BACKGROUND: Investment in health care information technology is resulting in a large amount of data electronically captured during patient care. These databases offer the opportunity to implement ongoing monitoring and analysis of processes with important patient care quality and safety implications to an extent that was previously not feasible with paper-based records. Thus, there is a growing need for analytic frameworks to efficiently support both ongoing monitoring and as-needed periodic detailed analyses to explore particular issues. One patient care process-the use of verbal orders-is used as a case in point to present a framework for analyzing data pulled from electronic health record (EHR) and computerized provider order entry systems. METHODS: Longitudinal and cross-sectional data on verbal orders (VOs) were analyzed at University of Missouri Health Care, Columbia, an academic medical center composed of five specialty hospitals and other care settings. RESULTS: A variety of verbal order analyses were conducted, addressing longitudinal-order patterns, provider-specific patterns, order content and urgency, associated computer-generated alerts, and compliance with institutional policy of a provider cosignature within 48 hours. For example, at the individual prescriber level, in July 2011 there were 14 physicians with 50 or more VOs, with the top 2 having 253 and 233 individual VOs, respectively. CONCLUSIONS: Taking advantage of the automatic data-capture features associated with health information technologies now being incorporated into clinical work flows offers new opportunities to expand the ability to analyze care processes. Health care organizations can now study and statistically model, understand, and improve complex patient care processes.
BACKGROUND: Investment in health care information technology is resulting in a large amount of data electronically captured during patient care. These databases offer the opportunity to implement ongoing monitoring and analysis of processes with important patient care quality and safety implications to an extent that was previously not feasible with paper-based records. Thus, there is a growing need for analytic frameworks to efficiently support both ongoing monitoring and as-needed periodic detailed analyses to explore particular issues. One patient care process-the use of verbal orders-is used as a case in point to present a framework for analyzing data pulled from electronic health record (EHR) and computerized provider order entry systems. METHODS: Longitudinal and cross-sectional data on verbal orders (VOs) were analyzed at University of Missouri Health Care, Columbia, an academic medical center composed of five specialty hospitals and other care settings. RESULTS: A variety of verbal order analyses were conducted, addressing longitudinal-order patterns, provider-specific patterns, order content and urgency, associated computer-generated alerts, and compliance with institutional policy of a provider cosignature within 48 hours. For example, at the individual prescriber level, in July 2011 there were 14 physicians with 50 or more VOs, with the top 2 having 253 and 233 individual VOs, respectively. CONCLUSIONS: Taking advantage of the automatic data-capture features associated with health information technologies now being incorporated into clinical work flows offers new opportunities to expand the ability to analyze care processes. Health care organizations can now study and statistically model, understand, and improve complex patient care processes.
Authors: David A Hanauer; Gregory W Hruby; Daniel G Fort; Luke V Rasmussen; Eneida A Mendonça; Chunhua Weng Journal: AMIA Annu Symp Proc Date: 2014-11-14