AHA 2018: Computerised-decision support has the potential to be a “powerful tool” for preventing cardiovascular events in AF patients

Gregory Piazza

Data presented at the 2018 American Heart Association (AHA) scientific session scientific sessions (10–12 November, Chicago, USA) indicate that the use of alert-based computerised-decision support significantly increases the rate of anticoagulation prescription among patients who are hospitalised with atrial fibrillation (AF). Furthermore, the use of such software is associated with a significant decrease in major adverse cardiac events.

Presenting the AF-Alert study at the AHA, Gregory Piazza (Division of Cardiovascular Medicine, Brigham and Women’s Hospital, Boston, USA) reported that he and his fellow investigators had previously found, in a study of patients hospitalised with AF, that “antithrombotic therapy was omitted in nearly 40% of those at risk”. He added: “Failure to prescribe antithrombotic therapy in these high-risk patients resulted in a 36% increase in major adverse events, including stroke and myocardial infarction.”

Noting that alert-based computerised decision support strategies “have been successfully implemented to improve underutilisation of venous thromboembolism prophylaxis in high-risk hospitalised patients”, Piazza stated that the objective of AF-Alert was to determine the impact of alert-based computerised-decision support on the prescription of anticoagulation among patients with AF. The hypothesis of the study, he explained, was that “alert-based computerised-decision support will increase prescription of anticoagulation in high-risk hospitalised patients with atrial fibrillation who were not being prescribed anticoagulation”.

In the study, 458 patients who were hospitalised for AF and who were not  already receiving anticoagulation were randomised to standard care plus alert-based computerised-decision support (248) or standard therapy alone (210). The primary efficacy endpoint was the rate of anticoagulation prescription during hospitalisation, at discharge and at 90 days, and the primary safety endpoint was the occurrence of major bleeding or clinically relevant non-major bleeding.

At all follow-up points (during hospitalisation, discharge, and at 90 days), the rate of anticoagulation prescription was significantly higher among the alert group patients. This meant that the rate of the primary efficacy endpoint, according to Piazza, “nearly tripled” in the alert group: 19.4% vs. 7.1% for the control group (p<0.001). There no significant differences in the rate of the primary safety endpoint between groups.

Also, the rate of the secondary efficacy endpoint—a composite of death, myocardial infarction, transient ischaemic attack or systemic embolic event at 90 days—was significantly lower in the alert group: 11.3% vs. 21.9% for the control group (p<0.002); a 50% reduction. Furthermore, aside from death, the individual components of this endpoint were significantly lower in the alert group. This meant that the alert group had significantly higher rates of freedom from major adverse cardiac events (MACE) and from major adverse events (MAE) at 90 days compared with the control group (p=0.004 and p=0.001 for the comparisons). Piazza commented that the reduction in MACE—“the most dramatic finding of our study”—was “beyond what was anticipated from the increase in prescription of anticoagulation”.

“Computerised-decision support has the potential to be a powerful tool in prevention of cardiovascular events in patients with AF,” Piazza concluded.

He told Cardiac Rhythm News that, in the study, the key reasons for not prescribing anticoagulation were “the perceived risk of bleeding and risk for falls”.


Please enter your comment!
Please enter your name here