A POINT OF CARE COMPUTER APPLICATION DESIGNED TO IMPROVE MEDICAL DECISION MAKING: A RANDOMIZED, CONTROLLED TRIAL

Monday, October 25, 2010
Sheraton Hall E/F (Sheraton Centre Toronto Hotel)
Farrell Lloyd, MD, MPH and Felicity Enders, PhD, Mayo Clinic, Rochester, MN

Purpose: Clinicians use online sources of medical knowledge at the point-of-care. Memory and learning science enhance the development and research of these applications. Our objective is to assess the efficacy of a point-of-care application to answering clinical prescribers' specific clinical questions versus other online applications (control).  The point-of-care application contains peer-reviewed answers to specific clinical questions.  There are currently over 3,000 questions and answers in the application.  The answers come from experts and are formatted to be concise and to-the-point.  Medical editors review and update the content.  The application is linked to the electronic medical record at clinical campuses in Arizona, Iowa, Minnesota, Florida, and Wisconsin.

Method: Design: Randomized, controlled, crossover trial. Setting:  Multi-campus, academic healthcare organization. Participants: Sixty practicing clinicians. Intervention(s): A desktop application that provides peer-reviewed answers to clinical questions was compared to a clinician’s choice of existing or familiar online applications. Main Measure(s): The primary outcome was the time to a correct response with at least 80% confidence in that response. 

Result: From 2/23/2008 to 3/14/2010, n = 1,474 clinical prescribers have accessed 6,842 frequently asked questions and made 2,625 expert contacts using the point-of-care computer application used in this study. Of the represented sample in this trial, clinicians using the point-of-care computer application scored about one additional question correct of seven than the control group (95% CI 0.5, 1.4; p<0.001). In the time-to-event analysis, the chance of finding a correct answer with at least 80% confidence was 2.3 times greater when using the point-of-care computer application than control (95% CI: 1.3, 4.6; p=0.008).

Conclusion: A point of care computer application designed to capture the gist of specific clinical questions from subspecialty experts and transfer this gist to experts outside their specialty may enable clinicians to answer questions more accurately, quickly, and with appropriate confidence than other online tools. Humans rely on memory of gist as they develop expertise. Gist persists longer than verbatim memory. This application may enhance decision making by improving the ability of less experienced experts to find the appropriate information and retrieve the appropriate memory for gist at the point-of-care.  This ability to “connect-the-dots” when taking care of patients is an essential component of logically coherent medical decision making at the point-of-care.