Specificity Improvement for Network Distributed Physiologic Alarms Based on a Simple Deterministic Reactive Intelligent Agent in the Critical Care Environment

Department of Anesthesiology and Critical Care, The University of Michigan Health Systems, 4172 Cardiovascular Center/SPC 5861, 1500 East Medical Center Drive, Ann Arbor, MI 48109-5861, USA.
International Journal of Clinical Monitoring and Computing (Impact Factor: 1.99). 02/2009; 23(1):21-30. DOI: 10.1007/s10877-008-9159-3
Source: PubMed


Automated physiologic alarms are available in most commercial physiologic monitors. However, due to the variability of data coming from the physiologic sensors describing the state of patients, false positive alarms frequently occur. Each alarm requires review and documentation, which consumes clinicians' time, may reduce patient safety through 'alert fatigue' and makes automated physician paging infeasible. To address these issues a computerized architecture based on simple reactive intelligent agent technology has been developed and implemented in a live critical care unit to facilitate the investigation of deterministic algorithms for the improvement of the sensitivity and specificity of physiologic alarms. The initial proposed algorithm uses a combination of median filters and production rules to make decisions about what alarms to generate. The alarms are used to classify the state of patients and alerts can be easily viewed and distributed using standard network, SQL database and Internet technologies. To evaluate the proposed algorithm, a 28 day study was conducted in the University of Michigan Medical Center's 14 bed Cardiothoracic Intensive Care Unit. Alarms generated by patient monitors, the intelligent agent and alerts documented on patient flow sheets were compared. Significant improvements in the specificity of the physiologic alarms based on systolic and mean blood pressure was found on average to be 99% and 88% respectively. Even through significant improvements were noted based on this algorithm much work still needs to be done to ensure the sensitivity of alarms and methods to handle spurious sensor data due to patient or sensor movement and other influences.

1 Follower
17 Reads

  • Critical care medicine 02/2010; 38(2):702-3. DOI:10.1097/CCM.0b013e3181bfe97f · 6.31 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The ongoing development of new sensors and parameters for intraoperative monitoring has outpaced the development of display design, leading to a gap between the load of information and the quality of its delivery. This is not a circumstantial problem, as a large portion of critical incidents is attributable to inadequate situation awareness and the failure to recognize readily monitored data. This review also addresses improvements of current threshold alarms. Research has focused on advanced integrated displays, drawing on the findings of human factor science and on the exploitation of alternative sensory pathways. Integrated displays, as well as auditory, vibrotactile and head-mounted displays have been shown to promote situation awareness and reduce cognitive workload. Intelligent alarm design can successfully reduce the number of false alarms. Improvement of the display of information in the operating room is warranted, and recent developments are promising. However, their introduction into mass market is not yet on the horizon, although the shortcomings of the traditional single-sensor-single-indicator principle are known for a long time. If manufacturers are reluctant to implement new techniques into their devices, they should at least facilitate access to monitoring raw data in order to allow independent development of displays.
    Current opinion in anaesthesiology 10/2010; 23(6):772-7. DOI:10.1097/ACO.0b013e32834045fd · 1.98 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To report an exploratory action-research process used during the implementation of continuous patient monitoring in acute post-surgical nursing units. Substantial US Federal funding has been committed to implementing new health care technology, but failure to manage implementation processes may limit successful adoption and the realisation of proposed benefits. Effective approaches for managing barriers to new technology implementation are needed. Continuous patient monitoring was implemented in three of 13 medical/surgical units. An exploratory action-feedback approach, using time-series nurse surveys, was used to identify barriers and develop and evaluate responses. Post-hoc interviews and document analysis were used to describe the change implementation process. Significant differences were identified in night- and dayshift nurses' perceptions of technology benefits. Research nurses' facilitated the change process by evolving 'clinical nurse implementation specialist' expertise. Health information technology (HIT)-related patient outcomes are mediated through nurses' acting on new information but HIT designed for critical care may not transfer to acute care settings. Exploratory action-feedback approaches can assist nurse managers in assessing and mitigating the real-world effects of HIT implementations. It is strongly recommended that nurse managers identify stakeholders and develop comprehensive plans for monitoring the effects of HIT in their units.
    Journal of Nursing Management 10/2011; 19(7):863-75. DOI:10.1111/j.1365-2834.2011.01295.x · 1.50 Impact Factor
Show more