Rapid‐Response Teams and Internal Medicine Training: Original Descriptive Summary of Rapid‐Response Team Implementation in a Community‐Based Residency Program

Background:

Rapid‐response teams (RRTs) are associated with decreases in unexpected inpatient death, code blue, and unexpected ICU transfer rates. There are many RRT designs, but little has been written about RRTs coordinated by internal medicine residents. As a result of the rapid growth of RRTs in the United States, academic institutions are struggling with incorporating residents into RRTs. After a recent cluster randomized trial failed to demonstrate a benefit, teaching programs have been critical of the use of RRTs in teaching institutions.

Purpose:

We implemented a rapid‐response team in our community‐based residency program and describe our experiences during the first 1015 interventions.

Description:

PDSA cycles began in April 2005 with the creation of activation criteria and process measure forms. Pilot studies began in September 2005 and continued until April 2006. Process improvements and resource allocation were addressed during the pilot studies. Community hospitalists joined our initiative in April 2006 in order to expand our scope to all general medical wards. Resident education was coordinated by a hospitalist educator. Mean arrival time was less than 5 minutes. Forty percent of the interventions were more than 45 minutes in duration. Of the interventions, 74.8% were coordinated by residents, and 6.7% were teaching panel patients. Reasons for activation included staff worried (24%), hypoxia (24%), tachypnea (22%), hypotension (18%), tachycardia (16.8%), mental status change (16.5%), and chest pain (9.7%). Interventions included aerosol treatment (26.9%), fluid boluses (25.3%), noninvasive ventilation (12.3%), intubation (8.4%), cardiopulmonary resuscitation (0.9%), and cardioversion (0.2%). Thirteen patients progressed to cardiac arrest, 9 of whom survived to be transferred (69% return of spontaneous circulation). Of those in the 2 categories used as surrogate markers of patients destined for transfer—“rapid deterioration” and “too much care for floor setting”—47% and 46.5%, respectively, were stabilized without ICU transfer. A minimum of 210 ICU transfers were averted. Fourteen percent of RRT interventions occurred less than 12 hours after emergency department triage, 49% of which required too much care for the floor setting. Approximately 85% of patients who received RRT interventions survived to hospital discharge.

Conclusions:

Our resident‐directed model, in concert with a hospitalist educator and community hospitalists, appears to be affecting the number of unexpected ICU Transfers. Of patients progressing to cardiac arrest, return of spontaneous circulation was 25% higher than that noted in the NRCPR database (69% vs. 44%). The high frequency of respiratory compromise and artificial ventilation suggests residents and hospitalists should be trained in advanced airway management. Involvement was time intensive. A substantial percentage of the interventions occurred on recent arrivals from the emergency department, suggesting opportunities for improving ward response time, ED continuity, or bed selection.

Author Disclosure:

B. G. Lee, None; F. Patel, None; S. Alkatib, None; G. Vallabhaneni, None; V.Abraham, None; H. Hardin, None; W. Harb, None; M. Lupu, None; P. Haydon, None; M. Donofrio, None.