Discussion Board

Seeking Participants for a Study to Evaluate a Diagnostic Decision Support System

  • 1.  Seeking Participants for a Study to Evaluate a Diagnostic Decision Support System

    Posted 01-29-2021 10:24

    Diagnostic Challenge Study Summary: 

    Conducted by Principal Investigator, Joe M. Bridges, a student in the program leading to a Doctorate in Health Informatics (DHI) at the School of Biomedical Informatics (SBMI) at the University of Texas Health Science Center in Houston (UTHealth), the study is a part of the Translational Project required by the DHI program. The study has received a waiver from the Institutional Review Board of the Committee for the Protection of Human Subjects at UTHealth (HSC-SBMI-21-0009 - JMB Diagnostic Challenge: Evaluation of a Diagnostic Decision Support System). 

    The study seeks participants for a "Diagnostic Challenge." Participants will be asked to diagnose a total of 46 cases, in sets of two cases each over approximately six weeks. The study commenced in mid-January 2021. Participants should be practicing physicians in areas of ambulatory care, such as primary care, internal medicine, family medicine, or similar practices. Those wishing to participate in the study may confirm by email to:


     Please feel free to send this recruitment request to any individual you think may be interested in participating. 

    Project Overview: 

    Using only the presenting elements of a patient case, evaluate the diagnostic retrieval accuracy of a commercially available diagnostic decision support system. Compare the system results to the results of physicians participating in the Diagnostic Challenge. Assess the improvement, if any, in the physicians' diagnostic accuracy after reviewing diagnostic suggestions from the system. Assess the participants' opinions about diagnostic suggestions in a questionnaire following the cases. 

    Diagnostic Challenge Format: 

    Participants will register and consent to participate, then diagnosis the cases sequentially and produce a differential diagnosis listing. Then, review the same case, now including diagnostic suggestions from the system, with the option to revise, reorder, or otherwise change the original list if desired. The cases will be presented in 23 sets of two cases per set by emailing the participants a link to a RedCap survey website. The sets will be presented to each participant by a daily email (or an email the next day following the completion of a set) until the participant completes all 23 sets.

    [Joe [Bridges]
    [School of Biomedical Informatics, UTHealth]