DO ALL HEART SCORES BEAT THE SAME: EVALUATING THE INTER-OPERATOR VARIABILITY OF A CHEST PAIN RISK STRATIFICATION TOOL IN A UK EMERGENCY DEPARTMENT. Issue 12 (18th November 2016)

• Record Type: Journal Article
• Authors: Niven, WG; Wilson, D; Green, S; Robertson, A; Goodacre, S
• Is Part Of: Emergency medicine journal. Volume 33:Issue 12(2016)
• Page Start: 913
• Page End: 913
• Publication Date: 2016-11-18
• Subjects: Trauma; Emergency medicine -- Periodicals; 616.02505
• DOI: 10.1136/emermed-2016-206402.28
• Languages: English
• ISSNs: 1472-0205

Abstract

Abstract : Objectives & Background: Patients presenting with chest pain represent a significant proportion of attendances to the emergency department (ED). The HEART score derived by Backus et al is a validated tool for the risk stratification of suspected ischaemic chest pain within the ED – image 1 below. Little is known about whether HEART scores can be reproduced with reliable agreement irrespective of the grade of nurse or doctor. The goal of this research was to establish the inter-operator reliability of the HEART score as performed in the ED by different grades of doctor and nurse. Methods: This was a prospective, cross-sectional study that was carried out at the Homerton Hospital – a district general hospital (DGH) with approximately 120 000 A+E attendances per year. Consented patients were interviewed separately by 4 clinicians of different grades. These were Senior doctor (ST4 and above), Junior doctor (FY2–ST3), Senior nurse (Band 6+), Junior nurse (band 5). Each clinician calculated the patient's HEART score independently. Data was entered electronically at the time of assessment. The intraclass correlation coefficient (ICC) was calculated using R statistical package as the primary measure of agreement. We calculated 86 comparisons would be needed to achieve a desired power of 80%. Results: 104 patients were consented for recruitment of which 88 complete comparisons were obtained. The mean HEART score was 3.7 with no significant difference between mean scores for each clinician group. The overall ICC was 0.9 (0.87, 0.93). The ICC was also calculated between each group of clinicians; junior nurse and senior nurse 0.88 (0.83, 0.92), junior nurse and junior doctor 0.92 (0.88, 0.94), junior nurse and senior doctor 0.91 (0.86, 0.94), senior doctor and senior nurse 0.9 (0.85, 0.93), junior doctor and senior doctor 0.89 (0.85, 0.93) and senior nurse and junior doctor 0.93 (0.9, 0.95). On analysis of subjective components of the HEART score, for history the ICC was 0.39 (0.27, 0.52), ECG interpretation 0.64 ( 0.54, 0.73), risk factor assessment 0.84 (0.79, 0.89). Conclusion: This study has demonstrated very strong overall inter-operator agreement between the 4 groups of clinicians studied. This suggests the HEART Score can be used reliably regardless of grade or role by both nurses and doctors as part of an early risk stratification process within the ED.Figure 1