🧠 Dementia Capacity Assessment Decision-Making Simulator

The Decision at Hand — Capacity Is Decision-Specific, Not Global

Decisional capacity is the clinical determination of whether a person can make a particular decision at a particular time. It is not a fixed trait, a diagnosis, or a global label — a patient with moderate dementia may lack capacity to manage complex finances while retaining full capacity to choose what to eat for dinner or to decline a blood draw. Every capacity assessment must begin by naming the exact decision in question.

• 4: Legal capacity domains (Appelbaum & Grisso, 1988)
• Decision-specific: Capacity is (not global or diagnosis-based)
• Unless proven otherwise: Adults presumed capacitated (legal default position)
• Any new decision: Reassessment trigger (capacity can fluctuate hour to hour)

Why "does this patient have capacity?" is the wrong question

Clinicians are often asked in a hallway consult: "does this patient have capacity?" — as if capacity were a single number stamped on a chart. It is not. The correct question is always: does this patient have capacity to make THIS decision, right now, given THIS level of risk and complexity?

A patient with moderate Alzheimer's disease may be fully capable of deciding whether to accept a flu shot (low complexity, low stakes, easily reversible) while being unable to weigh the risks and benefits of a high-risk cardiac surgery, or to manage a complex investment portfolio (high complexity, high stakes, hard to reverse). The same patient, same day, same diagnosis — different capacity outcomes depending on the decision.

This is the sliding-scale principle: the threshold for capacity rises with the seriousness and irreversibility of the decision. A trivial, easily-reversible choice requires only a low threshold of understanding; a high-stakes, irreversible choice (e.g. do-not-resuscitate status, major surgery, disposing of a home) demands a correspondingly higher and more rigorous demonstration of the four capacity domains.

A dementia diagnosis, or even a low Mini-Mental State Examination (MMSE) score, is never sufficient by itself to declare incapacity. Capacity must be assessed against the specific decision at hand — diagnosis indicates risk, not verdict.

The 4-domain legal standard (Appelbaum & Grisso)

The dominant framework used across US and much of international clinical practice comes from Paul Appelbaum and Thomas Grisso's 1988 synthesis of common-law competency doctrine into four testable abilities:

1. Understanding — can the patient register and recall the factual information relevant to the decision (diagnosis, nature of the proposed intervention, risks, benefits, and reasonable alternatives)?

2. Appreciation — can the patient apply that information to their own personal situation, acknowledging that the diagnosis and its consequences actually apply to them (not to someone else, and not through delusional denial)?

3. Reasoning — can the patient manipulate the information rationally, comparing options, weighing consequences against their own values, and arriving at a decision through a logical process (not necessarily the "right" outcome — a rational process, even one leading to an idiosyncratic choice, satisfies this domain)?

4. Expressing a choice — can the patient communicate a decision, and maintain it with reasonable stability, rather than reversing it unpredictably from moment to moment without any new information triggering the change?

All four domains must be intact, at a threshold matched to the decision's stakes, for the patient to be judged capacitated for that decision.

Common decision types assessed

Capacity assessments are decision-specific; typical triggers include:

• Medical treatment consent — accepting or refusing surgery, chemotherapy, resuscitation status, or any intervention with material risk • Financial decision-making — managing bank accounts, signing contracts, executing a durable power of attorney, making large gifts or investments • Living situation / placement — choosing to remain at home versus moving to assisted living or a nursing facility, often the most emotionally fraught assessment • Testamentary capacity — the specific, distinct standard for validly executing or modifying a will (lower threshold than medical or financial capacity in most jurisdictions) • Driving and independent living — informal but high-stakes safety judgments, often the first capacity question families raise

Each of these carries a different complexity and stakes profile, and therefore a different practical threshold on the sliding scale — even though the same four legal domains are being tested every time.

Understanding — Registering and Recalling the Relevant Facts

Understanding is the most basic and most commonly impaired domain in dementia: the ability to grasp and retain the factual information a reasonable person would need to make the decision — what is being proposed, why, what the risks and benefits are, and what the alternatives (including doing nothing) would be. Memory impairment, attention deficits, and reduced processing speed all erode this domain first and most severely.

