Association between increased Patients' out-of-pocket costs for apixaban and oral anticoagulant discontinuation in high-risk patients with atrial fibrillation - a retrospective Medicare claims study Free

Introduction:

Atrial fibrillation (AF) affects millions of adults in the United States and is a major risk factor for stroke, making stroke prevention with oral anticoagulants (OACs) a critical component of AF management. Apixaban, a direct OAC, is commonly prescribed to patients newly diagnosed with AF. However, upward shift in formulary tier placement for apixaban in 2017 and resulting increased out-of-pocket costs (OOPC) have created challenges for Medicare beneficiaries to access apixaban and adhere with OAC therapy.

Purpose:

To evaluate the impact of increased OOPC for apixaban on OAC treatment discontinuation in high-stroke-risk Medicare beneficiaries with AF.

Methods:

This retrospective cohort study used the Medicare fee-for-service claims data (01/01/2012–12/31/2019) and examined adults with AF and elevated stroke risk, as defined by the individual components of the CHA₂DS₂-VASc score. Patients were included in the study if they (1) began apixaban during 01/01/2013–12/31/2016; (2) maintained uninterrupted apixaban use through the end of 2016; (3) held continuous Medicare Parts A/B/D coverage ≥12 months before apixaban initiation and through the end of 2017; (4) were enrolled in a prescription drug plan (PDP) in 2016 that shifted apixaban to a higher tier the following year, but remained in the same PDP in 2017; (5) experienced a rise in apixaban OOPC from 2016 to 2017 and (6) were at a high-risk of stroke (males with a CHA₂DS₂-VASc score ≥2 and females with a CHA₂DS₂-VASc score ≥3). Monthly OOPC (2020 USD) were calculated from all apixaban claims in 2016 and 2017, respectively, and standardized to reflect a 30-day supply. Treatment discontinuation was defined as a >30-day gap in apixaban without switching to another OAC. Multivariable Cox proportional hazard models were used to evaluate the association between increased OOPC for apixaban (per $50 increments) and treatment discontinuation in 2017. Subgroup analyses were conducted for each individual stroke risk factor included in the CHA₂DS₂-VASc score.

Results:

The cohort comprised 1,128 high-risk Medicare beneficiaries treated with apixaban in 2016. The mean age was 78.8, 44.2% were women, 94.3% were White, and 35.2% were from the South, 29.2% Midwest, 23.0% Northeast and 12.7% from West region of the US. The mean duration of apixaban prior to 2017 was 9.8 months, including 9.2% with >2 years of use. All plans with tier 2 (2.7%) or tier 3 (97.3%) formulary coverage for apixaban in 2016 shifted apixaban to a higher tier in 2017, triggering a new prior authorization requirement in 2017 (99.9% of plans). Average monthly OOPC for apixaban almost doubled from $76.6 ± $40.8 in 2016 to $162.5 ± $60.0 in 2017, with a mean increase of $85.9 ± $57.2. In 2017, 27.8% of beneficiaries discontinued apixaban without switching to another OAC.

Overall, for every $50 increase in the average monthly OOPC from 2016 to 2017, the likelihood of treatment discontinuation significantly increased by 1.24 times (adjusted hazard ratio [aHR] = 1.24; 95% confidence interval [CI]: 1.12, 1.38). Subgroup analyses demonstrated that the association between increased OOPC and treatment discontinuation was more pronounced in women (aHR = 1.39; 95% CI: 1.16, 1.66) and patients with a history of stroke, transient ischemic attack, or thromboembolism (aHR = 1.43; 95% CI: 1.11, 1.84). Significant positive associations between increased OOPC and treatment discontinuation were also observed among male patients (aHR = 1.18; 95% CI: 1.02, 1.36), patients aged ≥75 years (aHR = 1.21; 95% CI: 1.05, 1.39), and those with a history of congestive heart failure (aHR = 1.29; 95% CI: 1.05, 1.59), or hypertension (aHR = 1.26; 95% CI: 1.13, 1.41).

Conclusions:

In these high-risk Medicare beneficiaries with AF, patients' OOPC nearly doubled following the shift of apixaban to a higher formulary tier. More than one in four of these patients subsequently discontinued treatment. Higher OOPC were significantly associated with an increased likelihood of treatment discontinuation. This association was consistently observed across stroke risk factors and was particularly pronounced among women and those with a history of stroke, transient ischemic attack, or thromboembolism. These findings suggest that patients with known conditions placing them at high-risk of stroke may be particularly vulnerable to non-medical barriers, highlighting the need for policies that ensure continued access to preventive treatments.