Programmatic Theme: Clinical Research Informatics

Abstract: Diagnostic errors can pose a serious threat to patient safety, leading to serious harm and even death. Efforts are being made to develop interventions that allow physicians to reassess for errors and improve diagnostic accuracy. Our study presents an exploration of misdiagnosis patterns mined from PubMed abstracts. Article titles containing certain phrases indicating misdiagnosis were selected and frequencies of these misdiagnoses calculated. We present the resulting patterns in the form of a directed graph with frequency-weighted misdiagnosis edges connecting diagnosis vertices. We find that the most commonly misdiagnosed diseases were often misdiagnosed as many different diseases, with each misdiagnosis having a relatively low frequency, rather than as a single disease with greater probability. Additionally, while a misdiagnosis relationship may generally exist, the relationship was often found to be one-sided.

Learning Objective: After reading this paper, the reader should be better able to
- Understand the serious threat posed by diagnostic errors and the importance of preventing such errors
- Learn about the benefits of physicians using diagnostic support systems
- Learn about possible ways of mining misdiagnosis patterns and how these patterns may fit into a diagnostic support system
- Learn about some general patterns of misdiagnosis among common diseases

Authors:

Cindy Li (Presenter)
Brown University

Elizabeth Chen, Brown University
Guergana Savova, Boston Children's Hospital and Harvard Medical School
Hamish Fraser, Brown University
Carsten Eickhoff, Brown University