Abstract
Inaccurate, untimely, and miscommunicated medical diagnoses represent a wicked problem requiring comprehensive and coordinated approaches, such as those demonstrated in the characteristics of learning health systems (LHSs). To appreciate a vision for how LHS methods can optimize processes and outcomes in medical diagnosis (diagnostic excellence), we interviewed 32 individuals with relevant expertise: 18 who have studied diagnostic processes using traditional behavioral science and health services research methods, six focused on machine learning (ML) and artificial intelligence (AI) approaches, and eight multidisciplinary researchers experienced in advocating for and incorporating LHS methods, ie, scalable continuous learning in health care. We report on barriers and facilitators, identified by these subjects, to applying their methods toward optimizing medical diagnosis. We then employ their insights to envision the emergence of a learning ecosystem that leverages the tools of each of the three research groups to advance diagnostic excellence. We found that these communities represent a natural fit forward, in which together, they can better measure diagnostic processes and close the loop of putting insights into practice. Members of the three academic communities will need to network and bring in additional stakeholders before they can design and implement the necessary infrastructure that would support ongoing learning of diagnostic processes at an economy of scale and scope.
Highlights
In its breakthrough 2015 report, Improving Diagnosis in Health Care, the National Academy of Medicine (NAM) argued for new approaches for health care organizations to “identify, learn from, and reduce diagnostic errors and near misses in clinical practice.”[1]
The potential of applying Learning Health System (LHS) methods to address the wicked problem of diagnostic error motivated program officials of the Gordon and Betty Moore Foundation to suggest an exploratory study of how this connection might be achieved, and how the characteristics of a functioning learning health system[11] can support the cultural and technical changes required to pursue diagnostic excellence
We proposed that the LHS methods employed by this community could provide a framework for the intersection of the IDx and machine learning (ML)/artificial intelligence (AI) communities
Summary
In its breakthrough 2015 report, Improving Diagnosis in Health Care, the National Academy of Medicine (NAM) argued for new approaches for health care organizations to “identify, learn from, and reduce diagnostic errors and near misses in clinical practice.”[1]. In particular—through the mandate to learn from every patient and their health experiences[12,13] and through the infrastructure-supported ability to rapidly deliver new knowledge into practice12—there are sound reasons to believe that LHS approaches hold the potential to catalyze efforts toward accurate and timely diagnoses
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
