Micros · Direct oral anticoagulants

One molecule, three prices

When apixaban lost exclusivity, England's price fell 95% in twelve months. Australia's fell 25% on a statutory timetable. America's kept rising, until federal negotiation, not competition, delivered the first cut. Same tablet, three systems.

In June 2023, apixaban, the anticoagulant sold as Eliquis, cost the NHS in England 89.5p a tablet, and at over £400m a year it was one of the largest single medicine bills in English primary care. Two years later the same tablet cost 4.4p. Over the same period the average US pharmacy's acquisition price rose from $8.97 to $9.69, roughly 170 times the English price for the same molecule.

Loss of exclusivity is usually modelled as a single number: a percentage erosion applied at a point in time. Watching one molecule cross the patent cliff in three pricing systems at once shows why that number is not a property of the drug. It is a property of the system.

Apixaban: price per market, indexed to June 2023 = 100

England £/tablet (dispensed) · US $/tablet (NADAC acquisition, brand) · Australia A$/prescription (PBS), monthly

England United States Australia

Sources: NHSBSA English Prescribing Dataset (actual cost ÷ quantity); CMS NADAC (national average drug acquisition cost, brand NDCs); PBS date-of-supply statistics (total cost ÷ prescriptions). England series to June 2025. Analysis: Altima Partners.

England: competition did the work

The English price fell in steps, and the steps were fast. From a stable 89p a tablet, the average dispensed price dropped to 74p in July 2023 as the first generics arrived, 26p by October, 8.7p by January 2024 and 4.7p by April: 95% erosion inside twelve months, settling at 4.4p by mid-2025. (The brief late-2024 rebound visible in the chart is consistent with the temporary price concessions the NHS grants when generic supply tightens; competition then resumed.)

The fiscal consequence: in the twelve months to June 2023 England spent £404m dispensing 8.4 million apixaban prescriptions. In the twelve months to June 2025 it spent £50m, on 9.8 million prescriptions. Spending fell 88% while volume grew 17%. That is roughly £354m a year released from one molecule, at no cost to access.

Same molecule, same month: 4.4p a tablet in England, $9.69 in the United States, about 170 times more.

United States: the price only went up

The US series is the mirror image. NADAC, the national average price pharmacies actually pay to acquire the drug, records brand apixaban at $4.30 a tablet in 2013, $7.98 in 2021 and $9.69 by 2025: a price that more than doubled across the product's life, with an increase almost every January. Patent litigation and settlement agreements have kept generic apixaban off the US market; through July 2026 the NADAC file still contains not a single generic listing.

Then, on 1 January 2026, the acquisition price fell 43% overnight, from $9.69 to $5.52. That is the day the first prices negotiated under Medicare's Drug Price Negotiation Program took effect, and apixaban was on the first list of ten drugs. The first meaningful price cut in the molecule's thirteen-year US history came not from generic competition but from statute, and even the negotiated price is still roughly 100 times the English generic price.

Australia: erosion by timetable

Australia's PBS sits between the two. Prices there fall through price disclosure: suppliers report the discounts they actually give wholesalers, and the reimbursed price is cut to match on fixed dates. The result is a staircase: a 4% step in April 2023, then a 25% step in August 2024 as post-generic discounting worked through the system, with gentle drift between steps. Net decline over five years: about 19%, delivered slowly, predictably, and without the volatility of the English market.

Three systems, three outcomes

Apixaban price by market, June 2023 versus latest

MarketMeasureJun 2023Latest ChangeWhat moved it
England£ / tablet0.895 0.044 Jun 25 −95%Generic competition
United States$ / tablet (NADAC)8.97 5.52 Jul 26 −38%Medicare negotiation; no generic yet
AustraliaA$ / prescription89.45 75.44 Jun 26 −16%PBS price disclosure

Note: units differ by market and are not directly comparable levels; the indexed chart above puts all three on a common base. US change occurred entirely on 1 January 2026.

Why this matters for forecasting

Revenue forecasts and budget impact models both need an answer to the same question: what happens to price and volume when exclusivity ends? The standard approach applies a generic erosion assumption, often a single global percentage borrowed from a published average. The apixaban data shows three ways that goes wrong:

The erosion rate is a system property, not a molecule property. The same tablet lost 95% of its price in England, 16% in Australia and, for two and a half years, nothing at all in the US. A forecast that applies one number across markets is not simplifying; it is wrong in every market at once.

The shape matters as much as the depth. England's erosion was front-loaded: three-quarters of the eventual price fall happened within six months of entry. Australia's arrived in steps over years. A net-present-value calculation, or a payer's five-year budget line, moves materially depending on which shape you assume.

Policy can move prices where competition cannot. The largest single-day price change in this dataset, the US cut of 43%, happened by statute, in the one market with no generic on sale. Forecasts built for US assets now need a negotiation scenario alongside the traditional patent cliff, with a different timetable and a different endpoint.

Nor is apixaban a one-off. Rivaroxaban, the other big oral anticoagulant, repeated the pattern a year later: England's price fell 95% in the twelve months to June 2025, while Australia's per-prescription cost halved over five years on the disclosure timetable. The molecule changes; the system signatures do not.

Every figure on this page comes from public data: England's prescribing dataset, the US NADAC file, Australia's PBS statistics. What it takes is assembling them at molecule level, aligning the timelines, and reading the price series the way a payer will.

Erosion curves for your forecast

We supply loss-of-exclusivity analogues from the market you are forecasting: real price and volume series, matched by molecule type and system. We build them into revenue and budget impact models. Fixed fees, senior-built.

Talk to us about your asset