Advancing the Learning Health System: One Scholar Sees Opportunities, Challenges
How can healthcare industry leaders help to advance the concept of the learning health system, system-wide? That’s a question that J. Michael McGinnis, M.D., M.A., M.P.P., a senior scholar at the National Academy of Medicine (NAM), has asked in an essay published on July 1 in The New England Journal of Medicine.
As McGinnis notes in “Advancing the Learning Health System,” McGinnis writes that “Accelerating developments that began in the 20th century have set the stage for a revolution in knowledge generation and application for progress in health. Evidence development that has been episodic, deliberate, and insulated is becoming more continuous, query-based, and relational — and the health system is both contributing and benefiting. The National Academy of Medicine (NAM),” he points out, “is working to facilitate this change. Recognizing that technological and methodologic advances could improve on the pace, generalizability, and costs of innovation in health and medicine, the NAM (previously the Institute of Medicine [IOM]) has envisioned and helped steward the evolution of a continuously learning health system (LHS),” going back as far as 2006—with the idea that “best practices and discovery [could become] seamlessly embedded in the delivery process, individuals and families as active participants in all elements, and new knowledge generated as an integral by-product of the delivery experience.”
As McGinnis explains it, “Underpinning this vision is an appreciation that health is shaped by multiple forces: genetic predispositions, social circumstances, physical environments, behavioral choices, and medical care.3 Engagement by myriad stakeholders — from medical care, public health, social, cultural, and research arenas — is therefore required to assess, understand, and improve individual and population health. Similarly, knowledge to guide those actions must derive from, and be useful to, all participants, especially the people whose health and well-being are the anchoring aim.”
McGinnis details a large number of elements that have until recently frustrated broad progress around creating a healthcare system-wide learning system. But, he notes, “The digital revolution has progressed in parallel and intertwined with advances in science and evidence development. Since the 1959 invention of the microchip, the 1983 establishment of the Internet, and the 1990 creation of the World Wide Web, efforts to develop automated health monitoring and health record storage devices have grown enormously. Many of the advances most useful for the LHS relate to developments in database management and assessment that constitute the basic infrastructure of the learning system, and companies such as Google, Apple, and others are bringing that capacity to the individual level. These include analytics involved in artificial intelligence and machine learning that could vastly increase the scale, speed, and sophistication of LHS operations.”
He cites examples such as the “REDUCE MRSA study,” which involved a “cluster-randomized trial demonstrated that a universal decolonization strategy led to a 44-percent reduction in all-cause central-line–associated bloodstream infections in intensive care settings. The ability to draw, in real time, on data from 74,000 patients in 43 hospitals yielded results in 18 months that, the authors noted, would have taken 64 years for a single hospital.”
McGinnis writes that the leaders of the National Academy of Medicine believe that broad collaboration between the NAM and a very broad range of stakeholders—including “professional societies, patient and citizen groups, health care organizations, public health leaders, payers, employers, product manufacturers, information technology companies, and government agencies”—will continue to move the healthcare system forward towards becoming a true learning health system over time.