Healthcare Technology Assessment Research Paper

Introduction

Health technology assessment (HTA) may be conceived as the science of what if. It is an attempt to lay out in a transparent and systematic way the likely consequences of adopting a particular healthcare technology in a specific healthcare system and social context (Banta and Luce 1993). Consequences of key interest include:

– How will it change the health prospects of individuals with particular disease conditions?

– How will it affect the daily practice of healthcare and the overall healthcare system, particularly in terms of resource utilization?

– What wider social, ethical, and legal implications may be anticipated?

Basically, answers to such questions should help to judge whether it is wise to proceed with the technology or not. Typically, the question of what if is asked at a time when evidence has been produced of the beneficial nature of a healthcare technology in a limited, controlled setting (e.g., randomized controlled trial), and a decision needs to be made whether it should be made more widely available. The latter may imply that it is raised as a standard of care, included into guidelines, or funded by third-party payers or from taxes, as in the National Health System in the United Kingdom.

Healthcare Technology Assessment: Facts And Values

It is important to note that, as such, Healthcare Technology Assessment involves a close entanglement of facts and values, for laying out the consequences of adopting a healthcare technology is a matter of identifying consequences that are likely to happen and are considered of interest. Whereas the former is largely an empirical issue, the latter is a function of values. An example may help to clarify this point. When clinical trials and modeling studies had produced evidence, suggesting the safety and clinical and cost-effectiveness of the human papillomavirus (HPV) vaccine, governments were faced with the need to decide whether some sort of HPV program should be implemented. At the time, it was pointed out that in Canada, women of lower income and literacy, new immigrants, and aboriginal women remained significantly more likely both to develop and die of cervical cancer. The question, therefore, was whether the introduction of an HPV program could help – at least partially – to redress this inequity. HTA, then, is not a matter of investigating the consequences of adopting a healthcare technology, but a matter of investigating consequences of adopting a healthcare technology that are considered plausible, relevant, and amenable to scientific inquiry. Whereas plausibility is a function of our current knowledge and understanding of the subject, relevance is a function of values, and amenability to scientific inquiry is a function of our methodological and epistemological standards. Conceivably, judgments of what is plausible, relevant, and amenable to scientific inquiry may vary across stakeholders. This is an important reason for stakeholder involvement, an issue that will be dealt with in more detail later.

The Rationale Of Healthcare Technology Assessment

Raising the question of what if and going to considerable lengths in answering it suggest that it is considered worthwhile to do so. Arguably, it answers to a specific standard of rationality in healthcare policy making: do not allow the adoption and wider use of a novel healthcare technology, unless its comparative value has been sufficiently established. Clearly, this assumes some type of governance with oversight and authority for making (dis)investment decisions. However, very few, if any, examples are known where Healthcare Technology Assessment has led to the decision not to adopt a healthcare technology. HTA, ten, should perhaps not be conceived as an inquiry supporting go/no-go decisions. An alternative goal of HTA might be to steer the development and implementation of a healthcare technology in such a way as to optimize outcomes. This distinction is usually referred to as summative versus formative evaluation. Whereas summative evaluation aims to estimate the overall value of a healthcare technology, formative evaluation seeks to identify or develop modes of optimizing its value. The latter fits better in a deliberative approach to the assessment of healthcare technology, engaging multiple stakeholders who seek to discover the meaning of a healthcare technology from their value perspectives. The potential role of HTA in such deliberative practices will be further elaborated below.

The Challenges Of Healthcare Technology Assessment

As the example of HPV programs already showed, predicting consequences of adopting a healthcare technology that matter may be quite challenging. This is partly, but only partly, due to the fact that some outcomes associated with the use of healthcare technology will become apparent only after prolonged period of time and more extended use. Another and possibly even more important source of uncertainty is how stakeholders respond to a novel healthcare technology. For instance, left ventricular assist devices (LVADs), which evolved from the heart-lung machine, were originally developed as a bridge to transplantation in patients with heart failure. Only when these devices were used as destination therapy or bridge to recovery, the implications of this healthcare technology became far more substantial, raising questions regarding the sustainability of this development. This suggests that HTA should not be regarded as a one-off activity, but rather as a sustained type of inquiry, accompanying a healthcare technology as it evolves over time. Another challenge is posed by the multiplicity of outcomes, differences in appreciation of these outcomes among various stakeholders, factors that determine their onset, and the interdependencies among these factors.

