Assessment of Undifferentiated Chest Pain in the Emergency Department (ED); an Observational Study of ED Nurse and Doctor ACS Risk Assessment and the Impact on Patient Outcomes

Background: Chest pain constitutes a large proportion of Australian emergency department (ED) presentations. Accurate risk assessment to rule out an acute coronary syndrome (ACS) is required. The aim was to measure accuracy and characteristics of clinicians assessing patients for suspected ACS and associations with outcome. Methods: A sub-analysis on 1,857 patient risk assessments was conducted from a randomised trial, comparing high-sensitivity troponin versus conventional troponin. Each participating nurse and doctor estimated the likelihood of an ACS diagnosis on their patient and recorded their professional characteristics. The primary outcome was nurse versus doctor impression of ACS in association with actual adjudicated ACS diagnosis. Concordance comparisons were conducted. Results: Six hundred and nine clinicians participated and 16% of patients were diagnosed with an ACS. There was no difference in the accuracy of ED nurses compared to doctors (C-statistic: 0.67 vs. 0.68 respectively; p = 0.35). There was minor discordance between nurse and doctor ACS-risk assessment. Nurse specialist qualification was associated with significantly higher accuracy than nurses without (65.4% vs. 55.3% respectively; p < 0.001). Consultant doctors were most accurate 138/186 (74.2%) vs. registrars 248/441 (56%) vs. residents 464/769 (60.3%) p < 0.001. Patients assessed as unlikely ACS versus definite/likely had a shorter median length of stay (4.8 h [iqr.3.2,7.9] v 5.8 h [iqr.4.1,8.0]). Conclusion: Nurses are similar to doctors in their clinical assessments, however both groups are suboptimal. Standard clinical pathways may be required to assist ED doctors and nurses to improve the assessment of suspected ACS in patients presenting with undifferentiated chest pain department (ED) presentations. Accurate risk assessment to rule out an acute coronary syndrome (ACS).