The Nine Domains of HTA: Why Only Four Are Clinical

The nine domains of HTA cover not only clinical evidence but also economic, ethical, organisational, social and legal questions, and only four of the nine are classified as clinical.

That proportion is not a curiosity. The distinction between clinical and non-clinical HTA domains is now built into the way European HTA operates, and it explains why some parts of an assessment can be performed jointly at EU level while others remain with national authorities.

What is the HTA Core Model?

EUnetHTA developed the HTA Core Model as a way for agencies in different countries to produce assessments that could actually be shared and reused rather than rewritten from scratch each time. It did three things: it set out a standard structure of research questions, it pointed to accepted methods for answering them, and it established a common reporting format.

The structure runs from domains to topics to issues, with the issue as the most granular question an assessor actually answers. Each issue is then framed as an assessment element, which is the unit the model uses when it talks about importance and transferability, which matters later.

Version 3.0, published in 2016, is the version most people mean when they refer to the EUnetHTA HTA Core Model. It is worth knowing that the nine domains did not originate with EUnetHTA: they follow the nine core criteria set out by Busse and colleagues in 2002 in the best-practice guidance produced for the European Collaboration for Assessment of Health Interventions, which the Core Model then formalised into an ontology of questions.

What are the nine domains of HTA?

The HTA Core Model identifies nine domains:

The first four are the clinical domains of HTA and the remaining five are the non-clinical domains. That distinction has become particularly important in European HTA because it now determines which parts of an assessment can be conducted jointly across the EU and which remain primarily national.

The four clinical HTA domains

1. Health problem and current use of the technology — CUR

This domain establishes the problem being addressed: the disease or condition, the affected population, its natural history and burden, and how patients are currently managed. Before asking whether a new technology performs better, an assessment has to establish what it is being compared with and what problem it is supposed to solve.

2. Description and technical characteristics — TEC

The TEC domain describes the health technology itself: what it is, how it works, how it is delivered or used, and what infrastructure, equipment or expertise it requires. It provides the technical context needed to interpret the evidence that follows.

3. Safety — SAF

The safety domain examines the harms associated with the technology, including adverse events and other safety outcomes, particularly in comparison with existing alternatives.

4. Clinical effectiveness — EFF

Clinical effectiveness asks what the technology actually does for patients relative to relevant alternatives. This is where evidence from clinical trials, observational studies and evidence synthesis is used to examine outcomes such as mortality, morbidity, symptoms, functioning and quality of life.

The five non-clinical HTA domains

The other five domains move beyond clinical evidence.

5. Costs and economic evaluation — ECO

The economic domain asks what resources the technology consumes and what value those resources produce. Depending on the assessment system, that may involve cost-effectiveness analysis, cost-utility analysis, budget impact and other forms of economic evaluation.

6. Ethical analysis — ETH

The ethical domain considers questions that cannot be answered by an effect estimate or an ICER alone. Who benefits, who might be disadvantaged, does using the technology raise questions about autonomy, fairness, dignity or access, and does introducing it change how limited healthcare resources are distributed?

7. Organisational aspects — ORG

A technology can be clinically effective and economically attractive and still be difficult to implement. The organisational domain looks at what introducing it would mean for the healthcare system itself: staffing, training, care pathways, infrastructure, capacity, and changes in responsibilities between healthcare professionals or institutions.

8. Patients and social aspects — SOC

The social domain examines the experience and perspective of patients and others affected by the technology. It can include acceptability, access, burden on patients and carers, impact on daily life, and differences in how the technology affects particular social groups.

9. Legal aspects — LEG

The legal domain considers the regulatory and legal consequences of using the technology. These may involve professional responsibility, consent, data protection, intellectual property, liability, or legal restrictions on how and where the technology can be delivered.

Together, these five domains explain why HTA is much broader than asking whether a treatment works.

A full HTA does not always mean all nine domains

The HTA Core Model was never intended to be applied in exactly the same way to every technology. Earlier versions were issued as separate applications for pharmaceuticals, medical and surgical interventions, diagnostic technologies and screening technologies; version 3.0 brought all four into a single model, from which an assessor selects the relevant subset. A further application covers rapid relative effectiveness assessment, or rapid REA, and that version focuses on only four domains: CUR, TEC, SAF and EFF. In other words, the clinical four.

