Skip to content
FonteumPublic-records evidence
FINANCIAL DISTRESS · ISSUE 057
cms-part-d-prescribersOriginal Research

The Most Expensive Prescription Drugs in Medicare Part D

Eliquis cost Medicare Part D $19.88 billion in 2024 — the single costliest drug in the program, yet only its 12th most-prescribed. That inversion defines Part D: brand-name drugs are 23.9% of prescriptions but 90.1% of the dollars, while cheap generics carry the volume and almost none of the cost.

BY FONTEUM LLC · JUNE 12, 2026 · 12 MIN READREVIEWED BY DR. JENNIFER MONTECILLO, MDSNAPSHOT 2026-04-04 · LAST UPDATED APRIL 4, 2026
CMS Medicare Part D Prescribers · 2026-04-04
Medicare's most-prescribed drugs are all cheap genericscms-part-d-prescribers · 2026-04-04
Atorvastatin
72.47
Amlodipine
49.24
Levothyroxine
42.47
Gabapentin
35.52
Losartan
35.42
Lisinopril
35.42
Built on CMS Medicare Part D Prescribers · snapshot 2026-04-04 · reproducible · re-derive the figures yourself
Key findings
Medicare Part D cost for Eliquis in 2024 — the single costliest drug in the program, yet only the 12th most-prescribed
cms-part-d-prescribers · CMS
of Part D drug cost goes to brand-name drugs, which are only 23.9% of all prescription claims
cms-part-d-prescribers · CMS
claims for atorvastatin, the most-prescribed Part D drug — at $749 million, costing less per claim than a single Eliquis fill costs the program many times over
cms-part-d-prescribers · CMS
average Part D cost per claim for the cancer drug Revlimid — among the highest in the program, on just 217,318 claims
cms-part-d-prescribers · CMS

The drug Medicare spends the most on and the drug it prescribes the most are almost never the same drug. In 2024, the single costliest drug in Medicare Part D was Eliquis, a blood thinner, at $19.88 billion — yet Eliquis was only the program's 12th most-prescribed drug. The most-prescribed drug, the generic statin atorvastatin, was filled 72.5 million times but cost the program just $749 million. That inversion is not an accident. It is the structure of the entire program.

Two prescribing economies in one program

Part D runs on two separate economies that barely overlap. One is built on volume: enormous numbers of cheap generic prescriptions for the chronic conditions of an older population — high cholesterol, high blood pressure, diabetes, reflux. The other is built on price: a smaller number of expensive brand-name prescriptions, many of them for cardiovascular disease, cancer, and immune conditions.

The split is stark. Across all of Part D in 2024:

Share of claimsShare of cost
Generic-dispensed drugs76.1%9.9%
Brand-name drugs23.9%90.1%

Three-quarters of every prescription is a generic, but those generics are only one-tenth of the spending. Brand-name drugs are the reverse: under a quarter of prescriptions, more than nine-tenths of the dollars. A program that looks, by prescription count, like a generic-drug program is, by dollars, almost entirely a brand-name program.

The drug Medicare spends the most on and the drug it prescribes the most are almost never the same drug.

The most-prescribed drugs are almost all generic

Rank Part D drugs by how often they are filled and the top of the list is a catalogue of inexpensive generics. Atorvastatin leads at 72.5 million claims, then amlodipine (49.2M), levothyroxine (42.5M), gabapentin (35.5M), losartan (35.4M), and lisinopril (35.4M). Every one of the 11 most-prescribed drugs is a generic. They cost a few dollars per fill, so even at tens of millions of claims their total spend stays modest — atorvastatin's 72.5 million claims cost less than $750 million.

The first brand-name drug does not appear until rank 12: Eliquis. That single fact captures the whole pattern — you can read eleven of the most common prescriptions in America before you reach a drug that moves the budget.

The costliest drugs are almost all brand-name

Re-rank the same data by total cost and a completely different list appears.

Rank by costDrug2024 Part D costClaimsMost-prescribed rank
1Eliquis$19.88B22,996,54312
2Ozempic$12.38B9,930,484
3Jardiance$10.68B10,654,599
4Mounjaro$5.83B4,694,896
5Xarelto$5.44B5,888,827
6Trelegy Ellipta$4.83B5,725,949
7Trulicity$4.79B3,816,376
The seven costliest Medicare Part D drugs in 2024 are all brand-name: Eliquis at 19.88 billion dollars, Ozempic 12.38 billion, Jardiance 10.68 billion, Mounjaro 5.83 billion, Xarelto 5.44 billion, Trelegy Ellipta 4.83 billion, and Trulicity 4.79 billion.
The costliest drugs in Medicare Part D are all brand-name products with no generic competition. None of them appears among the program's most-prescribed drugs except Eliquis, at rank 12. Source: CMS Medicare Part D Prescribers · data year 2024 · snapshot 2026-04-04.

