When James R. Duncan, MD, PhD, FSIR, walks into the hospital, he’s thinking about how to improve each and every step of his job. As a member of the IR staff at the Mallinckrodt Institute of Radiology and active member in quality improvement research, there is no questioning his dedication to improving the field of interventional radiology. He recently received a Radiological Society of North America (RSNA) Education Scholar Grant for his project, “Development and Implementation of a Radiology Improvement Leader Training Course,” and has published multiple articles in the Journal of Vascular and Interventional Radiology (JVIR) on quality improvement in interventional radiology. I recently spoke with Dr. Duncan on how he became involved with quality improvement and asked him for trainee level advice, including ideas for quality improvement projects (an Accreditation Council for Graduate Medical Education [ACGME] resident requirement).
Tabriz: How did you get started in quality improvement in IR? Have you found it fulfilling?
Duncan: I have had a longstanding interest in improvement. It started with a curiosity about how things work and realization that such knowledge should be used to continually improve anything and everything. I then pursued research training because I wanted to better understand how inborn errors of metabolism caused different disease phenotypes and what strategies might alleviate pain and suffering.
In contrast to medical therapy, where risk/benefit ratio of any agent depends primarily on its biochemistry, for IR, the risk/ benefit ratio of a procedure is highly dependent on physician performance. This encouraged me to study simulation as a means of building skills through deliberate practice. As my colleagues and I got into this work, we realized that we needed objective methods of assessing physician performance so that we could provide feedback to trainees. We began recording the simulated procedures and developing performance metrics. Since our goal was always to improve performance during actual procedures, we realized that a more direct route might be studying performance during actual procedures. This opened the door to ideas from other fields like Six Sigma, statistical process control as well as information theory, human psychology and team dynamics.
Caring for patients is a journey and I find it very fulfilling to have a career arc that combines my lifelong personal voyage of discovery with innumerable small daily journeys that I share with patients, their families and the rest of the health care team. For example, I currently spend a good percentage of my time caring for children. I view my preprocedure conversations with the parents as an opportunity to assure them that they can trust us as we care for their children. As a parent myself, it is not hard to step into their perspective and recognize that, while something like a PICC line might be a routine procedure to us, to them it is the most important thing happening in the world.
My work in quality and safety helped me understand that quality is all about meeting or exceeding the customer’s expectations. Thus I spend a lot of time trying to help parents develop reasonable expectations about the procedure including how long it will take and what they can expect in the coming hours and days. I also tell them that they should ask me how often I have done this procedure and while I can usually say “hundreds” or “thousands,” I then tell them that this is the first time I will be caring for their child and that their child might be different from all the prior children.
Tabriz: What key references would you recommend for a trainee looking for an overview of quality improvement methodology?
Duncan: Anyone interested in quality improvement should read The Best Practice, by Charles Kenney. Much of the book follows [former CMS administrator] Don Berwick on his journey, which includes the Institute of Medicine’s game changing “To Err Is Human” report and the creation of the Institute for Healthcare Improvement. For anyone wanting a deeper dive, the most influential book I have read in the last decade is An Introduction to Information Theory, by John Pierce. It describes Claude Shannon’s work on the mathematical basis of communication. Shannon’s work is the foundation of modern computing and telecommunications. That book helped me understand the importance of making predictions and processing feedback information in an attempt to continually improve my ability to make accurate predictions.
Tabriz: What are the hot topics for quality improvement in IR? What topics do you think could best be incorporated into a shortterm (less than three years) project to help residents fulfill their ACGME quality improvement requirement?
Duncan: Radiation use is clearly a hot topic for all of imaging. Lots of room for improvement, especially in IR, and many of the strategies for improvement are already fairly far advanced in CT. There’s every reason to apply them to fluoroscopic procedures.
Improving outcomes is another hot topic for IR. Regrettably, we have few instances of clear objective data on our outcomes. However, if companies like Google can use large datasets to measure and improve the performance of the predictive models they use for search, voice translation or the flu, we should be ready to do the same with our electronic datasets—be it patterns found in billing data or sections of the electronic medical record.
Teamwork is a third hot topic. Medicine is a team sport. Physicians need to recognize the importance of partnering with hospitals and patients to build “win-win” opportunities. Disease, suffering and lack of knowledge are the common enemies. We should keep this in mind as we try to build the shared mental models that underlie effective teamwork. While teamwork is a complex issue, we have found that improving timeout performance has been a great step towards promoting a safety culture for the team. It is a very frequent and observable activity that involves nurses, techs and physicians. It also resonates well with the hospital’s leadership.
Tabriz: What advice would you give IRs as they progress through training?
Duncan: “Always be thinking about what you will do if what you are currently trying fails.” In essence, it is an acknowledgment that no current plan is perfect and that, even though Plan A has worked 50 consecutive times, a Plan B is worth having because sooner or later, Plan A will fail. When Plan A does fail, having Plan B in your back pocket is your best defense against catastrophe. I find such contingency planning to be an incredibly interesting part of our specialty. Not only does it prompt me to identify leading indicators that Plan A is in trouble, but it also makes me predict what might cause Plan A’s failure and how that knowledge should be incorporated into my contingency plans.
