Digital alerts reduced AF-related stroke, MI rates

CHICAGO – High-risk hospitalized patients with atrial fibrillation (AF) whose doctors monitored them with a computerized alert system were more than twice as likely to be on anticoagulation and had significantly lower rates of death and other cardiovascular events, compared with patients on a standard admissions protocol, according to results of a randomized, controlled trial presented at the American Heart Association Scientific Sessions.

“Alert-based computerized decision support [CDS] increased the prescription of anticoagulation for stroke prevention in atrial fibrillation during hospitalization, at discharge, and at 90 days after randomization in high-risk patients,” said Gregory Piazza, MD, of Brigham and Women’s Hospital, Boston, in presenting results of the AF-ALERT trial. “The reductions in major cardiovascular events was attributable to reductions in MI and stroke/transient ischemic attack at 90 days in patients whose physicians received the alert.”

The trial evaluated 458 patients hospitalized for AF or flutter and with CHA2DS2-VASc scores of 1-8 randomly assigned to the alert (n = 258) or no-alert (n = 210) groups.

Dr. Piazza explained that for those in the alert group, the CDS system notified physicians when the patient’s CHA₂DS₂-VASc score increased. From there, the physician could choose to open an order template to prescribe evidence-based medications to prevent stroke, to elect to review evidence-based clinical practice guidelines, or to continue with the admissions order with an acknowledged reason for omitting anticoagulation (such as high bleeding risk, low stroke risk, high risk for falls, or patient refusal of anticoagulation).

“In patients for whom their providers were alerted, 35% elected to open the stroke-prevention order set, a very tiny percentage elected to read the AF guidelines, and about 64% exited but provided a rationale for omitting anticoagulation,” Dr. Piazza noted.

The alert group was far more likely to be prescribed anticoagulation during the hospitalization (25.8% vs. 9.5%; P less than .0001), at discharge (23.8% vs. 12.9%; P = .003), and at 90 days (27.7% vs. 17.1%; P = .007) than the control group. The alert resulted in a 55% relative risk reduction in a composite outcome of death, MI, cerebrovascular event, and systemic embolic event at 90 days (11.3% vs. 21.9%; P = .002). The alert group had an 87% lower incidence of MI at 90 days (1.2% vs. 8.6%, P = .0002) and 88% lower incidence of cerebrovascular events or systemic embolism at 90 days (0% vs. 2.4%; P = .02). Death at 90 days occurred in 10.1% in the alert group and 14.8% in the control group (P = .13).

One of the limitations of the study, Dr. Piazza noted, was that the most dramatic finding – reduction of major cardiovascular events – was a secondary, not a primary, endpoint. “CDS has the potential to be a powerful tool in prevention of cardiovascular events in patients with atrial fibrillation.”

Moderator Mintu Turakhia, MD, of Stanford (Calif.) University, questioned the low rate of anticoagulation in the study’s control arm – 9.5% – much lower than medians reported in many registries. He also asked Dr. Piazza to describe the mechanism of action for prescribing anticoagulation in these patients.

Dr. Piazza noted the study population was hospitalized patients whose providers had decided prior to their admissions not to prescribe anticoagulation; hence, the rate of anticoagulation in these patients was actually higher than expected.

Regarding the mechanism of action, “the electronic alert seems to preferentially increase the prescription of [direct oral anticoagulants] over warfarin, and that may have been one of the mechanisms,” Dr. Piazza said. Another explanation he offered were “off-target” effects whereby, if providers have a better idea of a patient’s risk for a stroke or MI, they’ll be more aggressive about managing other risk factors.

“There are a number of interventions that could be triggered if the alert prompted the provider to have a conversation with patients about their risk of stroke from AF,” he said. “This may have impact beyond what we can tell from this simple [Best Practice Advisory in the Epic EHR system]. I think we don’t have a great understanding of the full mechanisms of CDS.”

Dr. Piazza reported financial relationships with BTG, Janssen, Bristol-Myers Squibb, Daiichi Sankyo, Portola, and Bayer. Daiichi Sankyo funded the trial. Dr. Turakhia reported relationships with Apple, Janssen, AstraZeneca, VA, Boehringer Ingelheim, Cardiva Medical, Medtronic, Abbott, Precision Health Economics, iBeat, iRhythm, MyoKardia, Biotronik, and an ownership Interest in AliveCor.

CHICAGO – High-risk hospitalized patients with atrial fibrillation (AF) whose doctors monitored them with a computerized alert system were more than twice as likely to be on anticoagulation and had significantly lower rates of death and other cardiovascular events, compared with patients on a standard admissions protocol, according to results of a randomized, controlled trial presented at the American Heart Association Scientific Sessions.

“Alert-based computerized decision support [CDS] increased the prescription of anticoagulation for stroke prevention in atrial fibrillation during hospitalization, at discharge, and at 90 days after randomization in high-risk patients,” said Gregory Piazza, MD, of Brigham and Women’s Hospital, Boston, in presenting results of the AF-ALERT trial. “The reductions in major cardiovascular events was attributable to reductions in MI and stroke/transient ischemic attack at 90 days in patients whose physicians received the alert.”

The trial evaluated 458 patients hospitalized for AF or flutter and with CHA2DS2-VASc scores of 1-8 randomly assigned to the alert (n = 258) or no-alert (n = 210) groups.

Dr. Piazza explained that for those in the alert group, the CDS system notified physicians when the patient’s CHA₂DS₂-VASc score increased. From there, the physician could choose to open an order template to prescribe evidence-based medications to prevent stroke, to elect to review evidence-based clinical practice guidelines, or to continue with the admissions order with an acknowledged reason for omitting anticoagulation (such as high bleeding risk, low stroke risk, high risk for falls, or patient refusal of anticoagulation).

“In patients for whom their providers were alerted, 35% elected to open the stroke-prevention order set, a very tiny percentage elected to read the AF guidelines, and about 64% exited but provided a rationale for omitting anticoagulation,” Dr. Piazza noted.

The alert group was far more likely to be prescribed anticoagulation during the hospitalization (25.8% vs. 9.5%; P less than .0001), at discharge (23.8% vs. 12.9%; P = .003), and at 90 days (27.7% vs. 17.1%; P = .007) than the control group. The alert resulted in a 55% relative risk reduction in a composite outcome of death, MI, cerebrovascular event, and systemic embolic event at 90 days (11.3% vs. 21.9%; P = .002). The alert group had an 87% lower incidence of MI at 90 days (1.2% vs. 8.6%, P = .0002) and 88% lower incidence of cerebrovascular events or systemic embolism at 90 days (0% vs. 2.4%; P = .02). Death at 90 days occurred in 10.1% in the alert group and 14.8% in the control group (P = .13).

One of the limitations of the study, Dr. Piazza noted, was that the most dramatic finding – reduction of major cardiovascular events – was a secondary, not a primary, endpoint. “CDS has the potential to be a powerful tool in prevention of cardiovascular events in patients with atrial fibrillation.”

Moderator Mintu Turakhia, MD, of Stanford (Calif.) University, questioned the low rate of anticoagulation in the study’s control arm – 9.5% – much lower than medians reported in many registries. He also asked Dr. Piazza to describe the mechanism of action for prescribing anticoagulation in these patients.

Dr. Piazza noted the study population was hospitalized patients whose providers had decided prior to their admissions not to prescribe anticoagulation; hence, the rate of anticoagulation in these patients was actually higher than expected.

Regarding the mechanism of action, “the electronic alert seems to preferentially increase the prescription of [direct oral anticoagulants] over warfarin, and that may have been one of the mechanisms,” Dr. Piazza said. Another explanation he offered were “off-target” effects whereby, if providers have a better idea of a patient’s risk for a stroke or MI, they’ll be more aggressive about managing other risk factors.

“There are a number of interventions that could be triggered if the alert prompted the provider to have a conversation with patients about their risk of stroke from AF,” he said. “This may have impact beyond what we can tell from this simple [Best Practice Advisory in the Epic EHR system]. I think we don’t have a great understanding of the full mechanisms of CDS.”

Dr. Piazza reported financial relationships with BTG, Janssen, Bristol-Myers Squibb, Daiichi Sankyo, Portola, and Bayer. Daiichi Sankyo funded the trial. Dr. Turakhia reported relationships with Apple, Janssen, AstraZeneca, VA, Boehringer Ingelheim, Cardiva Medical, Medtronic, Abbott, Precision Health Economics, iBeat, iRhythm, MyoKardia, Biotronik, and an ownership Interest in AliveCor.

