A February 22 New York Times article entitled “Why Doctors So Often Get It Wrong” (registration required) deals with the problem of medical diagnostic error. Claiming that medicine suffers from a “misdiagnosis crisis” the article asks “How can this be happening? And how is it not a source of national outrage?”
We are told the story of a little girl who suffered consequences of delayed diagnosis. The New York Times piece states that the girl, diagnosed as chickenpox, actually turned out to have a flesh eating virus which her doctors failed to diagnose until her organs started shutting down. Hmmm----what’s a flesh eating virus anyway? Chickenpox is a virus. Did the doctors diagnose the wrong virus? A more accurate BBC report indicates that the girl actually had chickenpox complicated by necrotizing fasciitis, a bacterial infection.
With that little trifle out of the way let’s consider the article’s major premise about misdiagnosis. The article mentions autopsy data showing a 20% misdiagnosis rate, with JAMA as the source. Although the Times gives no citation, the JAMA article in question is probably this systematic review of autopsy-detected diagnostic errors. What the Times neglected to point out is that although the JAMA analysis found a 23.5% error rate, the rate of errors likely to have affected outcome was only 9%. Moreover, while the Times article claims that the error rate has not improved since the 1930’s the JAMA study found a continuous decline in the error rate between 1966 and 2002.
Although the Times article suggests that pay for performance and penalties for errors might solve the “crisis” the data suggest otherwise. Studies on Pay for Performance to date have failed to demonstrate improved quality. A spate of articles analyzing medical error indicates that promotion of a culture of blame by penalizing doctors for honest mistakes is counter productive.
The JAMA perspective is more nuanced: “However, it remains unclear to what extent clinically missed diagnoses represent errors per se, rather than acceptable limits of antemortem diagnosis in the face of atypical clinical presentations. In fact, because the vast majority of autopsy studies come from teaching hospitals, published autopsy series may be enriched for atypical cases.”