A health technology assessment weighs a new therapy's clinical effectiveness, cost-effectiveness measured in cost per QALY, and budget impact against a chosen comparator to decide whether it gets reimbursed. Outcomes differ by country because each body sets its own cost-effectiveness threshold, comparator, and weight on non-clinical evidence.
What is a health technology assessment?
A structured comparison of a new therapy against what a health system already funds, done by a national or regional body before it decides whether to pay for the therapy at all. The assessment is not a regulatory approval. A drug can clear safety and efficacy review and still fail an HTA, because the questions are different: not whether it works, but whether it works well enough, for whom, and at what price relative to the next best option.
That last clause is where most of the argument lives. An HTA body is not asking the manufacturer's question, which is usually framed against placebo. It is asking a health system's question, framed against whatever patients are currently getting, and the gap between those two framings is where a value dossier is won or lost.
The process is also slower and more procedural than approval, running through a submission, a clarification round, an independent modelling check, and often a public appraisal document before a decision is published. Manufacturers who treat it as a formality after approval tend to discover the gap in that timeline the hard way.
- 01
Map the comparator
Confirm what the target market actually treats as current standard of care.
- 02
Stress-test the model
Challenge your own utility values and extrapolation before a reviewer does.
- 03
Size the budget honestly
Project total population spend, not just the per-patient ratio.
- 04
Sequence submissions
Order country filings around where the dossier is strongest first.
What evidence gets weighed?
Three streams, evaluated separately before they are combined. Clinical effectiveness comes first: how much better is the therapy on outcomes that matter to patients, not just on a surrogate endpoint that was convenient to measure in a shorter trial. Cost-effectiveness comes second, built from a model that projects clinical benefit and cost over a lifetime horizon, not just the trial period.
Budget impact is the third stream and the one manufacturers most often underweight. A therapy can be cost-effective per patient and still be rejected, or reimbursed with restrictions, because the total spend across the eligible population would displace too much of the health system's budget in the near term. HTA bodies reason about both the per-patient ratio and the aggregate bill, and a dossier that only argues the first has not answered the question actually being asked.
Evidence quality is weighed alongside all three streams, not after them. A randomised trial with a long follow-up carries more weight than a single-arm study with an indirect comparison, and a reviewer will discount the whole submission if the underlying trial design leaves the effect size uncertain. Weak evidence on any one stream tends to drag the other two down with it, since the model inherits whatever uncertainty the clinical data started with.
How does the QALY logic work?
A quality-adjusted life year combines length of life and quality of life into one unit, so that a treatment adding two years at reduced quality can be compared against one adding one year at full quality. Cost per QALY gained is then compared to a threshold the body treats as its rough willingness-to-pay ceiling, though few bodies publish that number as a hard rule.
The QALY model is only as good as its inputs, and this is where reviewers spend the most time. Utility values (how quality of life is scored), discount rates, and the extrapolation from a trial's follow-up window to a lifetime horizon are all modelling choices, and a manufacturer's preferred assumptions on each one tend to look generous next to a reviewer's preferred assumptions. The committee's job is largely to referee that gap.
Severity and rarity often adjust the threshold rather than sit outside it. A therapy for a rare, severe condition with no existing treatment is frequently judged against a higher cost-per-QALY ceiling than one competing in a crowded category, on the reasoning that the health system's alternative to funding it is doing nothing at all. That adjustment is itself a policy choice, and it is one more reason the same ratio can pass in one country and fail in another.
| Regulatory approval | Health technology assessment | |
|---|---|---|
| Core question | Is it safe and does it work | Is it worth what it costs, relative to the alternative |
| Comparator | Usually placebo | Current standard of care in that market |
| Outcome | Approved or not, largely uniform across markets | Reimbursed, restricted, or rejected, and it varies by country |
Why do outcomes differ by country?
Because the inputs a country plugs into the same logic are not the same inputs. Cost-effectiveness thresholds vary with a country's income and how it values a QALY at the margin. Comparator choice varies with what a country's health system already funds as standard of care, which is not identical from one market to the next. A therapy compared against an outdated standard clears more easily than the same therapy compared against a newer one.
Willingness to weight non-clinical evidence also varies. Some bodies give real-world evidence and patient-reported outcomes meaningful weight in a borderline case; others hold close to randomised trial data and treat everything else as supportive at best. None of this is arbitrary. Each choice reflects the same underlying trade-off, of how a fixed health budget should be allocated, worked out from different starting conditions.
Governance structure adds a further layer. Some bodies issue a single binding reimbursement decision; others produce a recommendation that regional or hospital-level payers then apply with local discretion, which means a national HTA outcome does not always translate into uniform access even within one country. A manufacturer planning a launch sequence has to read the governance model as closely as the methodology, since a favourable national appraisal can still stall at the point where a regional budget holder decides how to implement it. Getting that sequencing right is really a question of market access strategy, decided alongside the dossier rather than after it clears review.
How do manufacturers prepare a value dossier?
By building the case for the reviewer's question rather than the trial's question. That starts well before submission, with payer interview research that surfaces which comparator a market actually treats as standard of care and which endpoints its committee weighs most heavily, the kind of ground-truth Nextyn's payer interview research is built to surface before a dossier is locked.
The dossier itself has to do three things at once: present the clinical effectiveness case cleanly, defend the cost-effectiveness model's assumptions against the obvious challenges a reviewer will raise, and size the budget impact honestly rather than optimistically. None of this happens in isolation from the pricing and sequencing decisions that sit around it, and it is usually decided together with the dossier, not after it.
Timing matters as much as content. A dossier finished after the pivotal trial reads, but built without early scoping of a market's likely comparator and threshold, tends to arrive with the wrong model already built, which is expensive to rebuild under a clarification deadline. Teams that scope the payer landscape while the trial is still running submit a dossier that answers the actual question on the first pass, rather than the one they assumed would be asked.