Eliquis is the rare drug that is both very common and very expensive — 23.0 million claims at a brand price. Its $19.88 billion is 8.8% of all Part D drug cost on a single product. Below it, the costliest drugs are brand cardiometabolic agents — the GLP-1 diabetes drugs Ozempic, Mounjaro, and Trulicity, and the SGLT2 inhibitors Jardiance and Farxiga — that we examine in the GLP-1 spending study.

When a tiny claim count produces a giant bill

The purest expression of the price economy is the specialty drug: a product with a small patient population and a list price high enough to land it among the costliest drugs in the program on almost no volume.

  • Revlimid (lenalidomide), a cancer drug, cost Part D $3.79 billion on 217,318 claims — about $17,431 per claim.
  • Vyndamax (tafamidis), for a heart-muscle disease, cost $2.50 billion on 98,896 claims — about $25,329 per claim.
  • Humira, for autoimmune disease, cost $3.80 billion on 437,072 claims — about $8,683 per claim.
For context, the average Part D prescription cost the program about $153 in 2024. A single Vyndamax claim costs roughly what 165 average prescriptions cost. These are the drugs that make total spending a poor guide to prescribing volume — a few hundred thousand claims can outweigh tens of millions.

These drugs never appear on a most-prescribed list and never will. They matter because they show why ranking drugs by cost and ranking them by volume produce two different maps of the same program — and why any conversation about Medicare drug spending has to be clear about which map it is reading.

How this connects to the rest of Part D

This two-economies pattern is the backbone of the whole Part D data graph. The Medicare Part D drug-spending overview lays out the program totals; the GLP-1 study zooms into the fastest-growing brand class; the prescribing-by-specialty study shows how the price economy concentrates in oncology and the volume economy in primary care; and the by-state study maps where the dollars land. Every figure across the family is a drug-, specialty-, or state-level aggregate — no individual prescriber is named.

Methodology

The two rankings come from two materialized views over the CMS Medicare Part D Prescribers "by Provider and Drug" 2024 file (snapshot 2026-04-04): part_d_top_drugs_by_cost_mv ranks drugs by sum(total_drug_cost), and part_d_top_drugs_by_claims_mv ranks them by sum(total_claims), each grouped by brand and generic name. The brand-versus-generic split follows the standard CMS convention: a row is generic-dispensed when its brand name equals its generic name (case-insensitive), brand otherwise; the two classes partition every row, so their claims and cost sum exactly to the program totals. Cost-per-claim is a drug's total cost divided by its total claims. The exact queries are in the reproducibility block below and trace to the Part D Prescribers dataset. Methodology version: part-d-costliest/v1.

Limitations

  • Gross cost, not net. All cost figures are point-of-sale totals before confidential manufacturer rebates. Net Medicare spending, especially on brand drugs, is lower than the gross amounts shown.
  • Brand-vs-generic by dispensing. The split classifies each prescription by whether it was dispensed as a brand or generic, the CMS convention. An authorized generic or a single-source brand can sit on the boundary; the partition is exact for accounting but is a billing label, not a clinical one.
  • A single year. This is the 2024 release. Rankings shift as patents expire, generics launch, and new brands ramp; treat it as a snapshot.
  • Cost is not appropriateness. A high total reflects price and volume, not whether any prescription was warranted. This study ranks spending; it does not evaluate prescribing.
  • Aggregate-only. Every figure is a drug- or program-level total. No individual prescriber is named, surfaced, or attached to any profile.

