The normativist-naturalist puzzle: Functions and assumptions of health assessment tools

Abstract

While there is no shortage in discussions of health assessment tools, little is known about health professionals’ experience of their practical uses. However, these tools rely on assumptions that have significant impacts on the practice of health assessment. In this study, we explore health professionals’ experiences with health assessment tools, that is, how they define, use, and understand these tools, and whether they take them to measure health and wellbeing. We combine a qualitative, interview-based study of the uses and understandings of health assessment tools among Danish health professionals with a philosophical analysis of these applications and perceptions. Our study shows that contrary assumptions are involved in the use of the tools, to the extent that one can speak of a normativist-naturalist puzzle: health professionals generally apply a normativist conception of health, find health assessment useful and valuable for their clinical practice, but believe that what the tools measure is basically not health proper but some proximal entity of a more naturalist kind. This result demonstrates the complexity of health assessment tools and suggests that they are used with care to ensure both that particular tools are used for the kinds of tasks they are most apt for, and that they are put to use in awareness of their limitations.

Keywords

health assessment tools naturalism normativism philosophical assumptions qualitative research

Background

Several strong trends in current health care support the development and use of tools for more nuanced and accurate assessments of health and health-related conditions like wellbeing, thriving, or vulnerability. The constant drive to increase health care efficiency has led to a quest for more precise, valid, and reliable measures. In recent years, there has been a shift from measuring productivity in the health sector in terms of sheer output (like the number of treatments performed) to focusing on the actual effects of interventions in terms of health, wellbeing, and personal satisfaction, known as “value-based health care” (EU Commission, 2019; Miller, 2009; With and Jensen, 2018). This tendency is connected to a widespread attempt to increase patient participation in health care (Castro et al., 2016; Carel, 2016: 180ff.), as well as the recognition of the complexity of most health conditions as determined by a multitude of interrelated factors (Hernandez et al., 2018; Ohrnberger et al., 2017). Striving for the overall best treatment for the individual patient (or “pre-patient”) engenders the need for more comprehensive and nuanced information about their condition.

Generic health assessment tools, also commonly known as generic patient-reported outcome measures (“generic PROMs”), are questionnaires designed for this exact purpose: evaluating the effects of certain conditions, interventions, social settings, among other things, on the overall health and well-being of patients (Krogsgaard et al., 2021). The tools originally derive from the field of health economics where they were developed for cost-benefit analyses to assist in making informed political prioritizations and resource allocation within health care (Nord, 1999). By having, for example, a patient evaluate their health and well-being before and after a surgery by answering a questionnaire, one gains a picture of the net gain or loss in health or quality of life, which can be aggregated with scores from other individuals on a standardized basis. The tools have since migrated to other fields and are currently routinely utilized in, for instance, clinical research, medical practice, and studies of populational health (Patrick and Deyo, 1989).

Despite the extensive use of these tools, they still give rise to methodological and philosophical questions that extend beyond concerns for validity, reliability, and psychometrics. Health assessment tools particularly build on assumptions about what defines health, on what basis health is to be evaluated, and how the relevant value of health can be measured. Though health assessment practice has been the object of several philosophical investigations and critiques (Hausman, 2015; Stegenga, 2015), and though philosophical conceptions of happiness underlying specific welfare-measuring instruments have been analyzed (Kusier and Folker, 2020, 2021; Landes, 2015), less is known about health professionals’ perceptions of and experiences with generic health assessment tools. Studies tend to focus on very specific issues, such as patients’ perceptions or the utility of generic assessment tools (Hickey et al., 2005; Nilsson et al., 2007) or psychometric traits and issues of various questionnaires (Coons et al., 2000). While some textbooks do mention fundamental challenges and limitations (Fayers and Machin, 2016), little attention has been devoted to philosophical assumptions and functions underlying generic health assessment practice according to health professionals and researchers. This is a key omission in the literature. To firmly understand central assumptions underlying the practical use of health assessment tools, we need a greater overview of the ways that generic tools are utilized and understood by their users.

In this study, we investigate how health professionals define and understand health assessment tools, how they use them in different parts of their work, and how they take these tools to contribute to the measurement of health and wellbeing. To gain insight into the complex but consequential relationship between assumptions and the actual function and use of the tools, we combine a qualitative, interview-based study of the uses and understandings of health assessment tools among Danish health professionals with a philosophical analysis of the underlying assumptions. From the analysis, we draw the following four-tier conclusion: (i) that health-professionals express a general optimism about the use and value of health assessment tools, (ii) that the definition of relevant health assessment tools is contextual, (iii) that there is a strong commitment among health-professionals to a form of normativism about health, and (iv) that health assessment tools do not adequately measure “health” as defined by this normativist conception. We conclude that these findings together express what we call the normativist-naturalist puzzle of health assessment practice.

What is a health assessment tool?

By a health assessment tool, we understand a tool for assessing generic health; that is, the overall health state of a person (this we call the comprehensiveness requirement). The process of assessing generic health differs in several important respects from disease-specific assessment, for example, radiology, blood tests, reflex tests etc., which do not qualify as health assessment tools in our sense. We therefore tentatively define health assessment tools as instruments intended for generic measurement of the overall health status of an individual, which can be used to monitor an individual’s health status over time, to compare health statuses within and between groups, and to measure the effect and quality of medical treatments and healthcare interventions. To illustrate this broad definition, we can point to the following paradigmatic tools: SF-12, SF-36, EQ-5D, NHP, 15D, QALY, DALY among others, which all fit the above definition. In this context, we knowingly generalize across several distinct instruments to investigate a shared set of functions and assumptions inherent to generic health assessment tools.

