Prospective evaluation of an ultrasonography training model for emergency physicians: Can we teach old dogs new tricks?

Abstract

Study objectives: Emergency bedside ultrasonography is gaining widespread use in emergency departments (EDs) throughout the country, and many guidelines exist for the training of emergency physicians in its use. One of the major challenges in training physicians in emergency bedside ultrasonography is the ability of new users to reliably recognize normal, abnormal, and nondiagnostic (technically limited) images. This study evaluated the retention of ultrasonographic knowledge and the accuracy of image interpretation 6 months after a standardized ultrasonographic course was completed. A novel approach to quality assurance in assessing interpretation skills, the recognition of technically limited studies, was tested. In addition, this study evaluated whether the training actually increases the use of bedside ultrasonography by attending-level emergency physicians. Methods: Emergency medicine attending faculty from 2 affiliated academic EDs completed an 8-hour introductory ultrasonographic course. The course consisted of 5 hours of lectures, with a printed syllabus covering the following topics: physics, focused abdominal sonography in trauma, cardiac, aorta, renal, gallbladder, pelvis. In addition, 3 hours of supervised hands-on sonography with live models was incorporated. Each participant completed the same pretest and posttest requiring the interpretation of 24 emergency bedside ultrasonography images. Four images from each of the 6 areas of emergency ultrasonography were included, with positive, negative, and nondiagnostic images represented in each area. The pretest was conducted the same day as the course, and the posttest was conducted 6 months later. A self-reporting survey of ultrasonography use in each of the primary indications was completed with each of the tests. Changes in test scores and the number of ultrasonographic examinations performed before and after the course were compared using the paired <i>t</i> test. Results: Nineteen emergency attending physicians (76% with no previous formal emergency bedside ultrasonography training) completed the introductory course and completed a pretest and posttest. The mean overall test score was improved from 57% before the course and 67% at 6 months (<i>P</i>=.003). Improvement in image interpretation accuracy was most pronounced in gallbladder (44% to 61%, <i>P</i>=.03) and cardiac (61% to 75%, <i>P</i>=.01) images. Recognition of true positive images improved (69% to 78%, <i>P</i>=.05); improvement in the recognition of true negative images approached significance (53% to 63%, <i>P</i>=.06). There was no significant improvement in the recognition of technically limited studies. There was a 68% increase in use of emergency bedside ultrasonography overall at 6 months that was not dependent on level of previous training. Conclusion: Consistent with previous studies, this evaluation demonstrates good retention of emergency bedside ultrasonography knowledge even 6 months after completion of a standardized 8-hour course. With minimal training of staff, the use of emergency bedside ultrasonography increased, and the accuracy of image interpretation improved, which may increase resident supervision and improve patient care. However, the recognition of technically limited studies did not improve. A common goal of quality assurance programs is to monitor the accuracy of emergency bedside ultrasonography within each of the 6 major indications; we recommend an ongoing educational and image review process to improve this outcome, and further research will test this prospectively.