The concept of a learning health system (LHS) is that every patient encounter is documented and studied, and the resulting insights employed to drive change. Last year, the University of Michigan (U-M) announced it would offer two new advanced degrees in Health Infrastructures and Learning Systems (HILS) to develop researchers and practitioners who can conceptualize and deliver innovative solutions. Recently, U-M announced two grant awards to innovative learning health system projects. Meanwhile, researchers at Duke University have published a report on their efforts to develop curriculum for an LHS training program for resident physicians.
Published in the online journal eGEMs (Generating Evidence & Methods to improve patient outcomes), the Duke researchers’ work highlighted both the opportunities and challenges in “teaching toward an ideal system that does not yet exist.”
The Duke program was created in 2013 with the help of grants from the Association of American Medical Colleges and Duke University Health System. It leverages the Duke Enterprise Data Unified Content Explorer (DEDUCE), a web-based query tool for the Duke clinical data warehouse that enables access to patient-level health data spanning three decades. Duke also had recently adopted and begun to optimize Epic, which researchers say presented both a motivation to invest in the training program and a use-case opportunity.
The first cohort has six internal medicine trainees. The second cohort was larger, with eight trainees from diverse specialty backgrounds, including surgery and neonatology.
The research team found that one particularly valuable exercise was to have trainees implement statistical concepts by constructing a small (five-patient) “mini-database” around a clinical question of their choice. “While the goal of the LHS is to operate at scale, clinical trainees are often unfamiliar with the characteristics of the available structured clinical data,” they wrote. “Having them begin with a small sample is intended to reinforce several skills: framing an appropriate clinical question, designing a data set that adequately addresses that question, developing a search strategy, getting and cleaning data, and reporting results using data visualization software. Once familiar with the relevant data, trainees are able to more effectively communicate their needs to data managers and statisticians for the final project.”
One such project, they note, has segued into a system-wide, health-system improvement effort “with a focus on improving EHR data quality and accuracy of nursing assessments, which will allow rapid quality reporting and eventually accumulate a body of evidence to support comparative effectiveness.”
The researchers also report that they have learned more about the informatics resources and data structures at Duke, including the strengths and limitations of existing data. “This has translated to trainee projects centered on improving existing data. It has also prompted us to identify colleagues who can grant us access to the best sources of data. We have become increasingly visible within the institution, and in this process are establishing stronger partnerships with others engaged in quality improvement work and clinical innovations where our trainees can see the impact of their work.”
The Duke researchers are exploring ways to disseminate their curriculum to other institutions and expand it to nurses and pharmacists, and measure its impact on trainees.
Learning health system projects awarded grant funding
Meanwhile, the University of Michigan Department of Learning Health Sciences and the university’s Office of Research named the first two awardees in a pilot grant program designed to accelerate development of a learning health system.
The first, “Scaling a Regional Learning Health System,” builds on the Collaborative Quality Improvement Learning Health System model established in Michigan. This team, including members of the Michigan Surgical Quality Collaborative, proposed scaling the regional system to a broader national platform. Under the interdisciplinary leadership of the Institute for Healthcare Policy and Innovation (IHPI), the Collaborative Quality Improvement Consolidation Center (CQIC2) would have the opportunity to emerge as a leader in accelerating a clinically led, practice-based, catalyst for scaling regional quality improvements on a national scale.
The second grant winner involves the “Integration of a Mobile Application for Heart Failure into the Learning Health System.” Clinicians and engineers at U-M are designing technologies that assist in monitoring heart failure status, increase adherence to evidence-based medications, promote health behavior modifications, and provide health information back to patients within a learning health system framework. They will use this award to supplement existing funding by addressing the following:
• Integrating motivational messages for heart failure patients that will evoke a behavior health change into a mobile application based on the Michigan Tailoring System;
• Integrating remote monitoring sensors data into the mobile application to enhance the use of the data within an algorithm;
• Creating an algorithm, using machine learning, that predicts clinical worsening of heart failure using remote monitoring;
• Developing knowledge objects to hold, manage, and facilitate deployment of the clinical worsening of heart failure predictive algorithm using the Knowledge Grid platform under development in the Department of Learning Health Sciences.