Philosophical issues and assumptions

Health assessment tools rely on assumptions about the definition and value of health as well as the validity of its measures. These assumptions are in most cases merely implicitly acknowledged and sometimes completely ignored. This is unfortunate since they can have significant impact on the practice and function of health assessment. In this section, we bring some of the most central assumptions out in the open.

Multidimensionality

Health assessment tools raise philosophical questions in part because they are necessarily multidimensional. Because they directly or indirectly target comprehensive and complex states of persons, they must cover several factors. This means that the factors must be weighted and taken together (though not necessarily “aggregated” in any technical sense; “taking them together” can consist simply in a patient making an intuitive, global judgment). Indeed, several of the best-known tools are systematically interconnected. EQ-5D, for example, is used to estimate QALYs. Since it is an open question to what extent the factors contributing to a state of health can be judged independently, health assessment tools raise questions about the extent to which “reductionistic” or more “holistic” views of health states are the most appropriate (Jewson, 2009), as well as about principles for weighting and combining factors.

Naturalism and normativism about health

It is an open question to what extent the tools actually measure health narrowly understood or wellbeing or something in between. This is not merely because the tools are used for numerous purposes, but also because it is a contentious issue whether health and disease are naturalistic or normative concepts. On a naturalist account, health is defined in reference to a normal level of functionality, while pathology, as the relevant counterpart to health, is seen as a failure of a bodily part to make its normal contribution to the species-typical goals of the organism as a whole (Boorse, 1997). Consequently, this early instantiation of naturalism is sometimes referred to as the biostatistical account of health. And while more recent developments within naturalism about health resists relying too heavily on statistical normality (Hausman, 2012), they still define health in reference to functional efficiency. Naturalism about health is conceptually narrow and value-neutral because it revolves around organismic capacities and thus leaves out broader issues of personal wellbeing. The ambition is to separate, at least conceptually, the assessment of health from broader normative issues relating to overall well-being, quality of life, or benefit (Hausman, 2012).

Normativists about health, on the other hand, understand health precisely in terms of such normative ends. On the normativist account, health is defined in reference to wellbeing—how good a person’s life is going—or the effective opportunity to achieve important goals in life (Engelhardt, 1974; Nordenfeldt, 1987). It follows that absence of health is necessarily bad and that health assessments must be open to various sources of quality-of-life-related information. Normativists vary on how to determine the relevant goals. Some tie health to the realization of objective human goals or functioning (Nordenfeldt, 1987; Venkatapuram, 2011), while others suggest a more subjectivist account of well-being (Carel, 2016). But health assessment, on all normativist accounts, must be sensitive to a broad spectrum of normative information regarding relevant life goals, subjective obstacles, and social conditions for any given person.¹ Normativism and contextualism thus often go hand in hand. (For a recent discussion of the practical advantages and limitations of naturalist and normativist conceptions for healthcare, see Klausen et al., 2022; for critical discussions of the use and understanding of the term well-being in contemporary health practice and discourse, see Coffey, 2022; Sointu, 2005).

The evaluative basis of health

Regardless of how the target state is defined, health assessment is almost always rooted in normative concerns, for example, the wish to measure the impact of a condition on wellbeing, to make informed prioritizations etc. This points to the next area of assumptions underlying health assessment tools: what the evaluative basis of health is—or, in other words, what determines the value of health. Most often, the value of health is simply generically and implicitly assumed in health assessment practice, although questions concerning how and why health is valuable in part determines how it should be assessed (Brock, 2002; Broome et al., 2002).

The question of the value of health carves a distinction between subjectivist, who believe subjective measures (e.g. positive experiences, elimination of pain and suffering, or preference satisfaction) ground the value of health, and objectivists, who determine the value of health in reference to objective standards (e.g. intrinsic values, opportunities, or capabilities). The issue of the value of health thus raises important philosophical questions about value-theory and wellbeing (Daniels, 2008; Griffin, 1986; Parfit, 1984; Sen, 1992; Sumner, 1996). But another, more particular, reason to be explicit about such assumptions for health assessments is that since different health states are difficult to compare directly, health professionals will often measure the effects of states of health rather than their states themselves. By what scale would one measure whether it is worse to suffer from an auto-immune disorder or a severe depression? Each of these states are invalidating and cause harm, but in different ways, and a ranking of which is worse seems implausible since they differ in essential respects from each other. In other words, there is no common scale on which to measure them. This is a well-known issue in the research (see Hausman, 2006, 2015), and it is a main reason why experts opt to measure the state of health through its effect on well-being or opportunity.

Measurement

This leads to the last area of assumptions, namely, how to measure the value of health in terms of its effect on wellbeing or opportunity. Surely, the assumptions made in this area often relate directly to assumptions about the definition and value of health—for example, a health assessment tool measuring health by self-reported life satisfaction would hardly make sense without the assumption that the value of health is its effect on subjective wellbeing—but it moreover carries several assumptions about how to appropriately measure the value of health. For example, there is a widespread assumption that the measure should be universally valid to be applicable for inter-health-state comparisons. How realistic this is, depends on the degree to which the effect or health on wellbeing or opportunity is mediated by contextual factors. An equally widespread, and no less controversial, assumption is that subjective preferences can be trusted as valid indicators for evaluations of health states. This assumption is found not just in practices that directly target patients’ perceptions but also underlying standard health economic measures like QALY and DALY, which rely essentially on eliciting people’s preference (for critical discussion of this, see Hausman, 2015).

Methods

Study design

The study design is a combination of qualitative, interview-based research with philosophical analysis of the underlying conceptualizations and perceptions of health, well-being, and the measurement thereof. The interview data provides insights into the health professionals’ perceptions of health assessment tools and their experiences using them in their own wordings. Subjecting the data to philosophical analysis enables us to elaborately interpret the assumptions underlying the practical use of health assessment tools. We thus deliberately depart from the tendency to use highly inductive study designs in qualitative research, using a more theory-driven framework (Macfarlane and O’Reilly-de Brún, 2012).

