• Temple EM

Errors in Diagnosis of Spinal Epidural Abscess in the Era of Electronic Health Records

The Article:

Errors in Diagnosis of Spinal Epidural Abscess in the Era of Electronic Health Records. Bhise V et al. Am J Med 2017 Aug; 130:957

The Idea: 

  1. Spinal Epidural Abscess is a difficult diagnosis to make; delayed diagnosis is associated with significant M&M

  2. There has not been a large EMR study conducted yet demonstrating reasons for diagnostic delay

  3. Can we demonstrate diagnostic delay in a nationwide EMR database and identify the factors behind diagnostic delay in order to develop strategies to mitigate patient harm

The Study: 

  1. 250 pt with diagnosis of SEA randomly selected from nationwide Veteran’s Affairs EMR

  2. Charts retrospectively reviewed for diagnostic error in diagnosis of SEA

  3. Diagnostic Error defined as: ““Missed opportunity to make the correct or timely diagnosis of SEA based on available evidence regardless of whether the patient experienced harm.”

  4. Translation: If there were clinical red flags (ACR Appropriateness Criteria for Back Pain), within 90 days prior to SEA diagnosis, were these factors documented and acted upon appropriately.

  5. Red Flags included: h/o cancer, Unexplained weight loss, immunosuppresion, UTI, IVDU,  prolonged use of corticosteroids, back pain not improved with conservative management

  6. Researchers also sought to classify the type of diagnostic error, and assess the degree of harm to patient’s who experienced diagnostic delay.

The Results: 

 131 patients excluded as diagnosis of SEA was not made at the VA. 66/119 (55.5%) of patients determined to have had experienced diagnostic error. There was no significant demographic or comorbid difference between the groups with and without diagnostic error (Table 1). If dx error was made: mean time to diagnosis was 12 days after recognition of red flag, compared to 4 days in the group without dx error. The most commonly missed red flags were Unexplained Fever (86%), Neuro Deficits (81%), Active Infection (81%) (table 2).  The most common type of diagnostic error was  provider – patient encounter. In 57.6% of cases of dx error, a red flag was documented however the appropriate test or referral was not provided. In 33% of cases, provider missed a red flag on H&P even though it might have been documented in previous or different office visit – eg history of IVDU documented, but not noted by provider.  In cases of dx error – MRI’s eventually ordered, but not marked as urgent, resulting in a median delay of 6 Days, which were not attributable to patient no show.  ESR/CRP only ordered in 27% of cases. 12% of dx error cases resulted in patient death. 48% resulted in severe harm (“interferes significantly with quality of life”).

Author Conclusions/Takeaways Points:

  1. There are lots of potential pitfalls in the dx of SEA

  2. It is difficult to make a complex diagnosis in complex patients with increasingly shorter patient interactions

  3. Know your Red Flags

  4. Where to go from here: Can we develop algorithms/clinical decision aids to risk stratify patients for SEA or utilize EMR pop-ups to help providers recognize patient’s with red flags who have not had the appropriate diagnostic work up?

#decisionrules #infectiousdisease

Recent Posts

See All

10th Floor Jones Hall

1316 W. Ontario Street

Philadelphia, PA 19140

T: 215.707.5435

F: 215.707.3494

E: TUHEMResidency@gmail.com