Sources

Frequently asked questions

What is the most expensive drug in Medicare Part D?
In 2024 it was Eliquis, a blood thinner, at $19.88 billion in total Part D drug cost. Ozempic was second at $12.38 billion and Jardiance third at $10.68 billion. All three are brand-name drugs with no generic competition, which is why they top the spending table.
What is the most prescribed drug in Medicare Part D?
Atorvastatin, a generic cholesterol statin, with 72.5 million claims in 2024 — far more than any other drug. It cost the program only $749 million, because as a generic it runs about ten dollars a fill. The 11 most-prescribed Part D drugs are all inexpensive generics.
Why isn't the most-prescribed drug also the most expensive?
Because price and volume run in opposite directions in Part D. The drugs prescribed most often are cheap generics that treat common chronic conditions. The drugs that cost the most are patent-protected brand-name products with high list prices. Eliquis is the costliest drug overall but only the 12th most-prescribed.
How much of Medicare Part D spending goes to brand-name drugs?
Brand-name drugs account for 90.1% of Part D drug cost — $204.4 billion of the $226.74 billion total — while making up just 23.9% of prescription claims. Generic drugs are the mirror image: 76.1% of claims but only 9.9% of the dollars.
Which drugs cost the most per prescription in Medicare Part D?
Specialty drugs with tiny patient populations. The cancer drug Revlimid averaged about $17,431 per Part D claim in 2024 and the heart drug Vyndamax about $25,329, versus roughly $153 across all Part D prescriptions. These drugs drive large totals on very few claims.
Does a high drug cost mean Medicare is overpaying?
Not on its own. These are gross costs before confidential manufacturer rebates, and a high total can reflect a large patient population, a high list price, or both. The data shows where the dollars concentrate; it does not judge whether any price or prescription was appropriate.
Can I reproduce these drug-cost rankings?
Yes. The costliest-drug and most-prescribed rankings come from two materialized views over the CMS Medicare Part D Prescribers 2024 file; the brand-versus-generic split comes from the overview view. The exact SQL is in the reproducibility block below and traces to a federal snapshot dated 2026-04-04.

Who uses this data

The source data behind this study is public

Compliance teams, journalists, and researchers work from the same federal source families cited above — queried by NPI or facility identifier through Fonteum’s open dataset pages and API. Every figure traces to a frozen, downloadable snapshot you can reproduce yourself.

Reproducibility

Every claim, reproducible

The SQL
most-expensive-medicare-part-d-drugs.sql
-- Most expensive vs most-prescribed Medicare Part D drugs, 2024 — reproducible query.
--
-- Source:   CMS Medicare Part D Prescribers, "by Provider and Drug".
-- Snapshot: cms-part-d-prescribers / data year 2024 / CMS release 2026-04-04.
-- Base:     public.cms_part_d_prescribers (~28.0M prescriber × drug rows).
-- Reads:    the frozen materialized views from
--           supabase/migrations/20260612150000_part_d_prescribing_research_views.sql.
--           The two top-drug views are each GROUP BY brand_name, generic_name over
--           the 2024 base rows, ranked by cost and by claims respectively; the
--           overview view carries the brand-vs-generic partition.
--
-- The whole study is the contrast between two orderings of the same drugs.
-- Audit observation: 2026-07-14. Source release remains 2026-04-04.

-- audit-claim: costliest-drug-vector
SELECT rank, brand_name, generic_name, claims, cost,
       round(cost::numeric / claims, 2) AS cost_per_claim
FROM public.part_d_top_drugs_by_cost_mv
ORDER BY cost DESC
LIMIT 12;
--  1 Eliquis    Apixaban       22,996,543  $19,876,304,796  $864     <- costliest drug; 12th by claims
--  2 Ozempic    Semaglutide     9,930,484  $12,375,986,475  $1,246
--  3 Jardiance  Empagliflozin  10,654,599  $10,678,957,372  $1,002
--  4 Mounjaro   Tirzepatide     4,694,896  $5,825,295,741   $1,241
--  5 Xarelto    Rivaroxaban     5,888,827  $5,436,701,739   $923
--  6 Trelegy …                  5,725,949  $4,832,494,643   $844
--  7 Trulicity  Dulaglutide     3,816,376  $4,786,817,612   $1,254
--  9 Humira(Cf) Adalimumab        437,072  $3,795,159,532   $8,683   <- giant bill on few claims
-- 10 Revlimid   Lenalidomide      217,318  $3,788,009,700   $17,431  <- specialty cancer drug
-- 15 Vyndamax   Tafamidis          98,896  $2,504,980,201  $25,329.44  <- specialty heart drug