Recruitment

The qualitative study consisted of 13 interviews. About 26 participants in total were invited for an interview and half of them consented to participate. The aim was not to gain an exhaustive overview of the generic health assessment practice but rather to glean insight into how select professionals in the field made use of the tools, and what theoretical conceptions laid behind this. We therefore opted for a greater breadth among the respondents in terms of professional backgrounds and areas of expertise (cf. Appendix I). While nine of the participants ended up having an affiliation to Danish universities and research, nearly all worked concurrently with health practice in some way. Several of the respondents were therefore familiar with both the theory and practice of health assessment, making them ideal interview candidates for a study about the general assumptions and functions of generic health assessment tools. The interviews can therefore be designated “elite” interviews. In order to not compromise identities, information regarding respondents except for employment status and areas of expertise are kept deliberately sparse since some of the communities are small, making the respondents easily identifiable. When the study was carried out, it was standard practice at the home institution to give advance approval to qualitative research projects with no patient involvement. The study did follow principles for human subjects’ research (see Appendix II).

Data collection

The interviews were conducted and recorded digitally via Zoom during the first lockdown in Denmark due to the COVID-19 pandemic in March and April 2020 by one of the authors along with three student assistants, who subsequently anonymized and transcribed the interviews. All participants signed a declaration of consent before the interview and were invited to attend a concluding workshop on the findings of the study to further discuss the topic.

The interviews were semi-structured and conducted using an interview guide that allowed for improvisations since the questions were formulated in an open-ended manner. The guide was refined upon after an initial probing interview and contained four overall branches of questions. The first branch related to specific instruments and uses of the practice, which progressively led into the more abstract and evaluative questions on the practice and rationale of generic health assessment. The interview guide did not specify specific tools at the outset but invited the respondents to share their thoughts on and examples of health assessment tools to elicit their immediate associations. In response to this, SF-36, EQ-5D, WHO-5, and QALY were mentioned as examples, which steered the direction of the conversation toward generic health assessment tools.

Data analysis

The data was divided into two-halves and through careful reading and rereading, important points were extracted, compared, and interpreted by two of the authors. Afterward, the entire author group interpreted the preliminary data analysis collectively and identified common themes in the material, which fitted the aims of this study. While the data collection process was mainly bottom-up, the interpretation of the data was informed by concepts and distinctions from philosophy of health such as naturalism, normativism, contextualism, holism, which helped us to identify and systematize points of relevance to the research question.

We neither expected nor found the philosophical concepts to perfectly match the understandings revealed by participants. Hardly any of the respondents expressed views that fitted neatly into theoretical positions such as the biostatistical theory of Boorse or the action-theoretic approach of Nordenfelt. Regardless, we maintain the usefulness of, for instance, the distinction between naturalism and normativism because it denotes archetypical and non-reducible elements of health and disease as phenomena, that is, biological functions or dysfunctions in contrast to well-being or lack thereof, which are sometimes combined to form hybrid accounts (Hofmann, 2002; Wakefield, 1992). These concepts proved helpful as an interpretive key in identifying different elements in the participants’ understandings of what the tools actually measure, how they achieve this, why the tools potentially fall short of this, and what the respondents generally thought about this practice.

Limitations of the study

The sample size makes it difficult to generalize the findings. However, among the respondents were several expert users, some who were deeply engaged in the development and translation of instruments themselves, and whose thoughts on the subject had a paradigmatic character. As the analysis did not comprise many “ordinary users,” there is a risk that the views expressed by the participants were not indicative of the way that many users apply and perceive health assessment tools. Furthermore, as almost all respondents were either from southern Denmark or the capital region, the study cannot conclude whether the views are representative of a general attitude either nationally or internationally toward the practice of health assessment.

Findings and analysis

Most participants were not familiar with the expression “health assessment tools” but preferred to speak instead of, for example, “health-based well-being measurement” or measurement of “health-related quality of life.” These denominations in themselves indicate a blending of normativistic and naturalistic concerns along with a comprehensive understanding of the purposes and functions of the tools. Despite this, there was a tendency toward more normative understandings as health is understood as closely connected to wellbeing. Subjective wellbeing, that is, “how one is doing” or “how one is feeling,” is assumed to function as the measure of health, the aim that health care services seek to promote, and the criterion by which the effectiveness of interventions can be measured. A general practitioner stated that:

The older you become in the medical profession, the better you understand the factors that . . . have to do with the holistic rather than the mechanical problem. . . . [W]hen you are inexperienced, you will simply think “this patient has pneumonia, which I must treat” and forget whether the patient is able to stand on her own two feet and take care of herself and what her quality of life is like.

A researcher in the field of mental health promotion saw both health assessment and certain tools as predominantly revolving around wellbeing:

We do not simply wish to measure the absence of symptoms but also to measure good mental health and wellbeing. . . . I think that [WHO-5] mainly measures how well one is doing, the emotional dimension of mental health.

And a doctor at an emergency department spoke interchangeably about “health condition” and “quality of life”:

[We assess] their health condition . . . after a month, after six months, and if it develops. It’s a study of older adults. Of whether they get a better or worse quality of life.

An associate professor of medicine, who has worked with the development of EQ-5D, similarly remarked:

Health in a narrow sense, that’s whether you are sick or not, whether you have symptoms or there are signs of disease, but I take it more broadly. Health is positive and negative health, so it is physical and mental, and especially also wellbeing.

