(426) Measuring Clinical Performance as a Benchmark for Enhancing It Through Digital Transformation

Introduction: Heterogeneity in disease presentation, physiologic responses to therapies, clinical decision making, and standards of care amongst a host of other biologic, pathologic, and socio-cultural factors make creating technologies for use in medicine difficult. Furthermore, failed delivery on promises by technology developers has fractured trust between clinicians and those seeking to implement new tools in clinical work domains. We seek to create new methods for measuring the impact of technology on clinical performance. We hypothesize that improving clinical performance with technology can reduce clinician task burden, improve care quality, enhance care efficiency, improve system performance and patient outcomes. This pilot study explores performance during combat casualty care (C3) for which there is a dearth of literature.

Methods: To measure baseline clinical performance for C3, we created standardized, high-fidelity, medical simulation (HFMS) models of trauma C3 patients. Each HFMS was tested to ensure researchers could administer models consistently in different clinical contexts (e.g., light, dark, smoky). We identified recognizable start and stop criteria for clinical tasks (e.g. life saving interventions, documentation) to successfully manage the C3 patient model. We collected data under IRB approved protocols using video and audio recordings of volunteer pre-hospital trained participants (e.g., emergency medical technicians, military medics) who were instructed to treat the “patient” to the best of their ability. Descriptive statistics of baseline C3 procedures under HFMS was performed.

Results: Ten participants completed C3 scenarios of four different trauma models. The most common baseline performance measures observed were (mean; SD in seconds): 20 documentation tasks (52; 38); 12 tourniquet applications (57; 15); 10 IV initiations (209; 69); 7 manage IV access events (28; 8); 5 pressure dressing applications (544; 476); 5 manage minor lacerations (285; 243) and 5 chest needle decompressions (49; 17).

Conclusions: HFMS models of C3 trauma patients can be used to measure medical performance of caregivers to establish a baseline to which performance with novel tools like artificial intelligence, clinical decision support, autonomous medical devices, and robotic assistances can be compared.