Lit Matters 3: Diagnostic error in the ED

Diagnostic error in the emergency department clusters around a small set of high-harm conditions, with neurologic disease leading the list. Stroke, myocardial infarction, aortic dissection, sepsis, and other time-sensitive diagnoses dominate both missed-harm patterns and malpractice narratives, but the evidence behind headline national estimates deserves careful scrutiny.

High-Risk Diagnostic Error Patterns

• Top missed-harm diagnoses: Stroke, myocardial infarction, aortic aneurysm or dissection, spinal cord pathology, venous thromboembolism, CNS infection, and sepsis account for much of the serious misdiagnosis burden seen in ED data.
• Neurologic disease concentration: Neurologic conditions made up 34% of the highest-harm group, underscoring how often subtle focal deficits, transient symptoms, and atypical syndromes drive emergency diagnostic failure.
• Stroke versus MI gap: Myocardial infarction had one of the lowest reported disease-specific error rates at 1.5%, while stroke was missed about tenfold more often at 17%, likely reflecting subjective syndrome recognition rather than ECG criteria.
• Extreme disease variability: Reported error rates varied widely across conditions, from low rates in MI to 56% in spinal abscess, a reminder that uncommon, protean presentations create disproportionate risk.
• Atypical presentation traps: Misses were more likely when disease appeared in the wrong age group or with mild, transient, or nonspecific symptoms. We get into the bedside pattern-recognition implications in the episode.

What the Review Claimed

• Per-visit error estimate: The review estimated that 5.7% of ED visits involve at least one diagnostic error, a headline figure with obvious policy implications and equally important methodological caveats.
• Preventable harm burden: Potentially preventable diagnostic adverse events were estimated at 2.0% of ED visits, with serious permanent disability or death reported at 0.3%.
• Mortality headline numbers: The paper extrapolated national totals suggesting hundreds of thousands of deaths and severe harms tied to ED misdiagnosis, numbers large enough to demand a close look at how they were derived.
• Malpractice signal: Malpractice claims linked 89% of diagnostic error cases to failures in clinical decision-making or judgment, though claim data carry major selection bias and cannot stand in for all real-world error.

Why the Estimates Are Contested

• Working diagnosis mismatch: Emergency care often aims for a diagnosis-informed disposition and treatment decision, not a final label, so later disagreement with the inpatient diagnosis is not automatically an ED error.
• Heterogeneous source studies: The review pooled non-US studies, mixed physician training backgrounds, and settings with different standards of emergency care, limiting how confidently the results translate to US emergency departments.
• Thin data foundations: Some widely quoted overall error and death estimates trace back to just a few studies, including older single-center data and work not originally designed to measure diagnostic error.
• Malpractice data limitations: Claims data reflect only filed cases and overrepresent allegations of cognitive error, which can distort the balance between individual judgment failures and the system contributors seen in broader patient-safety research.
• Methodology worth hearing: The review's intent was quality improvement, but several analytic choices may overstate certainty and scale. We walk through the most important flaws in the chapter.