Despite these indications of normativism, it is unclear to what degree the practitioners’ implicit and practically motivated conceptions of health match theoretical positions. It could also be argued that measuring conditions in terms of subjective quality of life is more of a pragmatic principle. Wellbeing can be used as a heuristic parameter that allows for comparability of qualitatively distinct phenomena. A general practitioner, for example, stated the following, which nuances the more clear-cut but implicit normative understandings delineated above:

Predominantly, our focus is whether there is something physically wrong . . . which is what I am looking for. . . . But I want to maintain that we need to know a lot more about a human being, about the whole, compared to today, because it tells us a lot more about what we have to do to help. It is rare for us in the emergency room to actively ask the patient: “what is your quality of life like right now?” I might ask them: “how is your quality of life? How are you really? What is your ordinary day like?,” but I only do that when I am sure that something does not add up. You can’t formalize that, you can’t use that in a questionnaire, at least we can’t since then it would become too fluffy and extremely hard to compare.

Statements like this indicate that more naturalistic understandings of health are also at play among the respondents. Likewise, the participant who insisted on speaking of “health-based wellbeing measurement” seemed to implicitly acknowledge the distinction between health and wellbeing, and to conceive of the former in functional terms, even though they insisted on their close relationship and the importance of measuring wellbeing.

This points to a view which can be more safely and widely attributed to the participants: They all seemingly endorsed a kind of contextualism about health. They are contextualist in the semantic sense (Preyer and Peter, 2005) since they tend to attach different meanings to “health” depending on the context, sometimes using the term in a naturalistic, sometimes in a more normativist sense. This reflects the fact that health professionals need the ability to shift between the somatic, mental, or holistic perspective according to the requirements of the patient and the situation. But the semantic contextualism often went hand in hand with an ontological contextualism, according to which health itself is a context-dependent state, and that it can only be adequately understood by taking the wider context of a condition or situation into account. A doctor of psychiatry continually expressed both his commitment to contextualism and some reservations about the word “context” and the associations it may give rise to:

You always have to see these scores [the aggregated scores yielded by the use of health assessment tools] in a context . . . these scales and assessment tools, you can’t take them out of their context. Although I don’t like the word, it has to be seen in a context. You mustn’t interpret such a number without taking into account the wider context.

The “wider context,” which the responder refers to, is generally assumed to comprise more than the patient’s physical condition. This trend toward holism and contextualism can also be seen from the fact that the participants express reservations toward quantitative approaches of the kinds that health assessment tools exemplify. They repeatedly emphasize that the assessments must be supplemented by, for example, disease-specific tools, consultations, individual screenings, or even self-made PROMs. The rationale seeming being that some facets of wellbeing and health are too complex to be captured solely through generic assessment. One participant remarked that:

You have to assess the whole, the entire human being (. . .) if you only consider the test in isolation, without placing it in its right context, then it has no value.

Generic assessment implies certain common traits to all conditions of health and wellbeing, which the instruments can operationalize. As a project leader at a large Danish hospital remarked:

The more prominent [generic assessment] becomes, the clearer it also becomes that each individual is individual, and even though you have the same disease, you don’t have the same symptoms or same perceptions, and you are not at the same point in life.

The assumption is that persons are affected functionally, cognitively, and emotionally in various ways and have diverse propensities for coping with certain physiological states. A person might be very susceptible to a condition and as a result experience a marked decrease in quality of life, while others will be less bothered by the same state. States of health, which might be very similar in terms of physical effects, affect everyone differently, which is bound to manifest itself in the readings of the tools. A professor at a health research department even said:

Well, I think that if you want to measure the same thing, then different tools might be exactly what you need in different cultures [“cultures” here referring to different groups of patients, contexts, and demographies].

To truly measure health and well-being, one would need different tools that are sensitive to the different ways that health and health-related conditions manifest themselves. The professor also found generic assessment to be too fixated on smaller details that might have no bearing on the individual’s well-being. To assess whether these details are relevant or not, they thought it necessary to conduct qualitative, conversation-based studies instead.

The contextual views of health and well-being are accompanied by a belief that the health professional’s attitude must likewise be holistic, and that the use of assessment tools should be qualified by a still more comprehensive and intuitive judgment. Several participants stressed that knowing how and when to apply generic assessment is an art in itself which require learning and experience. Yet when it comes to the question of whom or what is ultimately best able to gage the health and wellbeing of the patient, the participants almost overwhelmingly agreed that it must be the patient herself. As one researcher put it:

In principle, it is the patient that knows best about her quality of life.

This can be interpreted both as an endorsement of a more subjectivist understanding of health and wellbeing and as an epistemic point—that is, the patient knows best, regardless of whether her health and wellbeing are themselves subjectively constituted states. In any case, the patient is considered the arbiter of her health and determining the patient’s health profile is seen as matter of eliciting self-evaluations.

However, this is not to say that the professionals perceive self-evaluation as unproblematic. A researcher found, for example, that in cases of self-assessment of physical activity, people over- or underestimate their level of activity significantly, making self-perceived evaluations of physical health quite unreliable in their experience, thereby implicitly acknowledging a difference between health proper and self-perceived health. Likewise, a project leader at a Danish hospital stated that:

What makes [the tools] work less well is that the patients can figure out what they are responding to, which means that they can change their answers in such a way that they don’t have to talk about what is nagging them. So, there is a sort of schism in that it is up to the patient to fill out the form truthfully. I’m not saying that the patients are not truthful, but I also believe – or know, because I’ve interviewed many of them – that a lot of them actually avoid it.

The project leader also noticed that it is usually the most resourceful patients who have the means to complete self-administered questionnaires, which creates a bias in the overall assessments.

