BOSTON – The direct acting oral anticoagulants boost patient adherence compared with warfarin anticoagulation, but when patients did not adhere to their regimens, those prescribed a new, direct-acting oral anticoagulant had worse outcomes than did patients on warfarin – even patients poorly adherent with warfarin – based on data from more than 80,000 U.S. patients.
Among low-adherence patients, defined as those with adherence rates of 40%-80% based on prescriptions filled, patients with nonvalvular atrial fibrillation and a CHA2DS2-VASc score of at least 2 and treated with warfarin had a 3.37/100 patient-years rate of thromboembolic events–driven hospitalizations or emergency department visits, compared with a 4.05/100 patient-years rate among low-adherence patients receiving a direct-acting oral anticoagulant (DOAC),
In contrast, when patients were adherent, taking more than 80% of their prescribed drug, the performance of the DOACs generally surpassed that of warfarin. In an analysis adjusted for several demographic and clinical confounders, and when compared with patients adherent to a warfarin regimen, those adherent to a DOAC had a 7% lower rate of thromboembolic events, a 36% lower rate of hemorrhagic strokes, an 8% lower rate of any stroke, and a 10% lower rate of bleeds (excluding hemorrhagic strokes), all statistically significant differences, reported Dr. Lakkireddy, medical director of the Kansas City Heart Rhythm Institute in Overland Park, Kan.
The message from this analysis is the importance of maximizing patient adherence, Dr. Lakkireddy said.
“We should not make the false assumption that putting a patient on a DOAC will take care of everything. We need to make it a habit to make sure patients are taking their pills,” he said in an interview. “We were surprised. Our assumption was that the DOACs were more forgiving than warfarin” in poorly compliant patients. “We need to make talking about adherence with patients routine.”
The results also documented that adherence to therapy is better with a DOAC, with a 74% rate of good adherence among the nearly 41,000 patients in the database prescribed a DOAC compared with a 63% rate of good compliance among the more than 42,000 patients prescribed warfarin.
“It’s ironic that we might think low-adherence patients should go on warfarin. That’s sort of backwards,” said , a cardiac electrophysiologist and professor of medicine at the University of British Columbia in Vancouver. “It’s not biologically plausible” to predict that less adherent patients would do better on warfarin, he noted.
The study run by Dr. Lakkireddy and his associates used data collected by IBM Watson Health Market Scan from about 4 million Medicare patients and 47 million American residents with private insurance during 2012-2016. They focused on the more than 600,000 patients prescribed an anticoagulant during 2014 and 2015, and then narrowed the study group down to just over 83,000 adults with nonvalvular atrial fibrillation, a CHA2DS2-VASc score of 2 or more. The patients’ average age was about 74 years.
Dr. Lakkireddy has been a consultant to or has received research support from Biosense Webster, Boehringer Ingelheim, Bristol Myers Squibb, EstechPharma, Janssen, Pfizer, SentreHeart, and St. Jude. Dr. Krahn has been a consultant to Medtronic and has received research support from Medtronic and Boston Scientific.
SOURCE: Lakkireddy D et al. Heart Rhythm 2018, Abstract B-LBCT02-03.
What is clear from this analysis is that adherence to oral anticoagulant regimens is something we need to address. It would help if we could determine why patients are not well adherent, but regardless of the cause, changing patient behavior and improving adherence will require better patient education and better integration of medical care toward better adherence.
This study has several obvious limitations, including its reliance on an administrative database that does not allow for adjudication of outcomes. The analysis presented so far is also limited by not breaking down the direct-acting oral anticoagulant into individual drugs, and without propensity-score matching of the two treatment subgroups.
We must be very alert to the possibility for confounding by indication in a real-world dataset like this. For example, the stroke rate we see in the patients treated with a DOAC may be affected by clinicians who preferentially prescribed one of the direct-acting drugs to patients who had what they thought was a high stroke risk because they felt these drugs might work better for stroke prevention than warfarin.
What is also notable in the results is that, even among the patients with good adherence, the event and adverse effects rates remained high. Among the highly adherent patients the thromboembolic event rate was greater than 3% per year in both the warfarin and direct-acting drug groups, the stroke rate was also greater than 3% per year in both subgroups, and bleeding events occurred at rates of greater than 4% per year among the patients treated with direct-acting drugs and greater than 5% per year in the warfarin group.
, professor of medicine and chair of cardiology at the University of Antwerp, Belgium, made these comments as designated discussant for the study. He had no current disclosures.