 “Ethical Issues” In The Context Of Healthcare Technology Assessment

The development and use of healthcare technology can raise a host of ethical questions. For instance, extracorporeal membrane oxygenation (ECMO or heart-lung machine) in neonates with diaphragmatic hernia may improve survival but can also cause irreversible, severe disability, resulting from brain hemorrhage. Also, when the neonate fails to thrive while receiving ECMO support, the team may have to decide at some moment in time to discontinue treatment, which will result in the death of the child. Finally, the decision to start treatment needs to be made quickly, sometimes leaving insufficient time for the parents to make up their minds as to what treatment strategy they would prefer. The use of the ECMO machine in neonatal care results in situations where ethical values seem to guide us in opposite directions. Saving the life of a newborn child is, of course, a crucially important ethical value, but so are avoidance of serious harm, avoidance of futile treatment, and respect for parental autonomy. Thus, the development of a healthcare technology such as ECMO may profoundly change healthcare practice and result in “moral perplexity”: situations where it is not immediately obvious what, from a moral perspective, would be the appropriate thing to do (Brennan 1977). Is diaphragmatic hernia in neonates an indication for ECMO in the first place? What should be done if the treatment turns out to be unsuccessful? How should its provision be organized so as to observe parental autonomy? In the example of the HPV program, there is the prospect of further reducing morbidity and mortality associated with cervical cancer in women (beneficence), but it is unclear whether it also constitutes an effective means of reducing current inequity (justice).

Contingent Outcomes

Addressing ethical issues in the context of Healthcare Technology Assessment might be conceived as an attempt to answer precisely this question: is there any evidence suggesting that the use of the healthcare technology causes the sort of ethical dilemmas or moral perplexity as described above? This might be considered a largely empirical issue. When conceived in this way, three questions are in order: (1) what should a researcher do, in order to establish whether or not this is the case? This is a methodological issue, asking how comprehensive a researcher has been in this regard. What sources did she consult? Has she talked to various stakeholders? Has she made observations of various practice situations (“participant observation”), etc. (2) The second question is more fundamental in nature. It asks whether the contingent nature of the outcomes of such an inquiry is accepted. What is meant here is that the outcome of such an inquiry would depend on the moral competence of those who are involved and who have been queried in the context of the assessment. Is their moral experiences taken at face value, and how can we be sure that they have not overlooked certain moral aspects of situations, acts, or events? (3) The third and related question is: so what? What if there is evidence that the use of healthcare technologies causes ethical dilemmas? Should it be left to policy makers to decide whether this affects their decision regarding the use and wider adoption of the technology? In democratic societies, this might be considered to be appropriate, but policy makers may find themselves in an awkward position. Does what is being reported reflect moral preferences of some, but not others? Are policy makers forced to take sides in intractable moral conflicts? In the example of ECMO, does the use of ECMO, in fact, violate parental autonomy? Does it always do so? Is it perceived, as such, by the parents? By all parents? If parental autonomy is likely to be violated in certain circumstances, would that constitute a sufficient reason to refrain from using it in such cases? Clear answers to questions such as these have failed to materialize. This may have contributed to the fact that, so far, inquiry into ethical issues has played only a marginal role in HTA (ten Have 2004). To a large extent, this is due to a misapprehension about the relation between facts and values. Below, an alternative account will be presented, integrating empirical and normative analysis.

Different Values, Different Informational Requirements

The current practice of addressing ethical issues in the context of Healthcare Technology Assessment is usually overly simplistic and, in fact, conceptually incoherent. An HTA starts by presenting a more or less comprehensive description of the sort of consequences that may be expected from the wider adoption of a healthcare technology (the answer to the question what if ). Subsequently, it is asked whether any ethical issues may be anticipated, given some set of ethical principles, such as beneficence, non-maleficence, justice, and autonomy. What is wrong with this approach is that it assumes that a value-neutral description of consequences can be obtained, which may then be explored for its normative acceptability. This is not the case. In fact, it is rather the other way around. Value frameworks are operative at the stage of data collection, defining what is, and what is not, considered relevant. For instance, an important element in current HTA is the calculation of the incremental costs that are incurred by adopting a novel healthcare technology and the additional quality-adjusted life years (QALYs) that may be gained, as compared to usual care. These facts derive their relevance from utilitarian theory, stipulating that justice requires that for every gain in QALY, no matter how it is achieved or to whom it accrues, the same amount of resources should be spent (the criterion of Pareto optimality). However, if we would assess the consequences of adopting a healthcare technology from a different concept of justice, such as Rawlsian justice as fairness or Sen’s capability approach, we would need completely different data. In other words, different concepts of justice have different informational requirements. Acknowledging the close entanglement between facts and values has important consequences both for the practice of HTA and for the role of ethical inquiry in HTA. Ethical inquiry is, then, no longer an “end-of-the-day” reflection on the acceptability of the “given” consequences of adopting a healthcare technology in a particular healthcare system. Rather, it moves to the forefront of HTA, guiding deliberations on which aspects might be relevant, and for what reason, and how conflicting requirements might be practically resolved.

