From the point of view of the applied clinical informatics researcher, it is difficult to predict the unintended outcomes of Health IT. Software engineers work hard to make the EHR bug-free, testers perform their job and implementation teams successfully educate end-users. There is a disconnect between bedside care providers and EHR architects.
Four years ago our organization realized this and began to develop the AWARE platform (Ambient Warning and Response Evaluation). The project started small and was driven bottom-up by enthusiastic ICU clinicians, focused on clinical needs and patient outcomes. Importantly, the AWARE team partnered with IT architects and programmers who were committed to building this platform in very close collaboration with the clinicians. We worked shoulder to shoulder with a single principle to guide us: “this has to make care better for our patients.” The requirements were based on scientific evaluation of the information needs and workflow of frontline providers, as well as a sophisticated understanding of meaningful patient-centered outcomes. The clinicians drove the IT team to despair at times – beautiful code was ripped out and abandoned because the resulting tool violated the key principle of the collaboration.
Over time, the clinicians learned how to communicate specifications and requirements with programmers and programmers developed an understanding of the clinical environment, workflow and impact their tools were having on patients. AWARE grew up in a working ICU environment with careful oversight by the clinicians supported by clinical research methodology to identify risks to patients and clinical workflow.
Over the first few years a handful, and more recently hundreds, of clinicians drove changes and modifications to AWARE. The IT team became adept at anticipating how new features would improve AWARE for practice. Soon, our IT team had developed an architecture and platform that was highly responsive to the demands of clinical practice. Together we had turned the paradigm of “architecture first, clinical tools to follow,” upside down.
What followed was remarkable. In a simulated scenario AWARE demonstrated a positive effect in decreasing the number of medical errors, cognitive load of clinicians, and time needed to complete tasks . The Center for Medicare and Medicaid Innovation recognized the potential of AWARE and decided to support further platform innovation, implementation and study in Mayo Clinic Rochester and three other hospitals.
One day, I opened an intra-clinic envelope containing a signed petition from ICU clinicians on the Florida campus of the Mayo Clinic requesting implementation of AWARE in their practice. For a hospital at HIMSS stage level 7, it is unusual for clinicians to request one more EHR tool be added to their electronic environment. What we can learn from this story is described by a simple equation .
Clinical informatics is not “computer > human.” Rather, it is “computer + human + scientific method > human.”
- 1. Han YY, Carcillo JA, Venkataraman ST, et al. Unexpected increased mortality after implementation of a commercially sold computerized physician order entry system. 2005;116(6):1506–1512.
- 2. Ahmed A, Chandra S, Herasevich V, Gajic O, Pickering BW. The effect of two different electronic health record user interfaces on intensive care provider task load, errors of cognition, and performance. Crit Care Med. 2011;39(7):1626–34.
- 3. Hunter JS. Enhancing Friedman’s “fundamental theorem of biomedical informatics.” J Am Med Inf Assoc. 2010;17(1):112
About the Contributor
Dr. Herasevich, MD, PhD, MSc, CPHIMS is Assistant Professor of Medicine and Anesthesiology in Department of Anesthesiology at Mayo Clinic and Associate Program Director of newly established Clinical Informatics Fellowship program. His interest in the area of medical informatics extends back more than 17 years. With specific interest in clinical informatics in intensive care and science of healthcare delivery, Dr. Herasevich works on studying and development novel clinical integrated solutions including real-time dashboards, tasks specific viewers, “sniffers” (alerts) and analytical warehouses in support of patient-centered decision-making and outcome reporting. He has served HIMSS as member of the Physicians Committee.
Dr. Herasevich works in collaboration with Drs. Ognjen Gajic and Brian Pickering. Lacey Hart is Program Manager of ProCESSs AWARE grant. IT team includes David Zill, Troy Neumann and Ing Tiong. Clinical laboratory page: http://www.mayo.edu/research/labs/clinical-informatics-intensive-care/