CHICAGO – High-risk hospitalized patients with atrial fibrillation (AF) whose doctors monitored them with a computerized alert system were more than twice as likely to be on anticoagulation and had significantly lower rates of death and other cardiovascular events, compared with patients on a standard admissions protocol, according to results of a randomized, controlled trial presented at the American Heart Association Scientific Sessions.

“Alert-based computerized decision support [CDS] increased the prescription of anticoagulation for stroke prevention in atrial fibrillation during hospitalization, at discharge, and at 90 days after randomization in high-risk patients,” said Gregory Piazza, MD, of Brigham and Women’s Hospital, Boston, in presenting results of the AF-ALERT trial. “The reductions in major cardiovascular events was attributable to reductions in MI and stroke/transient ischemic attack at 90 days in patients whose physicians received the alert.”

The trial evaluated 458 patients hospitalized for AF or flutter and with CHA2DS2-VASc scores of 1-8 randomly assigned to the alert (n = 258) or no-alert (n = 210) groups.

Dr. Piazza explained that for those in the alert group, the CDS system notified physicians when the patient’s CHA₂DS₂-VASc score increased. From there, the physician could choose to open an order template to prescribe evidence-based medications to prevent stroke, to elect to review evidence-based clinical practice guidelines, or to continue with the admissions order with an acknowledged reason for omitting anticoagulation (such as high bleeding risk, low stroke risk, high risk for falls, or patient refusal of anticoagulation).

“In patients for whom their providers were alerted, 35% elected to open the stroke-prevention order set, a very tiny percentage elected to read the AF guidelines, and about 64% exited but provided a rationale for omitting anticoagulation,” Dr. Piazza noted.

The alert group was far more likely to be prescribed anticoagulation during the hospitalization (25.8% vs. 9.5%; P less than .0001), at discharge (23.8% vs. 12.9%; P = .003), and at 90 days (27.7% vs. 17.1%; P = .007) than the control group. The alert resulted in a 55% relative risk reduction in a composite outcome of death, MI, cerebrovascular event, and systemic embolic event at 90 days (11.3% vs. 21.9%; P = .002). The alert group had an 87% lower incidence of MI at 90 days (1.2% vs. 8.6%, P = .0002) and 88% lower incidence of cerebrovascular events or systemic embolism at 90 days (0% vs. 2.4%; P = .02). Death at 90 days occurred in 10.1% in the alert group and 14.8% in the control group (P = .13).

One of the limitations of the study, Dr. Piazza noted, was that the most dramatic finding – reduction of major cardiovascular events – was a secondary, not a primary, endpoint. “CDS has the potential to be a powerful tool in prevention of cardiovascular events in patients with atrial fibrillation.”

Moderator Mintu Turakhia, MD, of Stanford (Calif.) University, questioned the low rate of anticoagulation in the study’s control arm – 9.5% – much lower than medians reported in many registries. He also asked Dr. Piazza to describe the mechanism of action for prescribing anticoagulation in these patients.

Dr. Piazza noted the study population was hospitalized patients whose providers had decided prior to their admissions not to prescribe anticoagulation; hence, the rate of anticoagulation in these patients was actually higher than expected.

Regarding the mechanism of action, “the electronic alert seems to preferentially increase the prescription of [direct oral anticoagulants] over warfarin, and that may have been one of the mechanisms,” Dr. Piazza said. Another explanation he offered were “off-target” effects whereby, if providers have a better idea of a patient’s risk for a stroke or MI, they’ll be more aggressive about managing other risk factors.

“There are a number of interventions that could be triggered if the alert prompted the provider to have a conversation with patients about their risk of stroke from AF,” he said. “This may have impact beyond what we can tell from this simple [Best Practice Advisory in the Epic EHR system]. I think we don’t have a great understanding of the full mechanisms of CDS.”

Dr. Piazza reported financial relationships with BTG, Janssen, Bristol-Myers Squibb, Daiichi Sankyo, Portola, and Bayer. Daiichi Sankyo funded the trial. Dr. Turakhia reported relationships with Apple, Janssen, AstraZeneca, VA, Boehringer Ingelheim, Cardiva Medical, Medtronic, Abbott, Precision Health Economics, iBeat, iRhythm, MyoKardia, Biotronik, and an ownership Interest in AliveCor.