Diagnostic performance of telephone triage for nontraumatic chest pain in the emergency coordination center: a population-based study in Andalusia

Objective.

To assess the diagnostic performance of the Emergency Coordination Center (CCUE) in identifying priority one or emergent cases among calls reporting nontraumatic chest pain.

Method.

This retrospective diagnostic performance study (2020–2024) analyzed a total of 248,116 calls categorized as nontraumatic chest pain, of which 97,905 included a clinical diagnosis. The assigned priority level was compared with the clinical diagnosis established by the out-of-hospital medical team, which served as the reference standard. Diagnostic performance metrics were calculated according to sociodemographic, territorial, and methodological variables. Uni- and multivariate analyses were conducted to identify factors associated with under- and overtriage.

Results.

Undertriage occurred in 7.1% of cases and overtriage in 77.4%. Overall accuracy rate was 48.1%; sensitivity rate, 84.7%; and specificity rate, 40.6%. The positive predictive value was 22.6%, and the negative predictive value was 92.9%. Structured triage (GTTA23) showed lower sensitivity (78.9% vs 91.0%; P < .001) but higher specificity (44.2% vs 31.6%; P < .001), with a trend toward lower undertriage (6.6% vs 10.7%; P = .180) and higher overtriage (82.6% vs 61.8%; P < .001) vs direct assignment. Interprovincial differences were low to moderate. In multivariate analysis, undertriage was associated with older age, home care, and nighttime calls, while female sex was protective. Overtriage was more likely among women, in public settings, and when coordination time was shorter.

Conclusions.

The diagnostic performance of the CCUE for nontraumatic chest pain shows adequate sensitivity but low specificity, with acceptable undertriage and high overtriage. Structured triage improves the balance between safety and efficacy, although differences persist according to sex, age, and care setting.