Early discharge of patients with presumed opioid overdose: development of a clinical prediction rule

Abstract

Objective: To develop a clinical prediction rule to identify patients who can be safely discharged one hour after the administration of naloxone for presumed opioid overdose.

Methods: Patients who received naloxone for known or presumed opioid overdose were formally evaluated one hour later for multiple potential predictor variables. Patients were classified into two groups: those with adverse events within 24 hours and those without. Using classification and regression tree methodology, a decision rule was developed to predict safe discharge.

Results: Clinical findings from 573 patients allowed us to develop a clinical prediction rule with a sensitivity of 99% (95% CI = 96% to 100%) and a specificity of 40% (95% CI = 36% to 45%). Patients with presumed opioid overdose can be safely discharged one hour after naloxone administration if they: 1) can mobilize as usual; 2) have oxygen saturation on room air of >92%; 3) have a respiratory rate >10 breaths/min and <20 breaths/min; 4) have a temperature of >35.0 degrees C and <37.5 degrees C; 5) have a heart rate >50 beats/min and <100 beats/min; and 6) have a Glasgow Coma Scale score of 15.

Conclusions: This prediction rule for safe early discharge of patients with presumed opioid overdose performs well in this derivation set but requires validation followed by confirmation of safe implementation.