Should Assessments of Decision-Making Capacity Be Risk-Sensitive? A Systematic Review

Abstract

BackgroundThe concept of decision-making capacity (DMC) or competence remains controversial, despite widespread use. Risk-sensitive DMC assessment (RS-DMC)—the idea that the higher the risk involved in a decision, the greater the decisional abilities required for DMC—has been particularly controversial. We conducted a systematic, descriptive review of the arguments for and against RS-DMC to clarify the debate.MethodsWe searched PubMed/MEDLINE (National Library of Medicine), PsycInfo (American Psychological Association) and Philpapers, updating our search to February 15th, 2022. We targeted peer-reviewed publications in English that argue for or against RS-DMC. Two reviewers independently screened the publications and extracted data from each eligible manuscript.ResultsOf 41 eligible publications, 22 supported a risk-sensitive threshold in DMC assessment. Most arguments for RS-DMC rely on its intuitive appeal and practical merits. The arguments against RS-DMC primarily express concerns about paternalism and the seeming asymmetry between consent and refusal; critics of RS-DMC support epistemic, rather than substantive (i.e., variable threshold), risk-sensitivity; counterarguments responding to criticisms of RS-DMC address charges of paternalism and exhibit a notable variety of responses to the issue of asymmetry. Authors used a variety of frameworks regarding the definition of DMC, its elements, and its relation to decisional authority, and these frameworks were significantly associated with positions on RS-DMC. A limitation of our review is that the coding relies on judgment and interpretation.ConclusionThe review suggests that some of the debate about RS-DMC stems from differences in underlying frameworks. Most defenses of RS-DMC rely on its intuitive appeal, while most criticisms reflect concerns about paternalism or the asymmetry between consent and refusal. Defenses of RS-DMC respond to the asymmetry problem in a variety of ways. Further research is needed on the implications of underlying frameworks, the asymmetry problem, and the distinction between epistemic and substantive models of RS-DMC.