As many as 18 million errors and 360,000 adverse events occur annually in U.S. emergency departments. Even more risks are incurred by inappropriate overuse of diagnostic tests that may also contribute to unnecessary treatment delays, radiation exposure, and costs. In this busy and often chaotic environment, repeated deliberate decisions increase risks of errors and bias. This phenomenon, known as decision fatigue, may exacerbate risks for socio-economic disparities when clinicians, after exerting self-control through the course of their shift, find themselves treating persons of a minority race or background. A large number of preventable cognitive and diagnostic errors may be improved with scalable interventions that leverage decision psychology. This study builds on our prior multi-site studies of electronic medical record-based interventions and recent workflow and data analysis in over 23 Emergency Departments, and 1,154 clinicians and 3,047,113 visits showing evidence of decision failures and intervention opportunities. We are partnering with the largest electronic medical vendor in the world, an emergency medicine business intelligence firm, and experts and professionals in quality improvement, emergency medicine, and health and behavioral economics. The products of this work will include not only better academic understanding of decision failures, but reusable and scalable systems to improve care. In the R21 phase, we will deliver a comprehensive analysis of two large databases that include our intervention sites: sites showing performance improvement opportunities in emergency medicine, evidence of disparities, and evidence of decision fatigue. The byproducts of these results will include prioritized quality and safety goals, parameters for power analysis, and customized reports that will be used to engage stakeholders in planning. We will use these reports in a data-driven, systematic, multi-stakeholder consensus process. Clinical, Scientific, and Information Technology experts will review proposed interventions and optimize requirements for scientific impact, generalizability, and feasibility. Interventions will leverage electronic medical records and stakeholders will analyze other existing scalable channels for delivering or potentiating interventions, including continuing medical education, highly subscribed media venues targeting emergency physicians, and existing local channels for performance reporting and monitoring. A pilot site will implement the intervention to demonstrate feasibility. In this way, we expect to be prepared for a large scale trial. If the R21 phase is successful, we will have a high likelihood of achieving goals in the R33 phase, in which the final intervention will be tested in a multi-site, two-arm pragmatic randomized trial. We will assess intervention adherence, guideline adherence, and clinical outcomes, including guideline discordant radiation exposure, preventable readmissions, and mortality. The products of this work will include functional requirements and technical specifications, including locally defined parameters that can be reused and customized in other settings.

Public Health Relevance

By some estimates, 18 million errors and 360,000 adverse events occur annually in U.S. emergency departments. Most of these are preventable cognitive and diagnostic errors that may be influenced by behavioral interventions. We propose to apply decision science in scalable approaches to prevent cognitive errors and inappropriate overuse.

Agency
National Institute of Health (NIH)
Institute
National Institute on Aging (NIA)
Type
Exploratory/Developmental Grants (R21)
Project #
5R21AG057400-02
Application #
9566110
Study Section
Special Emphasis Panel (ZAG1)
Program Officer
Bhattacharyya, Partha
Project Start
2017-09-15
Project End
2019-08-31
Budget Start
2018-09-01
Budget End
2019-08-31
Support Year
2
Fiscal Year
2018
Total Cost
Indirect Cost
Name
University of Southern California
Department
Public Health & Prev Medicine
Type
Schools of Medicine
DUNS #
072933393
City
Los Angeles
State
CA
Country
United States
Zip Code
90089