Anticoagulant Drug Costs Exploded While the Number of Patients Barely Moved
Medicare spent $17.4 billion on apixaban in 2023. It spent $64.8 million on warfarin. The total number of anticoagulated patients didn't change much. What changed was the price tag per patient.
That gap, roughly 268 times more spending on apixaban than warfarin in a single year, is the defining feature of how anticoagulant therapy has evolved inside Medicare Part D over the past decade. It's not a story about more patients getting treated. It's a story about the same treatment costing dramatically more.
From $13 to $862: The Per-Claim Gap That Defines the Shift
Warfarin's cost per claim was $13.43 in 2015. In 2023, it was $13.58. Eight years, fifteen cents of inflation. For a drug that has anticoagulated patients for decades, the price has essentially been frozen.
Apixaban's cost per claim was $389.75 in 2015. By 2023, it had risen to $862.56, a 121% increase over the same period. Rivaroxaban moved in parallel, from $392.68 to $935.33. Both drugs more than doubled in per-claim cost while warfarin held flat.
The patient population shifted just as dramatically. Warfarin beneficiaries fell from 2,373,264 in 2015 to 662,192 in 2023. Apixaban beneficiaries grew from 283,207 to 4,191,162. That's a near-complete migration of patients from a $13-per-claim drug to an $862-per-claim drug, and Medicare's total anticoagulant bill reflects every step of that transition. Patients who moved from warfarin to apixaban didn't gain access to anticoagulation they didn't have before. They gained access to a drug that costs 63 times more per claim.
Cardiology Prescribers Drive the Highest-Cost Prescribing
Not all apixaban spending looks the same across specialties. Cardiologists accounted for 26.16% of total apixaban drug spending in 2023, with a cost per claim of $1,086.09. That's 47% higher than what internal medicine physicians generated per claim ($737.14) and 41% higher than family practice ($768.39).
| Specialty | Cost per Claim | Share of Apixaban Spending |
|---|---|---|
| Cardiology | $1,086.09 | 26.16% |
| Interventional Cardiology | $1,070.74 | 5.26% |
| Clinical Cardiac Electrophysiology | $1,060.25 | 4.17% |
| Nurse Practitioner | $840.08 | 14.59% |
| Internal Medicine | $737.14 | 20.05% |
| Family Practice | $768.39 | 17.54% |
The three cardiology-adjacent specialties all exceed $1,060 per claim, well above the program average. Internal medicine and family practice together account for 37.59% of apixaban spending, but at substantially lower per-claim costs. The specialty mix matters: as more apixaban prescribing flows through cardiology, the average cost per claim across the program rises with it.
$17.4 Billion for a Drug That Didn't Exist in Medicare a Decade Ago
Apixaban generated $24.6 million in total Medicare Part D spending in 2013. Ten years later, that figure reached $17.4 billion. No other single drug in the anticoagulant class comes close. Warfarin's entire 2023 spending, across 662,192 beneficiaries, totaled $64.8 million.
The scale of that comparison is worth sitting with. Medicare now spends more on apixaban every two days than it spends on warfarin in an entire year.
What makes this fiscally significant isn't that more patients are being anticoagulated. The warfarin beneficiary count fell by 1.7 million as apixaban's rose by 3.9 million, a substitution, not an expansion. Medicare is paying billions more per year to anticoagulate roughly the same pool of patients who would have been on warfarin a decade ago. The per-claim cost doubled. The patient count grew. The spending multiplied by a factor that warfarin's price history makes almost impossible to contextualize.
Warfarin's cost per claim remained essentially flat at $13.43 to $13.58 from 2015 to 2023, even as 1.7 million patients who once took it migrated to drugs costing 60 to 70 times more per claim. Whether clinical guidelines, prescriber habits, or drug manufacturer pricing strategies drove that migration, the fiscal outcome is the same: Medicare is now committed to a spending baseline for anticoagulation that has no precedent in the program's history.
Ask a follow-up
Run one of these against the same data. The question is filled in for you.
Explore the data yourself
Run your own queries against 240M+ rows of federal health data using natural language — powered by AI.
Start analyzing