Live from the AMIA 2012 Annual Symposium: Getting Crunchy on Co-Morbidity Issues

April 9, 2013
Four medical informaticists present findings and learnings from their research and development work in the area of providing clinical decision support to physicians treating patients with co-morbidities

At the AMIA 2012 Annual Symposium on Monday, Oct. 5, attendees got a good look at just how complicated clinical decision support is as an issue when it comes to patients with co-morbidities, in the educational session “Opportunities to Support Complex Medical Decisions through Informatics.” Presented by Mary K. Goldstein, M.D., MSc, Donna M. Zulman, M.D., MS, Roberto A. Rocha, M.D., Ph.D., and Mark A. Musen, M.D., Ph.D., the session offered its four presenters the opportunity to describe some of the efforts they are involved in at their patient care organizations and universities to explore the very common challenge of providing robust, meaningful clinical decision support to physicians treating patients with multiple medical conditions. As Goldstein, director of the Geriatrics Research Education & Clinical Center at the VA Palo Alto Health Care System in Palo Alto, Calif., and a professor of medicine at Stanford University, noted at the outset of the session that more than 50 percent of older adults in the U.S. suffer from three or more chronic diseases, and that these “multimorbid” patients, as they are sometimes called, account for a large percentage of annual healthcare spending. Yet using clinical decision support tools to help physicians diagnose and treat them remains a major challenge.

Each of the panelists in turn shed light on some of the learnings their colleagues are absorbing in this area. Zulman, a general internist and an investigator at the Center for Health Care Evaluation at the VA Palo Alto Health Care System, described in detail two bundles of challenges bedeviling clinicians in this area. First, she noted, there is a core patient self-management challenge, as multimorbid patients have to try to adhere to very complicated diet, exercise, and medication regimes. Second, she said, physicians treating such patients inevitably get caught up in the complexities of trying to take into account multiple treatment guidelines, adhere to multiple and sometimes-conflicting quality measures, and sort out individual patient needs, all at the same time. Informatics can help tremendously, she said, in navigating such situations. Among the typical examples she cited: physicians treating osteoarthritis are generally advised by CDS systems to prescribe NSAIDS for their pain, while those treating diabetes are told to prescribe aspirin. The combination of the two, of course, increases dramatically the potential for bleeding. So the opportunities to address co-morbidity interrelatedness, she said, will be among the richest areas of endeavor going forward.

Rocha, a senior corporate manager for knowledge management and clinical decision support in the Clinical Informatics Research and Development Group at the Boston-based Partners HealthCare system, described the work that he and his colleagues have been doing to define and manage complex patient states, including embedding about 340 action rules so far in their electronic health record (EHR), and working collaboratively to create what he called a “consistent representation of clinical context” in order for physicians treating multimorbid patients to look in more nuanced ways at their multimorbid patients.

And Musen, a  professor of medicine at Stanford University who conducts research related to intelligent systems, the semantic web, reusable ontologies and knowledge representations, and biomedical decision support, described the groundbreaking work that he and his colleagues have been involved in, in building the “GLINDA” system (Guideline Interaction Detection Architecture), which attempts to ameliorate the “messiness of clinical situations” by applying multiple clinical-practice guidelines, adjusting for patient co-morbidities, and adjusting for interactions among the individual recommendations providing by any CDS for specific situations, by “adjudicating” complex situations that trigger conflicting decision support advice.

The ability to develop synthesized clinical decision support, to develop an ontology of guideline interactions, and to “develop agents for detecting conflicts, repairing conflicts, prioritizing and integrating treatment recommendations,” he said, will be key to moving towards the next generation of clinical decision support.

In the end, the presenters agreed, it will take diverse, yet concerted, work on the part of many teams of medical informaticists to move clinical information systems forward towards the day when they will routinely be able to dispense the kinds of nuanced advice physicians need when caring for multimorbid patients. Doubtless, it will be researchers and developers with the insights and learnings gleaned by these informaticists and their colleagues, who will be in the forefront of that movement going forward.

Sponsored Recommendations

Northeast Georgia Health System: Scaling Digital Transformation in a Competitive Market

Find out how Northeast Georgia Health System (NGHS) enabled digital access to achieve new patient acquisition goals in Georgia's highly competitive healthcare market.

2023 Care Access Benchmark Report for Healthcare Organizations

To manage growing consumer expectations and shrinking staff resources, forward-thinking healthcare organizations have adopted digital strategies, but recent research shows that...

Increase ROI Through AI: Unlocking Scarce Capacity & Staffing

Unlock the potential of AI to optimize capacity and staffing in healthcare. Join us on February 27th to discover how innovative AI-driven solutions can revolutionize operations...

Boosting Marketing Efficiency: A Community Healthcare Provider’s Success Story

Explore the transformative impact of data-driven insights on Baptist Health's marketing strategies. Dive into this comprehensive case study to uncover the value of leveraging ...