system.

2014 John Wiley & Sons, Ltd. 457


Patient safety in radiology B.N. Bhaludin et al.

ensure the report for the radiograph is faxed to the referring clini- working environment the reporting radiologist was subject to
cian and in the case of suspected malignancy, the appropriate MDT during the analysis of a chest radiograph which may have lead to
coordinator. However, we found this form of communication has a delay in issuing a report or failure to state the need for a specialist
its flaws. This has been demonstrated in two of our cases, where follow-up.
there were failures in faxing the reports to the clinicians and the The field of radiological errors and use of the systems approach
MDT coordinator. Direct electronic communication between is currently limited. Future studies addressing the extent of delays
the radiologists and referring clinicians, on the other hand, has and factors leading to errors in radiological diagnosis in conditions
been implemented in other centres with good results. In the United will help to broaden this area of research. In order to minimize
States, Singh et al. [16] analysed the use of a computerized noti- such errors within a radiology department, efforts should be made
fication system of abnormal and critical imaging results. Their to include documentation of near misses, misinterpretations as
study revealed that the rate of lost follow-up imaging was 0.02% well as delays in diagnosis or faults in communication systems.
per outpatient visit, lower than that reported in other systems not Solutions that have proven to reduce delays and improve manage-
adopting this form of technology. Abujudeh et al. [17] studied the ment should be encouraged as well as issues relating to such
effectiveness of an important finding alert (IFA) system using changes in order that each hospital trust can predict the potential
email technology to alert referring physicians to important but hurdles in altering current services and systems.
non-urgent imaging findings. In a 3-year period 3.7% radiology
reports generated an IFA email, of which 74.6% were viewed
by the referring physicians. Although not entirely foolproof, the Conclusion
system was concluded as a potentially useful method of commu- Delays in the treatment and diagnosis of lung cancers in patients
nicating non-urgent but important imaging findings with email with abnormal chest radiographs are multifactorial in origin.
technology. Further recent research by Eisenberg et al. [18] has A systems approach to such delays reveals several causative
confirmed similar findings with a mean rate of 82.2% of important factors primarily poor methods for communication of findings as
imaging communications read by referring clinicians within a well as a lack of need for action despite acknowledgement of
48-hour time period with high levels (79%) of clinician satisfac- abnormalities.
tion levels with the service. Although implementation of such a Different solutions may be beneficial including increased
service may take some time, efforts are underway to introduce this awareness of radiologists to suggest follow-up for abnormal
form of communication at our local hospital. findings and more robust methods such as direct electronic
Our study highlighted areas whereby improvements can be communication to clinicians should be introduced to ensure radi-
made, but the full extent of radiological and systemic error may ology reports are read by the referring clinicians and acted on
not have been fully addressed in this study as we did not investi- appropriately.
gate the extent of interpretation errors by radiologists resulting in
false-negative chest X-ray reports. Only reports where abnormal
radiograph findings were identified were reviewed. The reason Acknowledgements
being that our aim was to identify delays and errors in diagnosis
and treatment after an abnormal chest radiograph had been issued, The authors would also like to thank the staff of Epsom and
and the communication errors surrounding this, rather than focus- St. Helier University Hospitals for their contribution in this study.
ing on the causes for radiological misinterpretation itself. Inclu-
sion of data on false-negative radiograph reports within this study
References
may highlight more areas for improvement among our own radio-
logical abilities and departmental settings, however, is outside the 1. Vincent, C., Neale, G. & Woloshynowych, M. (2001) Adverse events
scope of this study. in British hospitals: preliminary retrospective record review. BMJ
In addition, the extent of delays in communication of chest (Clinical Research Ed.), 322 (7285), 517519.
radiograph findings may not have been fully examined as we only 2. Berlin, L. (1996) Malpractice issues in radiology. Perceptual errors.
American Journal of Roentgenology, 167 (3), 587590.
reviewed reasons for delay in significant findings and not a failure
3. Robinson, P. J., Wilson, D., Coral, A., Murphy, A. & Verow, P. (1999)
to communicate a truly normal chest radiograph. The signifi- Variation between experienced observers in the interpretation of
cance of a normal chest radiograph may also change the man- accident and emergency radiographs. The British Journal of Radiol-
agement of the patient, although perhaps not in as dramatic a way ogy, 72 (856), 323330.
as one with a suspected malignancy. This was not conducted in our 4. Hamman, W. R. (2004) The complexity of team training: what we
study as it would be difficult to determine whether the report was have learned from aviation and its applications to medicine. Quality
acknowledged or not as both may result in no further treatment. and Safety in Health Care, 13 (Suppl. 1), i72i79.
Furthermore, reasons for communication delays were retrospec- 5. Calland, J. F., Guerlain, S., Adams, R. B., Tribble, C. G., Foley, E. &
tive and therefore it was not always clear to identify all the factors Chekan, E. G. (2002) A systems approach to surgical safety. Surgical
or the extent to which each factor played for delays and errors. Endoscopy and Other Interventional Techniques, 16 (6), 10051014.
6. Vincent, C., Moorthy, K., Sarker, S. K., Chang, A. & Darzi, A. W.
For example, where clinical notes and letters did not document
(2004) Systems approaches to surgical quality and safety: from
acknowledgement of an abnormal chest radiograph report finding, concept to measurement. Annals of Surgery, 239 (4), 475482.
it was assumed that the team had not read the radiological report. 7. Arora, S. & Sevdalis, N. (2010) Systems approach to daily clinical
This may not necessarily have been true if the report was read and care. International Journal of Surgery, 8 (2), 164166.
simply not acted upon. Alternatively, some factors are impossible 8. Pinto, A. & Brunese, L. (2010) Spectrum of diagnostic errors in
to determine retrospectively, such as the degree of disruption or radiology. World Journal of Radiology, 2 (10), 377383.

