Advanced Search

  • Topic
  • Specialty
  • Content Type


Also Related

< Back To Patient Safety
0 dislikes

< Hide

Comments For

Breakdowns in ED Diagnosis


< Shrink

Add Your Voice

All comments are posted anonymously. Your comment will be attributed to: "Anonymous user."

post comment


Are you sure you want to delete this comment?

Breakdowns in ED Diagnosis

By Tom A. Augello, CRICO

Related to: Ambulatory, Communication, Diagnosis, Emergency Medicine

Complexity is the hallmark for ED errors, which feature failures to order appropriate diagnostic tests, perform adequate histories and physical exams, and interpret tests correctly.


  • Allen Kachalia, MD, JD; Brigham and Women’s Hospital; Boston, MA
  • Richard Griffey, MD, MPH; Washington University Emergency Medicine; St. Louis, MO
  • Susan West, BSN, RN; University of Colorado Hospital; Denver, CO


Emergency departments represent some of the most challenging areas for the study and prevention of diagnostic errors. In academic settings, providers at various levels of experience, from multiple specialties, are involved in delivering emergency care. Patient status is often acute, relationships are frequently brand new, and follow-up is usually handled by providers in other settings. Add to this the challenges of higher and higher volumes and difficulty finding room to board emergency patients awaiting hospital beds.

“When there is a missed or delayed diagnosis in the ED, these diagnoses tend to have a complex etiology.”

Dr. Allen Kachalia is a hospitalist at Brigham and Women’s Hospital in Boston. Dr. Kachalia’s research into ambulatory diagnostic error led to findings about missed and delayed diagnosis in the emergency department that were published in the Annals of Emergency Medicine in early 2007.

The data were gathered as part of a larger study of medical malpractice claims from four liability insurers. Sixty-five percent of the 122 closed ED malpractice claims studied had an allegation of missed or delayed diagnosis.

The top two process breakdowns associated with those claims were failure to order an appropriate diagnostic test and failure to perform an adequate medical history or physical exam. But the complexity of the diagnostic problems in this setting suggests that one root cause is very elusive.

“There tend to be multiple steps that break down.  There tend to be multiple providers involved, and the diagnoses tend to be what one can think of as more severe or acute illnesses as opposed to something more chronic such as cancer.”

Another feature more common in the ED cases than in general ambulatory settings was a failure to interpret tests correctly. Thirty-seven percent of the malpractice cases in the study included an incorrect interpretation of a diagnostic test, making it the third-highest category of process break-downs for diagnostic cases in the ED setting.

“Our sense was that behind interpreting tests is frequently you would have either emergency room physicians or nonemergency medicine physicians because in some EDs you can sometimes have other specialties working in there.  You would have folks sometimes looking at radiological scans or looking at EKGs rather than the radiologist themselves being on call 24/7 and able to look at films or a cardiologist right there to look at the EKG.”

Co-author Richard Griffey is an emergency medicine physician and Associate Chief of Quality and Safety at Washington University Emergency Medicine Department in St. Louis. Dr. Griffey says the interpretation variable is likely to depend on the availability of ancillary services.

“In this study, and I imagine by and large this is the same nearly everywhere, these tended to be radiographic studies, whether they’re X-rays or whether they are ultrasounds.  I think those really made up the majority of problems with interpretation.  Of course, electrocardiograms and laboratory studies are also part of that. And I think that, depending on whether there is radiologist availability, whether there is real time or close to real time review of interpretations of these studies is an important factor in how many of these are going to take place.”

Dr. Griffey says approaches can vary significantly from hospital-to-hospital. But a key factor in minimizing diagnostic problems related to misinterpretation of test results is to have a clear process for review that is as timely as possible.

“Having a clearly defined review piece, if that’s over read of radiographs within a given time period, and having those systems in place that are locked tight, those would be helpful in at least getting at some of the problems found in the study.  It would be better to have 24/7 access to all services, but that’s sometimes just not possible.”

At University of Colorado Hospital in Denver, the emergency department was determined to address problems with preliminary film reads that were done by residents on nights and week-ends and were read by attendings the next day with different results. Susan West is Director of Clinical Excellence and Patient Safety.

“We would find that we’ve had some significant cases where the over read is different and we can’t get in touch with the patient, and it prompted us to realize that this is not good enough process. And we literally created an Emergency Room/Radiology Committee to review these cases short of being able to change staffing and have 24/7 attending coverage. We also decided to go to a telemedicine oversight by an off-site radiology program that real time would be reading the films with the residents and we would be getting within the hour rather than our previous process where we would have that delay.  That’s probably been the biggest significant change this year that has had a huge impact in some of the radiology issues that we’ve experienced.”

West says that her institution’s malpractice data show that diagnostic error is the top category in their ED. According to West, her hospital’s data reflects very similar break-downs identified in the Kachalia study.

Dr. Kachalia points to two approaches one could take to applying interventions. One would be to address cognitive or decision-making deficits that underlie many of the breakdowns and contributing factors. Potential efforts there could include checklists and protocols, or decision support within electronic records, or computerized order entry. Automated prompts for suggested tests, for example, can help widen the differential diagnosis.

“Another way they could do this is they could try to attack the issue of communication factors and try to find ways to improve hand offs within the ED. Or, likewise, they could attack other systems factors just looking at trying to improve supervision across the board. Of course this is a lot easier said than done, and the challenge is in trying to find a way to fix all of this while understanding that there are limited resources and it all has to fit.”

Dr. Kachalia, who is also an attorney, suggests that the liability associated with ED diagnostic claims is connected to the unique nature of the environment. For judges and juries who are asked to consider timeliness and accuracy of a given diagnostic case, all errors are not created equal. Dr. Kachalia says the high acuity of patients in the emergency department will be considered.

But physicians and hospitals will always be expected to do their best to prevent mistakes.

“The expectation is all determined by how a reasonable provider or a reasonable system would behave. I think that all depends on how the patient comes in, in terms of what level of care is provided. If someone comes in to an ED and all they’ve done is maybe they’ve fallen and they’ve fractured their arm, I think the expectation would be the same if they walked into the ED as opposed to walking into clinic as to determining whether to make the diagnosis of a fracture. The law is always expecting you to behave reasonably, so if you have somebody who is really sick in front of you and you’re trying to save them, I’m not necessarily sure I would say the law expects less…the law might give you more room in that aspect, but it won’t necessarily say well, you can be less reasonable. You’re still expected to be as reasonable as possible.”

May 1, 2007
0 dislikes

< Back To Patient Safety