Fluctuating and Diminishing Capacity

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

About 5 minutes. Free, in your language, no account needed.

Why this matters

Consent is usually treated as a single event, which works where capacity is stable.

For a substantial number of participants it is not. Capacity fluctuates within a day, improves as an illness resolves, or declines steadily over the course of a long trial. Treating consent as a moment, when capacity is a moving quantity, produces participants who consented validly and are no longer in a position to continue consenting, with nothing in the process designed to notice.

Fluctuation

Delirium is the clearest case, and it is common in hospital settings, particularly in older people, after surgery, and in critical illness. A person may be lucid in the morning and severely confused in the evening, and the fluctuation can be rapid.

Medication effects shift capacity across a dosing cycle: sedation, analgesia, and treatments for neurological conditions with an on-off pattern.

Pain and fatigue reduce the ability to attend and to weigh, without affecting understanding of individual facts.

Some psychiatric conditions fluctuate over days or weeks, and a person may have full capacity for most of a trial and be temporarily without it.

The practical consequence is that timing is part of the assessment. A person assessed as lacking capacity in the evening may have it in the morning, and consenting in a lucid interval is not a trick. It is the correct course, provided the interval is genuine, the assessment is made at that time, and the timing is documented.

Diminishing capacity

A trial in dementia, progressive neurological disease or advanced cancer may enroll a participant with capacity and continue for years while that capacity declines.

The person consented validly. The question is what happens when they can no longer do so.

Three obligations follow, and none is automatic.

Recognizing it. Nothing in a trial's ordinary process asks whether a participant still has capacity. Unless someone is looking, a participant who lost capacity eighteen months ago is still being treated as consenting.

Transitioning. When capacity is lost, continued participation requires whatever the jurisdiction provides: a legal representative, a court-appointed decision-maker, or in some cases discontinuation, because not every jurisdiction permits a representative to consent to every kind of research.

Honoring prior wishes. A person who expressed a view about continuing while they had capacity should have that view carry weight. This is the strongest argument for asking, at initial consent, what they would want to happen if they could no longer decide, and recording the answer.

Capacity returning

The mirror case, and the one most often missed.

A participant enrolled by a representative during critical illness recovers. They now have capacity, and their own consent must be sought. The representative's authority ended when the person regained the ability to decide.

Failing to seek it means the participant continues in a trial they never agreed to, on the authority of someone who no longer has it. This is a live issue in critical care research and it fails routinely, because the representative-consent process is designed as an entry mechanism with no exit.

The fix is a defined trigger: a scheduled review of capacity for every participant enrolled by a representative, at defined intervals, with an owner.

In practice

A trial in intensive care enrolls participants through representatives, since the participants are sedated and ventilated.

The coordinator builds a simple register: every participant enrolled by a representative, the date, and a weekly review of whether they have regained capacity.

Over four months, 14 participants are enrolled this way. Nine regain capacity. All nine are approached for their own consent, and the register is what makes that happen: the trial's own visit schedule contains no event that would have prompted it.

Two of the nine decline to continue. That outcome is the mechanism working. Both had been receiving a trial intervention they had not agreed to, lawfully, under a representative's authority, and both chose differently when asked.

What this looks like when it goes wrong

A trial in early Alzheimer's disease runs for three years. Participants consent for themselves at entry, and the protocol requires re-consent when the consent form is revised.

By year three, several participants have progressed substantially. One attends visits with his wife, who answers most questions and signs the revised consent form on his behalf.

She has no legal authority to do so. She is his wife, she knows his wishes, and she is acting entirely in good faith, and none of that is the same as being his legally designated representative under the applicable law.

The site accepted her signature because it was reasonable, because she had been present throughout, and because nobody had defined what to do when a participant's capacity declines mid-trial.

The participant has continued in a trial for over a year without valid consent from anyone entitled to give it, and the remedy at this point is limited.

Key takeaways