Single-parameter early warning criteria to predict life-threatening adverse events.

OBJECTIVES: Early warning criteria ("criteria"), used to activate rapid response system (RRS) teams, may improve patient outcomes by predicting life-threatening adverse events--urgent intensive care unit (ICU) transfers and cardiac arrests.

METHODS: We conducted a case-control study on medicine patients in an academic medical center from May 2005 to June 2006. Controls were matched to RRS activation patients by admission date and unit. Rapid response system activation patients were then excluded from analysis. Chart reviews identified positive criteria on non-RRS patients. For controls, physiologic data were collected from admission until the day of matched RRS activations. Data were collected in patients with adverse events for the 8 hours preceding events.

RESULTS: A total of 262 patients (2.8%) had 271 adverse events including 245 ICU transfers and 26 arrests. Positive criteria were found during 21.1% of control admissions (68/323). Positive criteria preceded 60.8% of the ICU transfers (158/260; 95% confidence interval, 54.8%-66.7%). However, 76.9% of patients with an arrest did not have previous positive criteria (20/26; 95% confidence interval, 60.7%-93.1%). Intensive care unit transfers with positive criteria were more likely to die than patients without criteria (35.4% versus 20.6%; odds ratio [OR], 2.1). Positive criteria most strongly associated with life-threatening adverse events were tachypnea (OR, 31.1) and 100% supplemental oxygen (OR, 13.7).

CONCLUSIONS: The early warning conditions used to activate RRS teams were only fair predictors of acute deterioration, although early signs of respiratory failure during routine monitoring were strongly associated with future life-threatening adverse events. Improved respiratory monitoring may improve outcomes from RRS interventions.