<h3>Introduction</h3> Data gathered from the biopsy database located at the University of Utah was used to analyze the accuracy of submitting clinicians' respective differential diagnoses. The main objective was to identify deficiencies in provider education to help guide presenters' selection of topics for future lectures to the dental professional community. <h3>Materials and Methods</h3> The database was analyzed by cross-sectional comparison of 7,463 submitted samples. Each submitted "Clinical History" section (which prompts clinicians to submit a differential) was evaluated in com- parison to the final histopathological result. Certain categories of submissions were excluded, such as lower lip minor salivary gland biopsies to evaluate for Sjogren's Syndrome. Results were compiled with keyword searches. Result:39% of submitted sample results were included within the differential diagnosis and 14% were similar in type to the final diagnosis. 7% were inaccurately predicted, 10% provided no information at all and 30% provided only descriptive information with no entities named. The most common misidentified lesion was the fibroma. Qualitative analysis revealed that when inaccuracies occurred it was often due to failure to identify key predicting information, such as identifying a fibroma after lower lip trauma had created a fluctuant mass. <h3>Conclusions</h3> With over half of submitted differentials having a high level of accuracy, clinicians are demonstrating success at providing differential diagnoses. The low percentage of inaccuracies reflects unexpected final diagnoses, severity biases, or failure to identify key predictors. The 30% of clinicians who only provided descriptive infor- mation, and the 10% who provided no additional information, represent a significant portion of clinicians where additional training could be targeted. The formation of a differential diagnosis is helpful in correlation with clin- ical appearance and for informing patients of possible biopsy diagnoses. Future lectures could focus more on the formation of reasonable differential diagnoses.
Read full abstract