Discussion Board

A major review of diagnostic error in the ED is now published in draft form on the AHRQ website.  This is a very important and well-done report, with a ton of data and analysis, that will serve as a landmark in this area.  A copy is attached.

The draft is open for comments until the end of the month, and I would encourage everyone to take this opportunity to provide feedback on the report.  In particular, I call your attention to one area that bothered me, concerning the recommendations section starting on Page 81.  The authors recommend that future interventions focus solely on disease-specific approaches.  Interventions relating to education, improving cognition and clinical reasoning, and patient-focused solutions aren't mentioned at all.  I would like to see a major report like this present a more balanced overview of the many possible interventions that could have impact in the ED, in addition to the disease-specific approaches.

You can provide feedback on the draft report here:  https://effectivehealthcare.ahrq.gov/products/form/diagnostic-errors-emergency

PEDIATRICS.

The paper also has some concerning blind spots when it comes to pediatric ED patients. For example, necrotizing enterocolitis is a disease primarily of premature neonates. This is not a condition that is at major risk of misdiagnosis in the ED. On the other hand, child abuse is. There is ample evidence that delayed diagnosis of child maltreatment and missed sentinel injuries is associated with more severe presentations later on. But these cases will almost NEVER end up in a malpractice case because of the inherent deceit in the diagnostic process on the part of caregivers (most child abuse is suffered in the home from family members and close contacts). Also, children, with fewer comorbidities, are less likely to suffer permanent harm resulting in lawsuits so they will be underrepresented in malpractice claims data. AHRQ, because of using skewed data, will perpetuate the existing de-prioritization of pediatric diagnostic error research because, as stated in the draft, they don't sue in the same numbers or have the same volume of bad outcomes – this is directly the result of using a non-representative, numerator only sample to define the problem for kids.  

WHAT CAN EDs RELIABLY BE EXPECTED TO DO.

The paper, at one point, acknowledges that the primary decision in the ED is the safe disposition of the patient rather than definitive diagnosis. While I appreciate that we would all like patients to get diagnosed correctly on their first (or second) visit to the ED, emergency medicine is about identifying and treating immediate life/limb-threatening conditions or those likely to become so imminently.

Yes, we can get just about any study/imaging test/consult needed rather quickly but that is not necessarily an appropriate use of resources in the ED. And all ED clinicians realize that many patients do not have a primary care provider to follow-up with. But shifting the responsibility of making timely diagnoses to the ED just because they have the resources and the patient in front of them is not an appropriate solution. We lack the ability to effectively follow-up with the patient and we risk overdiagnosis which is an equally insidious and harder to measure problem. We need work on diagnosing emergent conditions like strokes, MIs, sepsis, etc. We are already screening for many diseases like depression, STIs, trafficking, domestic violence and child abuse). Piling on more things for us to manage will not reduce diagnostic errors and runs the risk of diverting our attention from the patients most in need of it (the patient with sepsis, stroke, MIs etc). But expecting EDs to pick up on every indolent slowly progressing life-threatening disease is unfeasible and would also likely be very costly. A better solution is to address the lack of access to care that disproportionately affects the poor and people of color.