Impact of patient monitoring on the diurnal pattern of medical emergency team activation

Objective. To study the impact of time of day, day of week and level of patient monitoring on medical emergency team (MET) activation. Design: Retrospective observational study of all MET and cardiac arrest events between October 2001 and March 2005. Setting. University of Pittsburgh Medical Center Presbyterian Hospital, a tertiary care teaching facility in the United States. I nterventions. None. Measurements and Main Results. Cardiac arrest and MET event rate during the day Q am to 6:59 pm) and night (7 pm to 6:59 am) overall; for weekdays and weekends; and from unmonitored, monitored, and intensive care units (ICUs). There were 605 cardiac arrest and 4,072 MET events. MET event rate was higher during the day than at night in unmonitored units (62% day vs. 38% night; p <.001) and monitored units (59% day vs. 41 % night; p <.001) but not in ICUs (47% day vs. 53% night; p =.20). Unmonitored units had a greater daytime increase in MET event rate than monitored units (63% vs. 46%), whereas ICUs showed an 11% decline compared with night. The MET day vs. night difference was greater on weekdays (65% day vs. 35% night; p <.001) than on weekends (56% day vs. 44% night; p <.001). Cardiac arrest event rate showed no diurnal pattern in any unit setting but had a higher daytime event rate during weekdays (57% day vs. 43% night; p =.004). Conclusions., More MET events take place during the day. MET events in unmonitored units have a greater diurnal variability than those from monitored units. ICUs show no diurnal variation in MET event rate. Our results suggest a significant variability in the hospital ability to consistently detect patients who meet MET activation criteria.