The interviews also revealed a widespread awareness of the methodological limitations of the tools concerning reliability, validity, and applicability but lesser awareness of questions of ontology and values, which are often couched in concerns with methodological language and questions of psychometric reliability and validity. An example of this tendency to somewhat neglect fundamental questions about the nature of health and disease that the instruments rely upon and instead to focus on methodology is a doctor and professor, who initially remarked:

Well, first of all I can’t define the word health. And I haven’t read any particularly good definitions of it,

but then steered the conversation toward psychometric issues:

. . . that content validity is crucial . . . and secondly, what one terms the psychometric measurement properties: how good or reliable are the measurements of this tool?

While this focus on practical methodology is understandable, given the health professionals’ educational background and working conditions, the tendency to ignore or bracket conceptual and ontological questions about the nature and value of health might seem problematic in light of recent studies, which suggest that operationalizations of certain conceptions affect the resulting assessments fundamentally. The EQ-5D, for example, has no social component while the newly developed WALY grants social wellbeing a much larger role and yields markedly different readings compared to the EQ-5D regarding the effects of certain conditions on wellbeing (Johnson et al., 2016). Therefore, sound conceptualizations and operationalizations of health and wellbeing are required to extract accurate and objective readings of health conditions as such. Although the instruments can naturally be applied with a view to their special properties and limitations but with a more restricted utility.

Despite their reservations, the professionals generally find the tools to be valuable additions to their practice, though the understanding of their specific function and purpose varies according to professions and specializations. They are thought to be least reliable when applied at an individual level. A professor and doctor remarked:

You can make [generic health assessments] on a group level but never on an individual level.

The tools are considered more valuable if the same patient can be assessed at different times, as this allows for intra-personal comparison and so can help register improvements or reductions in health. A health service manager added:

Well, I use the tools on two levels. One concerns the citizen herself, for example, when making preventive home visits, i.e., to measure changes in scores over time. (. . .) and if there is a change [in scores], you can address it.

The respondents generally believe the tools to be sufficiently reliable when applied to larger groups since the contextual and idiosyncratic differences that spell trouble in individual cases are then believed to be leveled out. A doctor and professor added that, apart from being applied to patients and groups of patients, the tools are also used to assess treatments as well as to evaluate the work of health professionals and organizational units:

On the individual patient’s level, the assessments can be used to motivate the patient to do something or assess if the treatment needs to be adjusted. They can also be used on a departmental level for assessing individual doctors. . . . It can also concern the hospital as a whole, and, finally, societal levels, i.e. to assess what can be done on a broader scale.

Somewhat paradoxically, however, it emerges that the tools are most often used for assessment of individual patients, which was mitigated somewhat by their description of how they themselves tried to integrate them in a more comprehensive process involving considered judgment, professional and personal experience etc. Nevertheless, there was little trust among the participants that sufficient care and caution is generally taken. A researcher stated:

Most of my colleagues probably don’t use it [health assessment] as much, and if they do, they probably use it more uncritically,

afterward discussing situations where they personally had experienced misapplications of instruments among health professionals. One of the most pervasive and striking findings is that almost all participants seem to think that “others,” that is, other professionals who use of the tools, tend to use them too uncritically. That is, they assume a “naïve” understanding of the instruments as “really measuring health,” or of “health being what the instruments measure,” to be the norm in the medical profession.

This might seem paradoxical or simply as an expression of the typical self-confidence or superiority bias. It should be noted, however, that the participants pointed to different shortcomings and potentials of the tools and framed and characterized them quite differently. Hence it is not unlikely that most of them do have an experience of being fairly alone with their critical views, though our study indicates that this may be a superficial impression, and that many of the same insights and worries can be found even among different groups of health professionals. Moreover, because all the participants have, in virtue of our selection criteria, special knowledge, experience and interests related to generic health measurement, it is likely that they tend toward a more context-sensitive and holistic view of health, and are more reflective users of the tools, than their non-specialist colleagues who use the tools more occasionally and might take the results more at face value.

Discussion

Our findings reveal widespread agreement among health professionals on the following general points, (i) general optimism about the use and practical value of health assessment tools, (ii) a context-specific understanding of the definition of relevant health assessment tools, (iii) strong commitment to some form of normativism about health, and (iv) that health assessment tools are inadequate to measure “health” as defined by this normativist conception. Together these points paint an intriguing picture of what we call the normativist-naturalist puzzle of health assessment practice. Below we elaborate the four points in turn.

Optimism about the usefulness and value of health assessment tools

It is a general trend in our findings that the participants’ express optimism about the usefulness and value of health assessment tools. Although many also emphasize that health assessment tools should be used with care and only when well informed about their limitations, almost all respondents find their measures of practical value generally reliable. The participants accept that health assessment is a complex endeavor and that any standardized measure will inevitably be simplistic. For that reason, many admit that the health assessment tools that they are familiar with have issues with precision and validity, but our findings show that most find the common health assessment tools to be a useful if not necessary tool for clinical practice and research. We therefore conclude that health professionals are generally optimistic about the usefulness and practical value of health assessment tools. It is important to note, however, that while this optimism about the (ideal) use of the tools is widely accepted, participants are less optimistic about their actual use—that is, how far they are in fact used with sufficient care, how well standard medical training, and thinking supports an optimal use of them etc. In this way, the views of the participants correspond with widespread conceptions of the potentials and limitations of PROMs (Black, 2013; Churruca et al., 2021; Dawson et al., 2010).

