Erkelens DC, Rutten FH, Wouters LT, de Groot E, Damoiseaux RA, Hoes AW, Zwart DL. Limited reliability of experts' assessment of telephone triage in primary care patients with chest discomfort.
J Clin Epidemiol 2020;
127:117-124. [PMID:
32730853 DOI:
10.1016/j.jclinepi.2020.07.016]
[Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/26/2020] [Revised: 07/02/2020] [Accepted: 07/23/2020] [Indexed: 11/29/2022]
Abstract
OBJECTIVE
Root cause analyses of serious adverse events (SAE) in out-of-hours primary care (OHS-PC) often point to errors in telephone triage. Such analyses are, however, hampered by hindsight bias. We assessed whether experts, blinded to the outcome, recognize (un)safety of triage of patients with chest discomfort, and we quantified inter-rater reliability.
STUDY DESIGN AND SETTING
This is a case-control study with triage recordings from 2013-2017 at OHS-PC. Cases were missed acute coronary syndromes (ACSs, considered as SAE). These cases were age- and gender-matched 1:8 with the controls, sampled from the remainder of people calling for chest discomfort. Fifteen experts listened to the recordings and rated the safety of triage. We calculated sensitivity and specificity of recognizing an ACS and the intraclass correlation.
RESULTS
In total, 135 calls (15 SAE, 120 matched controls) were relistened. The experts identified ACSs with a sensitivity of 0.86 (95% CI: 0.71-0.95) and a specificity of 0.51 (95% CI: 0.43-0.58). Cases were rated significantly more often as unsafe than the controls (73.3% vs. 22.5%, P < 0.001). The inter-rater reliability for safety was poor: ICC 0.16 (95% CI: 0.00-0.32).
CONCLUSIONS
Blinded experts rated calls of missed ACSs more often as unsafe than matched control calls, but with a low level of agreement among the experts.
Collapse