Toward a Complete and Sustainable Learning Health System Charles P. Friedman, PhD Josiah Macy, Jr. Professor of Medical Education Chair, Department of Learning Health Sciences Professor of Information and Public Health University of Michigan July 27, 2015
Disclosure I am the chair of the Interim Steering Committee of the Learning Health Community, a grassroots not-for-profit organization. 2
Preamble I believe that people are naturally drawn to ideas that reflect imagination and vision. By the end of this talk, I hope to convince you that pursuit of a learning health system makes good sense for the nation, for states and public health, and for all health care delivery systems. And that PopMedNet has an important role to play in this pursuit. 3
Main Menu for Today 1. The why and what of the Learning Health System (LHS). The LHS transcends Big Data and Precision Medicine 2. The Learning Health System is infrastructure. PopMedNet is leading the way 3. Progress toward an LHS at local, state, and national levels. Seven channels of progress 4
The Problem in One Figure Expenditure Per Capita 5
And In Words 25% of the (unsustainable) ~$3 Trillion spent on health is wasted ~45 th in infant mortality. Japan and Sweden have a rate 40% of ours. To Err is Human Study (2000): ~100,000 deaths per year due to medical error. No improvement since then. 17 year latency between discovery of new knowledge and its widespread use The details differ but these are global problems 6
Enter The Learning Health System Health systems--at any level of scale--become learning systems when they can continuously study and improve themselves Perspective: Jan 3, 2013 Code Red and Blue Safely Limiting Health Care s GDP Footprint Arnold Milstein, M.D., M.P.H. U.S. health care needs to adopt new work methods, outlined in the Institute of Medicine s vision for a learning health system 7
A Health System That Can Learn Every consenting patient s characteristics and experience are available to learn from Best practice knowledge is immediately available to support decisions Improvement is continuous through ongoing study This happens routinely, economically and almost invisibly All of this is part of the culture 8
Macro View: An Ultra-Large Scale System All-Inclusive Large Scale Pharma Decentralized Reciprocal Insurers Patient Groups Tech Industry Governance Engagement Data Aggregation Analysis Dissemination Universities Healthcare Delivery Networks Government/Public Health 9 Research Institutes 9
10 10 Learning Systems Can Exist at Any Level of Scale Mono-organizational, inter-organizational, state/regional, national, global
A Learning System Routinely Enables: Pursuit of Best and Safer Care at Lower Cost: Communities of interest discover what interventions are most cost-effective and are supported in implementing them. Enhanced Public Health: During an epidemic, new cases are reported directly from EHRs, the spread of the disease is predicted, and clinicians are alerted. Consumer Empowerment: Patients facing difficult medical decisions discover the experiences of other patients like them. 11
How Learning Happens : Virtuous Cycles of Study and Change Interpret Results Analyze Data Assemble Experience Data A Problem of Interest Tailored Messages to Decision-Makers Take Action Decision to Study 12
The Afferent and Efferent Sides of the Learning Cycle Afferent (BD2K) A Problem of Interest Efferent (K2P) Learning = BD2K + K2P 13
Example of A Virtuous Learning Cycle Interpret Results: Are the results credible? What advice should be given? Analyze Data: What practices associate with lower fall rates? Assemble Data: How do we prevent falls? What is the fall rate? Reducing Falls in Nursing Homes Tailored Messages: Based on your current practice, you might want to consider Take Action: Change Current Practice: In whole or part Decision to study falls 14
The LHS and Big Data The LHS is bigger than Big Data Big Data addresses only the blue side of the learning cycle The LHS infrastructure must support complete learning cycles 15
The LHS Must Do This Interpret Results Analyze Data Assemble Relevant Data A Problem of Interest Deliver Tailored Message Take Action to Change Practice Decision to Study 16
Not This Interpret Results Journals? Analyze Data Assemble Relevant Data A Problem of Interest Deliver Tailored Message Take Action to Change Practice Decision to Study 17
Precision Medicine: An LHS Use Case Analyze Data: What geno/pheno types and therapies generate better responses? Assemble Data: Patient genotypes, phenotypes, therapies and responses. Interpret Results: Are the results credible? What advice should be given? What is the optimal therapy for EACH PATIENT? Tailored Messages: For this patient, you might want to consider Take Action: Administer recommended Rx: In whole or part Decision to study 18