Healthcare Technology Assessment: A Collocation Of Facts Guided By Moral And Nonmoral Notions

In the current practice of Healthcare Technology Assessment, ethical issues are treated quite differently from other aspects that are considered of interest when evaluating the impact of healthcare technologies, such as health impact and resource utilization. Whereas the latter are based on empirical investigations, ethical issues are usually presented as dilemmas that find their basis in diverging moral preferences . Accepted methods exist for establishing a technology’s true impact on health and resource utilization. No such methods seem to be available for the rational resolution of conflicting moral preferences . The unfortunate upshot of this has been that up until now, inquiry into ethical issues has played only a marginal role in HTA. This meticulous concern for separating facts from values in HTA is misplaced. An alternative account of how facts and values relate to each other will be presented, discussing its implications for the practice of HTA. It will amount to a plea for a more central role of normative analysis, which could substantially enhance the policy relevance of HTA. Getting normative and empirical analysis on a more equal footing has attracted considerable interest in the field of policy analysis generally (e.g., Fischer 2003) but has remained relatively underappreciated in the field of HTA so far. In the account that follows is based on the work by Julius Kovesi (1967) Moral Notions (referred to in the remainder of this paper as MN). One of the key points that Kovesi makes in this work is that meaningful description of a situation, event, or act cannot be obtained without using, implicitly or explicitly, some moral or nonmoral notion or set of such notions. A notion is something that people have, or not have, or have to some extent, about something. Notions organize people’s reasoning and thought and enable them to interpret and make sense of situations, events, or acts with which they are confronted. In the following quotation, Kovesi comments on the “ritual” of distinguishing between description and evaluation:

…the contemporary distinction between ‘evaluation’ and ‘description’ sometimes assumes that facts just are outside in the world waiting for us to recognize them; and that evaluation consists of selecting some facts on ‘purely factual grounds’ and then expressing our attitude towards them. Indeed, decisions and attitudes, insights, wants, needs, aspirations and standards do enter into our moral notions. But they do not enter from the top, they are part and parcel of our notions, and they are organized by the rational activity of conceptformation. Evaluation is not an icing on a cake of hard facts. It is not the case that we have ready-made facts, and that if we want to describe them we state them and say ‘yes’ about them, and if we want to evaluate them we state them and say ‘please’ about them.. .There is a point in bringing certain features and aspects of actions and situations together as being relevant, and by removing this point, by removing the ‘evaluative element’, we are not left with the same facts minus evaluation. (MN, pp. 24–25)

What Kovesi describes here reflects well the current practice in Healthcare Technology Assessment: facts are being collected, informing us about the possible or likely consequences of using a particular healthcare technology. When it is felt that the empirical evidence base is sufficiently robust and comprehensive, the evaluative question is raised: given these facts, how do we judge the overall value of this healthcare technology? It is this practice, which by its practitioners is considered so self-evidently correct, that is criticized by Kovesi. What is wrong about it is that it ignores the fact that moral and nonmoral notions are already operative at the stage of gathering evidence. They are necessary to determine what facts are relevant and that, for that reason, will be collected. Having a notion means having knowledge of something or a general understanding. If someone does not have a notion of something, this means that such person is not aware of its existence and significance, would be unable to recognize what it stands for, and would lack the words to conceptualize it and communicate about it. Kovesi makes a distinction between material elements and formal elements of notions. By material elements, Kovesi means any characteristics in which an object may vary without ceasing to be, for instance, a table. The formal element provides an answer to the question why we call a large variety of objects “tables” and refuse the word to others. Having a notion of something means that someone is able to cite or identify material manifestations of it and is able to explain why these qualify as such (the formal element of a notion; MN, pp. 16–17).

Kovesi distinguishes moral and nonmoral notions. He discerns commonalities and differences between them:

In the case of moral as well as other notions we bring together an unspecified group of features, aspects or qualities of things, situations, or objects. But the point of collocating these features, aspects or qualities, the reason for grouping them together, is different. (MN, p. 13; italics in the original text)

The difference, then, is to be found in their formal element:

If, on the other hand, the point of having a term in our language is not that we want to identify, buy and sell something [as, for instance, is the case with tables] but because we want to avoid or promote something, excuse or blame people for certain happenings or acts, then we have a different type of formal element shaping the life and use of our term.

