'I Don't Like Uncertainty, I Like to Know': How and why uveal melanoma patients consent to life expectancy prognostication.

Background: Technological advances have led to cancer prognostication that is increasingly accurate but often unalterable. However, a reliable prognosis of limited life expectancy can cause psychological distress. People should carefully consider offers of prognostication, but little is known about how and why they decide on prognostication. Using uveal melanoma (UM) patients, we aimed to identify (i) how and why do people with UM decide to accept prognostication and (ii) alignment and divergence of their decision-making from conceptualizations of a 'well-considered' decision.

Methods: UM provides a paradigm to elucidate clinical and ethical perspectives on prognostication, because prognostication is reliable but prognoses are largely nonameliorable. We used qualitative methods to examine how and why 20 UM people with UM chose prognostication. We compared findings to a template of 'well-considered' decision-making, where 'well-considered' decisions involve consideration of all likely outcomes.

Results: Participants wanted prognostication to reduce future worry about uncertain life expectancy. They spontaneously spoke of hoping for a good prognosis when making their decisions, but largely did not consider the 50% possibility of a poor prognosis. When pressed, they argued that a poor outcome at least brings certainty.

Conclusions: While respecting decisions as valid expressions of participants' wishes, we are concerned that they did not explicitly consider the realistic possibility of a poor outcome and how this would affect them. Thus, it is difficult to see their decisions as 'well-considered'. We propose that nondirective preference exploration techniques could help people to consider the possibility of a poor outcome.

Patient Or Public Contribution: This paper is a direct response to a patient-identified and defined problem that arose in therapeutic and conversational discourse. The research was informed by the responses of patient participants, as we used the material from interviews to dynamically shape the interview guide. Thus, participants' ideas drove the analysis and shaped the interviews to come.