DEVELOPMENT AND PRELIMINARY EVALUATION OF EMERGENCY MEDICAL ALLIANCE FOR TOTAL COODINATION IN HEALTHCARE (E-MATCH) TO RESOLVE MISMATCH BETWEEN PATIENTS NEEDS AND AVAILABLE RESOURCES

Background: When emergency medical technicians (EMT) in Japan make the initial call to hospitals to accept patients, they have difficulty identifying medical facilities that can provide appropriate care to match patient conditions. Such mismatches can be problematic and are a growing social problem, often resulting in the rejection of emergency patients from one hospital to another.

Purpose: To design, develop, and evaluate a decision support tool that resolves discrepancies between patient needs and available resources called “emergency Medical Alliance for Total Coordination in Healthcare (e-MATCH),” which runs on an iPad2 platform and promotes shared decision making between EMTs and medical facilities.

Method: A rule-based algorithm that integrates information on patient symptoms and signs, resource availability, and current status of each facility was implemented in e-MATCH. EMTs record relevant patient information at scene, which is then used for the algorithm and information sharing between EMTs and hospitals. We evaluated both the algorithm and human user interface design, as well as algorithm performance for patients in cardiac or respiratory arrest (CPA) transferred before or after algorithm implementation. Consecutive patients transferred from July 1 to July 31, 2010 (before implementation) and from January 31 to February 14, 2011 (after implementation) were analyzed to determine the proportion of patients accepted after a first call (1st time %) and after >4 calls (>4 %). Usability was also evaluated by Nielsen’s usability heuristics. Evaluated items included consistency, visibility, match, minimalist, memory, feedback, flexibility, message, error, closure, undo, language, control, and document.

Result: There were 56 and 46 CPA patients before and after algorithm implementation, respectively. The 1st time % before and after implementation were 64.3% (36/56) and 73.9% (34/46) (n.s.), and the >4 % were 10.7% (6/56) and 8.7％ (4/46), respectively (n.s.). While no problems leading to serious errors were identified, negative feedback was received from EMTs and physicians with respect to consistency, message, and error.

Conclusion: The e-MATCH system demonstrated a potential to reduce mismatches between patient needs and available resources. We have been improving the human user interface to satisfy EMT needs at scene based on heuristic evaluation.