Context-specific understanding and use of health-assessment tools

Most participants, when asked about their use of health assessment tools, express some uncertainty about the definition of health assessment tools and move on to give some more specific examples of particular tools they are themselves familiar with. Many, moreover, note either explicitly or implicitly that what a health assessment tool is must depend on the context in which it is used. It is thus a general trend among the health professionals that they perceive health assessment tools to be a broad and vaguely defined category. Our findings thus document a tendency to assume context-dependence on several different levels, ranging from the very understanding of health and wellbeing to the identification, definition, and practical use of a specific health assessment tool. The subgroup of participants engaged in both clinical practice and research draw a stark general distinction between the definition of health assessment tools for those different enterprises. The purpose, they believe, of health assessment tools for research is to provide generic measures of health to be used for comparative analysis in experimental trials and surveys, whereas in clinical practice the health assessment tools should serve as a proxy for measuring the patients’ health and wellbeing, and these purposes to a large extent require quite different tools or at least a different use of the same tools. This could be interpreted as a conviction that health, disease, and well-being are multidimensional and multimodular, as argued by certain researchers (Barnes, 2023; Hofmann, 2002; Twaddle and Nordenfelt, 1993), and that health care therefore serves manifold purposes, which is reflected in the utilization and perceptions of the instruments.

Normativism about health

Our findings report a general commitment among participants to some form of normativism about health, tying health to expressions of patients’ general quality of life or how well the patients are doing rather than physiological and functional performance. This normativist conception of health serves for many as a background assumption of health as a distinct entity or phenomenon with its own ontological status, which is what they believe health professionals should in principle be most concerned with and thus what health assessment tools ought, ideally, to capture.

While the assumption of a strong tie between health and wellbeing is quite pervasive, our data are inconclusive as to whether this expresses a commitment to full-scale normativism in the philosophical sense or merely a strong belief that the purpose of health assessment and interventions is to support wellbeing, and that they should be understood and used with this in mind. It should also be kept in mind that normativism comes in subjective as well as more objective forms, depending, inter alia, on how well-being is understood. The fact that the participants tend to distinguish well-being—and genuine health—from functional performance could be an indication that they lean toward a subjective understanding of well-being and thus a normativism more along the lines of contemporary phenomenology of illness (Carel, 2016) than that of Engelhardt (1974), Nordenfeldt (1987), or Venkatapuram (2011), which understands “doing well” in terms of functioning well and being able to realize human goals. It is interesting that the participants understand the health assessment tools to be biased toward objectivism and naturalism, whereas some philosophers of health have criticized them for implicitly relying on a subjectivist view of well-being (Kusier and Folker, 2020). It might seem that the health professionals have a dichotomous understanding of well-being and health on the one hand and functional performance on the other, the first being normative and the other naturalistic. They do not seem to consider the possibility that well-being and health could be both about functional performance and normative and context-dependent, as philosophers like Nordenfelt would have it. But since they generally have a flexible understanding of health and well-being and tend to think that different notions may be appropriate for different evaluative purposes, it is likely that they could be open for it as well.

In any case, participants are well aware that health on a normativist account is not easily measured by existing health assessment tools (and might never be) (as argued theoretically by Hausman, 2015). In certain cases, participants recall experiences of getting close to actually examining patients’ health in the sense of how they are actually doing, but interestingly, these experiences are reported to occur when leaving standardized tools aside and dedicating time and focus to dialog and comprehensive conversation with and about the individual patient. Since such conversation can both reveal facts about the patients’ own experiences and their ability to realize life goals, this does not in itself, however, suffice to determine which particular version of normativism (if any) they are committed to.

Health assessment tools are inadequate measures of health

While there is widespread agreement on the general value and usefulness of health assessment tools, it is also an emergent trend in our analysis that the interviewed health professionals do not consider them adequate measures of health. Put differently, although health assessment tools measure something (of good use to clinicians and researchers), they do in fact not measure health, at least according to the normativist understandings of the participants themselves (but also in agreement with more naturalist views, like that of Hausman, 2015). One plausible interpretation is that whereas health generally is understood as a normativist conception, the participants believe that health assessment tools are better equipped to capture naturalist properties in reference to functional capacities. Thus, however useful and valuable they are, they will inevitably serve as simplistic, and somewhat naturalist, proxies for health understood in a more holistic and normative sense.

Health assessment and health inequality

None of the respondents mentioned challenges related to health inequality. This is a significant negative finding, because health inequality is otherwise recognized as a serious problem in most fields of healthcare, not least in connection with user involvement, which is often subject to self-selection and risks further empowering resourced patients, rather than benefiting weak or marginalized groups (Hausman, 2015 and Schramme, 2017, both consider health definition and measurement out of a concern for health equality.). The lack of reflections on this aspect of the use of health assessment tools may be partly explained by the fact that these are used, in the health care fields under study, to target vulnerable groups, for example older adults living alone, and not in ways that could be exploited by resourced patients to gain disproportionate attention. It could, however, also point to a blind spot among the health professionals. It is possible that the tools more adequately capture the health and wellbeing on persons belonging to certain groups, in terms of gender, ethnicity, social class etc. It is beyond the scope of this paper to consider possible unjust or socially skewed consequences of the use of health assessment tools. But it should be noted that normativist understandings of health have been criticized for elevating narrow moral and political norms to general, and potentially oppressive, normality standards (Flew, 1973), and it certainly merits further reflection.

Conclusion

Together, the above points ground the normativist-naturalist puzzle of health assessment practices. The puzzle contributes to our understanding of why health professionals have ambivalent feelings toward these tools. On the one hand, the health assessment tools are considered generally useful as measures of functional capacity or lack thereof, but on the other hand, they do not in fact measure health in accordance with the conception of health widely applied by the health professionals themselves. This furthermore fits the general opinion that health assessment tools cannot be utilized in isolation and ought in most cases to be supplemented by comprehensive qualitative assessments of patients’ health and wellbeing.