The same pattern shows up in what agencies actually produce. A survey of INAHTA member agencies mapping common HTA product types found that a full HTA report will always include the first seven domains, a mini-HTA the first five, and a rapid review the first four, with the ethical, social and legal domains showing the lowest inclusion rates across every product type. So a full HTA can extend across all nine domains where they are relevant, but in practice the last two are the ones most likely to be absent even from a full report. A relative effectiveness assessment deliberately does not attempt them at all.

That distinction anticipated something much bigger that would later happen at EU level.

How the nine HTA domains became part of EU law

Regulation (EU) 2021/2282 took the distinction between clinical and non-clinical HTA domains and made it legally operative. Article 2(6) defines clinical assessment as the compilation and evaluation of the available scientific evidence on a health technology, compared with one or more others, across four clinical domains:

• the health problem addressed and the current use of other health technologies addressing it;
• the description and technical characterisation of the technology;
• its relative clinical effectiveness;
• its relative safety.

Article 2(7) defines non-clinical assessment as the part of an HTA based on the five non-clinical domains:

• cost and economic evaluation;
• ethical aspects;
• organisational aspects;
• social aspects;
• legal aspects.

The division is essentially the one found in the EUnetHTA Core Model, but it now has consequences for how European HTA work is divided.

Joint Clinical Assessment covers the clinical domains

The logic behind EU Joint Clinical Assessment (JCA) is relatively straightforward. Clinical domains rest primarily on a scientific evidence base, and the same trial does not produce a different hazard ratio because the assessment is being read in France rather than Poland. The underlying evidence on relative effectiveness and safety can therefore be assessed jointly.

Non-clinical domains are different. Costs vary between healthcare systems, care pathways and staffing models differ, ethical priorities can differ, and legal frameworks certainly differ. So for technologies that fall within the JCA framework, the clinical evidence is assessed jointly at European level while Member States retain responsibility for national interpretation, non-clinical assessment, pricing and reimbursement.

A survival analysis does not change when it crosses a border. Whether a technology fits a hospital’s staffing model does, as does whether reimbursing it creates an equity problem given who has access to care in that country, and whether the legal framework permits a particular delivery model. Those questions do not travel nearly as well.

The HTA Core Model already recognised the transferability problem

The authors of the Core Model were dealing with this problem long before it became part of the EU regulatory framework. Version 3.0 attaches judgements about importance and transferability to individual assessment elements, and it describes transferability as low for information highly specific to a particular region, country or health system, while noting that even non-transferable information can still be useful beyond its place of origin.

What makes that section unusual is its candour. The model itself states that the division of elements into groups, along with the associated data on importance and transferability, drew criticism from users and should be treated as an experimental feature rather than a mandate. Sorting knowledge according to how well it travels between health systems is genuinely difficult, and the model did not pretend otherwise.

What HTA reports actually cover

Here is where the picture becomes less tidy. The nine domains describe what a comprehensive assessment can cover; what actual assessments contain is another matter.

A 2025 scoping review of HTA methods for complex interventions gives one of the clearer recent snapshots. It screened over ten thousand references and analysed 35 records from twelve countries, of which 19 were methodological guidance and 16 were accounts of HTA carried out in practice. More than 94 per cent of the reports covered costs and economic evaluation, clinical effectiveness and organisational aspects. Emphasis was more moderate, around 75 per cent, on the technology’s description and technical characteristics, its current or intended use, safety, social aspects and patient preferences. Ethical and legal considerations came last.

The same review made a sharper observation. With the exception of the technology’s description, its current or intended use, economic impact and organisational aspects, the domains appeared more often in methodological guidance than in actual practice; safety, ethical, social and legal aspects were among them. Clinical evidence was the one domain always present in both. That gap between what the framework prescribes and what gets written is one of the most important things to understand about the nine-domain model.

Two caveats are worth keeping. The review looked specifically at complex interventions, such as screening programmes and cell and gene therapies, rather than at HTA as a whole, and the sample of actual assessments was small. Its authors note that on this point complex interventions are largely treated the same way as other technologies, and the pattern matches what the INAHTA product-type survey found a decade earlier, but these are percentages from a handful of documents rather than a census of the field.