The wish to avoid or promote something and to excuse or blame people for certain happening or acts is the “point” of moral notions. However, whether we are dealing with moral or nonmoral notions, facts play a critical role in judging whether or not something is the case:

…whether we make a judgment by using a descriptive term or a moral term, i.e., whether we make a judgment like ‘this is a table’ or ‘this is murder’, the justification for our judgment lies in the presence or absence of certain relevant facts. (MN, p. 62–63)

In both cases, notions play a key role in determining which facts are relevant and which facts are not relevant. To put it very briefly, moral notions do not evaluate the world of description but describe the world of evaluation (MN, p. 119)! If Kovesi is right, this should have far-reaching consequences for the practice of Healthcare Technology Assessment. HTA should then no longer be considered a matter of collecting the facts which then serve as a basis to make up our mind about the value of what is being evaluated. Instead, it should start by asking what moral and nonmoral notions seem to determine what type of events and what features of situations are considered relevant when judging the merits and demerits of a healthcare technology. Is there general agreement among stakeholders in this respect and is there a common understanding of the material and formal elements of these notions? This, then, would open up the possibility for an inquiry which is fully amenable to reason. It helps to explain how ethical issues, as currently perceived, emerge when some people have no notion of certain things whereas others do or where people have quite different notions of things, and where, as a result, people have quite different ideas as to what is, and what is not, relevant to the evaluation task at hand. The following example serves to illustrate how Kovesi’s model can shed new light on public debates on the ethical propriety of specific healthcare technologies.

Notions In The Debate On Cochlear Implants For Prelingually Deaf Children

A healthcare technology that has stirred a lively debate on its moral propriety is the cochlear implant for prelingually deaf children (Reuzel 2001). The assessment of the technology was conducted along two lines. On the one hand, empirical evidence was collected on the impact of the device on the recipient’s hearing capacity, development of spoken language, ability to pursue mainstream education, quality of life, etc. On the other hand, ethical issues were explored, notably the resistance from deaf organizations, claiming that the technology was a threat to deaf culture. In Kovesi’s model, such completely separate inquiries into facts and values are unwarranted. In both cases, the key question is whether relevant facts have been collected, substantiating the claim that cochlear implants can, and do, lead to the alleged outcomes. The difference being that different moral and nonmoral notions were employed, yielding a different set of facts as relevant to the assessment task.

In the case of the cochlear implants for prelingually deaf children, it should not be concluded that there are certain facts about the technology on the one hand (e.g., its impact on a child’s proficiency in spoken language) and some ethical issues on the other hand (e.g., the demise of deaf culture). Instead, different events and situations are being associated with the use of cochlear implants in deaf children, giving rise to different descriptions. In Kovesi’s model, these descriptions can be traced down to a different set of moral and nonmoral notions that are brought to bear on the situation. In Kovesi’s words:

When one makes a moral decision one .. .works out the proper description of the situation on the basis of the relevant facts, and understands the significance of the description. (MN, p. 157)

The key task, then, is to develop a description of the situation or events that does justice both to the actual circumstances and moral and nonmoral notions that jointly make up our world view and value orientation. This is a complicated task for a number of reasons. Firstly, neither the situation in which we find ourselves nor the moral and nonmoral notions which we hold are fully determined. Although this will usually be experienced as a problem, it might, in fact, be conceived as an important asset. For if they were fully determined, there would be no opportunity for humans to exercise their freedom and develop their identity (Arendt 1998). In this regard, Kovesi calls notions “open-textured” (MN, p. 8; see also Porter 1995). It would be a serious mistake of course to infer from this that reality is simply what we want it to be. The appropriateness of our description will usually become apparent over time, depending on the consequences of our choices and actions. The latter will of course also be subject to description, but at least they can be compared with the expectations that we had on the basis of our previous description. The second reason which complicates this task is that working out the proper description of a situation or some event is inherently a public activity (MN, p. 55). People within a community will probably share a number of notions, but only to a certain extent. In this respect, it is interesting to ask what it means for someone not to have a notion of, say, justice, or freedom, for the person herself and for other members of the community who do have such notions. It also raises the question where we get our notions from and what we can do in order to maintain and foster them. Pursuing these questions would be beyond the scope of this paper; Kovesi’s model does, however, provide a compelling argument for communities not to neglect this task. A third complication is the dynamic character of the process: by accepting a specific description of a situation, the situation assumes a specific significance, and, likewise, the associated notions may assume a different meaning.

The development and use of healthcare technologies can create novel situations, the meaning of which we try to establish using our current repertoire of moral and nonmoral notions. Inevitably, the material element of our notions evolves over time and so may their formal element. Indeed, novel technologies may give occasion to further specification of existing notions, leading to the emergence of novel terms.