458 2014 John Wiley & Sons, Ltd.


B.N. Bhaludin et al. Patient safety in radiology

9. Vincent, C., Taylor-Adams, S. & Stanhope, N. (1998) Framework 121). Available at: http://guidance.nice.org.uk/CG121 (last accessed
for analysing risk and safety in clinical medicine. British Medical 16 April 2014).
Journal, 316 (7138), 11541157. 15. National Patient Safety Agency (2007) Safer practice notice 16. Early
10. Physician Insurers Association of America (2005) Lung Cancer Identification of Failure to Act on Radiological Imaging Reports.
Claims Study. Rockville, MD: Medical Newsmagazine. London: NPSA.
11. Turkington, P. M., Kennan, N. & Greenstone, M. A. (2002) 16. Singh, H., Arora, H. S., Vij, M. S., Rao, R., Khan, M. M. & Petersen,
Misinterpretation of the chest ray as a factor in the delayed diagnosis L. A. (2007) Communication outcomes of critical imaging results in a
of lung cancer. Postgraduate Medical Journal, 78 (917), 158 computerized notification system. Journal of the American Medical
160. Informatics Association, 14 (4), 459466.
12. The Royal College of Radiologists (UK) (2002). Framework for risk 17. Abujudeh, H. H., Kaewlai, R., Choy, G., Whelton, D. G. & Rosenthal,
management. D. I. (2009) Important imaging finding e-mail alert system: experience
13. The Royal College of Radiologists (UK) (2008). Standards for the after 3 years of implementation 1. Radiology, 252 (3), 747753.
communication of critical, urgent and unexpected significant radio- 18. Eisenberg, R. L., Yamada, K., Yam, C. S., Spirn, P. W. & Kruskal, J. B.
logical findings. (2010) Electronic messaging system for communicating important,
14. National Institute for Health and Clinical Excellence (2011). The but nonemergent, abnormal imaging results. Radiology, 257 (3), 724
diagnosis and treatment of lung cancer (update). (Clinical guideline 731.

2014 John Wiley & Sons, Ltd. 459


Copyright of Journal of Evaluation in Clinical Practice is the property of Wiley-Blackwell
and its content may not be copied or emailed to multiple sites or posted to a listserv without
the copyright holder's express written permission. However, users may print, download, or
email articles for individual use.