A method for evaluating medication error recovery processes in the ICU

The medication management process is a common source of errors in the hospital Intensive Care Unit (ICU). Many efforts are underway to redesign systems to prevent these errors. One example is the implementation of health information technology (health IT) to improve information availability and provide decision support at the point of care. Mechanisms that lead to the detection and correction of errors after they occur are also important, since it is not possible to prevent all errors from occurring (Kontogiannis 1997; 1999). Although health IT will affect both planned and unplanned error recovery, error recovery mechanisms do not typically receive the attention that error prevention does when evaluating the impact of health IT on medication safety, nor have they been systematically evaluated in hospital ICUs. We describe a method for identifying and evaluating error recovery pathways in the medication management process in two adult ICUs before and after electronic health record (EHR) implementation. This study was part of a larger study designed to evaluate the impact of EHR implementation in ICUs on patient safety, quality of care and end-users ( https://www.w3.org/1999/xlink" xlink:href=" https://cqpi.engr.wisc.edu/cpoe_home "> https://cqpi.engr.wisc.edu/cpoe_home ). The goals of this study were (1) to identify whether medication management errors were recovered from and (2) to characterize the recovery pathways. We collected as much information about the recovery pathway as possible to better understand the impact of EHR implementation on medication errors and error recovery pathways. Specifically, we sought to understand how medication errors were detected and corrected, based on retrospective review of the patient record (paper or electronic), available related artifacts that may not be part of the patient record (e.g., communications between providers about errors), the hospital&s;s event reporting system, and information volunteered to the research team by clinicians. A mixed-methods evaluation was completed using data from 1) a job-task analysis of physicians and nurses, 2) ethnographic observations of nurses, physicians and pharmacists, and 3) interviews and team meetings with ICU nurses and managers, ICU pharmacists, and physicians. This information was supplemented by a literature review on theoretical and applied error recovery in the medication management process and in other industries (Kanse et al., 2006; Senholzi & Gottlieb, 2003), along with the application of data collection protocols for evaluating the impact of computerized provider order entry on medication errors (Bates, et al., 1998, 1999). From these sources we generated a flow chart of the medication management process and the known mechanisms of error recovery during medication management (the process of medication ordering to medication monitoring). Each recovery mechanism was assessed according to who was involved, the mechanism by which the error occurred, when error recovery occurred in relation to the occurrence of the error, and the stage of the medication-management process stage during which the error and recovery occurred. We identified 10 categories of persons involved in error recovery and 11 mechanisms of error recovery (error detection and/or correction) which could be identified by available sources. Challenges in the collection of data regarding error recovery and in the identification of error-recovery methods are discussed. Estimates of the reliability of our methods for identifying error-recovery methods are presented.