Implications Of Kovesi’s Model For The Practice Of Healthcare Technology Assessment

For proponents of the cochlear implant, the key (and obvious) problem is that deaf children cannot hear. As a consequence, they will be unable to develop spoken language, cognition, emotion, etc. Since cochlear implants, when followed by appropriate rehabilitation, restore, albeit imperfectly, hearing to a certain extent, they will also support these other developments. An underlying notion might be the notion of normal functioning, and the importance of maintaining or restoring normal functioning, both for the individual and for society at large. Other notions include the critical role of language to cognitive development and a general concern for the well-being of children who are born deaf. Opponents of the cochlear implant conceded the importance of language to cognitive and emotional development, but challenged the idea that it should be spoken language that fulfills this function. Their argument was that sign language is a fully fledged language, which can serve this function equally well. Given the poor and highly variable outcome of the cochlear implant, they preferred excellent performance within a deaf community over constant under performance in a hearing community. A key underlying notion here might be recognition: acknowledge the value of deaf culture, with sign language as one of its most distinguishing features (Honneth 1995).

In terms of Kovesi’s model, the cochlear implant for deaf children is an example where two different descriptions were presented:

…when I ask whether I should do this or that, by ‘this’ or ‘that’ I mean acts of different descriptions and not this or that particular act that fall under the same description. When one makes a moral decision one does not choose the good instance of an act that falls under a certain description but works out the proper description of the situation on the basis of the relevant facts, and understands the significance of the description. (MN, p. 157. Italics in the original text)

For the cochlear implant, two different descriptions were developed, with different notions (e.g., normal functioning, recognition) at work. So how to proceed in such cases? One obvious strategy would be to test among the various stakeholders whether these are, indeed, the notions that seem to summarize their concerns well or whether still other notions are at work. Also, the material elements of these notions may be explored, e.g., what other instances do we know where recognition seems to be an important issue in the context of healthcare? Similarly, the formal element of these notions may be explored: what is their point, and why do we have a notion like “recognition” in the first place? Such exercise may or may not produce a different perspective of the situation. When we probed the various stakeholders for their views of the situation, it turned out that proponents and opponents of the technology differed in yet another respect (Reuzel 2001). Proponents reasoned – implicitly – from the assumption that acquisition of spoken language and sign language is mutually competitive: what a child gains in one modality, he/she will lose in the other. In contrast, opponents of the technology reasoned from the assumption that acquisition of the two linguistic modalities may actually be mutually supportive, analogous to what is known from bilingualism in spoken language. For them, the problem is not so much that deaf children cannot hear. Rather, the problem is that deaf children, during the first months of their lives, usually do not get the sort of sensory input that is appropriate for them. Sound, any sound, is inappropriate for these children, in the sense of not being useful for their development. In contrast, visual input is highly appropriate. Having uncovered these different views, empirical evidence on this issue became crucially relevant. At the time, none of the stakeholders was able to provide convincing empirical support for their position. When we searched the literature for relevant research findings, we found reports from studies that had been conducted in Sweden, which favored the mutual support hypothesis, rather than the mutual competition hypothesis (Preisler et al. 2002). Accepting the validity and relevance of this evidence paved the way for agreement among stakeholders on how to proceed with the cochlear implants program for deaf children in the Netherlands. There was agreement on the importance of the early identification of deafness among newborn children. Also, there was agreement on the importance of the early usage of sign language by the parents in their communication with their child, irrespective of the question whether the child would be put forward as a candidate for a cochlear implant. Thus, a potential practical solution was developed by exploring moral and nonmoral notions underlying stakeholders’ judgments, in conjunction with relevant evidence. Clearly, practical implications need then to be further explored (e.g., what is necessary in order to guarantee access to such services), and details need to be elaborated (e.g., is the use of sign language compulsory? Is a certain proficiency in sign language on the part of the child a necessary condition for being eligible for cochlear implantation, etc.). Also, the resource requirements of such measures need to be worked out, allowing for an assessment of their affordability and opportunity costs, introducing the notion of scarcity. This, in turn, allows for a judgment whether the commitment of the requisite resources can be justified, introducing for instance Sen’s notion of a fair distribution of capability. Clearly, introducing a notion such as capability brings along its own informational requirements: we need to know certain things about deafness, about the position and prospects of deaf people in our societies, and about the outcomes of cochlear implants, in order to be justified to reach certain conclusions in this particular respect.

