Abstract

Objectives: To evaluate the performance of a simple screening tool for chest radiography for identification of community-acquired pneumonia (CAP) among emergency department (ED) patients who present with respiratory-related complaints. Our screening tool is a modification of a previously published guideline, which relied only on the presence of vital-sign abnormality (97% sensitivity, 19% specificity). We included respiratory symptoms to improve the specificity, defining our screening tool as the presence of any one respiratory symptom (cough, chest pain or shortness of breath) and any abnormality of the vital signs (temperature >38 °C, heart rate >100 beats/min, respiration rate >20 breaths/min, or pulse oximetry <95%).

Methods: This was a 3-month retrospective chart review of all ED visits from an inner city teaching hospital. CAP was defined as the presence of a new radiographic infiltrate compatible with CAP. Patients with asthma were excluded.

Results: Of 8811 patient visits evaluated, 1948 presented with a respiratory symptom. Of these, 198 had definitive CAP. Sensitivity, specificity, positive and negative predictive values of the ED screening tool were 90% (95% CI 85% to 94%), 76% (95% CI 74% to 78%), 30% and 99%, respectively. Positive and negative likelihood ratios were 3.72 (95% CI 3.38 to 4.09) and 0.13 (95% CI 0.08 to 0.19), respectively.

Conclusions: A simple screening tool with high sensitivity and specificity was used in an urban inner city ED to decide on the requirement for chest radiographs for patients with respiratory symptoms for identification of CAP. Validation studies are required to determine the utility of this screening tool for improving time to diagnosis and treatment.