The Definition of Health Technology Assessment: How HTA Took 44 Years to Define

What makes it stranger is that this is a discipline whose central promise is explicitness: publish your methods, state your assumptions, let anyone check the reasoning. For most of its history, HTA could not point to one agreed sentence describing what it was.

The story of the HTA definition is therefore more than a question of terminology. It shows how the field’s understanding of evidence, value and decision-making has developed, and why defining health technology assessment proved so difficult.

HTA definitions before 2020

There was never a shortage of definitions of health technology assessment. There were too many.

David Banta, among the first to use the term in the 1970s, described early technology assessment as a form of policy research examining the short- and long-term consequences of applying a technology, the goal being to give policy makers information on the alternatives open to them. Read it today and something is missing: no value, no comparator, no cost-effectiveness, no reimbursement. The discipline that grew out of that sentence is not the one the sentence describes.

In the 1990s, Goodman defined healthcare technology assessment as the systematic evaluation of the properties, effects or impact of medical technology, with the purpose of informing technology-related policy making in health care. Closer, but still broad.

Then in 2005 Draborg and colleagues did something more useful than proposing another definition: they compared how HTA was defined internationally with how it was actually practised, and concluded that the two did not match. The field described itself as a multidisciplinary field of policy analysis studying the medical, social, ethical and economic consequences of healthcare interventions, while real assessments were frequently far narrower than that. The gap between the stated scope and the delivered scope was documented twenty years ago, and anyone who has read a rapid assessment recently will recognise it.

Most practitioners eventually converged on the version posted in the HTA Glossary, a joint initiative of INAHTA and HTAi with other partner organisations. It described HTA as the systematic evaluation of the properties and effects of a health technology, addressing the direct and intended effects of that technology as well as its indirect and unintended consequences, and aimed mainly at informing decision-making about health technologies. An accompanying note added that the work is carried out by interdisciplinary groups using explicit analytical frameworks that draw on a variety of methods. Perfectly serviceable. Meanwhile, ISPOR, the WHO, EUnetHTA, HTAsiaLink and RedETSA each used their own versions.

Why a shared definition of HTA mattered

Here is the honest part: the central concepts of HTA were consistent across these definitions. Nobody was genuinely confused about what HTA was for.

The problem was different. The existing definitions were technical, built out of words and phrases that do not translate easily out of English, and they were neither memorable nor aspirational. For a field expanding into health systems where English is not the working language, and where HTA has to be explained to finance ministries rather than to fellow analysts, that is a real constraint. There was also no single definition that the broader HTA community had ever been formally consulted on.

How the 2020 HTA definition was developed

In 2018, the Board Chair of INAHTA and the President of HTAi created a feasibility task group to gauge interest in developing a new and internationally accepted definition. It was co-chaired by Brian O’Rourke for INAHTA and Wija Oortwijn for HTAi, and invitations to join went to EUnetHTA, HTAsiaLink and RedETSA, all of which accepted. The group met once by teleconference in January 2018 and reached consensus not only that a new definition was feasible, but that the broad international collaboration itself had value for the HTA community, since nothing on that scale had been undertaken in this space before.

The full joint task group was then convened under the same two co-chairs, with representatives of EUnetHTA, HTAsiaLink, RedETSA and the HTA Glossary Committee participating. Further invitations followed: ISPOR became a member and the WHO acted as an observer.

Its first order of business was not the definition itself. It was a set of guiding principles for how to write one, including the requirement that the result be simple, short and memorable. The group then met on several occasions from May 2018 until September 2019, completing several jointly agreed rounds of consultation and review.

From consultation to consensus

Version 1 was drafted by the task group and forwarded to the leadership and approval authorities of the participating organisations for early input, which the group then discussed and integrated. Version 2 was the one circulated for open consultation with the broader HTA community, electronically, through the membership and contact lists of the member organisations. It drew 172 responses, which the group sorted into four groupings:

That last point is unsurprising. Value is context-related and largely conditioned by the perspective taken in the assessment, which is precisely why the field argues about it.