Our findings point to both theoretical and practical implications. On the theoretical part, the normativist-naturalist puzzle should not just be taken to show that health professionals are confused or apply different understandings in different contexts. It also demonstrates the complexity of the issue, and the possibility to simultaneously entertain—and act on—views that otherwise might seem difficult to reconcile. For example, the observation that the same health state (narrowly conceived) may influence a patients’ wellbeing very differently, depending on the context, may seem to express a normativist commitment; this did indeed seem to be how the participants intended it. But it is no less compatible with naturalism, inasmuch as it posits a contingent link between health (narrowly conceived) and wellbeing.

In terms of practical implications, our findings suggest, firstly, that health professionals view health assessment tools not as independent of other sources of information about a patients’ health, but rather as an integral part of a more comprehensive assessment practice. Secondly, that health assessments tools are used carefully to ensure both that particular tools are used for the kinds of tasks they are most apt for, and that they are put to use in awareness of their limitations. Finally, while admitting that the vast variety of different tools is not unproblematic, our findings may be taken to imply that there is a need to further develop tools that (like WALY) take non-physical factors sufficiently into account, provided that health professionals really want tools that measure health-related quality of life or wellbeing.

Research Data

sj-docx-1-hea-10.1177_13634593241255006 – for The normativist-naturalist puzzle: Functions and assumptions of health assessment tools

sj-docx-1-hea-10.1177_13634593241255006 for The normativist-naturalist puzzle: Functions and assumptions of health assessment tools by Thor Hennelund Nielsen, Lasse Nielsen and Søren Harnow Klausen in Health

This article is distributed under the terms of the Creative Commons Attribution 4.0 License (http://www.creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).

Supplemental Material

sj-docx-2-hea-10.1177_13634593241255006 – Supplemental material for The normativist-naturalist puzzle: Functions and assumptions of health assessment tools

Supplemental material, sj-docx-2-hea-10.1177_13634593241255006 for The normativist-naturalist puzzle: Functions and assumptions of health assessment tools by Thor Hennelund Nielsen, Lasse Nielsen and Søren Harnow Klausen in Health

Supplemental Material

sj-docx-3-hea-10.1177_13634593241255006 – Supplemental material for The normativist-naturalist puzzle: Functions and assumptions of health assessment tools

Supplemental material, sj-docx-3-hea-10.1177_13634593241255006 for The normativist-naturalist puzzle: Functions and assumptions of health assessment tools by Thor Hennelund Nielsen, Lasse Nielsen and Søren Harnow Klausen in Health

Supplemental Material

sj-docx-4-hea-10.1177_13634593241255006 – Supplemental material for The normativist-naturalist puzzle: Functions and assumptions of health assessment tools

Supplemental material, sj-docx-4-hea-10.1177_13634593241255006 for The normativist-naturalist puzzle: Functions and assumptions of health assessment tools by Thor Hennelund Nielsen, Lasse Nielsen and Søren Harnow Klausen in Health

Footnotes

Author’s note

Thor Hennelund Nielsen is now affiliated to Aarhus University, Denmark.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding for this article was provided by the University of Southern Denmark’s Human Health Project.

ORCID iD

Lasse Nielsen

Supplemental material

Supplemental material for this article is available online.

Notes

Author biographies

Thor Hennelund Nielsen is a postdoc in philosophy at Aarhus University, Denmark. His research covers topics such as philosophy of health and medicine, epistemic injustice, and phenomenology. During his PhD, he was affiliated with the Human Health project.

Lasse Nielsen is Associate Professor of philosophy at University of Southern Denmark. He works on ethics, philosophy of health and political philosophy.

Søren Harnow Klausen is professor of philosophy at the University of Southern Denmark. His broad research interests include wellbeing, philosophy of health and social epistemology. He was head of the SDU Human Health/Health and Wellbeing Assessment project 2018-2022.

References

Barnes

(2023) Health Problems: Philosophical Puzzles About the Nature of Health. New York, NY: Oxford University Press.

Black

(2013) Patient reported outcome measures could help transform healthcare. BMJ 346: f167.

Boorse

(1997) A rebuttal on health. In: Humber

Almeder

(eds) What is Disease? Biomedical Ethics Review. Totowa, NJ: Humana Press, pp.3–134.

Brock

(2002) The separability of health and well-being. In: Murray

CJL

Salomon

Mathers

, et al. (eds) Summary Measures of Population Health: Concepts, Ethics, Measurement and Applications, vol. 115. Geneva: World Health Organization, pp.115–121.

Broome

Murray

Salomon

, et al. (2002) Measuring the Burden of Disease by Aggregating Well-Being. Summary Measures of Population Health: Concepts, Ethics, Measurement and Applications. Geneva: World Health Organization, pp.91–113.

Carel

(2016) Phenomenology of Illness. Oxford: Oxford University Press.

Castro

Van Regenmortel

Vanhaecht

, et al. (2016) Patient empowerment, patient participation and patient-centeredness in hospital care: A concept analysis based on a literature review. Patient Education and Counseling 99(12): 1923–1939.

Churruca

Pomare

Ellis

, et al. (2021) Patient-reported outcome measures (PROMs): A review of generic and condition-specific measures and a discussion of trends and issues. Health Expectations 24(4): 1015–1024.

Coffey

(2022) Assembling wellbeing: Bodies, affects and the ‘conditions of possibility’ for wellbeing. Journal of Youth Studies 25: 67–83.

10.

Coons

Rao

Keininger

, et al. (2000) A comparative review of generic quality-of-life instruments. PharmacoEconomics 17(1): 13–35.

11.

Daniels

(2008) Just Health: Meeting Health Needs Fairly. Cambridge: Cambridge University Press.

12.

Dawson

Doll

Fitzpatrick

, et al. (2010) The routine use of patient reported outcome measures in healthcare settings. BMJ 340: c186.

