Eschewing Unsystematic Eclecticism

“Eclecticism,” in philosophical contexts, refers to a theoretical approach that draws ideas from a variety of different philosophical theories and traditions. There is a trend in the HTA literature to promote the use of eclectic approaches to ethics analyses in HTA (8;Reference Heitman11;Reference Hofmann13;Reference Saarni, Hofmann and Lampe23). The EUnetHTA Core Model endorses an unsystematic eclectic approach explicitly, describing at least six different methods stemming from divergent theoretical areas, any combination of which HTA producers are encouraged to use in their ethics analysis.

Other authors on HTA and ethics have defended the use of an eclectic approach on the grounds that it provides analyses that are “typically a more comprehensive evaluation of ethical reasoning behind and consequences of various courses of action than is possible with only one analytic method” (Reference Heitman11). Still others have contended that such an approach avoids being vulnerable to the objections to any single method (Reference Hofmann13). Despite their putative appeal, eclectic approaches pose two major problems. First, they can result in the choice of ethical approaches that carry conflicting philosophical assumptions. Second, and perhaps more problematically, with its inattention to theory, eclecticism is vulnerable to resulting in unsystematic approaches to ethics in HTA, unless the various pieces of divergent theory are united within a single logically coherent system.

Approaches that do not engage adequately with theory, like the one recommended in the EUnetHTA core model, seem less likely to produce the benefits that ethics analysis is thought to bring to HTA. Such an approach is unlikely to produce transparent HTAs because an unguided application and interpretation of moral theory greatly reduces the chances for logical and clear record keeping and reporting. An eclectic approach may enable comprehensiveness by providing several strategies for uncovering ethical issues, but it does not show how they can be systematically addressed. The usefulness of an HTA that uses this approach will depend greatly on how the individual HTA producers record the process and analyze the results. Without systematic guidance, HTA producers would need to rely heavily on their own resources and discipline to generate useful ethics discussions in their HTA report.

Better Discussion of Which Methods to Use, Why, and How

Most HTA reports fail to describe how the chosen theoretical approach and associated analytic tool was selected for the ethics analysis. If the absence of such a description is due to a lack of active consideration on the part of HTA producers, this creates several problems. First, the analytic tool often determines which ethical issues related to a particular technology end up being the focus of the analysis. Without considering other approaches, HTA producers risk overlooking an issue that is significant for the technology under consideration. Such an omission puts the comprehensiveness of the analysis at risk. Furthermore, a failure to acknowledge and explain the decision about which theoretical approach to take and why limits HTA producers' capacity to provide a transparent account of the assessment.

Second, the usefulness of the ethics analysis tool varies with the nature of the technology and with the context in which it is being applied. A clear consideration of the tools in a specific context is required to understand the ways in which a particular approach (even an eclectic one) maps onto the technology and the HTA environment (Reference Lehoux and Williams-Jones17). Third, a failure to discuss how a particular ethics approach was selected eliminates the possibility of discussion about the consideration and rejection of alternative approaches. In the process of an HTA, even a brief discussion of which methods were considered and why the chosen methods were selected adds transparency to the process.

Failure to Provide Ethical Direction

The extent to which an ethics analysis renders an HTA more useful depends in part on the HTA user's context. To a government policy maker, a report is useful when it uses clear language and provides succinct information that is relevant to the policy-maker's context. Moreover, the useful report does as much work as possible, providing analysis and basic direction that the advisor can easily include in communication documents for other decision makers within the organization.

The HTA literature fails to direct HTA producers to analyze ethical issues in a way that generates this basic direction. In fact, some literature remains silent about the necessity of doing an ethics analysis at all (Reference Heitman11;Reference Hofmann13). Raising ethical issues without analysis results in a cataloguing of issues. An HTA report that simply lists the relevant ethical issues requires end users to do the additional work of weighing issues, prioritizing underlying values, and identifying points of contention to fully understand the ethical implications of the technology. Guidance that fails to provide even minimal instruction on how to analyze ethical issues (as is the case with the dominant eclectic approach in the HTA literature) results in HTAs that are unlikely to achieve the expressed goal of usefulness for HTA users.

There Is no Expertise, so None Is Required

The belief that no expertise is required for conducting ethics analysis rests on the assumption that the subject matter of ethics analysis resists the development of expertise or that a lack of consensus regarding a philosophical approach suggests there is no real or useful expertise.

