Performance of computerized self-reported medical history taking and HEAR score for safe early rule-out of cardiac events in acute chest pain patients: the CLEOS-CPDS prospective cohort study

Abstract

Aims: A simplified version of the history, electrocardiogram, age, risk factors, troponin (HEART) score, excluding troponin, has been proposed to rule-out major adverse cardiac events (MACEs). Computerized history taking (CHT) provides a systematic and automated method to obtain information necessary to calculate the HEAR score. We aimed to evaluate the efficacy and diagnostic accuracy of CHT in calculating the HEAR score for predicting MACE.

Methods and results: Prospective study including clinically stable adults presenting with chest pain at the emergency department (ED) of Danderyd University Hospital (Stockholm, Sweden), in 2017-19. Participants entered their medical histories on touchscreen tablets using CHT software. The HEAR and HEART scores were calculated from CHT data. Thirty-day MACE and acute coronary syndrome (ACS) outcomes were retrieved, and the diagnostic accuracy was assessed. Logistic regression was used to determine the most predictive components of the HEAR score. Among 1000 patients, HEART and HEAR scores could be calculated from CHT data in 648 and 666 cases, respectively, with negative predictive values [95% confidence interval (CI)] of 0.98 (0.97-0.99) and 0.99 (0.96-1.00). Two patients with HEAR score <2 experienced a 30-day MACE. The age [odds ratio (OR) 2.75, 95% CI 1.62-4.66] and history (OR 2.38, 95% CI 1.52-3.71) components of the HEAR score were most predictive of MACE. Acute coronary syndrome outcomes provided similar results.

Conclusion: The HEAR score acquired by CHT identifies very-low-risk patients with chest pain in the ED, safely ruling out ACS and MACE. This highlights the value of computerized history taking by patients, which may reduce unnecessary tests and hospital admissions.

Trial registration: ClinicalTrials.gov NCT03439449.

Keywords: Artificial intelligence; Chest pain; Coronary artery disease; Medical history taking; Medical informatics; Risk assessment.

Graphical Abstract

Figure 1

User interface and example questions in the computerized history-taking programme.

Figure 2

Study flow diagram. ED, emergency department; ECG, electrocardiogram; CHT, computerized history taking; Ref test, reference test. During the study period, 13 044 patients presented at the emergency department with chest pain, of which 1012 provided informed consent and met the inclusion criteria when research staff were on duty. Twelve patients were excluded, and 1000 patients were ultimately included in the study. Insufficient computerized history-taking data indicate that the patient did not complete the computerized history taking interview far enough to provide the necessary data for HEAR or HEART score calculations (see Supplementary material online, Figure S1 for details).

Figure 3

The distribution of HEAR scores in patients based on the outcome of major adverse cardiac event, showing a correlation between increasing HEAR scores and the frequency of major adverse cardiac event.

Figure 4

Receiver operating characteristic curves for HEART and HEAR score, for a 30-day MACE. Area under the receiver operating characteristic curve levels for both HEART and HEAR score are generally considered as excellent; however, the area under the receiver operating characteristic curve for HEART score was slightly higher (P < 0.01). AUROC, area under the receiver operating characteristic curve.