Pennsylvania Patient Safety Advisory Communication of Radiograph Discrepancies between Radiology and Emergency Departments ABSTRACT of technologic services (e.g., picture archiving and A radiograph ordered in the emergency department communication systems [PACS], electronic medical (ED) may not be reviewed immediately by a radi- records [EMRs]). Accordingly, discrepancies may be ologist for a number of reasons, including limited handled by means such as electronic- and paper- availability of radiology services after hours and tracking systems. Regardless of the method, consistent the increasing demand on radiology services due and reliable communication between the ED and the to growing ED volume. In 2008, facilities submit- radiology department is essential to ensure timely and ted 195 reports to the Pennsylvania Patient Safety adequate follow-up of any discrepancy. Authority identifying a discrepancy between an ED physician’s preliminary radiograph finding and the Clinical Literature results of a radiologist’s final reading. Processes for A review of the literature found that discordance communicating radiograph readings from the radiol- between ED physician and radiologist interpreta- ogy department to the ED vary among facilities due tions of radiographs has been reported in a number to factors including the availability of radiology ser- of studies as ranging from 0.3% to 17%.3-6 The vices during off hours and availability of technologic majority of studies focus on rates of discrepancies; services such as picture archiving and communica- however, few studies evaluate the clinical impact of tion systems. When discrepant interpretations occur discrepancies on patient care. Not all discrepancies between the preliminary reading by an ED physician have the same degree of clinical significance. A 2003 and the final reading by a radiologist, communicat- study comparing ED physician and senior radiologist ing the radiologist’s findings to the ED and patient for interpretations of 509 chest radiographs investigated follow-up is essential to ensure that the patient has the effects of misinterpretation of chest radiographs received appropriate care. This article examines risk on discharge recommendations.7 The study showed reduction strategies, including standardization of sys- that, when classified by level of clinical significance tems for communicating and reconciling radiograph (i.e., mild, moderate, high), the highest sensitivity of discrepancies between the radiology department and the ED physicians’ interpretation (60%) was found ED that will promote optimal patient care. (Pa Patient in the group with highly significant clinical findings Saf Advis 2010 Mar;7[1]:18-22.) (e.g., consolidation, congestion, pleural effusion, mediastinal widening). While this study found that the missed findings were of a minor nature, another Introduction study found that follow up of ED radiographs detects clinically important abnormalities that may have been The number of emergency department (ED) visits overlooked. During a six-month study period, 19,468 in the United States increased substantially between ED visits generated 11,749 radiographic examina- 1995 and 2005, from 96.5 million to 115.3 mil- tions. Discrepancies were detected in 175 patients lion. Along with an increase in ED volume is an (1.5%). Of these 175 patients, 136 (78%) were subse- increase in radiologic examinations performed on quently shown to have been incorrectly interpreted in ED patients.1 According to the National Center for the ED (i.e., false negatives), with 40 patients (23%) Health Statistics, in 2005, radiographic imaging was undergoing a change in management as a result. In ordered in 43.7% of ED visits, representing at least the remaining 39 discrepancies, the ED interpretation 50.3 million radiographs performed that year.1 A was evaluated to be correct, with 16 patients requiring radiograph ordered in the ED may not be reviewed additional investigations or visits to the ED to con- immediately by a radiologist for reasons including firm the radiographic finding.8 limited availability of radiology services after hours and the high demand on radiology services due to While the literature is inconclusive about the impact growing ED volume. Typically, unless an immedi- of discrepancies on patient management, reports sub- ate consult is required, the radiologist reviews the mitted to the Pennsylvania Patient Safety Authority radiograph and generates a final report within 24 show that discrepancies occur often and may have an hours.2 A process must be in place so that, if there is impact on patient safety if not communicated by the a discrepancy between the ED physician’s preliminary radiology department to the ED. interpretation and the radiologist’s subsequent inter- pretation, it is communicated to ED providers so that Authority Reports the patient will receive appropriate follow-up care. The Authority received 3,173 reports from June 2004 Processes for communicating the radiograph readings to December 2008 related to discrepancies between from the radiology department to the ED vary among the ED physician interpretation of a radiograph and facilities because of factors such as the availability of the final reading by a radiologist. The Authority radiology services during off hours and availability received 2,699 of these reports over a two-year period Page 18 REPRINTED ARTICLE - ©2010 Pennsylvania Patient Safety Authority Vol. 7, No. 1—March 2010 Pennsylvania