The task group incorporated the consultation feedback into version 3 and sent it back to the participating boards for approval. Most approved it, some offered further suggestions for improvement, and the resulting version 4 is the one that carried. It was approved by all participating networks and organisations, and by leadership from the EuroScan International Network. It appeared in June 2020 in the International Journal of Technology Assessment in Health Care and, in parallel, in Value in Health.

The 2020 definition of health technology assessment

The agreed definition runs to two sentences rather than one. The first describes HTA as a multidisciplinary process that uses explicit methods to determine the value of a health technology at different points in its lifecycle. The second gives the purpose: to inform decision-making in order to promote an equitable, efficient and high-quality health system. Four accompanying notes carry the clarifying detail.

Several choices inside those two sentences are worth understanding, because they explain things that might otherwise look arbitrary.

Why particular words were chosen

Why the definition uses “determine” rather than “assess”

In the first sentence, the word assess was replaced by the word determine, for two reasons. It avoids using the same word in the definition as in the term being defined. And it avoids duplicating concepts once the definition is translated: in Spanish, assess the value becomes evaluar el valor. This is what it looks like when a definition is written to survive being translated.

“Best available evidence” rather than “all available evidence”

Note 2 exists because of the consultation. It was created to hold some of the words that were stripped out of the definition itself in the drive to make it shorter, which is worth knowing before treating the notes as optional extras. Within that note, all available evidence was then changed to the best available evidence, bringing it into line with a position document published by EUnetHTA. It sounds like a small edit, and it is not: the two phrases imply different obligations for an assessor.

Why the HTA definition does not list every decision it can inform

The definition does not specify the different levels of decision-making where HTA can play a role, and the stated reason is the guiding principle of keeping it simple and short. The phrase about informing decision-making is meant to reflect the full range of decision contexts in low-, middle- and high-income countries: formulary coverage or reimbursement decisions, including disinvestment; clinical practice guideline development; defining emergency kits, disaster planning, benefit packages and essential medicine lists; medical device and equipment procurement planning; negotiating prices for health technologies; and other contexts at national, regional or local level, including hospitals. A fixed list would have privileged some of those and quietly demoted the rest.

Why “equitable, efficient and high-quality” are not defined

The closing phrase of the definition is left unspelled out for a related reason. Health systems of democratic societies often share these values and principles, but they can apply them differently, so the concepts have to be defined in the context of the health system under consideration, including how they are applied by the HTA agencies involved and how decision makers interpret them. The task group pointed to the HTA Glossary entries for equity, efficiency and quality as a starting point.

The group also chose not to reopen the definition of health technology itself. Several respondents proposed changes; the task group kept the wording already in the HTA Glossary and referenced it in Note 1, and passed the suggested edits, mainly aimed at aligning the concept with the intervention-based vocabulary used in low- and middle-income settings, to the Glossary Committee for consideration.

The notes to the HTA definition are where the substance sits

People often quote the two sentences and skip the notes. That is a mistake, because Note 3 in particular is where many of the most important questions in HTA live.

Note 1: What counts as a health technology?

Note 1 takes the scope of health technology straight from the HTA Glossary. A health technology is an intervention developed to prevent, diagnose or treat medical conditions, to promote health, to provide rehabilitation, or to organise healthcare delivery. The intervention can be a test, device, medicine, vaccine, procedure, programme or system.

Note 2: How HTA should be conducted

Note 2 sets out how the work is expected to be done. The process is:

• formal;
• systematic;
• transparent;
• conducted with state-of-the-art methods;
• built on consideration of the best available evidence.

The fourth item is the one most often dropped when the note is quoted. Methods are expected to be current, not merely explicit, which is a standard an assessment can fail while remaining perfectly transparent about how it failed.

Note 3: What does “value” mean in HTA?

Note 3 does two things that are easy to miss.

First, it says the dimensions of value may be assessed by examining the intended and unintended consequences of using a health technology compared with existing alternatives. The comparator sits inside the definition of value itself, which is why a technology cannot be valuable in the abstract.

