In the Literature
Aug 2021

“Perspectives” in Health Technology Assessment

Anthony J. Culyer, DEcon
AMA J Ethics. 2021;23(8):E619-623. doi: 10.1001/amajethics.2021.619.


This article considers 2 types of standard by which health technology assessment (HTA) studies should be judged: methodological and social. Methodological desiderata specify characteristics of a good quality analysis and should be met regardless of context. Transparency about an HTA study’s perspective (eg, specifying whose costs and whose benefits from an intervention should be counted) is one such desideratum. Whether any particular perspective is the right one is, by contrast, contingent upon conditions in which the analysis is to be applied. A perspective ought always to be treated as context sensitive. Recently, it has been advocated that an HTA study’s perspective should always be “societal” (ie, including consequences, good or bad, for anyone affected in any way by a technology’s use). This article argues that this is a mistake, ethically attractive though it might appear.

Health Technology Assessment

Health technology assessment (HTA) is a widely used way of thinking about setting priorities in health care investments, selecting treatments for inclusion in or exclusion from insurance benefits packages, or prioritizing the order in which a public health measure is to be rolled out across various population groups.1 Besides standard textbooks,2,3 there are several specific guides to HTA best practice in journals.4,5,6 It is commonly accepted as good practice for HTAs and related methods like cost-effectiveness analysis and cost-benefit analysis to state the “perspective” from which analyses are to be, or have been, conducted.2,3,4,5  Perspective defines the kinds of effect, their distributions, and changes in them that are likely to result from health care investment decisions. More specifically, a societal perspective was described by Gold et al as one in which “the analyst considers everyone affected by the intervention and counts all significant health outcomes and costs that flow from it, regardless of who experiences the outcomes or costs.”2 As examples, among the benefits of  a new effective treatment for a chronic disease would be the relief afforded to informal family carers of patients; among the costs of transferring hospital resources from their usual functions to meet urgent needs arising from a pandemic are the health losses associated with the usual treatments no longer available. This article argues that it is a mistake to treat a societal perspective as a general methodological desideratum, ethically appealing though it may appear.

Two Forms of Advocacy

A distinction can be made between methodological advocacy and social advocacy in the design and conduct of HTA. Methodological advocacy promotes HTA as a tool of analysis with wide applicability and specifies general, context-free standards by which HTA studies may be designed and judged under any decision-making context. Social advocacy in the design and conduct of HTA requires—but does not constitute—the introduction of specific and universal social value judgments, such as the measure of health gain or loss that should be used, the types of consequence that should be considered, the concept of equity that should be used, and the choice of population group that use of the technology impacts. Transparency makes these critically important value judgments clear.3,4,5 It is always a desideratum in HTA. It is a universal requirement, regardless of context.

Analysts are also, however, sometimes urged to adopt a particular perspective, usually labelled, as in Gold et al, societal. As Drost et al note: “The societal perspective in economic evaluations is important because of its higher decision-supportive power to optimize resource allocation.”7 Urging analysts to adopt the societal perspective for the good of society as a matter of procedure is an example of social advocacy. Proponents of the societal perspective include Byford and Raftery,8 Johannesson et al,9 Jönsson,10 and Walker et al.11 These authors maintain that taking a societal perspective, as defined by Gold et al, is context-free because it is independent both of the interests of the commissioners of the study, whose interests may well be more limited than a societal perspective, and of the political and social character of the society for which the study is intended.

My suggestion is that all perspectives are context sensitive (ie, appropriate or inappropriate in the context for which a study is intended). The value judgments people hold about health care and its manner of access and delivery vary greatly internationally, and even intranationally, according to circumstance and time and should therefore be treated as context dependent in HTA studies. Specifying a particular perspective, even one as general as societal, ought always to be considered explicitly in each study in the expectation that the context in which the study results might possibly be applied differs from other contexts in ethically important ways.11 There are many cases in which the appropriate perspective may not be societal. Examples include studies of workplace health in which decision makers must consider safety interventions conducted from (a) the employer’s perspective, (b) the trade union’s perspective, or (c) a third party payer’s perspective; pharmaceutical interventions assessed from (a) the patients’ perspective, (b) the patients’ caregivers’ perspective, or (c) a health agency perspective; or a family planning intervention assessed from (a) a Roman Catholic perspective, (b) a demographic perspective, or (c) a low-income country’s traditional healers’ perspective. The perspective, in short, ought to match the character of the question any HTA is to address.