Why the ethical domain is often missing

The ethical domain has its own literature on this problem. A Basque Office for Health Technology Assessment study published in 2018 randomly retrieved 101 HTA reports from the INAHTA database covering 2004 to 2009 and found explicit ethical analysis in 30 per cent of reports from 2004 to 2005 and 43 per cent from 2006 to 2009, a difference that was not statistically significant. A descriptive review of Canadian HTA reports by DeJean and colleagues put the figure at 17 per cent, and concluded that where ethics was addressed it was usually addressed superficially. More interestingly, that review found implicit ethical questions running through reports that made no claim to include ethical analysis at all.

That gets close to the real problem. Ethics in HTA is difficult partly because ethical questions do not sit neatly inside one section of a report; they run through the entire assessment. Which outcomes count? Whose perspective matters? How much inequality is acceptable? Does improving average population health justify making access worse for a particular group? These are ethical questions even when nobody labels them as such.

Several explanations have been proposed for the weak representation of formal ethical analysis: there is no single agreed methodology for integrating ethics into HTA, ethicists are still outsiders in many assessment teams, and ethical questions are pervasive rather than sectional. The Basque authors added two more prosaic ones, a lack of budget for hiring the relevant experts and a lack of the relevant skills among HTA staff, and argued that one-off measures will not shift the underlying tendency towards reductionism. The combined effect is that ethical questions resist being confined to a chapter of their own while also risking disappearing entirely if nobody is explicitly responsible for them.

The methodological gap between clinical and non-clinical HTA

Something similar applies across the five non-clinical domains. Clinical effectiveness has decades of methodological infrastructure behind it: systematic review methodology, risk-of-bias tools, meta-analysis, GRADE, and established approaches to interpreting clinical outcomes and uncertainty. Economic evaluation also has a substantial methodological literature.

Organisational, social, ethical and legal analysis is less standardised. That does not make those domains less important; it makes them harder to do consistently. And that distinction matters more under the new European HTA framework, not less.

Why the five non-clinical domains matter more under EU JCA

There are two ways to read the European reform. The first is optimistic: instead of multiple countries repeatedly assessing much of the same clinical evidence, a joint clinical assessment reduces duplication, and national agencies can concentrate more of their capacity on the areas that genuinely depend on national context: economics, organisation, ethics, social consequences and law. That is exactly the kind of division of labour the nine-domain framework seems to support.

But there is another reading worth considering. Joint clinical assessment gives Member States a common treatment of the four domains that HTA systems have historically been best equipped to assess, and leaves national systems with greater responsibility for the five domains where methods tend to be thinner, expertise scarcer and reporting less consistent. The domains that travel worst are also some of the domains HTA has historically handled least consistently.

That creates a test for the new system. Freed capacity could genuinely move into stronger economic, organisational, ethical, social and legal assessment, and some agencies will do exactly that. Or the non-clinical domains could remain thin while the clinical side becomes increasingly standardised, widening the methodological gap between the two halves of HTA.

Which way it goes will become visible in national assessments over the coming years. It is one of the more interesting things to watch in European HTA, and it receives considerably less attention than JCA timelines and procedural requirements.

The nine domains are more than a checklist

The nine HTA domains are easy to memorise as a list, and that misses the point. The HTA Core Model was an attempt to describe the full set of questions that might matter when a healthcare system decides what to do with a technology.

Clinical effectiveness and safety are central, but they are not the whole assessment. A treatment can work and still be poor value. It can be cost-effective and still be impossible to implement. It can be clinically and economically attractive while creating an ethical, social or legal problem. That is why there are nine domains rather than two, and it is why understanding the five non-clinical domains may become increasingly important as EU Joint Clinical Assessment takes responsibility for more of the clinical side.

For readers new to the field, our introduction to what Health Technology Assessment (HTA) is covers the basics, and the article on the definition of HTA examines the words the field agreed on. The nine domains are covered in full in the HTA Course, including the five that many introductions mention once and then quietly drop.