Abdominal pain is among the most challenging complaints to evaluate. Diagnosis is complex, uncertain, and error-prone. A diagnostic path, as defined by the authors, consists of the sequence of diagnostic steps taken from initial presentation until a diagnosis is obtained or the evaluation ends for other reasons. Diagnostic processes followed by different physicians for the same presenting complaint are highly variable. Our objective was to map and mine these processes as portrayed in the EHR, seek to establish patterns, and to interview a number of primary care physicians to gain insight into their diagnostic practice in relation to undifferentiated abdominal pain and examine the extent to which the perceived process matches the record that emerges from the EHR.

Describe the new knowledge and additional skills the participant will gain after attending your presentation.: 1. Learn and evaluate a new concept: "diagnostic path" and differentiate it from pathways, guidelines, etc.
2. Understand the use of process mining and associated techniques in creating and validating practice-based evidence.
3. Understand an aspect of the learning health system.
4. Develop ideas on how to exploit their own EHR to discover usable knowledge.


Anthony Solomonides (Presenter)
NorthShore University HealthSystem

Yiye Zhang, Weill Cornell
Gordon Dri, University of Chicago
Goutham Rao, Case Western Reserve
Rema Padman, Carnegie Mellon University
Monica Vuppalapati, University of Chicago
Leibao ("Nick") Qi, University of Chicago
Victoria Bauer, NorthShore University HealthSystem
Arnab Bose, University of Chicago
David Newman-Toker, Johns Hopkins
Katherine Kirley, American Medical Association
Paul Epner, Society to Improve Diagnosis in Medicine

Presentation Materials: