Knowing how to assess capacity to consent is a four-step conversation: check whether the patient can understand the decision, retain it long enough to work with it, use or weigh the relevant information, and communicate the choice — always for one specific decision, at one specific moment. If all four hold, the patient has capacity, whatever their diagnosis.
Capacity assessment is one of the most examinable skills in psychiatry — it appears in MRCPsych CAMPs, RANZCP written papers, USMLE vignettes, and everyday ward practice. This guide walks through the legal framework, the four pillars with exactly how to test each one, a structured conversation format, documentation wording you can reuse, and what to do when capacity is lacking. Read it once and you can run a defensible capacity assessment tomorrow morning.
What capacity actually means in law and practice
Capacity is decision-specific and time-specific. A patient with well-treated schizophrenia may lack capacity to consent to a complex financial decision while retaining full capacity to consent to clozapine monitoring. A person in a delirious fog at 3 a.m. may regain capacity by lunchtime when the infection is treated. There is no global, permanent "has capacity / lacks capacity" label — only answers to the question capacity for which decision, right now?
The legal test varies by jurisdiction, so check your local legislation. In England and Wales, the Mental Capacity Act 2005 sets the framework: capacity is presumed unless it is established that the person cannot decide for themselves (section 1), and the four-pillar test sits in section 3. Scotland operates under the Adults with Incapacity (Scotland) Act 2000, Ireland under the Assisted Decision-Making Act 2015, and US practice generally follows state-level statutes with the Oakley framework — an assessable, teachable structure very close to the four pillars. In research contexts, regulations such as the Common Rule and ICH-GCP add their own consent standards.
The consistent core everywhere: the burden of proof rests with the person asserting incapacity, and a psychiatric diagnosis alone never settles the question. A person with a delusion about a specific treatment may still weigh options coherently; a person with no diagnosis may be incapable of one decision while drunk, terrified, or overwhelmed.
The four pillars of capacity — and how to test each
Most statutory tests decompose into four abilities. The art of the assessment is testing each one with a concrete question rather than a vague "do you understand?" Here is the table to memorise — pillar, what it means, how to test it, and phrasing for your notes.
| Pillar | What it means | How to test it | Documentation phrasing |
|---|---|---|---|
| Understand | Comprehend the information relevant to the decision | "In your own words, what does this treatment involve?" | "Able to describe the proposed procedure and its main risks in her own words." |
| Retain | Hold the information long enough to use it (minutes to hours, not days) | Re-ask a key fact after 5–10 minutes of discussion | "Retained the information about alternatives across a 15-minute discussion." |
| Use or weigh | Deliberate — consider risks, benefits, and alternatives against each other | "What matters most to you in this decision? What would change your mind?" | "Weighed the risks of surgery against continued physiotherapy and articulated a rationale." |
| Communicate | Express a choice by any means (speech, writing, signing, gesture) | Accept non-verbal responses; do not require speech | "Communicated a clear choice to decline, consistently over two conversations." |
The subtle failure point is "use or weigh." Patients often repeat information flawlessly — they understand and retain — yet cannot weigh it because a fixed belief distorts the calculus ("the consultant is poisoning me, so the risks are irrelevant"). The test is not whether the patient reaches the same conclusion as you, but whether they can engage with the information at all. Rejecting an option for reasons you find irrational is not incapacity; being unable to consider the option's pros and cons is.
Exam trap
Capacity is decision-specific — a patient can refuse ECT yet consent to blood tests, refuse one antipsychotic while accepting another. And diagnosis ≠ incapacity: "she has schizophrenia, so she can't consent" is wrong on both counts. Examiners fail candidates who assess global capacity instead of capacity for the decision in front of them.
How to structure the capacity conversation
Treat the assessment as a short, deliberate clinical interview rather than an interrogation. A workable structure:
- Set the frame. "I'd like to make sure you have enough information to decide about X. I'm going to ask you some questions, and this helps me understand how you're thinking about it — it's not a memory test."
- Check the basics first. Orientation, arousal, and hearing/vision aids in place. A patient who cannot hear you has not failed comprehension — you have failed the setup.
- Give the information. Explain the decision, options, main risks and benefits, and what happens if nothing is done — in plain language, then ask for teach-back.
- Test each pillar explicitly with the questions from the table above.
- Probe the reasoning. "What is your understanding of what happens if you refuse?" separates irrational preference from distorted weighing.
- Invite questions, then confirm the choice. Check once more whether new information would alter the decision.
If any pillar fails for this decision, capacity is absent — and you stop relying on the patient's consent and move to the legal route for substitute decision-making. If all four pass, the decision stands, even if you strongly disagree with it. Autonomy protects choices you would not make yourself.
Assessing capacity for treatment vs research consent
Treatment consent and research consent are tested with the same four pillars, but the information burden differs. Research consent requires understanding of voluntariness, randomisation, the right to withdraw without affecting care, and what data will be used. Capacity for treatment does not imply capacity for a complex trial — a patient may consent to a medication yet not grasp placebo allocation or a 12-month double-blind phase.
For research, most ethics frameworks additionally require an independent capacity assessment for higher-risk studies, and consent must be re-checked at each study phase when the condition fluctuates. When capacity is lacking for research, most jurisdictions permit a legal representative to consent on the person's behalf for minimal-risk studies only, with the participant's assent still sought.
Documenting the assessment — and what to do when capacity is lacking
Your note is the assessment's lasting record; if the decision is challenged later, the note is the evidence. Write it immediately, structured around the pillars: the decision assessed, information provided, each pillar tested with the patient's actual responses, your conclusion, and the next step. Avoid vague lines like "patient has capacity" with no supporting detail — state which decision, at what time, and why.
When capacity is lacking, three moves follow. First, treat the decision as unmade by the patient and move to the lawful route — best-interests decision-making under the MCA's framework, a lasting power of attorney or court-appointed deputy where one exists, or an advance decision to consider. Second, treat any reversible cause: delirium, infection, intoxication, severe agitation, or medication side effects frequently restore capacity within hours to days, so for urgent but deferrable decisions, treat the cause and reassess. Third, escalate when the decision is serious or disputed — serious medical treatment usually warrants a second opinion or court involvement depending on jurisdiction. Patients who lack capacity for a decision still participate: involve them, respect prior wishes, and never let the assessment become a reason to stop talking to them.
If you are studying for MRCPsych or RANZCP, capacity is a recurring theme across ethics, law, and clinical stations — Preppsy's concept library links it to mental-health legislation, risk, and the therapeutic relationship so you revise it in context rather than in isolation.
Frequently asked questions
What are the four pillars used to assess capacity to consent?
The four pillars are understanding, retention, using and weighing the information, and communicating a decision. You test each with a direct question: ask the patient to explain the decision in their own words, re-check a key fact later in the conversation, ask what matters most to them in deciding, and accept any reliable way they express a choice, spoken or not.
Does a psychiatric diagnosis mean a patient lacks capacity to consent?
No. Capacity is decision-specific and time-specific, and diagnosis never substitutes for assessment. A patient with schizophrenia can retain full capacity for most decisions, while a person with no psychiatric diagnosis may lack capacity for a single decision during intoxication, delirium, or acute distress. The burden of proving incapacity always rests with the person asserting it.
How do you document a capacity assessment in medical notes?
Record the specific decision and date, the information you gave, the patient's responses pillar by pillar in their own words, your conclusion with reasoning, and the next step taken. Concrete quotes — such as a patient declining with "I understand I may need a transfusion but I'd rather try iron tablets" — make the note defensible far more than the bare statement that capacity was present.
Is capacity to consent to treatment the same as capacity to consent to research?
The four-pillar test is the same, but the information burden differs. Research consent requires understanding voluntariness, randomisation, withdrawal rights, and data use, so a patient may have capacity to consent to treatment yet not to a complex trial. Higher-risk studies typically demand an independent capacity check and ongoing assent from the participant.
Key takeaways
- Capacity is decision-specific and time-specific — never a global label attached to a diagnosis.
- The four pillars are understand, retain, use or weigh, communicate; test each with a concrete question, not "do you understand?"
- The usual failure point is using and weighing — irrational conclusions are allowed, distorted deliberation is not.
- Document the decision, the information given, the patient's responses, and your conclusion — the note is the evidence.
- If capacity is absent, move to the lawful substitute route, treat reversible causes first, and escalate serious or disputed decisions.
- Research consent demands more information than treatment consent; check your local legislation (MCA 2005 in England and Wales).
Master the four pillars, structure the conversation, and document every step — then run the assessment exactly as you would in an exam station. If you want the full ethics-and-law syllabus mapped into a revision plan, see Preppsy's MRCPsych programme and build your capacity to consent knowledge into an adaptive study schedule.
Cover image: “Assessing Capacity to Consent (cover)” by Preppsy Editorial Team (Original artwork) via Preppsy.
