Online Program


Detours on the Road to a Correct Diagnosis: The Role of Clinical Certainty
Carol L. Link, New England Research Institutes 
Lisa Marceau, New England Research Institutes 
John B. McKinlay, New England Research Institutes 
*Rebecca Shackelton, New England Research Institutes 

Keywords: Decision Making, Differential Diagnoses, Diagnostic Uncertainty

Clinical uncertainty may cause some physicians to take diagnostic detours while travelling the road to a correct diagnostic destination. These detours, at key junctures along the diagnostic pathway, can make it difficult to return to the correct diagnosis. In this study we examine the contribution of physician certainty to diagnostic detours, despite the identical presentation of the signs and symptoms of diabetes. As part of a factorial experiment we presented primary care physicians in three countries, the United States (US), n=192, the United Kingdom (UK) n=128, and Germany, n=64) with clinically authentic filmed vignettes of different “patients” presenting with identical signs and symptoms strongly suggesting diabetes. Physicians in the UK were the most likely to list diabetes in their differential diagnosis (89.1%)—physicians in the US and Germany correctly identified diabetes only 60.9% of the time (p < 0.0001). Physicians in the US were more likely to list a mental health diagnosis as the most likely condition (79.2%) than either UK (66.4%) or German (53.1%) physicians (p = 0.0002) even though the signs and symptoms of a mental health condition were not imbedded in the vignette. Physicians in the US were significantly more certain of their most likely diagnosis (64.0 out of a 0-100 scale) than physicians in the UK (57.5) or Germany (56.2), which may in part lead to their reluctance to consider other probable diagnoses (including the correct diagnosis of diabetes). For each 10% increase in certainty of the most likely condition, the probability that the physician will eventually consider diabetes decreases by the same amount. In a filmed vignette factorial experiment we identify junctures along the diagnostic pathway at which physicians may be deflected from a correct diagnosis of diabetes. It appears that if physicians entertain alternative diagnoses with greater certainty (particularly a mental health diagnosis) it is less likely that they will ultimately return to the correct diagnosis of diabetes. We focus on the alternative diagnosis of a mental health condition since mental health conditions are particularly difficult to rule out. There is no definitive test to rule out a mental health condition which makes it particularly difficult to return to the diagnostic pathway leading to a correct diagnosis.