Healthcare Technology Assessment As Democratic, Reflective Discourse

Acknowledging the importance of moral and nonmoral notions in producing different descriptions of a situation, an event, or, as in our case, the sequelae of adopting a healthcare technology and understanding the implications for the demarcation of relevant evidence support a close involvement of stakeholders in Healthcare Technology Assessment. HTA, then, is primarily seen as a means for democratic, reflective discourse. Or, as the late Alex Farrell put it:

Assessment processes are embedded in different sorts of institutional settings, within which scientists, decision-makers, and advocates communicate to define relevant questions for analysis, mobilize certain kinds of experts and expertise, and interpret findings in particular ways. (Farrell et al. 2001)

The example of the cochlear implant also points to the transformative potential of HTA: a solution emerged from the process, which none of the stakeholders had brought to the evaluation. Conceivably, such a transformation can only occur when it is accompanied by a transformation of the conceptualization of the issue. In other words, HTA may be seen as an important means of social learning in a technological environment (Grin and Van de Graaf 1996).

A Brief Characterization Of The Model Of (Non)Moral Notions

The upshot of the analysis so far is that whatever answer is given to the question of what if, it is imbued with a complex set of moral and nonmoral notions. Indeed, it is this feature that makes Healthcare Technology Assessment interesting. For only then, the point of the specific collocation of facts that any HTA in fact is can be better understood: why were these facts selected as being of particular interest and relevance to answer the question of the value of this healthcare technology? And only then, we are in a better position to understand many disputes about ethical issues. A debate such as the one on cochlear implants for prelingually deaf children can be understood as a disagreement on the question what constitute the relevant facts. This, in turn, derives from a difference in the set of moral and nonmoral notions that are brought to bear, implicitly or explicitly, on the case. This, then, opens up opportunity for addressing such issues in a rational way, as indicated above. Kovesi’s work is highly relevant to HTA, since it provides such a thoughtful analysis of how the distinction between description and evaluation (facts and values) is usually conceived and how this has prevented a constructive integration of empirical and normative analysis. Since Kovesi’s work is relatively unknown, it may be helpful to try to “locate” his work with respect to some longstanding debates in the field of moral philosophy.

Cognitivism Versus Noncognitivism

Kovesi’s work should be considered cognitivist. The key concept in his work is “(moral) notion.” A notion is something that someone has (or has not, or has to some extent). Having a notion means having knowledge of something or a general understanding. It derives from the Latin word noscere, which means “coming to know” and “becoming acquainted with.” Not having a notion of something means that someone is not aware of its existence and significance, would be unable to recognize it, and would lack the words to conceptualize it and communicate (from the Latin word communicare: to share; literally, to make common) about it. Having a notion of something means that someone is able to cite or identify material manifestations of it and is able to explain why these qualify as such (the formal element of a notion). This model, consisting of open-textured (moral) notions, their material and formal elements and the relation between material and formal elements (they do not entail each other), and the associated nondeductive but analogical mode of argumentation, opens up a possibility for reasoning about the moral significance of situations, acts, or events in a systematic and transparent way. Indeed, Kovesi’s work could be conceived as a comprehensive and sustained rebuttal of David Hume’s conclusion “but ‘tis the object of feeling, not of reason” (MN, p. 70). This does not, of course, mean that people are emotionally indifferent to situations, events, or acts that seem to challenge our moral notions. On the contrary, however, such emotional involvement does not preclude a rational inquiry of the moral significance of a situation, event, or act.

Realism Versus Idealism

It is harder to locate Kovesi’s work on this “axis.” Kovesi is quite clear in acknowledging the existence of a world which exists, independent of our conceptualization of it. However, in a way, this is not our world. It is not the same world as the one that we appropriate through language. Kovesi:

In an important sense, in the world there is no value and there no murders, tables, houses, accidents or inadvertent acts. But our language is not about that world in which there is no value or no tables, houses, accidents or inadvertent acts. That world, the world of raw data, cannot be described for the sense of that world lies outside it and the very description of it, likewise, lies outside of it. Thereof one really cannot speak. (MN, p. 19; italics in the original text)

So, does a world exist, independent of our conceptualizations? Yes, it does. However, it is relatively uninteresting, since we have no means to talk and think about it. For, as soon as we do, it becomes a different world. To be sure, nature talks back. However, it uses a language different from our own.

