Objective: To study the rate of documentation of vital signs in the period before the occurrence of an adverse event or emergency team call and to measure the effect of introducing the medical emergency team (MET) system on the rate of such documentation.
Methods: During a cluster, randomised trial of the MET in 23 Australian hospitals, we collected the data on lowest systolic blood pressure, highest and lowest respiratory rate and heart rate from 15min to 24h before an adverse event (cardiac arrest, death or unexpected intensive care unit admission) or emergency team call. We derived the document of these vital signs (yes/no) from the numerical values recorded. We used analytically weighted and random-effect regression models to examine the association between non-documented (missing) vital signs, hospital characteristics and MET allocation, and to examine their trend over time.
Results: We found marked variability in documentation, with a high proportion of missing vital signs in some hospitals. Close to 77% of patients suffering adverse events had at least one vital sign missing immediately before the event. Allocation to a MET system was associated with significantly increased documentation of respiratory rate and blood pressure before emergency team review (P<0.01) as well as an improvement in documentation over time (P<0.01). At all stages and for both MET and control hospitals, the respiratory rate was the least commonly documented vital sign (P<0.01).
Conclusions: The documentation of vital signs in the period before adverse events was commonly incomplete with a particular deficiency in the documentation of the respiratory rate. Introduction of a MET system was associated with improvement in the rate of documentation of vital signs.