-- audit-claim: most-prescribed-drug-vector
SELECT rank, brand_name, generic_name, claims, cost,
       round(cost::numeric / claims, 2) AS cost_per_claim,
       lower(brand_name) = lower(generic_name) AS is_generic
FROM public.part_d_top_drugs_by_claims_mv
ORDER BY claims DESC
LIMIT 12;
--  1 Atorvastatin Calcium  72,474,420  $748,890,955   $10.33  generic
--  2 Amlodipine Besylate   49,241,894  $310,756,194   $6.31   generic
--  3 Levothyroxine Sodium  42,465,073  $456,346,665   ...     generic
--  ...  (ranks 1-11 are ALL generic)
-- 12 Eliquis   Apixaban     22,996,543  $19,876,304,796        brand   <- first brand on the list

-- audit-claim: brand-generic-partition
SELECT
  generic_claims, brand_claims,
  generic_cost,   brand_cost,
  round(100.0 * generic_claims / total_claims,    1) AS generic_claims_pct,  -- 76.1%
  round(100.0 * generic_cost   / total_drug_cost, 1) AS generic_cost_pct,    --  9.9%
  round(100.0 * brand_claims   / total_claims,    1) AS brand_claims_pct,    -- 23.9%
  round(100.0 * brand_cost     / total_drug_cost, 1) AS brand_cost_pct       -- 90.1%
FROM public.part_d_prescribing_overview_mv;
-- generic + brand claims sum exactly to total_claims; generic + brand cost to
-- total_drug_cost — the partition is by lower(brand_name) = lower(generic_name).

-- audit-claim: vyndamax-example
SELECT
  rank,
  brand_name,
  generic_name,
  claims,
  cost,
  round(cost::numeric / claims, 2) AS cost_per_claim
FROM public.part_d_top_drugs_by_cost_mv
WHERE lower(brand_name) = 'vyndamax';
The snapshot
dataset_idcms-part-d-prescribers
snapshot_date2026-04-04
The JOINs
top_by_cost   = part_d_top_drugs_by_cost_mv   order by cost desc   -- Eliquis $19,876,304,796 leads
top_by_claims = part_d_top_drugs_by_claims_mv order by claims desc -- Atorvastatin 72,474,420 claims leads
brand_cost_share  = overview_mv.brand_cost / overview_mv.total_drug_cost   -- 204,355,041,438 / 226,740,902,131 = 90.13%
brand_claim_share = overview_mv.brand_claims / overview_mv.total_claims     -- 353,740,012 / 1,479,628,807 = 23.91%
generic vs brand split via lower(brand_name) = lower(generic_name)
The pipeline version
methodology_versionpart-d-costliest/v1

Reproduce this

Run the exact query against the frozen 2026-04-04.

-- Most expensive vs most-prescribed Medicare Part D drugs, 2024 — reproducible query. -- -- Source: CMS Medicare Part D Prescribers, "by Provider and Drug". -- Snapshot: cms-part-d-prescribers / data year 2024 / CMS release 2026-04-04. -- Base: public.cms_part_d_prescribers (~28.0M prescriber × drug rows). -- Reads: the frozen materialized views from -- supabase/migrations/20260612150000_part_d_prescribing_research_views.sql. -- The two top-drug views are each GROUP BY brand_name, generic_name over -- the 2024 base rows, ranked by cost and by claims respectively; the -- overview view carries the brand-vs-generic partition. -- -- The whole study is the contrast between two orderings of the same drugs. -- Audit observation: 2026-07-14. Source release remains 2026-04-04. -- audit-claim: costliest-drug-vector SELECT rank, brand_name, generic_name, claims, cost, round(cost::numeric / claims, 2) AS cost_per_claim FROM public.part_d_top_drugs_by_cost_mv ORDER BY cost DESC LIMIT 12; -- 1 Eliquis Apixaban 22,996,543 $19,876,304,796 $864 <- costliest drug; 12th by claims -- 2 Ozempic Semaglutide 9,930,484 $12,375,986,475 $1,246 -- 3 Jardiance Empagliflozin 10,654,599 $10,678,957,372 $1,002 -- 4 Mounjaro Tirzepatide 4,694,896 $5,825,295,741 $1,241 -- 5 Xarelto Rivaroxaban 5,888,827 $5,436,701,739 $923 -- 6 Trelegy … 5,725,949 $4,832,494,643 $844 -- 7 Trulicity Dulaglutide 3,816,376 $4,786,817,612 $1,254 -- 9 Humira(Cf) Adalimumab 437,072 $3,795,159,532 $8,683 <- giant bill on few claims -- 10 Revlimid Lenalidomide 217,318 $3,788,009,700 $17,431 <- specialty cancer drug -- 15 Vyndamax Tafamidis 98,896 $2,504,980,201 $25,329.44 <- specialty heart drug -- audit-claim: most-prescribed-drug-vector SELECT rank, brand_name, generic_name, claims, cost, round(cost::numeric / claims, 2) AS cost_per_claim, lower(brand_name) = lower(generic_name) AS is_generic FROM public.part_d_top_drugs_by_claims_mv ORDER BY claims DESC LIMIT 12; -- 1 Atorvastatin Calcium 72,474,420 $748,890,955 $10.33 generic -- 2 Amlodipine Besylate 49,241,894 $310,756,194 $6.31 generic -- 3 Levothyroxine Sodium 42,465,073 $456,346,665 ... generic -- ... (ranks 1-11 are ALL generic) -- 12 Eliquis Apixaban 22,996,543 $19,876,304,796 brand <- first brand on the list -- audit-claim: brand-generic-partition SELECT generic_claims, brand_claims, generic_cost, brand_cost, round(100.0 * generic_claims / total_claims, 1) AS generic_claims_pct, -- 76.1% round(100.0 * generic_cost / total_drug_cost, 1) AS generic_cost_pct, -- 9.9% round(100.0 * brand_claims / total_claims, 1) AS brand_claims_pct, -- 23.9% round(100.0 * brand_cost / total_drug_cost, 1) AS brand_cost_pct -- 90.1% FROM public.part_d_prescribing_overview_mv; -- generic + brand claims sum exactly to total_claims; generic + brand cost to -- total_drug_cost — the partition is by lower(brand_name) = lower(generic_name). -- audit-claim: vyndamax-example SELECT rank, brand_name, generic_name, claims, cost, round(cost::numeric / claims, 2) AS cost_per_claim FROM public.part_d_top_drugs_by_cost_mv WHERE lower(brand_name) = 'vyndamax';