The belief that the subject matter of ethics analysis precludes the development of specific expertise may be based on the ambiguity of “expertise” or on the failure to recognize the distinctly philosophical nature of the questions being asked and of the approaches being advocated. For example, Saarni et al. (Reference Saarni, Hofmann and Lampe23) appear to reject a place for ethics expertise, claiming that all that is necessary for an adequate ethics analysis is a broad literature search and the inclusion of relevant “stakeholders.” However, both the INAHTA and EUnetHTA approaches acknowledge ethics expertise. The Core Model specifies the need for “philosophical analysis of the logic and coherence of the argumentation” (8). And the use of wide reflective equilibrium, another recommended approach, requires philosophical arguments to bring out the relative strengths and weaknesses of the alternative sets of principles (or competing moral conceptions) (Reference Daniels5). The role of ethics expertise and the philosophic dimension of the questions are clear.

The belief has also been expressed that the lack of consensus among moral philosophers regarding potential approaches indicates the impossibility of expertise (Reference Saarni, Hofmann and Lampe23). To philosophers, the kind of disagreement over approaches highlighted in the HTA literature is considered philosophically healthy and not, as is suggested, a reason for resignation. Furthermore, it is unclear why it might be thought that philosophers ought to agree that one theory or set of concepts is sufficient to understand and appreciate the ethical import of a particular technology.

Expertise May Exist, but It Is not Necessary or Helpful

Even when it is acknowledged that the questions being asked are philosophical in nature and presuppose philosophical expertise, the belief that such expertise is not necessary to conduct an ethics analysis has been stated explicitly by some of the authors of the EUnetHTA core model (Reference Saarni, Hofmann and Lampe23). This belief leads to the equally contentious assertion that the eclectic method of the EUnetHTA core model offers a tool that is usable by HTA organizations, regardless of their methodological expertise (Reference Saarni, Hofmann and Lampe23). Underlying this belief is one or more of the following assumptions: (i) the nature of an ethics analysis suggests that there ought to be methods for the nonphilosopher; (ii) some ethics analyses might require philosophical expertise, but not all do; (iii) the methods and vocabulary of professional philosophers would make the results of an ethics analysis incomprehensible to the intended audience of the HTA; and (iv) experts in other disciplines can easily learn the tools and methods required to conduct an adequate ethics analysis.

(i) The claim that ethics analysis can be conducted without a professional ethicist (Reference Saarni, Hofmann and Lampe23) is supported by the rather bewildering claim that there ought to be methods for ethics analysis that are applicable for nonphilosophers. Epidemiologists, statisticians, economists, and lawyers are not required to produce methods for nonexperts, so it is unclear why this demand should apply uniquely to moral philosophers. The demand that there be such methods is perhaps based on the erroneous belief that the apparent simplicity or ordinariness of philosophical questions entails equally simplistic or quotidian methods.

(ii) There is the belief that expertise and ethics analysis, in general, are most appropriate for obviously morally contentious technologies (Reference Hofmann12) with the implication that expertise is not required for all ethics analyses. This belief undercuts the original motivation for conducting ethics analysis: Analysis is required because the values implicit in and related to the adoption of a technology and that may challenge our ethical norms are not easily recognized. The ability to make these values explicit and to articulate and respond to such challenges requires a knowledge of philosophical approaches, concepts, and arguments.

(iii) Some authors have expressed the view that, even if professional philosophers were to become involved, ethicists' professional vocabulary would only hinder the use of ethics analysis (Reference Hofmann13). This assumes, without justification, that ethicists cannot discuss ethical issues in a way that is clear to the nonexpert. Every expert involved in HTA, without question, makes use of a specialized vocabulary to conduct her respective analysis, yet is still expected to write in a way that is accessible to the nonexpert. As with (i) above, it is unclear why a problem that logically applies to all experts involved in an HTA is restricted to ethicists.

(iv) The conduct of ethics analysis in HTAs and the literature discussing ethics analysis suggest that experts in other disciplines can easily learn the methods required to conduct an adequate ethics analysis. Possible reasons for this belief include the lack of obvious technical apparatus in philosophy and the erroneous assumption that doing philosophy well does not require special training or aptitude (Reference Benatar2). The depth of the ethical questions confronting researchers is hidden, as it is in many philosophical questions, by their apparently ordinary nature and by the relative stability of the intellectual landscape within which researchers work (Reference Heitman11). But we see these facts providing still more reason for needing ethics expertise in this area and not as a reason for championing the view that the relevant knowledge and skills can be picked up along the way.

