Assessing Graduating Resident Confidence in Essential Emergency Medicine Procedures

Invasive procedures are routine in Emergency Medicine (EM) but carry risk of complications. Although acquisition of procedural skills can prevent harm, the current numerical standard for procedural competency in EM is not backed by evidence. Due to a lack of alternative metrics, self-reported confidence levels have been used as a proxy for procedural competency. Procedural proficiency peaks at the end of residency, making this population ideal for studying skill acquisition. This study’s objective was to characterize confidence levels for EM index procedures among graduating residents.

In Spring 2025, a cohort of graduating EM residents representing a diverse national sample was polled using a 25-item, cross-sectional survey developed via literature review and expert input. Anonymous responses were collected using a 5-point Likert-type scale, and collected data were analyzed using regression and cluster analysis.

Overall, 103 residents (41%) from 22 programs (61%) responded, including 79 (77%) from 3-year and 24 (23%) from 4-year EM programs. As expected, few respondents reported high confidence for high-acuity-low-occurrence (HALO) procedures (cricothyrotomy, 16.3%; pericardiocentesis, 7.7%). Residents were highly confident in central line placement (91.3%) and intubations (89.4%) but had a variable response for lumbar puncture (LP) (45.2%), and small- and large-bore chest tube insertion (49%, 48.1%). Multilevel regression analysis of other program factors, such as ED volume and training length, did not show a significant effect. Every program used simulation to complete requirements.

Despite achieving ACGME thresholds, graduating residents reported inconsistent confidence in common EM procedures, namely LP and chest tube insertion. High confidence levels were noted for intubations and central lines, suggesting numerical thresholds may have validity for these procedures. The resident experience is variable, but national trends indicate a decline in common procedure numbers. Length of training did not have a significant impact on confidence. Better standards are needed to assess procedural proficiency, and training programs will have to create opportunities to increase confidence levels. Understanding the factors that influence procedural confidence can inform improvements in training curricula, evaluation, and simulation resources.