The ruler we never questioned.

The intuitive answer: what a QALY is worth.

Here's the answer almost everyone reaches for first.

A QALY — a year of full health — is valuable. Society should be willing to pay something for it. Ask what a year of healthy life is worth to people (through their taxes, their insurance, their stated preferences) and you get a number: a willingness-to-pay threshold. Set the line there, and "cost-effective" means "worth it, in society's eyes."

It's an appealing story. It makes the threshold a statement of values — of how much we care about health. And it's the way the threshold is almost always described in the press: "the price the NHS puts on a year of life."

Hold onto it, because we're about to watch it fall apart.

Why it breaks: the budget is fixed.

The willingness-to-pay story quietly assumes you can find the money — that if a QALY is worth £50,000 to society, the system can simply pay £50,000 for it.

But a tax-funded health system doesn't work that way. Its budget is fixed, set in advance, and already fully spent on things that are already helping patients. There is no reserve of unspent money waiting for a QALY worth buying.

So when a new technology is funded, the money doesn't come from nowhere. It comes from somewhere — and that somewhere is the rest of the health service. This single fact — the budget is fixed — is what breaks the willingness-to-pay answer. It doesn't matter how much a QALY is "worth" in the abstract if paying that price means taking the money out of care that was already producing health.

Spending displaces.

Follow the money. To fund a new drug, the system has to free up cash — by not doing something else it would otherwise have done. Those displaced activities weren't waste; they were producing health for other patients. So funding the new technology doesn't just add health. It also removes health, elsewhere, invisibly.

That gives us the real threshold. Every health system converts money into health at some rate — and at the margin, where the squeeze happens, there's a rate at which the last, least-efficient pound currently spent produces QALYs. Call the cost of one of those marginal QALYs k.

k is the true cost-effectiveness threshold — the opportunity cost of health spending. Not because it's what a QALY is worth to us — but because it's what a QALY costs us, in health given up, every time we spend. Adopt a technology whose ICER is below k and you buy health more cheaply than you displace it: the population gains. Adopt one above k and you pay more per QALY than the care you're displacing — you lose more health than you create.

The threshold isn't a measure of how much we value health. It's a measure of how efficiently our budget already produces it.

The displacement calculator.

You've decided to fund a new technology for a group of patients, at a total extra cost of £60,000. The left slider is its ICER — the price it charges per QALY. The right slider is k, the cost of a QALY in the care that £60,000 would otherwise have funded. Watch what the same money buys, both ways.

Slide the ICER past k and the net turns negative — that crossover, ICER = k, is the threshold. Everything to the right of it is a technology that looks like progress on its own but leaves the population worse off. That's the whole argument in one moving line.

Two QALYs lost for every one gained.

Now put real numbers on it. The most-cited empirical estimate of the NHS's marginal rate — the cost of one of those displaced QALYs — is around £13,000. Not £30,000. Not £20,000. About thirteen.

So take a technology approved at an ICER of £30,000 per QALY:

Read that again. Approving at £30,000 per QALY, in a system whose marginal QALY costs about £13,000, removes roughly two QALYs from other patients for every one it delivers. The patients who benefit are visible — named in the trial, championed by a charity. The patients who lose are invisible: they're whoever would have been treated by the services that got squeezed. But the arithmetic doesn't care who's visible.

Now you.

So where does NICE's number come from?

If the real threshold is the opportunity cost — around £13,000 for the NHS — then where does NICE's famous range come from?

The uncomfortable answer: not from either theory. The £20,000–£30,000 range wasn't derived from a rigorous willingness-to-pay study, and it wasn't derived from an estimate of opportunity cost. It emerged in NICE's early years as a rough working range and simply stuck — unchanged for more than two decades. It's often described as a willingness-to-pay figure, but it was never actually measured as one.

And here's where it gets sharp. The best empirical work on the NHS's opportunity cost puts the real threshold below NICE's range — nearer £13,000 than £20,000. If that's right, NICE has for years been approving technologies at prices that displace more health than they create.

Then, in April 2026, the threshold moved for the first time in over twenty years — but not toward the evidence. The UK government raised the standard range to £25,000–£35,000 per QALY, as an explicit act of industrial policy tied to a pharmaceutical trade deal. It was a ministerial decision, not a methods update, and it pushed the threshold further from the opportunity-cost estimate, not closer.

That's the real lesson of this screen. The threshold isn't a scientific constant waiting to be measured precisely. It's a policy lever — and which way it gets pulled depends on who's pulling.

The threshold isn't one number.

Even within a single system, there is no single threshold. NICE layers on adjustments, each one a deliberate value judgement:

• Severity modifiers. For more severe conditions, incremental QALYs are multiplied by 1.2 or 1.7 before the ICER is judged — in effect, a higher threshold for severe disease. Society choosing to pay more per QALY when the stakes are higher.
• Ultra-rare conditions. Highly specialised technologies for very rare diseases are judged against a far higher range — up to £100,000–£300,000 per QALY — reflecting a choice to pay a premium for treating the few.

These aren't economics; they're ethics, made explicit and bolted onto the number. The threshold encodes what a society values, not just what its budget can produce.

And step outside the UK and the picture fragments further. An explicit threshold is actually unusual: of 36 high-income countries surveyed, only around eight have one written into law or guidance. Germany's IQWiG rejects the idea of a single threshold altogether, comparing technologies within a therapeutic area instead. Reasonable systems don't just disagree about the number — they disagree about whether to have one at all.

Taking the opportunity-cost view seriously, what is the catch?

Why this matters for HTA

The threshold is where an appraisal stops being arithmetic and starts being politics — and where a manufacturer's dossier is most likely to lean on a number the reader hasn't examined.

• "Below the threshold" is a claim about a contested line, not a fact of nature. Knowing the threshold is a policy choice — recently moved on industrial grounds, and arguably above the true opportunity cost — tells you an ICER of £28,000 isn't safely "cost-effective." It's cost-effective against a line someone chose.
• The invisible losers never file a submission. Every approval above opportunity cost displaces care for patients who have no dossier, no charity, no name. Part of the assessor's job is to remember they exist when everyone in the room is focused on the visible beneficiaries.
• Modifiers are where values live. When a case turns on a severity multiplier or a rarity premium, the real debate is ethical, not technical — and it should be argued as such, not smuggled in as a modelling assumption.

A threshold is not the price of health. It's the price of everything you didn't buy instead. Read it that way and half the political noise around "access" resolves into a single question: what got displaced?

The threshold, in one breath.

• The threshold is not what a QALY is worth to us (willingness to pay). In a fixed-budget system it's what a QALY costs us elsewhere — the opportunity cost, k.
• Fund something with an ICER below k and the population gains health; above k and it loses health, because the displaced care was producing QALYs more cheaply.
• NICE's range (£20,000–£30,000 for decades, £25,000–£35,000 from April 2026) was never derived from theory; empirical opportunity-cost estimates for the NHS sit lower, around £13,000.
• There's no single threshold: severity modifiers, an ultra-rare premium (£100,000–£300,000), and wide international disagreement mean the "line" is plural, value-laden, and political.

The ICER asks what a technology costs per QALY. The threshold asks what that money would have bought instead. Cost-effectiveness is just those two questions, held side by side.

Next, the final piece of Module 7. We've now got the ICER (a technology's price per QALY) and the threshold (the price of what it displaces). One number combines them — and, unlike the ICER, it never loses its signs or flips between quadrants. That's net benefit.