Objectives: We examined whether a previously validated verbal dyspnoea rating scale, and/or other demographic and clinical parameters, could predict ED departure status, among ED patients presenting with shortness of breath.
Methods: In this prospective observational study, a convenience sample of patients presenting to an inner urban adult tertiary hospital ED with shortness of breath were assessed at triage using objective and subjective breathlessness parameters. These included respiratory rate, oxygen saturation, heart rate, systolic blood pressure and verbal dyspnoea scores. A verbal dyspnoea score for worst dyspnoea during the current episode and basic demographic and presentation characteristics were also collected. These variables were assessed as predictors of ED departure status (inpatient admission or ED discharge) using logistic regression.
Results: From a sample of 253 participants, verbal dyspnoea scores > or =8 predicted inpatient admission 89% specificity (95% confidence interval [CI] 82.1-93.4), and scores < or =3 predicted discharge with 95% specificity (95% CI 89.5-98.0). For patients with shortness of breath as the primary complaint, the combination of verbal dyspnoea score > or =6, heart rate > or =94 bpm at triage and ambulance arrival predicted admission with 90% sensitivity (95% CI 82-95%) and 84% specificity (95% CI 73-92%). These same variables predicted admission for all patients with 84% sensitivity (95% CI 75.8-89.2) and 79% specificity (95% CI 71.5-85.5).
Conclusion: Verbal dyspnoea score, alone and in combination with heart rate and arrival transport, can accurately predict admission. Once validated they might be useful in assessing, prioritizing and making rapid site of care decisions for breathless patients presenting to the ED.