James S. Todd Memorial Research Award:
Automated Maintenance of Problem-Drug Matches in the Electronic Medical Record to Promote Patient Safety
Chiang S. Jao, PhD, University of Illinois College of Medicine
The advent of electronic medical records (EMRs) with computerized physician order entry (CPOE) makes real-time clinical decision support systems (CDSSs) feasible. The purpose of this research is to test the hypothesis that a CDSS that automates the matching of ordered drugs to problems on the problem list can enhance the maintenance of both the drug list and the problem list. Enhanced maintenance of both the drug and problem lists can permit the exploitation of advanced decision support strategies that will yield higher patient safety. We have designed a decision support tool that automatically matches drugs ordered to problems on the problem list. We plan to debug this decision support tool on 100 preliminary cases and then assess the performance of this decision support tool on 140 additional cases. We will specifically compare the ability of this decision support tool to match drugs-to-problems with the performance of human domain experts who will perform detailed chart audits to match drugs-to-problems. The decision support tool will be designed as a standalone system in order to more rapidly assess its potential benefit. We consider its construction a proof of concept. If this standalone system is useful in making problem-drug matches, maintaining or encoding the problem list, or identifying extraneous or unnecessary drugs, it could be converted in the future into an embedded decision support system that operates transparently within the EMR-CPOE system.
Post-Surgical Patients Receiving Patient Controlled Analgesia (PCA)
Frank J. Overdyk, MD, Medical University of South Carolina
This study investigates the incidence and etiology of respiratory compromise in post-surgical patients receiving patient controlled analgesia (PCA). It seeks to quantify and stratify the risks of this therapy, and develop prognostic algorithms that may be incorporated into PCA pump alarms system. Respiratory compromise during PCA may be transient or progress to respiratory arrest unless diagnosed and treated in a timely manner. This study will use a new technology, PCA pumps with integrated, telemetric, oximetry and capnometry modtiles, to continuously measure respiratory parameters of post-surgical patients. The only clinical site worldwide that has this equipment currently implemented is the venue for this study. Although well described in the operating room and procedural suites, this monitoring combination has never been applied to patients receiving PCA. The respiratory data collected, synchronized with postoperative clinical events, will allow the incidence of respiratory compromise to be measured and characterize the respiratory and heart rate patterns that precede these events. Analysis of pilot study data suggests that heuristic pattern recognition algorithms may be developed that will predict impending respiratory compromise and aid in the determination of its causes. A multidisciplinary research team with expertise in pain management, pharmacology,
Developing a Knowledge Base for RN Stacking: A Critical Patient Safety Strategy for Nursing Care Delivery
Patricia Ebright, DNS, RN, Indiana University School of Nursing
Preliminary research by this research team identified a critical safety-related aspect of registered nurse (RN) work not previously addressed in the literature, nor taught in schools of nursing. Called stacking, it is a workload management strategy for adapting and coping with the demands of workplace variability and complexity. Increased understanding of stacking has the potential to affect the education and care delivery of this bolus of neophytes likely to be swelling the ranks of nursing over the next few years. A first step to reaching that goal is to increase our understanding of the phenomenon of stacking and to identify critical attributes of stacking management that lead to safe delivery of patient care. Our specific aims are to explore the following questions to develop a knowledge base about stacking.
- What activities are stacked by RNs in the context of real-life care situations?
- What factors, including goals and constraints, contribute to RN decisions surrounding what to stack and stacking strategies?
- What strategies do RNs use to manage the stack, particularly those that reduce the potential for and consequences from erroneous actions, unexpected situations and complicating factors?
We will collect data from a larger and more diverse RN sample than in previous studies across multiple healthcare settings. Multiple methods of data collection will be used to capture rich data about RN stacking experiences to begin building a useful knowledge base. Findings from this study will inform future curricular development in nursing education and healthcare facility RN orientation programs, both of which strive to provide basic preparation for the safe management and delivery of patient care. In addition, findings may be particularly important for informing future healthcare system and information technology designs that support RN work.