Supported Decision-Making for People with Dementia Should Focus on Their Values

This article refers to:

Supported Decision Making With People at the Margins of Autonomy

This article is referred to by:

Supported Decision Making with People at the Margins of Autonomy: Response to Commentaries

In their thoughtful and rigorous article, Peterson and colleagues (2021) extend an account of supported decision-making that was originally developed for people with static cognitive impairments, to show how such regimes can promote the autonomy of people with dynamic cognitive impairments like traumatic brain injury and dementia. We share the view that, for many people with dementia and related conditions, well-designed supported decision-making approaches would be preferable to traditional surrogate decision-making arrangements. However, as the authors acknowledge, there are important gaps remaining in their account of supported decision-making (and in the prior literature that their account draws from). In this commentary we aim to show that these gaps can be addressed by focusing on the important role of beneficiaries’ values. On our view, two crucial elements of decisional support are promoting beneficiaries’ values and promoting their ability to value.

Much work in the philosophy of agency is premised on some form of a distinction between what people value and what they merely desire or prefer (Seidman 2009; Watson 1975). While people sometimes act out of desires that they are estranged from or disavow (as in cases of addiction or weakness of will), full-fledged agency involves acting in accordance with what one values. In part, what we value (which in this commentary we will gloss as “our values”) constitutes the standpoint from which the agent can act on reasons. Accordingly, in cases of conflict, we give normative priority to agents getting what they value over what they merely desire. Personal values also figure prominently in bioethical discourse; e.g., advance care planning aims at care concordant with patients’ values (Sudore et al. 2017), and shared decision-making unites clinicians’ biomedical expertise with patients’ knowledge of their own values (Barry and Edgman-Levitan 2012).

While Peterson and colleagues critique individualistic accounts of autonomy that emphasize individual capacities such as “being able to value,” they also insist that rational capacities and agency should still play a central role in decision-making. Their contention is instead that agency and its prerequisites can be distributed across people. Therefore, if philosophical accounts of agency (and bioethical constructs like shared decision-making) that assign a central role to personal values are correct, then accounts of relational autonomy should elaborate how valuing can be distributed across a beneficiary and their supporters.

Incorporating personal values into relational autonomy also addresses two major gaps that the authors acknowledge in their account. The first concerns the influence that trustees can exert over the motivations and thus the resultant choices of beneficiaries. A potential weakness of supported decision-making is that if it requires only decisions in accordance with beneficiaries’ preferences, then unscrupulous trustees could manipulate beneficiaries to embrace preferences that benefit the trustee at the expense of the beneficiary. Peterson and colleagues note that the potential for abuse and exploitation also exists in surrogate decision-making, but this is an unsatisfying response. The more fundamental problem is not the practical potential for exploitation, but instead that the authors’ account offers no criterion for even distinguishing exploitation from other, justified forms of influence that trustees should exert over beneficiaries’ personal goals.

On our view, agency includes regulating and revisiting one’s own “mere” preferences and desires in the service of one’s values. However, these features of agency often draw upon cognitive faculties such as prospection, perspective-taking, working memory, and metacognition that are typically compromised relatively early in the progression of dementia. In the context of supported decision-making, this then suggests that trustees would legitimately intervene to shape a beneficiary’s motivations and choices when this is necessary to promote the beneficiary’s values. Such interventions might include redirecting a beneficiary from unattainable goals to more realistic ones that align with their values, helping a beneficiary to more fully imagine and assess possible outcomes according to their impact on what the beneficiary values, and steering a beneficiary away from other preferences that conflict with what they value. While some interventions of these types are common in friendly relationships among healthy people, they require special care in the context of cognitive impairment, where the risks of manipulation are heightened. Furthermore, given the central role of valuing in agency, it would be legitimate for trustees to intervene by shaping a beneficiary’s motivations and choices when necessary to conserve or recover the beneficiary’s waning capacity to value. For example, a trustee could legitimately entice the beneficiary to choose a residential facility whose staff is more skilled at soliciting information about what the residents care about and in buttressing the residents’ confidence that their values still matter.

Our emphasis on the need for a criterion of appropriate support may have been anticipated in Silvers and Francis’s (2009) work on collaborative “thinking about the good,” which encompasses parts of the role that valuing plays in our account. Our account departs from theirs in focusing on beneficiaries’ contemporaneous capacity to value (even if diminished), but shares a concern for “the authenticity of a trustee’s expression of the good” for a beneficiary: “the subject is the sole inspiration for the conceptualization the trustee advances.” While Silvers and Francis view this aspect of support as “an interactive process that refine[s] the disabled subject’s goals,” we add the key element that the subject-centered sources of inspiration for a beneficiary’s goals are not all on an equal footing. Instead, on our view this process should focus on beneficiaries’ values and on supporting beneficiaries as valuers.

As a second gap, Peterson and colleagues also acknowledge the need for a “trigger” identifying when trustees can go beyond only facilitating beneficiaries’ decisions to instead override beneficiaries’ decisions. (While Peterson and colleagues describe this as a transition from supported to surrogate decision-making, in practice we see these activities as often occurring in parallel; after all, many beneficiaries will be more capable of participating in some decisions than others.) We believe that reference to the values of people with dementia can help address this problem as well. While judgments of reasonableness or harmfulness would be third-personal assessments external to an agent’s normative standpoint, we can instead assess whether an agent’s actions promote their own values. To take an example previously discussed in Jaworska (1999), a person with dementia might strongly desire a new truck, but might fail to recognize that such a purchase would jeopardize other things that they value, such as their grandchildren’s financial security. Appropriate support in such a case would begin with persuasion, cajoling, and other efforts to arrive at a mutual decision. However, if this purchase genuinely would undermine the beneficiary’s values, but the beneficiary cannot be dissuaded, then a trustee not only may but should override such a decision.

In previous work, Jaworska (1999) has addressed what valuing is, and related problems in distinguishing values from mere preferences in people with diminished agency. Some accounts of valuing or related constructs are highly cognitively demanding, requiring “principles and ends which [one] articulates as definitive of the good, fulfilling, and defensible life” (Watson 1975) or “a sense of a whole life, a past connected to a future, that could be the object of any evaluation or concern as a whole” (Dworkin 1993). On such highly cognitivist accounts, it may be implausible that people with moderate dementia can have contemporaneous values or meaningfully participate in a dialogical process of valuing. However, on our preferred account, valuing has an emotional and motivational core that is preserved in many people with Alzheimer’s disease and other dementias, even after abilities to plan and reason are severely impaired. This dissociation helps us to recognize both why supported decision-making can be necessary, and also how trustees can aid in articulating values that are still genuinely the beneficiary’s own.

To conclude: we regard appropriate support as involving the promotion not only of beneficiaries’ values, but also of their ability to value. Peterson and colleagues review evidence that surrogate decision-making arrangements can contribute to isolation and stigma, and preliminary evidence that supported decision-making arrangements increase beneficiaries’ confidence. By promoting active engagement and recognizing beneficiaries’ agency to the fullest practicable extent, supported decision-making arrangements hold promise for supporting beneficiaries as ongoing valuers, with their own distinctive normative standpoints that can and should be engaged in deliberations about their care.

Additional information

Funding

WC’s contribution to this work was supported by the National Institute of Mental Health [R01MH126997].