It is neither necessary nor desirable for all studies to be standardized to take the same perspective, but all should state their perspective clearly. 

As these examples suggest, context includes factors such as disease burden, demography, culture (including religion), traditions, history, wealth, decision-making capacity, data availability, and the degree of risk-aversion in public decision making. It also includes local understandings of health and of fairness and equity, the social structures of a society, and the extent to which members of the community in question have shared understandings. In some cases, an HTA might properly adopt the societal perspective, for example, because it is required by a ministry protocol. In practice, however, even health ministries typically adopt a less-than-societal perspective—for example, by considering only those costs borne by the ministry’s budget and those benefits received by insured persons. The choice of perspective is thus rarely context free. Any particular choice is loaded with value judgments12 and is therefore, as Drummond and Brandt have argued, context sensitive: “the value judgments made in economic evaluations could, quite legitimately, vary from setting to setting.”4 It is neither necessary nor desirable for all studies to be standardized to take the same perspective, let alone any specific one, but all should state their perspective clearly. Perspective should, in short, always be explicit but also always be context sensitive.

Social Advocacy Assumptions in HTA

Assumption 1: Information costs of HTA can be safely ignored. Conscientiously to search out the most precise estimates of all conceivable costs and consequences of a decision, which is what the societal perspective requires, is to presume that the value of the expected improvement in the quality of the decision in question (somehow measured) is always and everywhere greater than the cost of acquiring the additional information. This presumption is so evidently irrational that it scarcely needs further elaboration. But any compromise on the comprehensiveness of the data set necessarily makes the analysis, to a greater or lesser extent, less than fully societal. A less-than-societal approach is therefore inevitable.

Assumption 2: Analysts make better social value judgments than other people. For analysts to stipulate a specific perspective for general use is presumptuous and implies an inherent unearned ethical authority. Stipulating a universal perspective is not a task for which HTA analysts are equipped by technical training, by their ethical rectitude, or by political authority granted through a due process. Analysts are often quite good at eliciting the implicit perspectives and values of decision makers and other stakeholders, which is a useful—indeed, highly desirable—early step in any HTA study, but to elicit perspectives and values is not to stipulate them. In eliciting them, analysts may also encourage decision makers to reconsider their own presumptions and even to weigh the case for adopting a societal perspective. But the process is not, or ought not to be, one of persuading decision makers to accept the value judgments that happen to be those preferred by the analyst.

Underlying the social advocacy of some analysts is often a specific philosophical view: one that is consequentialist, that is based upon preferences, and that is individualistic. This approach seeks to aggregate the preferences of all individuals in a society over all the possible consequences of the decision in question in order to make a preference ranking. I will not argue against the careful consideration of individuals’ preferences, but preferences need to be judged and carefully weighed. (Are ill-informed and well-informed preferences to count equally? Are inconsistent and consistent preferences to count equally?)

Assumption 3: It is ethical to ignore the political, historical, and constitutional contexts of health policy. What is politically acceptable, culturally conditioned, and economically possible varies according to national and regional context. In virtually all jurisdictions, and for reasons well-rehearsed by health economists over many years,13 policy and legislative arrangements have been adopted to combat the antisocial consequences of unregulated health care finance and provision: inequity of financial burdens, externalities, imperfect agency, monopoly, transaction costs of insurance, among others. In most jurisdictions, one consequence is the creation of ministries of health with ministers appointed by a due process and accountable—at least in democracies—to a parliament or generally elected assembly of society’s representatives. Governments characteristically set budgets across broad categories of economic activity (eg, health, education, the environment) and also set rules determining how those budgets are to be spent, the consequences to be taken into account in allocating expenditures, and the processes of accountability for decisions taken. One conspicuous consequence of these rules is that decision makers in such ministries nearly always adopt a less-than-societal perspective. Two questions therefore demand an answer: is it reasonable when, and by what moral argument do, nonelected, unaccountable, analysts set themselves above elected and accountable public officers? An embarrassingly bold answer to this question was given years ago by a brilliant, but I think misguided, economist of public policy: “the value-judgments made by economists are, by and large, better than those made by non-economists.”14