13.

Engelhardt

Jr (1974) The disease of masturbation: Values and the concept of disease. Bulletin of the History of Medicine 48(2): 234–248.

14.

EU Commission (2019) Defining Value in “Value-Based Health Care”. Report of the Expert Panel on Effective Ways on Investing in Health. Luxembourg: European Union.

15.

Fayers

Machin

(2016) Quality of Life. The Assessment, Analysis and Reporting of Patient-Reported Outcomes. Oxford: Wiley.

16.

Flew

(1973) Crime or Disease? London: MacMillan.

17.

Griffin

(1986) Well-Being: Its Meaning, Measurement and Moral Importance. Oxford: Oxford University Press.

18.

Hausman

(2015) Valuing Health: Well-Being, Freedom, and Suffering. Oxford: Oxford University Press.

19.

Hausman

(2006) Valuing health. Philosophy & Public Affairs 34(3): 246–274.

20.

Hausman

(2012) Health, naturalism, and functional efficiency. Philosophy of Science 79(4): 519–541.

21.

Hernandez

Bassett

Boughton

, et al. (2018) Psychological well-being and physical health: Associations, mechanisms, and future directions. Emotion Review 10(1): 18–29.

22.

Hickey

Barker

McGee

, et al. (2005) Measuring health-related quality of life in older patient populations: A review of current approaches. PharmacoEconomics 23(10): 971–993.

23.

Hofmann

(2002) On the triad disease, illness and sickness. Journal of Medicine and Philosophy 27(6): 651–673.

24.

Jewson

(2009) The disappearance of the sick-man from medical cosmology, 1770-1870. International Journal of Epidemiology 38(3): 622–633.

25.

Johnson

Jenkinson

Stinton

, et al. (2016) Where’s waly?: A proof of concept study of the ‘wellbeing adjusted life year’ using secondary analysis of cross-sectional survey data. Health and Quality of Life Outcomes 14: 126.

26.

Klausen

Christiansen

Engelsen

(2022) Health, disease and wellbeing. In: Di Nucci

Lee

J-Y

Wagner

, et al. (eds) The Rowman & Littlefield International Handbook of Bioethics. Lanham: Rowman & Littlefield Publishers, pp.16–26.

27.

Krogsgaard

Brodersen

Christensen

, et al. (2021) What is a PROM and why do we need it? Scandinavian Journal of Medicine and Science in Sports 31(5): 967–971.

28.

Kusier

Folker

(2020) The well-being index WHO-5: Hedonistic foundation and practical limitations. Medical Humanities 46(3): 333–339.

29.

Kusier

Folker

(2021) The satisfaction with life scale: Philosophical foundation and practical limitations. Health Care Analysis 29(1): 21–38.

30.

Landes

(2015) Building happiness indicators some philosophical and political issues. Les ateliers de l’ethique 10(2): 4–37.

31.

Macfarlane

O’Reilly-de Brún

(2012) Using a theory-driven conceptual framework in qualitative health research. Qualitative Health Research 22(5): 607–618.

32.

Miller

(2009) From volume to value: Better ways to pay for health care. Health Affairs 28(5): 1418–1428.

33.

Nilsson

Wenemark

Bendtsen

, et al. (2007) Respondent satisfaction regarding SF-36 and EQ-5D, and patients’ perspectives concerning health outcome assessment within routine health care. Quality of Life Research 16: 1647–1654.

34.

Nord

(1999) Cost-Value Analysis in Health Care: Making Sense Out of QALYS. Cambridge: Cambridge University Press.

35.

Nordenfeldt

(1987) On the Nature of Health: An Action-Theoretic Approach. Dordecht: D Reidel Publishing Company.

36.

Ohrnberger

Fichera

Sutton

(2017) The relationship between physical and mental health: A mediation analysis. Social Science & Medicine 195: 42–49.

37.

Parfit

(1984) Reasons and Persons. Oxford: Oxford University Press.

38.

Patrick

Deyo

(1989) Generic and disease-specific measures in assessing health status and quality of Life. Medical Care 27(3 Suppl): S217–S232.

39.

Preyer

Peter

(eds.) (2005) Contextualism in Philosophy. Oxford: Clarendon.

40.

Schramme

(2017) Theories of Health Justice: Just Enough Health. London: Rowman & Littlefield Publishers.

41.

Sen

(1992) Inequality Re-examined. Oxford: Oxford University Press.

42.

Sointu

(2005) The rise of an ideal: Tracing changing discourses of wellbeing. Sociological Review 53: 255–274.

43.

Stegenga

(2015) Measuring effectiveness. Studies in History and Philosophy of Science Part C: Studies in History and Philosophy of Biological and Biomedical Sciences 54: 62–71.

44.

Sumner

(1996) Welfare, Happiness, and Ethics. Oxford: Clarendon Press.

45.

Twaddle

Nordenfelt

(1993) Disease, Illness and Sickness: Three Central Concepts in the Theory of Health: A Dialogue Between Andrew Twaddle and Lennart Nordenfelt, vol. 18. Linköping: Linköping University.

46.

Venkatapuram

(2011) Health Justice: An Argument From the Capabilities Approach. Cambridge: Polity Press.

47.

Wakefield

(1992) The concept of mental disorder: On the boundary between biological facts and social values. American Psychologist 47(3): 373–388.

48.

With

Jensen

(2018) Værdibaseret sundhed i Danmark. Anbefalinger for vejen frem. Danske Regioner. Available at: https://www.regioner.dk/media/11353/anbefalinger-for-det-fremtidige-arbejde-med-vaerdibaseret-sundhed.pdf (accessed 21 May 2024).

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.19 MB

0.02 MB

0.01 MB

0.02 MB