A scalable service to improve health care quality through precision audit and feedback

点击翻译按钮获取中文摘要

All health care delivery organizations measure care quality and outcomes, increasingly via electronic clinical quality measures 1 and dashboards 2,3. However, these organizations lack evidence-based strategies for putting their quality and outcome data to work to improve performance 4,5. The most common approach is audit and feedback (A&F), the delivery of clinical performance summaries to providers, which demonstrates potential for large effects on clinical practice 6–8. But A&F too often produces negligible effects 5,9, creating little more than distraction for providers who are fatigued by information chaos 9–11. As currently implemented, A&F is a blunt, “one size fits most” intervention. Each provider in a care setting typically receives identical metrics in a common format, despite a growing recognition that “precisionizing” interventions holds significant promise to improve their impact 12–15. A precision approach to A&F would prioritize display of information in the single metric that, for each recipient, carries the highest value for improving performance, such as when the metric's level drops below a peer benchmark or minimum standard for the first time, revealing an actionable performance gap 16–19. Furthermore, precision A&F would employ an optimal message format (including framing and visual displays 20–24), based on what is known about the recipient and the intended gist meaning being communicated, to improve message interpretation while reducing cognitive processing burden 25–28. Well- established psychological principles, frameworks, and theoretical mechanisms provide a knowledge base to achieve precision A&F 16–19,29–33. From an informatics perspective, precision A&F requires a knowledge-based system that uses psychological theory at its core, but which enables mass customization by giving precedence to configurable knowledge about recipients at the group and individual levels. A precision A&F service employs this knowledge as requirements (necessary characteristics for message acceptability) and preferences (the relative importance of message characteristics) to generate messages that are more likely than a “one size fits most” report to positively influence clinical decision-making and practice. An equally important informatics challenge is to enable widespread improvement through a service for precision A&F at scale. A scalable precision A&F service must function as infrastructure compatible with a wide range of computing environments and supporting a wide range of clinical domains. In his previous NLM K-award, the principal investigator developed and tested a prototype knowledge-based system for precision A&F in email messages in anesthesia care. Preliminary data show that provider preferences are not uniform, suggesting that a platform for computable knowledge is necessary to support scalable precision A&F. The Knowledge Grid platform, developed at the University of Michigan, has been shown to support “precisionizing” for clinical decision support systems 34–36. Based on our prior work, the proposed project will advance the creation of more general services for precision A&F, applying the service in anesthesia care as a demonstration domain.