Artificial intelligence-based clinical decision support may reduce sex disparities in suspected ACS: sub-analysis of a multicentre cluster-randomised trial
R Rebecca, K Lambrakis, E Khan, T Briffa, L A Cullen, J W Verjans, J Karnon, C Papendick, D P ChewAbstract
Background
Women presenting to the emergency department (ED) with suspected acute coronary syndrome (ACS) experience higher rates of misdiagnosis and adverse outcomes than men. Existing clinical decision aids improve risk stratification but have sex-specific limitations that may disadvantage women. Artificial intelligence (AI)-based clinical decision support at the point of care has the potential to reduce this inequity, though its real-world impact remains uncertain.
Purpose
To determine whether AI-based decision support narrows sex-based disparities in care and outcomes among patients with suspected cardiac chest pain.
Methods
We conducted a sex-specific subgroup analysis of a multicentre cluster-randomised controlled trial across 12 South Australian EDs which deployed real-time AI-based clinical decision support in the ED for suspected ACS. Analyses were conducted both on the overall cohort, and among those with a predicted 30-day risk of death or myocardial infarction (MI) of >1%. The primary outcome was 12-month cardiovascular (CV) death, MI and unplanned CV readmission. Time-to-event outcomes were analysed with Cox proportional hazards models, adjusted for clustering.
Results
Among 14,131 participants (49% women), AI-based decision support reduced the 12-month primary composite endpoint in women (HR 0.80, 95% CI 0.70-0.92; p = 0.001), with the greatest effect in the elevated-risk group (HR 0.76, 95% CI 0.69–0.85; p < 0.001). No significant benefit was observed in men, suggesting that the intervention did not impact the cohort uniformly.
Conclusion
AI-based decision support reduced major CV events in women presenting to the ED with suspected cardiac chest pain, with effects most evident in those at higher baseline risk. This may represent a novel strategy to address sex-based inequities in ACS care.