Download the data: SQL · JSON · CSV

Cite this study

Citation-ready for researchers and AI.

Fonteum Research Bureau (2026). The Most Expensive Prescription Drugs in Medicare Part D. CMS Medicare Part D Prescribers, snapshot 2026-04-04. https://fonteum.com/research/most-expensive-medicare-part-d-drugs

Check the chain

The cited snapshot identifies the federal file. Where an attestation is present, its file hash can be re-derived; figures are not individually signed.

1
Snapshot
cms-part-d-prescribers · 2026-04-04
2
Supplied file hash
Not supplied
3
Snapshot attestation
Inspect supplied artifact
Figures are not individually signed · check it in Attest →

Federal source citations

  1. CMS Medicare Part D Prescribers · snapshot 2026-04-04 · federal source family · US-Government-Works

Fonteum LLC · June 12, 2026 · All figures trace to the frozen federal-data snapshot cited above.

What’s on file, by the numbers

Platform snapshot · 2026-07-30

13.4Mproviders & companiesProviders, organizations, owners, and facilities on file
26.2Msource-linked factsSource-linked field facts in the dated platform snapshot
90sources with dataDistinct snapshot source IDs with at least one positive record count
78fresh sourcesDistinct source IDs whose latest positive-data snapshot falls within the preceding 45 days
111sources integratedActive registry rows; integration does not establish a load
13state Medicaid jurisdictionsDistinct states represented in the state-exclusions serving table

Integrated, with-data, and fresh-observation counts are separate. No platform-wide source-completeness count is published. Completeness is source-specific and must be evaluated against the named source's expected scope. State coverage is a separate jurisdiction measure.

Source authority is record-specific

Use the issuer named on the record.

Fonteum spans federal, state, and global public publishers. A source page or returned record identifies its issuer and dataset where that metadata is available. A platform registry count does not assign every page to one authority or establish loaded, fresh, or complete coverage.

Browse source records and their stated limitations →

Reproducible by design

Inspect the evidence each published figure actually supplies.

Source and date

Research pages expose the named public file and observation date where those fields are available. Source-file SHA-256 coverage is separate; facts do not currently link deterministically to signatures.

Available derivation

Studies with a retained release and committed derivation link the SQL or method used. Other studies state the evidence and reproduction limits they actually have.

Daily observations

Dated table row-count observations can detect local drift. They do not imply that an upstream publisher released or Fonteum ingested new data that day.

Named medical review

Reviewed by Jennifer Montecillo, MD, medical reviewer. Non-practicing medical reviewer.

Read the full provenance and attestation methodology →

Request access