Developing a Continuous Monitoring Infrastructure for Detection of Inpatient Deterioration

摘要

One of the fundamental problems underlying all initiatives to reduce the incidence of failure to rescue (FTR) on non-critical medical/surgical units is the detection of the deteriorating patient by clinical staff, who must address the needs of multiple patients in a fast-paced and task-rich environment. As we have described elsewhere, previous approaches to the study and prevention of FTR events have been limited by the sparse physiologic data collected in manual vital-signs checks. The application of continuous monitoring for all patients into this setting-termed surveillance monitoring to differentiate it from condition monitoring, which is employed on a selective basis-is a promising new approach. In addition to the benefits of immediate detection, the implementation of surveillance monitoring creates a new opportunity, through the use of automated systems of data collection and storage, to better understand the physiologic dynamics of inpatient deterioration. Collections of patient physiologic data have been shown to be useful in other areas of medicine. For example, there is a long history of gathering and distributing cardiac dacasets. A compilation of 48 two-channel electrocardiograph recordings was distributed starting in 1980 by researchers at the Beth Israel Hospital (Boston) arrhythmia laboratory. The European ST-T database consists of 90 records from 1985 through 1992 featuring hundreds of ST segment and T wave changes. Together with the American Heart Association database, collected between 1978 and 1980, these early archives enabled great steps forward in the development of arrhythmia detection systems. More recently, Jager et al. published a description of a database containing more than 80 electrocardiograph records that enabled them to explore a variety of ST segment changes. This archive has been used extensively in the study of ischemia.