RESUMO
OBJECTIVE: To determine whether a novel, service-centric, medical emergency team (MET) model can impact cardiac arrest (CA) rates. METHODS: A retrospective pre- vs. postintervention analysis was performed on patients ≥ 18 years who had a CA between 2007 and 2012. A service centric MET model was initially implemented on the inpatient cardiology service and expanded hospital wide during 2008-2009, maturing to 10 teams in 2010. Service centric is defined as a medical or surgical service-specific MET team based on the location of the patient. RESULTS: The rate of CA (per 1,000 hospital days) in the year 2007 prior to the initiation of MET was compared to rates during program maturation (2008/2009) and after full maturation to 10 teams (2010-2012). A total of 1,140,233 hospital-care days were analyzed between 2007 and 2012, with 745 CAs recorded (0.65 events per 1,000 hospital days). The overall CA rate was higher prior to MET initiation (0.84 in 2007) compared to postinitiation (0.59 in 2008/09) and maturation to 10 teams (0.64 in 2010-12) (P < 0.003 for both pre- vs postcomparisons). No differences in CA rates were detected between either post-MET initiation time frames (P = 0.342). Similar trends in CA rates were observed in the intensive care unit (ICU) (3.96 vs 2.14 vs 2.68 per 1,000 hospital days in 2007, 2008/2009, and 2010-2012 respectively, with P < 0.001 for both pre- vs postcomparisons). CONCLUSIONS: A service-centric MET program was associated with a reduction in the rate of CAs both hospital wide and in the ICU. These observations maybe explained by the earlier intervention in care of unstable patients by an expanded group of caregivers.
