Measuring Resident Experience
with Intraoperative Hemorrhage
Matthew Caldwell, MD
University of Michigan
Department of Anesthesiology
Disclosures
• Foundation for Anesthesia Education and Research
– FAER-REG 08-15-2023
• Michigan Medicine Graduate Medical Education Innovation Grant Program
– Award U078433
Agency for Healthcare Research and Quality, Rockville, MD.
https://www.ahrq.gov/learning-health-systems/about.html
Current data for anesthesiology GME
How might MPOG data help improve Graduate Medical Education?
How might MPOG data help improve Graduate Medical Education?
Rubin et al. Anesth Analg. 2020 Apr;130(4): 1026-1034.
INSPECT: Improvement in the Scope and Precision of Educational Cases for Trainees
• Aim 1: Quantify and describe the experience of anesthesiology residents with severe
intraoperative hemorrhage resuscitation.
• Aim 2: Determine trainee and program factors associated with low (<25
th
percentile)
resident exposure to severe hemorrhage resuscitation.
• Aim 3: Develop and optimize data visualization tools for use by Program Directors to
improve resident education.
Study population
• 37 MPOG centers with anesthesiology residency training programs
– 28 centers with sufficient resident data for inclusion in study
• 4 resident cohorts from graduation years 2020-2023
• 1640 total residents
Study outcomes
• Primary outcome severe hemorrhage resuscitation (equivalent of 4 units of pRBCs)
• Secondary outcomes
– pRBC transfusion management (transfusion of any volume of red blood cells)
– Non-pRBC transfusion
– Pediatric transfusion
• Subgroup analyses
– Emergency status
– Surgical subgroups
• Factors associated with resident participation in severe hemorrhage resuscitation
Distribution of residents across centers (preliminary data)
Resident cases with severe hemorrhage (preliminary data)
Per resident number of cases with severe hemorrhage
Median
Range
(min, max)
IQR
(Q1, Q3)
6
(1,35) (3,10)
Single Center Data: Severe Intraoperative Hemorrhage
Median
Range
(min, max)
IQR
(Q1, Q3)
10 (2, 22) (8,12)
*Single center data
Timing of first exposure
*Single center data
How might we use this data to improve education
and ultimately patient care?
• Educational scaffolding
• Reflective practice
• Smooth variability across residents
• Intentional scheduling to increase likelihood of specific clinical scenarios
• Identify high value low frequency scenarios for simulation
Key Points Summary
• MPOG data can be used to support GME
–Can provide normative data
–Responsive to shifts in clinical practice
–Precision education
• May allow for optimized experiential learning
• Rich opportunity to further explore
Mentors and Collaborators
• University of Michigan
– Douglas Colquhoun, MB ChB
– Sachin Kheterpal, MD
– Michael Mathis, MD
– Graciela Mentz, PhD
– Norah Naughton, MD
– Lara Zisblatt, EdD
– Xinyi Zhao, MS
• Oregon Health & Science University
– Zheyan Jenny Chen, MD, PhD
– Amy Miller Juve, EdD
• University of California San Francisco
– Christy Boscardin, PhD
– Kristina Sullivan, MD
• University of North Carolina
– Greg Balfanz, MD
– Alexander Doyal, MD
– Fei Chen, Phd
Matthew Caldwell
mdcaldwe@umich.edu
• University of Oklahoma
– Christine Vo, MD
– Casey Windrix, MD
• Washington University St. Louis
– Avi Dobrusin, MD
– Muthuraj Kanakaraj, MD
– Rachel Moquin, EdD
• Yale University
– Sadhvi Khanna, MPH
– Daniel Kinney, MD
