P074: Clinician gestalt in the evaluation of pulmonary embolism risk factors: the CEPERF study

Abstract

Introduction: Pulmonary Embolism (PE) is a difficult to diagnose presentation associated with significant morbidity and mortality. Despite development of risk stratification tools (RST), physician gestalt continues to play a large role in the diagnostic evaluation of PE. Implicit in this gestalt is the evaluation of PE risk factors (RF). It is unknown, however, if physicians are similar and accurate in their assessment of known PE RF. Methods: An online survey presented paired comparisons (n=55) of 11 known PE RF to active Emergency Physicians (n=20), Family Doctors (n=11), and Residents (Family Medicine [n=20]; Emergency Medicine [n=5]). The Bradley-Terry Model converted the paired comparisons to rank order lists for the cohorts and these lists were compared. The perceived efficacy and use of RST and gestalt was also assessed across the cohorts. Results: The response rate was 72%. Emergency Physicians had the highest perception of gestalt as an effective method of risk stratification (7.4±1.4 out of 10) while Family Medicine Residents had the lowest (5.1±1.9). More than 95% of Emergency Physicians and Residents employed RST (PERC and Wells) compared to 46% of Family Physician respondents. Those who used RST utilized the tools in the majority of their clinical encounters (>75% of the time). There was good agreement between the cohorts in regards to their rank order lists (Tau-b ≥ 0.71). Age was identified as a RF which was consistently ranked lower than literature reported values amongst the cohorts. Conclusion: Physicians in various practice settings and levels of training rank PE risk factors similarly when forced to compare them. There are important RF, most notably age, which were identified in the current study that were consistently undervalued. This finding may highlight how RST are shaping perceptions of PE RF through their use and how age as a PE RF may warrant more attention in education and clinical assessments.