The Daily Dose • Friday, April 8, 2022

Improving Best Practices in Perioperative Data and Patient Care

An accomplished investigator in improving perioperative care for patients with heart failure, Michael Mathis, MD, made the case for why collaborations with clinicians and informaticians are key to advance research during his presentation on “Collaborations with Clinicians and Informaticians: The MPOG Perspective,” held Saturday, April 2 during the IARS, AUA, SOCCA and eSAS Scholars’ Day. His presentation was one of the thought-provoking explorations of how to succeed in academic anesthesiology during the session, “Expanding the Footprint of Academic Anesthesiology.”

Kimberly Rengel, MD, Assistant Professor, Anesthesiology and Critical Care Medicine at Vanderbilt University Medical Center and Co-Chair of the IARS Meetings Committee and Co-Vice President of the Early-Stage Anesthesiology Scholars (eSAS), moderated this session and led a robust discussions with all three speakers following the session.

Michael Mathis, MD, an Assistant Professor of Anesthesiology in the Division of Adult Cardiac Anesthesia and Director of Cardiothoracic Research at the University of Michigan and the Research Director for Multicenter Perioperative Outcomes Group (MPOG), revealed some of the important lessons he has learned throughout his research career. He described how collaborations outside your department can enhance the impact anesthesiologists are able to impart on public health as a whole. He framed this discussion around developing and sustaining an academic career.

Many of the opportunities that advanced his academic career grew from his meaningful supporters over the years. Great opportunities come from great supporters, each of whom offers something required for success, Dr. Mathis explained. Mentors help individuals develop career goals, plans and aspirations. Sponsors serve as advocates and enable financial security. Sometimes mentors and sponsors can overlap. Coaches help perfect specific skills. Role models are those you wish to emulate, even if you have never met them.

Additionally, he highlighted a national initiative in the field, the Raising Anesthesiology Diversity and Anti-Racism (RADAR) initiative, that is challenging what success and role models representing that success should look like. He encouraged the scholars in attendance to reflect on who these individuals may be for them and consider having a conversation with them about their potential value to their career growth.

Dr. Mathis transitioned into a discussion of how anesthesiologists can collaborate with informatics experts to build an academic career. Focusing on improving clinical care and research can be accomplished in a Learning Health System, which is traditionally thought of as an academic medical center. However, he explained that this could apply to any medical center. The concept of a Learning Health System is to marry the processes of research and quality improvement into a cycle, beginning with a hypothesis and ending with iterating and disseminating the results. The second half of this cycle is often the most difficult, including taking that knowledge and determining how to update best practices based on this new knowledge, engaging providers and implementing these improved practices. That is better known as Quality Improvement (QI), he explained.

He offered MPOG as an example of one of those Learning Health Systems. MPOG is an international consortium of over 50 hospitals (academic and private practice) working in the US and united by a shared goal, using anesthesiology data to improve care.

First, data quality is more important than data quantity. As the Research Director of MPOG, he oversees research that is proposed by investigators at participating institutions. He shared an example from his own research to illustrate how informaticians can help clean and triangulate data to lead to more effective results. Through this experience, he has found that data must be examined more closely to remove artifactual data through digital phenotyping. Phenotypes are a composite representation of individual observable traits. He referenced the phenotype knowledgebase (, which has validated phenotype algorithms.

He emphasized that ideas without strategies are doomed. Quoting Bill Gates, “A bad strategy will fail no matter how good your information is. Lame execution will stymie a good strategy. If you do enough things poorly, you’ll go out of business.” Clinicians are destined for failure if we don’t think through our processes, he stressed.  Informatics is 10% of medicine, 10% of technology and 80% of sociology (people agreeing on best practices and embracing new standards).

How do we make big data available to do research for clinicians who aren’t necessarily computer programmers? Dr. Mathis offered one tool developed by MPOG called DataDirect. The idea is to democratize access to data for these clinicians. MPOG created this intuitive interface for any physician to create research queries for cases at local hospitals or across MPOG. Supporting data for research and quality improvement, the ASPIRE QI Dashboard shows individual clinicians statistics and performance outcomes.

To conclude his presentation, Dr. Mathis reemphasized takeaways for attendees to contemplate. Collaboration opportunities require career supporters to achieve advancement in the specialty. Perioperative data is messy and abundant but informaticians are a valuable resource to harness this data and use it for good. Following through is key, impact comes from collaboration and commitment. Commit to improving best practices.