Advance Medical Decision-Making Differs Across First- and Third-Person Perspectives

Abstract

Background

Advance healthcare decision-making presumes that a prior treatment preference expressed with sufficient mental capacity (“T1 preference”) should trump a contrary preference expressed after significant cognitive decline (“T2 preference”). This assumption is much debated in normative bioethics, but little is known about lay judgments in this domain. This study investigated participants’ judgments about which preference should be followed, and whether these judgments differed depending on a first-person (deciding for one’s future self) versus third-person (deciding for a friend or stranger) perspective.

Methods

A vignette-based survey was conducted (N = 1445 US Americans; gender-balanced sample), in a 3 (relationship: self, best friend, stranger) × 2 (T1 preference: treat, do not treat) × 2 (T2 contrary preference: ambiguous, unambiguous) design.

Results

Participants were more likely to defer to the incapacitated T2 preference of a third-party, while being more likely to insist on following their own T1 capacitated preference. Further, participants were more likely to conclude that others with substantial cognitive decline were still their “true selves,” which correlated with increased deference to their T2 preferences.

Conclusions

These findings add to the growing evidence that lay intuitions concerning the ethical entitlement to have decisions respected are not only a function of cognition, as would be expected under many traditional bioethical accounts, but also depend on the relationship of the decision to the decision-maker’s true self.

Keywords:

• Medical decision-making
• advance directives
• experimental bioethics
• self/other
• true self

Acknowledgments

The authors would like to thank the members of the Oxford BioXPhi Lab, a workshop audience at the Yale-Oxford-Jagiellonian BioXPhi Conference in Kraków, and three anonymous peer reviewers, whose comments and feedback immeasurably improved this paper.

Disclosure statement

The authors report no conflicts of interest.

Author contributions

All listed authors helped conceive and implement this study, and draft this paper.

Ethics approval

This study was approved by University of Oxford CUREC, reference R80692/RE001.

Notes

1 As Lewis and Holm (2022) discuss, the terms “competence”, “capacity”, “mental capacity”, and “capacity for autonomy” have tended to be treated as equivocal in the fields of bioethics and medical ethics. In general, these terms are used to refer to a person’s cognitive functioning to comprehend, understand, retain, use, and weigh up information and communicate their decision (Foddy and Savulescu 2006; Holroyd 2009; Schaefer et al. 2014; Lewis 2021). In addition to this prevailing understanding, theorists have recently proposed alternative conceptions of capacity less linked to cognitive functioning (Mirza and Appel 2024; Lewis and Holm 2023; Appel 2022a; Toomey 2022). Other accounts of capacity hold that the only relevant cognitive ability required for capacity is the ability to decide and express decisions (Shiffrin 2004). Throughout, we use “capacity” to refer to the still-prevailing cognitive conceptualization, and the “standard” or “traditional” bioethical account to refer to approaches that hold that some threshold of capacity higher than the ability to merely express a decision is required for decisions to hold substantive normative weight in medical decision-making contexts.

2 Sometimes known as “bioxphi,” experimental philosophical bioethics employs experiment-based designs drawn from experimental psychology and other cognitive sciences to advance debates in bioethics (Earp et al. 2020; Lewis 2020; Earp et al. 2021; Lewis, Demaree-Cotton, and Earp 2023). It is thought that experimental evidence about the moral intuitions of ordinary people is particularly significant in healthcare contexts because it is ultimately ordinary people who must make these sorts of real-life decisions with real-life stakes and consequences (Earp, Latham, and Tobia 2020, 72; Earp et al. 2020; Earp et al. 2021; Lewis, Demaree-Cotton, and Earp 2023).

3 In the Best Friend case, the participant was first prompted to enter the initials of “someone you are close with in real life, within 5 years of your age,” which we refer to as the participants’ “Best Friend.” These initials were piped throughout the vignettes.

Additional information

Funding

This work was supported by the Uehiro Center for Practical Ethics at Oxford University, and the Spanish Ministry of Science and Innovation (grant number PID2020.119791RA.I00).