Key result
Computer ECG interpretation improves paramedic STEMI accuracy ~80% when correct but worsens it when incorrect.
Why the study?
Does the visibility of computer interpretation messages affect the accuracy of paramedics' diagnosis of STEMI?
RCT (n=254)
Double-blind
Block randomisation
Does the visibility of computer interpretation messages affect the accuracy of paramedics' diagnosis of STEMI?
Incorrect computer ECG interpretations significantly negatively influence paramedics' diagnostic accuracy for STEMI, highlighting the need for caution when relying on automated algorithms.
Caution advised with computer ECG interpretations for paramedic STEMI diagnosis; reinforces limitations of automated algorithms in prehospital care.
Background Timely diagnosis and appropriate management of patients with ST-segment elevation myocardial infarction (STEMI) depends on accurate interpretation of the 12-lead ECG by paramedics. Computer interpretation messages on ECGs are often provided, but the effect they exert on paramedics' decision making is not known. The objective of this pilot study was to assess the feasibility of a trial using an online assessment tool, to determine the effect of computer interpretation messages on paramedics' diagnosis of STEMI. Methods The RESPECT pilot study is a cluster-randomised crossover trial using a bespoke, web-based assessment tool. Participants were randomly allocated 12 of 48 ECGs, with an equal mix of correct and incorrect computer interpretation messages, and STEMI and STEMI-mimics. These were viewed in two phases, with message visibility reversed between phases. Results 254 paramedics consented into the study, 205 completing the first phase and 150 completing phase two. The data were analysed in two subsets: computer message correct and computer message incorrect. In the subset of correct computer interpretations, accuracy was 84% (message hidden) and 87% (message visible). The subset of incorrect computer interpretations resulted in an accuracy of 77% (message hidden) and 71% (message visible). For the subset of correct computer interpretation, the unadjusted OR was 1.31 (95% CI 1.01–1.71), adjusted OR, 1.80 (95% CI 0.84–4.80), the ICC for participants was 0.36 and for ECGs, 0.18 Incorrect computer interpretations had an unadjusted OR of 0.76 (95% CI 0.61–0.93), adjusted OR, 0.58 (95% CI 0.41–0.80), the ICC for participants was 0.06 and for ECGs, 0.01. Conclusion A randomised crossover trial to determine the effect of computer interpretation messages is feasible. Pilot data have provided an indication about expected rates of discordance and suggest that incorrect computer messages have a stronger influence across participants and ECGs.
No takes yet. Share an insight, caveat, or question.
Pilbery et al. (2015) conducted an RCT in ST-segment elevation myocardial infarction (STEMI) (n=254). Computer interpretation messages on ECGs vs. No computer interpretation message (hidden) was evaluated on Correct paramedic interpretation of ECGs when the computer interpretation was correct (OR 1.80, 95% CI 0.84-4.91). Computer interpretation messages influenced paramedic ECG interpretation, improving accuracy when the computer was correct (OR 1.80) and significantly worsening accuracy when incorrect (OR 0.58).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: