Now I See It‐Building Expertise in Ocular Point of Care Ultrasound

Objectives Ocular point‐of‐care ultrasound (O‐POCUS) is a valuable tool in the assessment of ocular and neurological conditions. However, there is currently no standardized training or competency assessment for O‐POCUS. We evaluated the effectiveness of an O‐POCUS image interpretation education intervention on emergency physician diagnostic accuracy. We also determined the rate of skill development, the participant image review processes associated with a correct diagnosis, and the imaging findings associated with diagnostic error. Methods This was a multicenter prospective cohort study conducted on an online educational intervention for O‐POCUS interpretation. Participating emergency physicians deliberately practiced diagnosing 190 unique O‐POCUS cases, receiving immediate feedback after every case. They practiced on as many cases as necessary to achieve a pre‐defined performance standard. Results We enrolled 152 emergency physicians and derived 17,944 total case interpretations. From participant initial 25 to final 25 cases, there were learning gains in diagnostic accuracy (+11.3%; 95% CI 9.4, 13.2), sensitivity (+8.6%; 95% CI 6.9, 10.3), specificity (+16.1%; 95% CI 14.5, 17.7), pathology location accuracy (+12.9% 95% CI 10.9, 14.9) and specific pathological diagnosis (+18%; 95% CI 15.2, 20.7). Of the 125 participants with analyzable data, 94 (75.2%) achieved the performance standard in a median of 94 cases (IQR 64, 132) or 29 min (IQR 19.7, 40.7). Participants had lower odds of correct diagnosis when they spent less time reviewing a case (OR = 0.96; 95% CI 0.94–0.98), but higher odds when they were certain of their response (OR = 4.25; 95% CI 3.73–4.84). The imaging findings most at risk of diagnostic error were ocular masses, cataracts, and indicators of raised ICP. Conclusions This study demonstrates that an online and performance‐based education intervention can meaningfully improve O‐POCUS image interpretation among emergency physicians and identifies sonographic pathology prone to diagnostic error.