An assessment of the triage performance of the efferent arm of the rapid response system

Objectives: To measure the triage performance of the efferent arm of a rapid response system (RRS) by assessing the 24 h outcome of patients triaged to remain on the ward after rapid response team (RRT) review. Methods: We performed a retrospective observational study of all consecutive RRS activations between August 2005 and December 2011 in a university-affiliated hospital. Calls involving patients with documented limitations of medical therapy (LOMT) orders were excluded. We determined patients who were triaged to stay on the ward at the end of their first (index) call and analyzed their vital status and location 24 h later. Finally, we reviewed medical charts of patients triaged to remain on the ward and had a cardiac arrest and/or died within 24 h of RRT review. Results: We studied 8304 RRT calls. We excluded 1794 calls involving patients with LOMT, 2165 that were repeat calls, 20 where data was missing, 650 where patients were immediately transferred to a high dependency (HDU) or an intensive care unit (ICU) and 92 where calls were rapidly upgraded to cardiac arrest calls. Thus, we identified 3583 index calls at the end of which patients were triaged to remain on the ward. Within 24 h, 454 (12.7%) of those had a repeat RRT activation and 378 were transferred to HDU/ICU. 12 (0.3%) suffered a cardiac arrest on the ward. Altogether, 14 (0.4%) patients died within 24 h of the index RRT activation. Of those 6 had LOMT applied after the call, 4 had been admitted to ICU in a further call and 6 (0.2%) patients had unexpected cardiac arrest on the ward. Conclusions: The rate of unexpected cardiac arrest in the 24 h following RRT activation is very low for patients triaged to stay on the ward. Major triage errors by the RRT appear uncommon. (C) 2012 Elsevier Ireland Ltd. All rights reserved.