Assessing consent capacity in participants with cognitive impairment

Research into conditions like dementia and mild cognitive impairment faces a genuine tension at the consent stage: the population most relevant to the research is, by definition, a population where decisional capacity can't simply be assumed the way it would be for a cognitively unimpaired adult. A meta-analysis of decisional capacity in mild cognitive impairment, alongside work applying structured capacity assessments in practice, gives a clearer picture of how this actually plays out, and it's more nuanced than a simple capable-or-not binary.

Why capacity isn't a single yes-or-no judgement

Decisional capacity for research consent typically involves several distinct components: understanding the information presented, appreciating how it applies to one's own situation, reasoning through the relevant considerations, and being able to communicate a choice. A person with mild cognitive impairment may retain full capacity in some of these dimensions while showing genuine difficulty in others. Treating capacity as one overall judgement, rather than assessing each component, risks either wrongly excluding someone who's genuinely capable of making an informed decision, or wrongly including someone who's struggling in a specific dimension the assessment didn't check for.

What the evidence shows about mild cognitive impairment specifically

The meta-analysis found that decisional capacity in people with mild cognitive impairment is meaningfully variable, some individuals show capacity indistinguishable from unimpaired peers, others show clear impairment in specific components, particularly appreciation and reasoning, even when understanding of the basic facts remains largely intact. This variability is itself the key finding: a blanket policy either assuming full capacity or assuming impairment for everyone with an MCI diagnosis would misclassify a substantial number of people in either direction.

What a structured assessment approach actually looks like in practice

Work applying decisional capacity assessment specifically for older research participants with cognitive impairment points toward a few consistent practical elements:

• Assessing each component of capacity separately, rather than a single global judgement, so a specific difficulty with reasoning doesn't automatically override otherwise clear understanding and appreciation.
• Using a structured, validated assessment tool, rather than an informal judgement based on a general conversation, which introduces inconsistency between different assessors.
• Involving a trained assessor, not necessarily the recruiting researcher, to reduce the risk that enthusiasm for enrolment influences the capacity judgement.
• Building in a process for supported decision-making, involving a trusted family member or carer in the consent conversation without that person making the decision on the participant's behalf, where the participant retains capacity but benefits from support.
• Reassessing capacity over time for longer studies, rather than treating a single assessment at enrolment as valid for the study's entire duration, particularly for progressive conditions.

Why getting this right matters for both ethics and the evidence base

Excluding people with mild cognitive impairment from research on the assumption that they can't meaningfully consent doesn't just risk being unfair to individuals who retain genuine capacity. It also risks systematically excluding exactly the population a study on cognitive conditions needs to include to produce valid, generalisable evidence. A properly designed capacity assessment process protects both the individual's autonomy and the scientific validity of research into conditions where cognitive status is central to the research question itself.

The practical takeaway

A diagnosis of mild cognitive impairment is a reason to assess capacity carefully, not a reason to assume it either way. Structured, component-based assessment, applied consistently and revisited over the course of a longer study, is what actually protects both the participant and the integrity of the research they're taking part in.