Consequentialism, Deontology, And Virtue Ethics

Again, it is hard to characterize Kovesi’s work in this respect. One of the examples of moral notions that Kovesi discusses is murder. There are many material manifestations of murder that, purely empirically, have little in common (strangulation, poisoning, stabbing, shooting, or pushing someone over a cliff). What enables someone to recognize all these acts as instances of murder is an understanding of the formal element of murder: the intentionally taking of the life of someone who is innocent, with the aim of personal gain or satisfaction (MN, p. 4). This does not mean that for a proper use of the concept of murder, it suffices to have knowledge of its material manifestations and an understanding of its formal elements. Indeed, “features and aspects of situations and things, intentions and expectations, or perhaps some aspect of a legal, scientific or conceptual framework” may be needed, too (MN, p. 22). So far, this reasoning seems largely deontological, in the absence of any reference to adverse consequences associated with committing murder. However, as repeatedly pointed out by Kovesi, (moral) notions have a point. There are reasons why (moral) notions came into being in the first place and survived. Kovesi said:

…the naming of a colour and the subsequent use of that colour-word involves the existence of a way of life where there is a need for talking about colours and a language in which there is a place for colourwords. (MN, p. 6)

…the life and use of the word ‘table’ is shaped not only by the way we use the object table, but also by all those other activities in the performance of which we have to use the word ‘table’. (MN, p. 15)

Without the need for ‘tak’ in a way of life we will not start forming the notion or using the word, and the word will not acquire meaning. (MN, p. 40)

…if we had not had the need to blame or excuse ourselves and others, we would not have begun to form the notion of an inadvertent act, and thus have started excusing ourselves and others. (MN, pp. 15–16)

…so we selected certain features that need to be present for an act to be an act of murder, because we recognize that the presence of these would render someone liable to blame and we need to blame and to discourage certain sorts of act. (MN, p. 21)

In all of these quotations, the emergence and continued use of words, concepts, and moral and nonmoral notions are situated in particular social practices, where there is a need for such words, concepts, and notions. They fulfill a function. In case of moral notions (the latter two quotations), they support a specific way of communal life. This, then, seems to introduce a consequentionalist element, although it could also be argued that putting these moral notions into practice is constitutive of the relevant communal ways of life. It is interesting to note the Darwinistic or, perhaps better, Lamarckian elements (Gould 2002) in Kovesi’s account of moral notions: they come into being and survive because they fulfill some specific function. In fact, experimental evidence has been recently obtained supporting the hypothesis that hominin reliance on stone tool making generated selection for teaching and language (Morgan et al. 2015). Also, there is an element of reflexivity here (MN, pp. 49–50): once these notions have come into being, they, in turn, influence the social context from which they emerged (Murray and Holmes 2013). In addition to comprising both deontological and consequentialist elements, Kovesi’s model may also be associated with virtue ethics: having certain notions and acting and reasoning in accordance with them may be considered as an important element of one’s identity.

Methodology Of Moral Argument

There are certain similarities between Kovesi’s model and casuistry (Brennan 1977). In both models, notions (concepts) are considered to be open-textured, precluding deductive argumentation and preferring reasoning by analogy. What is called the material element of notions may be considered as the equivalent of the explication of moral concepts in casuistry: an answer to the question what follows from our commitment to a particular moral concept. Likewise, the rationale of a moral concept (an answer to the question why it is important in the first place) may be considered the equivalent of the formal element of a moral notion. In Kovesi’s model, there is an important unity between the material and formal element of a moral notion; similarly, in casuistry, the rationale constrains the range of appropriate explications of a moral concept. However, there is an important difference, too, for, in Kovesi’s model, analogy between material manifestations of a moral notion plays a less prominent role as compared to reasoning by analogy in casuistry: “far from knowing the meaning of “good” already [that is, disposing of a repertoire of paradigmatic cases], we are trying to elucidate what it is.” MN, p. 114. Finally, there are some commonalities with Richardson’s method of specification of norms (Richardson 1994). In fact, the example that Kovesi elaborates, suggesting that “misticket” might be an appropriate term for erroneously printed bus tickets, can be conceived as a specification of the more general notion of mistake.

Kovesi’s argument can be considered as an attempt to achieve maximal coherence among the varied and often conflicting moral notions to which we are committed. (see Richardson 1994, pp. 143–152).

