“The effect of ED crowding on education”

My heart almost stopped when I read this article title in Amer J of Emerg Med. This was the premise of my recently completed study – using a prospective, time-motion methodology. I’m in the process of writing the manuscript. Did I get scooped by my friends at U Penn?

Whew. Fortunately, no. Different methodology.

This study was a cross-sectional study looking at learner assessment of education, using a validated tool called the ER (Emergency Rotation) Score. The results are interesting.

The problem

We know that ED crowding negatively impacts clinical care. How does it impact our teaching of medical students and residents? The ED is traditionally known as a great place for learning how to resuscitate high-acuity patients, to manage and risk-stratify undifferentiated cases, and to perform procedures. Experientially, I feel like I teach less when it gets extremely crowded.

Methodology

Over a 5-week period, 43 residents and 3 medical students prospectively assessed 34 attendings using a simple ER Score tool. There were 352 separate encounters. This validated tool assessed the attending based on 4 domains (teaching, clinical care, approachability, helpfulness) with each domain assessed on a 5-point scale. The scores were correlated with crowding measures (waiting room number, occupancy rate, number of admitted patients, and patient-hours).

What was their enrollment scheme?

Upon arrival, the research assistant selected the patient with the most recent admission order where the learner-attending pair was still present in the ED. The learner was asked to fill out the ER Score tool. For each admitted patient case, the research assistant also enrolled a non-admitted patient with a similar triage intake time. The learner for this non-admitted case was also asked to fill out the ER Score tool. The study group intentionally structured this methodology to oversample admitted patients, which they assumed impacted education more than non-admitted patients.

Results

The median score was 16 of 20. ED crowding levels were NOT associated with ER scores or their individual domains.

How fascinating that learners still felt that the quality of teaching and learning in the ED was maintained despite the ED being overwhelmed beyond capacity.

The next step is to follow Kirkpatrick’s model in conducting educational research. In this model, satisfaction/reaction-based studies are the first (lowest) tier. Such studies inherently have flaws based on bias, recall, and halo effect. The next study is to look at more objective measures assessing the impact of crowding on education. Hmm, I better get going on my manuscript.

Kirkpatrick’s 4-tiered model to evaluate training and education

Reference
Pines JM, Prabhu A, McCusker CM, Hollander JE. The effect of ED crowding on education. Amer J Emerg Med (2010) 28, 217–220.