Loading field map
Health technology assessment (HTA) began as a practical question for publicly funded health systems: which drugs, devices, and procedures should the state pay for? The question demanded systematic evidence, but it also demanded a method for turning evidence into a decision. From the start, HTA has been a field defined by methodological disagreement. Different schools of thought have offered competing answers about what counts as value, whose preferences matter, and how technical analysis should relate to political deliberation. The history of HTA is not a story of steady refinement toward a single correct method; it is a story of persistent pluralism, with each framework narrowing, extending, or challenging the ones that came before.
The earliest systematic frameworks for HTA came from welfare economics. The Cost-Benefit Analysis (CBA) School, which emerged in the 1970s, treated health technologies like any other public investment. CBA required analysts to monetize all consequences—lives saved, pain avoided, productivity gained—and compare them against costs in a single monetary metric. If benefits exceeded costs, the technology was worth adopting. CBA had a clear theoretical foundation in welfare economics, but it ran into a practical wall: putting a dollar value on a year of life or on relief from suffering was ethically contentious and methodologically fragile. Health decision-makers found CBA's monetization requirement hard to defend in public, and the framework gradually lost ground in health-specific settings.
The Cost-Effectiveness Analysis (CEA) Paradigm responded to CBA's monetization problem by narrowing the metric. Instead of converting everything into dollars, CEA measured health outcomes in natural units—life-years gained, cases detected, blood-pressure points reduced. Costs were still measured in money, but the result was a ratio (cost per life-year gained) that avoided putting a price on health itself. CEA was more palatable to clinicians and policymakers, but it had a limitation: it could only compare technologies that affected the same natural outcome. A cancer drug and a heart-disease program produced different natural units and could not be ranked against each other.
The Cost-Utility Analysis (CUA) School, which took shape in the 1980s, solved that comparability problem by extending CEA rather than replacing it. CUA introduced the quality-adjusted life year (QALY), a composite metric that combined length of life with a preference-based weight for health-related quality of life. By expressing every health outcome in QALYs, CUA made it possible to compare any two technologies on a single scale, regardless of the disease or body system involved. The QALY became the signature tool of HTA, and agencies such as the UK's National Institute for Health and Care Excellence (NICE) built their appraisal processes around cost-per-QALY thresholds. CUA did not reject CEA; it absorbed CEA's logic and generalized its metric. Yet the single-metric approach also attracted criticism. Reducing health to a single number, critics argued, erased dimensions of value that patients and communities cared about—fairness, autonomy, rarity of disease, or the severity of the condition being treated. That critique opened the door for the value-pluralist schools that followed.
By the early 2000s, a new generation of frameworks began to challenge CUA's dominance. These schools did not reject efficiency analysis outright, but they argued that HTA should incorporate multiple kinds of value, not just QALY maximization.
The Multi-Criteria Decision Analysis (MCDA) School offered the most direct technical alternative to CUA. MCDA frameworks ask decision-makers to specify a set of criteria—clinical benefit, cost, equity, feasibility, severity of disease, and so on—and then weight each criterion according to its importance. Technologies are scored against each criterion, and the weighted scores are aggregated into a single ranking. MCDA differs from CUA in a crucial way: it does not assume that health-related quality of life is the only relevant dimension of value. In practice, MCDA has been used in settings where CUA's single-metric threshold seemed too narrow, such as orphan drug assessments and hospital formulary decisions. Yet MCDA has also faced criticism: the choice of criteria and weights can be arbitrary, and different weighting methods can produce different rankings from the same data. The reproducibility debate between MCDA and CUA remains unresolved.
The Deliberative Health Technology Assessment School took a different path. Instead of building a more comprehensive technical model, deliberative HTA argued that the legitimacy of an assessment depends on the process by which it is produced. Stakeholders—clinicians, patients, industry representatives, policymakers—should deliberate together, discuss evidence, and arrive at a judgment through reasoned dialogue rather than through a pre-specified algorithm. Deliberative HTA does not reject CUA or MCDA as inputs; it insists that the final decision should emerge from group discussion, not from a formula. Agencies such as the Canadian Agency for Drugs and Technologies in Health (CADTH) and some European HTA bodies have incorporated deliberative elements into their appraisal committees. The deliberative school challenges the technocratic assumption that HTA is primarily a technical exercise. Its weakness is that deliberation can be slow, resource-intensive, and vulnerable to capture by the most vocal participants.
The Patient-Centered Health Technology Assessment School emerged alongside the deliberative movement but with a different emphasis. Patient-centered HTA argues that the evidence base for assessments should include patient-reported outcomes, real-world data, and qualitative accounts of lived experience. It also insists that patients should have a role in setting research agendas, selecting outcomes, and interpreting results. Organizations such as the Patient-Centered Outcomes Research Institute (PCORI) in the United States have institutionalized this approach, funding studies that prioritize patient-identified questions and that use participatory research designs. Patient-centered HTA overlaps with deliberative HTA in its concern for stakeholder involvement, but the two schools are not identical. Deliberative HTA focuses on the decision-making process; patient-centered HTA focuses on the content of evidence and the framing of research questions. They can complement each other—a patient-centered evidence base can feed into a deliberative committee—but they can also pull in different directions when patient priorities conflict with the efficiency logic of CUA.
All six frameworks remain active today, and no single school has displaced the others. CUA, anchored by NICE and similar agencies, is still the dominant framework for national reimbursement decisions in many countries. Its strength is that it produces a clear, comparable metric that can be linked to a budget constraint. Yet even within CUA-dominant agencies, the newer schools have reshaped practice at the margins. NICE now incorporates patient-reported outcomes and has introduced a severity modifier that departs from strict QALY maximization. MCDA has been adopted by the European network EUnetHTA for certain types of assessments, and deliberative processes are standard in many appraisal committees.
What the leading frameworks agree on is that HTA must be systematic, evidence-based, and transparent. They disagree on what counts as evidence, whose values should be reflected in the assessment, and how much discretion decision-makers should have. CUA assumes that a single preference-based metric can capture social value; MCDA assumes that value is inherently multi-dimensional and must be weighted explicitly; deliberative HTA assumes that value is constructed through dialogue; patient-centered HTA assumes that value is best understood from the perspective of those who experience the technology. These are not merely technical disagreements about which formula to use. They are competing philosophical commitments about the nature of value and the proper role of expertise in democratic decision-making.
The Credibility Revolution in health economics—the broader movement toward transparent research practices, pre-registration, and sensitivity analysis—has influenced all HTA schools. Every framework now faces pressure to make its assumptions explicit, to test the robustness of its conclusions, and to disclose conflicts of interest. This cross-cutting development has not resolved the disagreements between schools, but it has raised the standard of evidence that each school must meet.
Health technology assessment is not a field that has converged on a single method. It is a field that has learned to manage methodological pluralism. The early efficiency frameworks—CBA, CEA, and CUA—established the principle that resource allocation should be guided by systematic comparison of costs and outcomes. The value-pluralist schools—MCDA, deliberative HTA, and patient-centered HTA—expanded the range of considerations that an assessment can include. Today, practitioners select among frameworks based on the decision context, the values of the stakeholders, and the institutional setting. The tension between technical efficiency and participatory deliberation is not a problem awaiting a solution; it is the defining condition of the field.