Abstract: Nosocomial (i.e., hospital-acquired) infections
(NI) is a major cause of morbidity and mortality in hospitals. NI
rate is higher in intensive care units (ICU) than in the general
ward due to patients with severe symptoms, poor immunity,
and accepted many invasive therapies. Contact behaviors
between health caregivers and patients is one of the infect
factors. It is difficult to obtain complete contact records by
traditional method of retrospective analysis of medical records.
This paper establishes a contact history inferential model
(CHIM) intended to extend the use of Proximity Sensing of
rapid frequency identification (RFID) technology to
transferring all proximity events between health caregivers and
patients into clinical events (close-in events, contact events and
invasive events).The results of the study indicated that the
CHIM can infer proximity care activities into close-in events
and contact events.
The infection control team could redesign and build optimal
workflow in the ICU according to the patient-specific contact
history which provided by our automatic tracing system.
Abstract: Researchers of drug-drug interaction alert systems
have often suggested that there were high overridden rate for alerts and
also too false alerts. However, research about decreasing false alerts is
scant. Therefore, the aim of this article attempts to proactive
identification of false alert for drug-drug interaction and provide
solution to decrease false alerts. This research involved retrospective
analysis prescribing database and calculated false alert rate by using
MYSQL and JAVA. Results of this study showed 17% of false alerts
and the false alert rate in the hospitals (37%) was more than in the
clinics. To conclude, this study described the importance that
drug-drug interaction alert system should not only detect drug name
but also detect frequency or route, as well as in providing solution to
decrease false alerts.