Expertise

Acknowledging and using appropriate expertise will help to ensure that an adequate ethics analysis is conducted. A common, although erroneous, assumption is that because a subject matter requires philosophical analysis, empirical evidence, for example, the work of clinical investigators, sociologists, and anthropologists, is not important. Good ethics begins with good facts, although the extent that empirical work should be used to inform an analysis of ethical issues is a matter of debate among philosophers (Reference Molewijk, Stiggelbout, Otten, Dupuis and Kievet20;Reference Schleidgen, Jungert and Bauer24) and other researchers (Reference Solomon25). Answering practical ethical questions, such as the ones raised by HTA, requires both philosophical and nonphilosophical reasoning; however, exactly how the answers to nonphilosophical questions are to be woven into answering ethical questions is part of ethical reasoning and, thus, falls within the domain of moral philosophy (Reference Benatar2).

We have been asked what difference there may be between an ethics analysis conducted by someone with ethics expertise and someone without such expertise. The question appears to assume that one can talk meaningfully about differences between clinical and economic evaluations conducted by experts and nonexperts but that any such differences diminish or disappear for ethics analysis. Our opinion is that someone with expert knowledge and skills in any area is more likely to provide an adequate analysis and to present the results in such a way that they are useful to the reader than is someone without such expertise. Because a trained ethicist can draw on many conceptual tools and has a broad knowledge of the potential issues, the differences would likely be parallel to those in other disciplines involved in HTA: The use of expertise allows for a broader and deeper (richer) analysis and makes it more likely that the results will be valid and useful.

We would be remiss if we did not recognize the limitations of organizational capacity to do ethics analyses in HTAs. Procuring staff with expertise in ethics issues is often a great challenge, especially when resources are scarce (Reference Lehoux, Tailliez, Denis and Hivon18). It would likely be beneficial if HTA producers became more familiar with ethics analyses and qualitative methods to conduct primary research to address ethical issues. However, it is essential that ethicists be recruited into agencies (either permanently or on a contract basis) (15;Reference Lehoux and Williams-Jones17). We acknowledge that ethicists may be difficult to identify. HTA producers might first look to professional bioethics and ethics associations to begin to develop future contacts. We hope our argument serves as a call to both HTA organizations and ethicists to work together to promote a culture that rises to the challenge of integrating ethics analysis in HTAs.

Guidelines for Conduct and Reporting and Methodological Quality

The EUnetHTA report on best practices in HTA states that HTAs must provide a detailed, reproducible, and transparent description of the scientific methods (Reference Busse, Orvain and Velasco3). The INAHTA and EUnetHTA models are a similar step in this direction for ethics analysis. There are a variety of methods used to conduct and report on ethics analysis; however, there is little reason to think that reporting cannot be standardized in a manner similar to systematic reviews (Reference Liberati, Altman and Tetzlaff19) and economic evaluations (Reference Drummond and Jefferson7). Such reporting emphasizes transparent description and a detailed protocol; it does not prohibit the flexibility of methods and judgment required to conduct ethics analysis. If ethics analysis is not conducted or is restricted in some way (e.g., by focusing exclusively on the impact of the technology on patient autonomy or on the patient-caregiver/health professional relationship) this should be explained and justified.

To help end users make use of the results of ethics analysis, analysts should collaborate to develop methods to evaluate the appropriateness and strength of ethics analysis. Presently, the best that can be done is to have the relevant expertise involved in peer review of the protocol and report.

Untapped Resources

Review of the literature suggests that many of the issues of ethics in HTA have been discussed in other, largely untapped, areas. We encourage researchers to look to this literature. Examples of such resources include discussions of expertise (Reference Benatar2;Reference Weinstein29) and the moral authority of consensus decisions (Reference Moreno21) in the bioethics literature and debates about technological determinism (Reference Baylis and Robert1;Reference Hofmann14) in bioethics and philosophy of technology. Examining issues with analogous technologies is one important way of identifying and exploring ethical issues, so an awareness of areas that discuss similar issues is particularly important.