The Four Abilities That Constitute Capacity

Part of Consent and Vulnerable Populations: Capacity and How It Is Assessed.

About 6 minutes. Free, in your language, no account needed.

Why this matters

Capacity is routinely treated as a property a person either has or lacks, established by their diagnosis or by an impression formed in conversation.

Neither is right, and both produce errors in both directions: excluding people who could decide for themselves, and enrolling people who could not.

Capacity is decision-specific and it is assessable. There is a structure to it, and the structure is what makes assessment something other than an impression.

The four abilities

A person has the capacity to make a particular decision if they can do four things.

Understand the information relevant to the decision. What the trial involves, what will happen to them, the main risks and benefits, and that they may decline.

Retain that information long enough to use it. Long enough to weigh it and reach a conclusion. Someone who cannot retain information for the duration of a conversation cannot use it, even if they understand each piece as it is presented.

Weigh it as part of making a decision. Use the information in a reasoning process. This is the ability most often impaired in subtle ways, and it is the hardest to assess. A person may understand and retain every fact and be unable to bring them to bear, because of a fixed belief, a compulsion, or a condition affecting reasoning itself.

Communicate their decision. By any means. Speech, writing, gesture, or an assistive device. A person who can do the first three and cannot speak has capacity and needs a means of expression, not a representative.

All four are required. Failing any one means the person lacks capacity for this decision.

Decision-specific, not global

A person may have capacity to decide one thing and not another at the same moment, and this is ordinary rather than exceptional.

The threshold rises with the complexity and the consequences of the decision. Deciding whether to give a blood sample is a simpler decision than deciding whether to enter a trial of an experimental intervention with uncertain risks.

So the question is never "does this person have capacity." It is "does this person have capacity to make this decision, now." A capacity assessment recorded without naming the decision it concerned has recorded an opinion rather than a finding.

What capacity is not

It is not the same as a diagnosis. Many people with dementia, intellectual disability or serious mental illness retain capacity for many decisions. Inferring incapacity from a diagnosis is both wrong and a substantial injustice, and it excludes from research precisely the populations who most need evidence about their own conditions.

It is not agreement with the clinician. A person who understands, retains, weighs and communicates a decision the clinician thinks unwise has capacity. The right to make a decision others consider foolish is the point of the concept. A capacity assessment triggered only by refusal is not an assessment.

It is not fixed. Capacity fluctuates with illness, medication, time of day, fatigue, pain and delirium.

It is not all or nothing in practice. Many people with partially impaired capacity can decide with support: information presented differently, more time, a familiar person present, several shorter conversations.

Supporting capacity before replacing it

The obligation is to help a person decide for themselves before concluding they cannot.

Practically: simpler language, one idea at a time, visual aids, more than one conversation, choosing a time of day when the person is at their best, a supporter present, and repetition without impatience.

A person who cannot decide when presented with a 20-page form and a single 20-minute conversation may be entirely able to decide across three short conversations with a plain-language summary. The first approach did not find their capacity absent; it failed to reach it.

In practice

A coordinator assesses an older participant with early dementia for a trial.

The first conversation goes poorly. The participant becomes confused, cannot repeat what the trial involves, and defers to their daughter.

Rather than concluding incapacity, the coordinator changes the approach. A one-page summary in large type. A second conversation in the morning rather than late afternoon. The daughter present but asked not to answer. One idea at a time, with a pause after each.

In the second conversation the participant describes, in their own words, that they would receive a medicine that might or might not help, that they would come to the hospital more often, and that they could stop whenever they wanted. They ask whether the extra visits would be on the same day as their regular appointment.

That is capacity. The first assessment measured the method, not the person.

What this looks like when it goes wrong

A site enrolls participants from a stroke rehabilitation unit. Capacity is assessed by asking whether the participant understands the trial, and recording their yes.

One participant has receptive aphasia. She understands far less than her responses suggest, and her social fluency is intact, so she answers appropriately to the shape of questions without following their content.

She says yes. The assessment records that capacity was confirmed. She is enrolled and does not understand what is happening to her.

The assessment asked the one question that a person with her presentation would reliably answer correctly regardless of understanding. Asking her to describe the trial in her own words would have revealed it immediately.

Key takeaways