AI positive significance in clinical decision-making does not change patient outcomes

A study conducted in London at ESC Congress 2024 shows that AI does not help in better outcomes for patients with heart attacks in Emergency departments. The results of RAPIDx AI showed no difference in the outcomes of patients treated with AI-assisted care and patients treated with normal care.  However, it does assist in quick and apt diagnoses. 

“Our large cluster-randomized trial involving patients presenting to the ED with suspected cardiac conditions across South Australia did not improve clinical outcomes, however, it did highlight the ability of real-time AI to influence clinical decisions and practice towards evidence-based care.”
Lead author Kristina Lambrakis at the Victorian Heart Hospital

About 14, 131 were tested in a trial in South Australia. These patients were tested for heart conditions through hs-cTn (high-sensitivity troponin). Twelve hospitals were involved in this trial. Six hospitals were asked to provide standard care practice to the patients while the other 6 were using AI-assisted decision-making in the process. The study was conducted over the period of 6 months.

Up to 3,029 patients were studied in the primary analysis, number of deaths by heart attack. The results showed 26% deaths in AI-assisted care and 26.4% deaths in the standard care group, showing no significant difference.

Study shows AI’s significance in making clinical decisions and using evidence-based treatment.

On the other hand, the study shows the positive use of AI as well. The angiography which would have been redundant was avoided by 47%. Antiplatelet therapy and statins were also prescribed which are proven treatments. The early discharge decision of patients from ED was also made for safety. 

The research study proved AI’s significance in making clinical decisions and using evidence-based treatment. However, it did not impact any change on the death rate as compared to the patients with normal care.