Conclusion

Healthcare is provided because it allows us to do a lot of good. It allows us to relieve suffering, to restore functioning, to prevent untimely death, to expand capabilities, and to reduce inequities. Healthcare is provided collectively, because we feel that all people should stand an equal chance of benefiting from it, according to their needs. Not infrequently, however, this pursuit creates situations where certain moral values seem to be compromised. It would be fitting for Healthcare Technology Assessment, being the science of what if, to include this dimension in its inquiries. So far, such type of analysis played only a marginal role in HTA. This is largely due to an incorrect conceptualization of what ethical issues, in fact, are. Currently, facts are being presented (for instance in terms of an incremental cost utility ratio), serving as a basis to make up our mind about the comparative value of the technology under investigation. This comes down to asking whether the incremental cost utility ratio is below the threshold of what is generally considered acceptable (e.g., 80.000 Euro per quality adjusted life year) and whether there are any ethical issues that need to be taken into account. Such distinction between description and evaluation is unwarranted. It completely ignores the fact that the result of any HTA is a specific collocation of facts, driven by specific notions. Whether these notions are moral or nonmoral, our judgments always depend on the presence of relevant data. The model opens up novel opportunities for rational inquiry into values in the context of Healthcare Technology Assessment. Questions of interest would include:

• What moral notions do we, as a community, have?
• What do we consider material manifestations of these notions?
• Why do we have these notions in the first place, what is their point? (formal element)
• Can we discover relations between these notions? A hierarchy?
• What role do moral notions play in our experience of the world, of ourselves, and of others?
• Where do we get our notions from? Can notions be fostered?
• What would it be like, not to have a notion of, for instance, justice, or suffering?
• How do moral notions determine what we consider relevant, because valuable?
• How do nonmoral notions determine what we consider relevant, because plausible?
• What is it that is rational about having moral notions, using them, being aware of them, making them explicit, deliberating about them, etc.?

The results of such an inquiry could serve as a background or foundation for conducting HTAs of specific technologies, integrating normative and empirical analysis. The key reason for doing this would be to reestablish the connection between what we do and what we value. According to Hannah Arendt (1998), this is how we exercise our freedom and develop our identity. No more, no less.

Bibliography :

1. Arendt, H. (1998). The human condition (2nd ed.). Chicago: The University of Chicago Press.
2. Banta, H. D., & Luce, B. R. (1993). Healthcare technology and its assessment. An international perspective. Oxford: Oxford University Press.
3. Brennan, J. M. (1977). The open-texture of moral concepts. Lanham: Rowman and Littlefield.
4. Farrell, A., VanDeveer, S. D., & Jager, J. (2001). Environmental assessments: Four under-appreciated elements of design. Global Environmental Change, 11, 311–333.
5. Fischer, F. (2003). Reframing public policy. Discursive politics and deliberative practices. Oxford: Oxford University Press.
6. Gould, S. J. (2002). The structure of evolutionary theory. Cambridge: Belknap Harvard.
7. Grin, J., & van de Graaf, H. (1996). Technology assessment as learning. Science, Technology & Human Values, 21, 72–99.
8. Honneth, A. (1995). The struggle for recognition. The moral grammar of social conflicts. Cambridge: Polity Press.
9. Kovesi, J. (1967). Moral notions. London: Routledge and Kegan Paul.
10. Morgan, T. J. H., Uomini, N. T., Rendell, L. E., ChouinardTully, L., Street, E., et al. (2015). Experimental evidence for the co-evolution of hominin tool-making teaching and language. Nature Communications, 6, 6029. doi:10.1038/ncomms 7029.
11. Murray, S. J., & Holmes, D. (2013). Towards a critical ethical reflexivity. Phenomenology and language in Maurice Merleau-Ponty. Bioethics, 27(6), 341–347.
12. Porter, J. (1995). Moral action and Christian ethics (pp. 23–36). Cambridge: Cambridge University Press (1999 Edition).
13. Preisler, G., Tvingstedt, A. L., & Ahlstrom, M. (2002). A psychosocial follow-up study of deaf preschool children using cochlear implants. Child: Care, Health and Development, 28(5), 403–418.
14. Reuzel, R. P. B. (2001). Health technology assessment and interactive evaluation: Different perspectives. Thesis, Radboud University Nijmegen.
15. Richardson, H. S. (1994). Practical reasoning about final ends. Cambridge: Cambridge University Press.
16. ten Have, H. A. M. J. (2004). Ethical perspectives on health technology assessment. International Journal of Technology Assessment in Health Care, 20(1), 1–6.
17. Hofmann, B. M. (2008). Why ethics should be part of health technology assessment. International Journal of Technology Assessment in Health Care, 24(4), 423–429.
18. Shrader-Frechette, K. S. (1985). Science policy, ethics and economic methodology. Dordrecht: Reidel.
19. Wilt, G. J., van der Reuzel, R., & Grin, J. (2014). Technology, design, and human values in health care. In P. Vermaas & I. van den Poel (Eds.), Handbook of ethics, values and technology design (pp. 1–18). New York: Springer.
20. ten Have, H. A. M. J. (1995). Medical technology assessment and ethics: Ambivalent relations. Hastings Center Report, 25(5), 13–19