Doctors Incorrectly Diagnose Symptoms 10-15% of the Time

Although we all like to think doctors are infallible... we are far from it. Studies on doctors' diagnostic skills have consistently shown error rates around 10-15% based on different approaches. Why aren't we better?

It's because human biology is complex and doctors are human.

When a patient comes in with a symptom... let's take dizziness... there are numerous possible causes for this particular symptom: stroke, medication side effect, alcohol intoxication, BPPV, Meniere's, vestibular neuritis, cerebral vascular insufficiency, head trauma, concussion, etc.

Beginning with a medical history, we try to narrow down the possible causes down to a few and ideally one before making a recommendation for treatment. Some are easily eliminated, but others are more challenging to rule-out.

That's when testing is performed to further narrow down the last few possibilities. Using the example of dizziness, numerous tests are performed including hearing tests (asymmetric hearing loss?), MRI scan (brain tumor vs stroke), and ENG (peripheral vestibular dysfunction?). Doctors will also have to take into consideration that NO test is 100% accurate and the testing is only as good as the interpretation of the test results which brings another human element where error may crop up.

In any case, where does this 10-15% physician error rate come from? It is mainly from the research of Arthur Elstein, a cognitive psychologist who spent his entire lifetime studying how doctors think. However, subsequent studies by others have confirmed this error rate.

Autopsy studies which are considered the "gold standard of tests" reveal diagnostic errors of 10-20%. That's part of the reason why autopsies are so important when an "unexpected" death occurs as it reveals cause of death and serves to "teach" where physicians may have gone wrong.

Survey studies have been conducted where doctors as well as patients are questioned whether they personally or know of somebody who was incorrectly diagnosed. Such studies reveal an error rate around 30%.

"Fake Patient" secret shopper studies where actors pretend to have a particular illness have revealed error rates of about 13%

All this serves to remind healthcare utilizers that physicians are not infallible. Utilizing 2nd opinions help to minimize error rates as well as seeing specialists, though such steps may increase other types of errors if not confusion if different opinions are rendered.

References:
Clinical reasoning in medicine. In: Higgs J, ed. Clinical reasoning in the health professions. Oxford, England: Butterworth-Heinemann Ltd, 1995;49–59

Diagnostic errors in three medical eras: a necropsy study. Lancet 2000;355:2027–31

Views of practicing physicians and the public on medical errors. N Engl J Med 2002;347:1933–40

Assessing the accuracy of administrative data in health information systems. Med Care 2004;42:1066–72.

The incidence of diagnostic error in medicine. BMJ Qual Saf 2013;22:ii21-ii27 doi:10.1136/bmjqs-2012-001615