Patient Safety Advisory from the same facility, possibly reflecting a targeted days later found a radiology report addendum from quality-improvement project. In 2008, facilities submit- the previous day showing a dislocation that was not ted 195 reports of this event type, which Authority reported to ED. analysts reviewed individually. None of these reports were submitted as a Serious Event. However, 68 (35%) Practice Guidelines of the reports indicated that the discrepancy involved The Joint Commission National Patient Safety Goal a potentially significant clinical finding, as follows: for improving communication among caregivers ■ Fracture: 50 addresses critical test results by requiring that facilities have a process in place for verbal and telephone com- ■ Pneumonia: 14 munication of such results. The 2009 communication ■ Appendicitis: 4 goal requires that facilities “measure, assess, and, if The information in the reports does not convey how appropriate, take action to improve the timeliness or when the discrepancies in the radiology findings of reporting, and the timeliness of receipt by the were communicated between the radiology department responsible licensed caregiver, of critical test results and ED, although 55 (28%) of the reports indicate and values.” Accredited facilities are expected to apply that communication was a factor contributing to the this goal not only to laboratory tests, but also to all event. Also, the reports do not indicate that an error diagnostic tests (e.g., imaging studies, arterial blood occurred in every case in which a discrepancy occurred. gas assessments, electrocardiograms).9 The Joint Com- The reports do reflect the potential for patient harm if mission requires organizations to define an acceptable a discrepancy is not communicated between the radiol- length of time between when critical tests are first ogy department and ED in a timely manner. ordered and when critical results are reported.9 Examples of Authority reports related to such com- The American College of Radiology (ACR) guideline munication issues follow. on communication maintains that the radiologist is to provide imaging services to patients seen in the ED, Communication of Radiology Discrepancies including interpretation and appropriate communica- to the ED tion.10 The guideline emphasizes that interpretation A patient presented to the ED with the complaint of should be timely to facilitate decisions regarding a seizure. The patient had a seizure and fell. A pre- treatment, although it does not specify a time frame liminary reading of a CT [computed tomography] scan in which radiology results should be communicated. was reported as negative, and the patient was dis- The ACR guideline also addresses discrepancies in charged from the ED. A review of the radiographs the interpretation between a preliminary and final radiol- next day showed the patient had compression of the ogy interpretation. Changes between preliminary and spine. The results were not conveyed to the ED physi- final interpretation should be reported in a manner cians. The patient returned to the ED several days that reasonably ensures timely receipt by the referring later and was admitted for neurosurgical intervention. or treating physician when such changes could impact [A radiology staff member left a voice-mail message] patient care. regarding x-ray discrepancy for ED support staff. The The American College of Emergency Physicians voice mail was listened to later the next day. The (ACEP) endorses that the interpretation of diagnostic support staff discussed [the discrepancy] with the phy- studies ordered for the immediate evaluation of and sician. The physician stated the patient must return management of ED patients should be done contem- to the ED. Voice mails should not be left on ED sup- poraneously with the ED visit.11 If the ED physician port staff phone. If [there is] no answer, [the caller] believes that urgent consultation is needed for the MUST [sic] notify charge nurse. interpretation of a diagnostic study, the radiologist Communication of Radiology Reports to the ED must be immediately available. The interpretation An x-ray was done and the report was signed 45 min- of the diagnostic study, both preliminary and final, utes later. The ED physician/department was not must be documented in writing and entered into the notified of a result of subdural hematoma. patient’s medical record. A patient was admitted from the ED. The physician Risk Reduction Strategies reported several hours later that the patient had a Although Joint Commission, ACR, and ACEP do dissecting aneurysm. A review of the chart showed not provide specific guidelines related to how a patient had a CT done in the ED. The overnight facility should communicate discrepancies, a num- radiology service report of CT showed dissection of ber of methods for following up on ED/radiology aneurysm. [The aneurysm was] not documented in discrepancies have been suggested. The process has ED notes and was not treated. been referred to as an “information chain,” start- A patient was diagnosed with a sprain. The patient ing with image generation, proceeding with image was discharged and instructed to follow up with interpretation, and ending with communication of orthopedics. The patient returned later that day the interpretation.12 The goal of the entire process is with pain. Radiology report was reviewed and was to follow up on any clinically significant discrepan- negative. A quality