The Errors of Individualistic Public Health Interventions: Denial of Treatment to Obese Persons Comment on “Denial of Treatment to Obese Patients—the Wrong Policy on Personal Responsibility for Health”

Abstract

I agree entirely with Nir Eyal’s perspective that denying treatment to obese patients is morally wrong. However, the reasons for this belief differ in some ways from Eyal’s analysis. In this commentary, I will try to explain the similarities and differences in our perspectives. My primary claim is that the denial of treatment to obese patients is wrong principally because (i) it eschews a whole-population approach to the problem of poor nutrition and is therefore likely to be ineffective; (ii) it is likely to expand obesity-related health inequities; and (iii) it is likely to intensify stigma against already-marginalized social groups. I shall consider each in turn, and explore the extent to which Eyal would be likely to agree with my claims.

The ineffectiveness of individualistic interventions

First, denying clinical treatment to obese patients on grounds of their obesity reflects the methodological individualism that characterizes dominant approaches to health promotion in the US at least (1). As I have argued, such individualism is an enormous problem insofar as it is ineffective, is more likely to expand health inequalities than an alternative whole population approach, and is likely to intensify stigma against obese persons, who disproportionately belong to groups that already experience higher levels of discrimination and stigma. Rose’s whole population approach suggests that making structural changes across an entire population is more likely to improve the overall health outcomes and compress inequalities than targeting high-risk groups with individualized health promotion interventions (2).

Eyal acknowledges that a focus on more upstream social support is essential, but argues that such “does not touch on the question whether policies that engage individual patients and their incentives directly make sense as well, alongside these social supports” (3).

This is of course logically true; the fact that attention to macrosocial determinants of obesity and undernourishment/malnutrition may be a preferable public health policy intervention to methodologically individualistic approaches does not in and of itself demonstrate that the latter is morally wrong. However, the conclusion that individualized interventions are highly likely to be ineffective across the target population is not best construed as an airtight proof. Such interventions are rather ethically deficient, and therefore should receive significantly less resources and attention than an alternative whole population approach (1). Of course, as Eyal points out, this implies that individualized approaches might well have a (small) place in the bundle of policies and levers used to diminish obesity incidence and prevalence. Nevertheless, where there is reasonable evidence that such individualized approaches to health promotion are prioritized in the US at least, such is at the very least ethically suboptimal, if not categorically wrong (1).

It is also worth pointing out here that attention to upstream macrosocial determinants of poor nutrition is not justified solely because doing so constitutes “institutional encouragement of healthy individual choices.” Although Eyal does not use the term, such resembles descriptions of the so-called “nudge” in public health. As Crawshaw has recently pointed out, the problem with the nudge is that it remains methodologically individualist (4). The end goal remains reform of individual health behaviors rather than the structural conditions. The primary normative goal of attention to the social determinants of health ought not to be seen as the production of less risky individual health behaviors, even if the former is likely to have such effects. Although there is significant ongoing debate regarding the extent to which health behaviors determine health outcomes, (1) there is little disagreement that structural conditions affect health outcomes in ways beyond their influence on health behaviors.

Individualistic approaches such as denial of treatment may expand health inequities

An additional problem with denying treatment to obese patients is that such methodologically individualistic interventions are significantly more likely than a whole population approach to expand obesity-related health inequities. Capewell and Graham term methodologically individualistic approaches “agentic” insofar as the extent of benefit capture depends on the resources which the individual agent is able to bring to bear (5). Unsurprisingly, affluent individuals generally benefit more from agentic interventions than the least well-off. Because such implies that health inequities between rich and poor will increase even if the intervention successfully improves absolute health, this is an enormous ethical problem. As applied to the case at hand, one would be justified in predicting that the ‘stick’ of denying treatment to obese patients is accordingly more likely to produce behavior change among the affluent than the poor. Unfortunately, a literature search does not reveal studies that have directly evaluated this hypothesis.

Eyal cites Sir Michael Marmot for the same point, but argues that “when unhealthy choices are more common among the poorer or less educated... then the collective positive impact on the poor and less educated could remain greater than the one on the rich and educated.” I either do not understand this claim or do not agree with it. First, it is unclear how Eyal construes “collective positive impact,” which matters morally because one of the fighting issues is the extent to which we should permit an expansion of socioeconomic status (SES)-related inequities if such includes a small absolute improvement in health among the least well-off. Thus, it does not follow from the mere fact that the least well-off obtain any health benefit from an agentic obesity intervention that such intervention is morally permissible. Second, it is of course possible that the collective positive impact could be greater for the least well-off than that enjoyed by the affluent, but Marmot’s (and my) point is precisely that the weight of the evidence suggests that such an occurrence is unlikely via an agentic approach. And although the question itself awaits a full empirical assessment, there is little justification for predicting as such regarding obesity when the evidentiary trends for other agentic interventions in public health contexts suggest the contrary (5,6).

Denial of treatment may intensify the stigma

The third and arguably most morally problematic aspect of denying treatment to obese patients is the likelihood that doing so may intensify stigma against already-marginalized populations. Although of course individuals may experience stigma, it is most accurately understood as a group-level phenomenon. In public health contexts, stigma is defined by an in-group’s assessment of difference on the basis of an identifiable characteristic shared among the out-group, and then by the in-group’s assignation of deviance to the out-group on the basis of that characteristic (7,8). Because the in-group must by definition be enfranchised (otherwise it would be unable to mark the out-group as different and attribute deviance), stigma is inextricably linked to structures of power and privilege.

Moreover, the focus on individual responsibility for health has a long history in American culture, and there is little dispute that it remains a powerful framework through which notions of desert and culpability are assigned to illness sufferers in the US (1). By virtue of their relentless focus on the individual, agentic interventions are as a category more likely to intensify stigma (1). This is especially problematic insofar as risky health behaviors are disproportionately prevalent among the least well-off. This means that health stigma is more likely to be directed at the least well-off, who are already more likely to experience stigma and discrimination in their respective social lives.

There is a sense in which denying treatment to obese persons exacerbates the moral problem. Exhorting the most materially deprived to consume less sugar-sweetened beverages may in general be more likely to create or intensify pre-existing stigma, but it nevertheless seems qualitatively different from the denial of service at issue in the present case. And where such denial is disproportionately more likely to be directed at groups who have historically been disenfranchised and denied from services, the denial of treatment becomes increasingly difficult to defend. Note that this is a normative critique, and does not even address the overwhelming evidence that stigmatizing obese persons is not only likely to be ineffective, but may actually worsen obesity.

Eyal agrees that the specter of obesity stigma is “substantial,” although his commentary does not emphasize the moral problems raised by stigma (9).

Conclusion

Ultimately, Eyal and I agree that the denial of treatment to obese persons is morally illegitimate, although our rationales for thinking so are at least somewhat different. Perhaps the takeaway is that there is a plethora of reasons to find the practice morally dubious. Other public health policy responses to the problems posed by obesity-related disease are vastly superior.

Citation: Goldberg DS. The errors of individualistic public health interventions: denial of treatment to obese persons; Comment on “Denial of treatment to obese patients—the wrong policy on personal responsibility for health”. International Journal of Health Policy and Management 2013; 1: 237–238.