Second, it gives the dimensions that value often includes. The word is often, not always, so the list illustrates rather than closes:

That is an admission, written into the field’s own definition, that value is not simply a property sitting inside a technology waiting to be measured. It depends on who is asking and why, and every disagreement between a manufacturer and an agency, along with many disagreements between two agencies, can be traced back to that idea.

Note 4: HTA across the technology lifecycle

Note 4 confirms that HTA can be applied at different points in the lifecycle of a health technology:

This note also answered an objection raised during the consultation, that the definition might exclude activities and institutions spanning the appraisal, decision and selection space. The task group held that the point was already sufficiently addressed, and noted that HTA is a process with several interrelated components: governance and structure, scoping, assessment, appraisal and implementation, and monitoring, as set out in the ISPOR HTA Council working group report on good practices in HTA. HTA is therefore not restricted to a single reimbursement decision at one point in time.

Criticism of the new HTA definition

Not everyone was pleased with the result. In 2022, Anthony Culyer and Don Husereau published a comment in the same journal, and their argument had two parts.

On process, they held that the development had been ill-described: the reader is not told how the organisational representatives were selected, how often the group met, what was deliberated, where members disagreed, or how differences of opinion were settled, which they considered hard to square with a community that treats transparency as a core value. On the outcome, their objection was lexicographic. A definition ought to differentiate the entity being defined from other entities, and this one, in their reading, states traits that are true of HTA without being unique to it: other activities are interdisciplinary and use explicit methods too. They added that examples of HTA’s use had been embodied in the definition, which risks precluding uses not mentioned.

That second point sits directly against the task group’s own reasoning. The group treated the absence of a list of decisions as a virtue; the commentators read the closing clause about an equitable, efficient and high-quality health system as exactly the kind of embedded example that narrows a definition instead of delimiting it. Both positions are defensible, and a reader who has followed the drafting history can now see why.

The task group replied in the same journal and rejected the criticism. It argued that the commentators were working from a narrow and outdated understanding of contemporary HTA methods, processes and practices; that the guiding principles and the drafting process had been communicated to consultees, and the open consultation was open to patients, clinicians, industry and academics alike; and that lexicographic objections of this kind had already been raised during that consultation and weighed during drafting. Details of individual methods and disciplines were left out deliberately, because those vary between health systems and between technologies. Transparency of method to the decision maker does not.

Both pieces are short and both are worth reading. The exchange is a better introduction to how this field argues than almost any summary of it.

The definition of early HTA came even later

The vocabulary of health technology assessment is still being built. The internationally agreed definition of HTA dates from 2020, while the agreed definition of early HTA was published only in 2025. An HTAi working group produced it through a five-stage process that included a two-round Delphi survey, with 133 respondents in the first round and 99 in the second.

In the second round, 86 of the 99 panellists, 87 per cent, agreed or strongly agreed with it.

Two details are worth holding onto. The technology may be conceptual or already in existence, so early HTA is not confined to ideas on paper. And the purpose is to inform decisions, which is a deliberately broader framing than the binary go or no-go tool that some earlier definitions described.

The exercise was necessary because early HTA, early advice, early dialogue, early awareness and development-focused HTA had sometimes been used interchangeably, even though they do not mean the same thing.

So the shared language of this discipline is not ancient and settled. Large parts of it were written in the last few years, by people who are still working, after arguments that remain on the record. Anyone learning HTA now is not arriving late to a finished subject.

Why the definition of HTA still matters

A definition can look like a minor academic issue. In HTA, it is not. The 2020 definition establishes several ideas that shape how the field works today.

Those points are much more important than memorising a single sentence, because they explain why two organisations can examine the same technology and still reach different conclusions without either assessment necessarily being wrong.

For readers new to the field, our introduction to what Health Technology Assessment (HTA) is covers the basics. This article goes one level deeper: into how the definition itself was built, why particular words were chosen and what those choices tell us about modern HTA. That is also where the HTA Course begins, with the 2020 definition taken apart word by word, including the words the task group argued about.