Analytical Humility

What’s wrong with taking one’s moral authority from a publicly accountable authority rather than from the preference utilitarianism upon which much of HTA still rests? What’s wrong with designing a study according to the objectives set by a client? Why should a study not be designed to tackle only part of a problem? Why should a study not examine consequences for only deprived groups of the population? Why should a study not be designed to identify only the likely losers from a decision (in order perhaps to consider appropriate compensation)? Using HTA to answer these questions requires taking a perspective that can be clearly stated but is not societal.

I recommend analytical humility. As Keynes wrote: “If economists could manage to get themselves thought of as humble, competent people, on a level with dentists, that would be splendid!”15 That quotation should apply to all analysts—the servants, not the masters—of decision makers.


  1. O’Rourke B, Oortwijn W, Schuller T; International Joint Task Group. The new definition of health technology assessment: a milestone in international collaboration. Int J Technol Assess Health Care. 2020;36(3):187-190.

  2. Gold MR, Siegel JE, Russell LB, Weinstein MC, eds. Cost-Effectiveness in Health and Medicine. Oxford University Press; 1996.

  3. Drummond MF, Sculpher MJ, Torrance WG, O’Brien BJ, Stoddart GL. Methods for the Economic Evaluation of Health Care Programmes. 3rd ed. Oxford University Press; 2005.

  4. Drummond M, Brandt A, Luce B, Rovira J. Standardizing methodologies for economic evaluation in health care. Practice, problems, and potential. Int J Technol Assess Health Care. 1993;9(1):26-36.
  5. Drummond MF, Jefferson TO; BMJ Economic Evaluation Working Party. Guidelines for authors and peer reviewers of economic submissions to the BMJBMJ. 1996;313(7052):275-283.

  6. Wilkinson T, Claxton KP, Sculpher MJ, et al. The International Decision Support Initiative reference case for economic evaluation: an aid to thought. Value Health. 2016;19(8):921-928.
  7. Drost RMWA, Paulus ATG, Evers SMAA. Five pillars for societal perspective. Int J Technol Assess Health Care. 2020;36(2):72-74.
  8. Byford S, Raftery J. Perspectives in economic evaluation. BMJ. 1998;316(7143):1529-1530.
  9. Johannesson M, Jönsson B, Jönsson L, Kobelt G, Zethreaus N. Why should economic evaluations of medical technologies have a societal perspective? Office of Health Economics briefing 51. October 2009. Accessed June 9, 2021.

  10. Jönsson B. Ten arguments for a societal perspective in the economic evaluation of medical innovations. Eur J Health Econ. 2009;10(4):357-359.
  11. Walker S, Griffin S, Asaria M, Tsuchiya A, Sculpher M. Striving for a societal perspective: a framework for economic evaluations when costs and effects fall on multiple sectors and decision makers. Appl Health Econ Health Policy. 2019;17(5):577-590.
  12. Culyer AJ. Assessing cost-effectiveness. In: Banta HD, ed. Resources for Health: Technology Assessment for Policy Making. Praeger; 1982:107-120.

  13. Culyer AJ. The normative economics of health care finance and provision. Oxf Rev Econ Policy. 1989;5(1):34-58. Article republished with modifications in: McGuire A, Fenn P, Mayhew K, eds. Providing Health Care: The Economics of Alternative Systems of Finance and Delivery. Oxford University Press; 1991:65-98.

  14. Turvey R. Present value versus internal rate of return: an essay in the theory of third best. Econ J (Lond). 1963;73(289):93-98.
  15. Keynes JM. Essays in Persuasion. Macmillan; 1931.


AMA J Ethics. 2021;23(8):E619-623.



Conflict of Interest Disclosure

The author(s) had no conflicts of interest to disclose.

The viewpoints expressed in this article are those of the author(s) and do not necessarily reflect the views and policies of the AMA.