review of radiology report two cies with the patient. The way the process operates Vol. 7, No. 1—March 2010 REPRINTED ARTICLE - ©2010 Pennsylvania Patient Safety Authority Page 19 Pennsylvania Patient Safety Advisory will depend on the availability of technology such ■ Develop a consistent method to reconcile the as PACS, voice-recognition dictation systems, and radiographic interpretation with the actual care EMRs. However, with any system, it is important to provided.13,14 A consistent system for identifying the do the following: clinical significance of the finding is essential. The ED physician may find that (1) the discrepancy ■ Develop a system for interpreting radiographs and has no clinical importance, (2) the patient has communicating the interpretations that can be already been admitted and the subsequent treating implemented regardless of the time of day or day of physician needs to be notified of the finding, (3) the week.13,14 A hospital may have separate processes the patient has been transferred to another facility for each shift for handling radiograph interpreta- and the subsequent treating physician needs to be tion, depending on availability of radiology services. notified of the finding, (4) the patient has received In the case of plain radiographs, a common sce- appropriate treatment in the ED and requires nario is interpretation of ED radiographs by the no follow-up, (5) the finding was missed and the radiologist during normal business hours and inter- patient requires a follow-up contact, or (6) follow- pretation by the ED physician during after-hours up studies are required for equivocal findings.15 shifts, with a radiologist overreading the radiograph ■ Develop a consistent method for timely commu- the next day. Although rates of discrepancies nication of radiographic readings to the referring between ED physician and radiologist interpre- or subsequent treating physician and the patient tations vary in the literature, standardizing the as appropriate. One approach described in the method of identifying discrepancies and the action literature for ensuring that radiologic findings are plan for responding to them—for all shifts—will avoid communicated in a timely manner is direct com- confusion related to the use of multiple systems.13 munication of the findings by the radiologist or ■ Implement a standardized method for informing radiology facility to the patient.20 the radiologist of the ED physician’s interpre- tation.13,14 If the hospital uses a paper-based Conclusion system, the ED physician can document his or As Authority reports indicate, discrepancies may her interpretation for requisition by the radiolo- occur between the ED physician’s interpretation of gist.15 Another paper-based approach involves the a radiograph and the final interpretation of the radi- radiology department maintaining a log in the ologist. A discrepancy may be clinically significant, ED to document all radiographs. Radiology staff and a system must be in place to communicate the are responsible for logging the patient’s name and discrepancy to the ED. Every ED needs a system to views taken. The ED physician can make a nota- ensure that once a discrepancy is communicated to tion of his/her reading in the log. The log can then the ED, the discrepancy is correlated with the patient be taken to the radiologist for review.15 If PACS record to determine whether follow-up is necessary. technology is available, methods for integrating Although systems may vary depending on factors such notations into the system from the ED physician as availability of an electronic record, the system of and the radiologist have been described in the communicating discrepancies should be simple and literature.16-18 For example, one facility successfully broadly applicable across all hours and days of the implemented a PACS that includes a preliminary week. Finally, open communication among ED and note window. The window contains two text radiology providers will help promote patient safety boxes—one for the ED physician’s preliminary by ensuring that the patient will receive timely and interpretation and the other for the radiologist’s appropriate follow-up care should a discrepancy occur. interpretation.16 Notes ■ Implement a standardized system for communica- 1. National Center for Health Statistics. National Hospi- tion of the radiologist’s interpretation of the ED tal Ambulatory Medical Care Survey: 2005 emergency radiograph to the ED in a timely manner.13,14 If a department summary. Advance data from vital and discrepancy occurs between the ED physician’s and health statistics; no. 386. Hyattsville (MD): National radiologist’s interpretations, it is important that Center for Health Statistics; 2007. the ED receives this information. A 2008 survey of 2. Stolberg HO. Re: “Improving the quality of radiology current ED imaging practices showed that the most reporting: a physician survey to define the target.” J Am commonly used method of communicating urgent Coll Radiol 2004 Sep;1(9):700-1. findings or a discrepancy is verbal communica- tion between practitioners.19 Documentation of 3. Brunswick JE, Ilkhanipour K, Seaburg DC, et al. Radio- any verbal communication in the patient’s record graphic interpretation in the emergency department. Am is essential. Voice-recognition dictation systems J Emerg Med 1996 Jul;14(4):346-8. can expedite the availability of a radiologist’s final 4. Mayhue FE, Rust DD, Aldag JC, et al. Accuracy of report, but they do not eliminate the need for a interpretation of emergency department radiographs: consistent method to transmit the report to the effect of confidence levels. Ann Emerg Med 1989 ED in a timely manner.16-18 Aug;18(8):826-30. Page 20 REPRINTED ARTICLE - ©2010 Pennsylvania Patient Safety Authority Vol. 7, No. 1—March 2010 Pennsylvania Patient Safety Advisory 5. Espinosa JA, Nolan TW. Reducing error made by 12. Berlin L. Standards for radiology interpretation and emergency physicians in interpreting radiographs: longi- reporting in the emergency setting. Pediatr Radiol 2008 tudinal study. BMJ 2000 Mar 18;320(7237):737-40. Nov;38 Suppl 4:S639-44. 