Summary Points Learning health systems can exist at any level of scale: local to ultra-large Learning occurs through complete learning cycles, each addressing a health problem. Learning cycles have an afferent (blue) side and an efferent (red) side. Learning = BD2K + K2P 19
Main Menu for Today 1. The why and what of the Learning Health System (LHS). The LHS transcends Big Data and Precision Medicine 2. The Learning Health System is infrastructure. PopMedNet is leading the way 3. Progress toward an LHS at local, state, and national levels. Seven channels of progress 20
How Do We Convert? A set of learning loops, each addressing a different problem: Problem 1 Problem 3 Problem 2 Into an efficient large scale system: 21
We Build an Infrastructure! Learning loops enable learning but do not create a Learning Health System If you want to get 350,000 people per day across a river, do you build 350,000 rowboats? No, you build a: 22
LHS Infrastructure A Single Socio-Technical Platform Supports Multiple Simultaneous Learning Cycles Different Problems Slower Cycle People Technology Policy Process Rapid Cycle SUPPORTING PLATFORM
Why a Platform? Without a platform: Every cycle requires its own agreements, technology, staffing, analytics, dissemination mechanisms No economy of scale Cost of 10 cycles = 10 x (Cost of one) With a platform: All cycles are supported by the infrastructure Big economy of scale Cost of 10 cycles << 10 x (Cost of one) 24
25 So the Scale of the System = The Scope of the Platform
So What s in a Complete Platform? Mechanisms for managing communities of interest 26
PopMedNet is a Key Platform Component Mechanisms for managing communities of interest 27
PopMedNet and LHS Infrastructure The Harvard Pilgrim team got the platform concept from the outset: PopMedNet was built to support multiple learning efforts PopMedNet supports federated networks PopMedNet addresses policy, process, and technology PopMedNet seeds other components 28
Main Menu for Today 1. The why and what of the Learning Health System (LHS). The LHS transcends Big Data and Precision Medicine 2. The Learning Health System is infrastructure. PopMedNet is leading the way 3. Progress toward an LHS at local, state, and national levels. Seven channels of progress 29
The LHS Can t be Framed Purely as a Technical Problem = 30
Elements of a National Learning Systems are Assembling 1. Words: calls and reports 2. Digital data 3. Learning Islands 4. Data federations and networks 5. Grant programs 6. A grassroots coalition of the willing 7. An emerging research community 31
Victor Dzau et al. Transforming Academic Health System for an Uncertain Future, Sept 2013. In the Big Data era, AHCs should strive to become learning health systems by making clinical data research grade and lowering the costs of data acquisition and knowledge generation. 10-Year Agenda: The Learning Health System 32
Progress: Health Data are Increasingly in Digital Form 33
Progress: Learning Islands Organizations that have become Learning Health Systems at their level of scale. But don t routinely connect with other islands. 34
Progress: Inter-organizational Data Federations and Networks 35
Progress: Grant Programs NIH Big Data to Knowledge PCORI Research Networks NSF Smart and Connected Health 36
Progress: Grassroots Movement National Summit convened in May 2012 to envision LHS as set of shared beliefs A Dumbarton Oaks conference for the LHS ~ 70 organizations represented at the National Press Club Resulted in 10 consensus Core Values 73 organizations have formally endorsed Giving rise to a Learning Health Community 37
78 Endorsements of the LHS Core Values* (As of 5/19/2015) Veterans Health Administration Office of Informatics & Analytics Program in Health Informatics, SONHP The Center for Learning Health Care SecureHealthHub, LLC Siemens Health Services GE Healthcare IT Department of Primary Care and Public Health *To be included on the www.learninghealth.org website.
Progress: A Research Community A national workshop to explore the research challenges inherent in realizing a high functioning LHS April, 2013 in Washington, DC 55 attendees: 45 invited participants plus federal liaisons Many disciplines represented: computer science to epidemiology to economics 39
Outcomes of the 2013 Workshop Reflected in report and JAMIA article: A vision of a high-functioning system 106 research questions organized into 4 broad categories and 19 sub-categories A sense that a scientific community of interest might form around the LHS A science of cyber-social ecosystems? 40
In Closing 1. The why and what of the Learning Health System (LHS). The LHS transcends Big Data and Precision Medicine 2. The Learning Health System is infrastructure. PopMedNet is leading the way 3. Progress toward an LHS at local, state, and national levels. Seven channels of progress 41
Thanks & Write to Me cpfried@umich.edu lhs.medicine.umich.edu 42