Economic analysis of an AI-enabled ECG alert system: impact on mortality outcomes from a pragmatic randomized trial

Abstract

Findings from a previous study (ClinicalTrials.gov: NCT05118035) demonstrated that an AI-enabled electrocardiogram (AI-ECG), combining AI reports and physician alerts, effectively identified hospitalized patients at high risk of mortality and reduced all-cause mortality. This study evaluates its cost-effectiveness from the health payer’s perspective in Taiwan over a 90-day post-intervention period. Cost data were obtained from electronic health records of participating hospitals, and incremental cost-effectiveness ratios (ICERs) per death averted were calculated. Non-parametric bootstrap techniques were used to address uncertainty. Among 15,965 patients, 90-day all-cause mortality was 3.6% in the intervention group versus 4.3% in controls. Medication and ICU costs were higher in the AI-ECG group, but overall medical cost was similar ($6204 vs. $5803). The ICER was $59,500 (95% CI: $-4657 to $385,950) per death averted. The cost-effectiveness acceptability curve showed that 95% of the probability mass lies below a willingness-to-pay threshold of $409,321, supporting favorable cost-effectiveness despite uncertainty.