6. Campbell SG, Murray DD, Hawass A. Agreement 13. Janiak BD, Pawl RJ. Imaging interpretation discrepan- cies: “Danger, Will Robinson!” ED Legal Letter 2009 between emergency physician diagnosis and radiolo- Feb;20(92):13-6. gist reports in patients discharged from an emergency department with community-acquired pneumonia. 14. Glauser J. The legal aspects of emergency radiol- Emerg Radiol 2005 Jun;11(4):242-6. ogy: a system must be in place to notify patients if a discrepancy exists between the EP’s original read 7. Gatt M, Spectre G, Paltiel O, et al. Chest radiographs in and the radiologist’s overread. Emerg Med News 2002 the emergency department: is the radiologist really nec- Oct;24(10):16. essary? Postgrad Med J 2003 Apr;79(930):214-7. 15. Lawrence JD. Procedures for handling diagnostic 8. Benger JR, Lyburn ID. What is the effect of reporting all error will keep you out of court. ED Legal Letter 2002 emergency department radiographs? Emerg Med J 2003 Oct;13(10):109-20. Jan;20(1):40-3. 16. Mates J, Branstetter BF, Morgan MB, et al. ‘Wet Reads’ in the age of PACS: technical and workflow consider- 9. Joint Commission. Accreditation programs: hospital. ations for a preliminary report system. J Digit Imaging National Patient Safety Goals [online]. 2009 [cited 2007 Sep;20(3):296-306. 2009 Sep 6]. Available from Internet: http://www. jointcommission.org/NR/rdonlyres/31666E86-E7F4- 17. DeFlorio R, Coughlin B, Coughlin R, et al. Process modification and emergency department radiology ser- 423E-9BE8-F05BD1CB0AA8/0/HAP_NPSG.pdf. vice. Emerg Radiol 2008 Nov;15(6):405-12. 10. American College of Radiology. ACR practice guide- 18. Gorman A, Donnell L, Hepp H, et al. Improving com- line for communication of diagnostic imaging findings munication and documentation concerning preliminary [online]. 2005 [cited 2009 Sep 30]. Available from and final radiology reports. J Healthc Qual 2007 Mar- ? Internet: http://www.acr.org/SecondaryMainMenu Apr;29(2):13-21. Categories/quality_safety/guidelines/dx/comm_diag_ 19. Thomas J, Rideau AM, Paulson EK, et al. Emergency rad.aspx. department imaging: current practice. J Am Coll Radiol 11. American College of Emergency Physicians. Interpreta- 2008 Jul;5(7):811-6e2. tion of imaging diagnostic studies [online]. 2006 [cited 20. Berlin L. Communicating results of all outpatient radio- 2009 Sep 30]. Available from Internet: http://www.acep. logic findings directly to the patients: the time has come. org/practres.aspx?id=32874. Am J Roentgenol 2007 Dec;189(6):1275-82. ? Self-Assessment Questions The following questions about this article may be useful for internal education and assessment. You may use the following 2. Which of the following statements about the potential impact of radiographic discrepancies between the radiol- examples or come up with your own. ogy department and the ED on patient management is 1. Risk reduction strategies to ensure timely and adequate INACCURATE? communication and reconciliation of radiograph dis- a. Discordance between ED physician and radiologist crepancies between the radiology department and the interpretations of radiographs has been reported in a emergency department (ED) include all of the following number of studies as up to 20%; however, not all dis- EXCEPT: crepancies have the same degree of clinical significance. a. Eliminate the need for a consistent method to transmit b. The majority of studies regarding discordance between a report from the radiology department and the ED ED physician and radiologist interpretations of radio- by the implementation of a voice-recognition dictation graphs focus on rates of discrepancies; however, few system. studies evaluate the clinical impact of discrepancies on b. Develop a consistent method to reconcile the radio- patient care. graphic interpretation with the actual care provided. c. A 2003 study comparing ED physician and senior c. Implement a standardized method for informing the radiologist interpretations chest radiographs showed radiologist of the ED physician’s interpretation. that, when classified by level of clinical significance, the highest sensitivity of the ED physicians’ interpreta- d. Standardize the method of identifying discrepancies tion was found in the group of chest radiographs with and the action plan for responding to them—for all highly significant clinical findings. shifts—to avoid confusion related to the use of multiple systems during different shifts. d. Studies about the effects of misinterpretation of chest radiographs in the ED on discharge recommendations have consistently shown that discrepancies are of a minor nature and subsequently have little to no effect on patient management. Vol. 7, No. 1—March 2010 REPRINTED ARTICLE - ©2010 Pennsylvania Patient Safety Authority Page 21 Pennsylvania Patient Safety Advisory 3. All of the following are potential barriers to the com- 5. A 56-year-old patient presented at 11 p.m. to the ED with the munication of radiograph readings from the radiology complaint of a seizure. The patient had a seizure in the ED department to the ED EXCEPT: and fell off the stretcher, striking his head on the floor. A a. Limited availability of radiology services after hours preliminary reading of a computed tomography (CT) scan of b. Different communication processes according to the the patient’s head was reported as negative, and the patient shift or day of the week was discharged from the ED. An overread of the CT scan the c. Reliance on paper-based communication systems next day showed the patient had a subdural hematoma. The d. Lack of documentation of any verbal communication results were not conveyed to the ED physicians. The patient between the ED and the radiology department in the returned to the ED several days later with a severe headache patient record and was admitted for neurosurgical intervention. 4. Which of the following statements is INACCURATE Predict which of the following risk reduction strategies would according to accrediting bodies and organizational guidance NOT help prevent the recurrence of this type of event. about communication of radiographic results between the a. Communicate findings by means of a telephone conver- ED and radiology department? sation between the radiologist reviewing the preliminary a. The Joint Commission requires organizations to define reading the next morning and the ED physician who an acceptable length of time between when critical tests, was on duty. which include all diagnostic studies, are first ordered b. Report changes between preliminary and final inter- and when critical results are reported. pretation in a manner that reasonably ensures timely b. The American College of Radiology guideline on com- receipt by the referring or treating physician when such munication maintains that the radiologist is to provide changes could impact patient care. imaging services to patients seen in the ED, including interpretation and appropriate communication within a c. Expedite the availability of the radiologist’s final report time frame defined by the facility. by using a voice-recognition dictation system (recalling c. The American College of Emergency Physicians (ACEP) that there needs to be a consistent method to transmit endorses that the interpretation of diagnostic studies the report to the ED). ordered for the immediate evaluation and management d. Ensure that the hospital has consistent processes for of ED patients should be done contemporaneously with each shift for handling radiograph interpretation. the ED visit. d. ACEP endorses that if the ED physician believes that urgent consultation is needed for the interpretation of a diagnostic study, the radiologist must be immediately available. Page 22 REPRINTED ARTICLE - ©2010 Pennsylvania Patient Safety Authority Vol. 7, No. 1—March 2010 PENNSYLVANIA PATIENT SAFETY ADVISORY This article is reprinted from the Pennsylvania Patient Safety Advisory, Vol. 7, No. 1—March 2010. The Advisory is a publication of the Pennsylvania Patient Safety Authority, produced by ECRI Institute and ISMP under contract to the Authority. Copyright 2010 by the Pennsylvania Patient Safety Authority. This publication may be reprinted and distributed without restriction, provided it is printed or distributed in its entirety and without alteration. Individual articles may be reprinted in their entirety and without alteration provided the source is clearly attributed. This publication is disseminated via e-mail. To subscribe, go to https://www.papsrs.state.pa.us/ Workflow/MailingListAddition.aspx. To see other articles or issues of the Advisory, visit our Web site at http://www.patientsafetyauthority.org. Click on “Patient Safety Advisories” in the left-hand menu bar. THE PENNSYLVANIA PATIENT SAFETY AUTHORITY AND ITS CONTRACTORS The Pennsylvania Patient Safety Authority is an independent state agency created by Act 13 of 2002, the Medical Care Availability and Reduction of Error (“Mcare”) Act. Consistent with Act 13, ECRI Institute, as contractor for the Authority, is issuing this publication to advise medical facilities of immediate changes that can be instituted to reduce Serious Events and Incidents. For more information about the Pennsylvania Patient Safety Authority, see the Authority’s Web An Independent Agency of the Commonwealth of Pennsylvania site at http://www.patientsafetyauthority.org. ECRI Institute, a nonprofit organization, dedicates itself to bringing the discipline of applied scientific research in healthcare to uncover the best approaches to improving patient care. As pioneers in this science for nearly 40 years, ECRI Institute marries experience and independence with the objectivity of evidence-based research. More than 5,000 healthcare organizations worldwide rely on ECRI Institute’s expertise in patient safety improvement, risk and quality management, and healthcare processes, devices, procedures and drug technology. The Institute for Safe Medication Practices (ISMP) is an independent, nonprofit organization dedicated solely to medication error prevention and safe medication use. ISMP provides recommendations for the safe use of medications to the healthcare community including healthcare professionals, government agencies, accrediting organizations, and consumers. ISMP’s efforts are built on a nonpunitive approach and systems-based solutions.