• 6 domains: MacCAT-T understanding items (disclosed & rated 0-2 each)
• Plain language: Typical disclosure format (6th–8th grade reading level)
• "Tell it back to me": Paraphrase test (gold-standard comprehension check)
• Necessary, not sufficient: Understanding alone (must combine with other 3 domains)

What understanding actually requires

Understanding is tested by disclosing the relevant facts and then asking the patient to paraphrase them back in their own words — not simply to say "yes, I understand." A patient who nods along but cannot restate the core facts has not demonstrated understanding.

The facts a competent disclosure must cover, matched to the decision:

• The nature of the current condition or problem • The proposed intervention or decision itself, in concrete terms • The material risks of proceeding • The material benefits of proceeding • Reasonable alternatives, including the option of doing nothing • The likely consequences of each alternative, including non-action

Information should be delivered in short, plain-language segments, ideally supplemented by simple diagrams, and checked incrementally rather than in one long monologue — dementia patients often have reduced working memory capacity, so recall degrades sharply as the amount of simultaneously-presented information increases.

The MacCAT-T — a structured, validated assessment tool

The MacArthur Competence Assessment Tool for Treatment (MacCAT-T), developed by Grisso and Appelbaum, is the most widely validated semi-structured instrument for operationalizing the four legal domains at the bedside. It walks the clinician through:

• Disclosure of condition, treatment, risks, benefits, and alternatives — tailored to the individual patient's actual clinical situation (not a generic script) • Understanding items — patient paraphrases each disclosed element; scored 0 (no understanding), 1 (partial), or 2 (adequate) • Appreciation items — patient acknowledges the disclosed information applies to their own situation, and gives a reason for any denial of applicability • Reasoning items — patient compares options, offers consequential and comparative reasoning, and generates a logical (not necessarily "correct") conclusion • Choice — patient states a clear preference and it is checked for stability

The MacCAT-T is not a pass/fail test with a numeric cutoff — it produces a structured clinical picture across the four domains that a clinician integrates with the specific decision's stakes to reach a judgment, ideally alongside psychiatric or neuropsychological consultation for borderline or high-stakes cases.

The MacCAT-T deliberately avoids a single "capacity score." A patient can score imperfectly on recall detail yet still be judged capacitated for a low-stakes decision — the instrument structures the domains, but the sliding-scale judgment remains clinical.

Optimizing conditions to maximize demonstrated understanding

Clinicians have an ethical obligation to maximize a patient's capacity before concluding it is absent — a low score obtained under poor conditions is not evidence of incapacity, it is evidence of a bad assessment. Techniques include:

• Timing — assess during the patient's best hours (many dementia patients show "sundowning," with cognition markedly worse in late afternoon/evening) • Environment — quiet, low-distraction room; hearing aids and glasses in place; adequate lighting • Pacing — short segments of information, frequent checks for comprehension, repetition without condescension • Format — written summaries, simple diagrams, teach-back method ("tell me in your own words what we just discussed") • Treating reversible contributors first — delirium, pain, depression, medication effects, and sensory deficits can all masquerade as incapacity and must be ruled out or treated before a durable capacity determination is made

Appreciation & Reasoning — Applying Facts and Weighing Consequences

Understanding facts is not enough. Appreciation asks whether the patient accepts that those facts genuinely apply to them — not through denial, delusion, or anosognosia. Reasoning asks whether the patient can manipulate the information logically: comparing options, weighing risks against personal values, and tracing a chain of consequences to a conclusion. These two domains are where dementia-related executive dysfunction does its most damage, even when memory for isolated facts is relatively preserved.

• ~40–50%: Anosognosia in dementia (reduced illness awareness, varies by stage)
• Process, not outcome: Reasoning tested by (an unusual choice can still be rational)
• Early & sensitive marker: Executive dysfunction (often precedes memory loss in some dementias)
• Compare to prior stated values: Values-consistency check (family / advance directive input)

Appreciation — "this applies to me"

Appreciation is the bridge between abstract facts and personal reality. A patient can correctly recite "if I refuse the surgery, the infection may spread and could be fatal" (understanding) while genuinely believing "but that outcome will not happen to me" (impaired appreciation) — sometimes due to denial, sometimes due to a delusional belief system, and sometimes due to anosognosia, a neurologically-based unawareness of one's own deficits that is common in frontotemporal and moderate-to-severe Alzheimer's dementia.

Appreciation is probed by asking the patient directly: "do you believe this diagnosis/treatment applies to you? Why or why not?" A patient who denies the applicability of the facts for reasons rooted in impaired insight (rather than a legitimate second opinion or informed disagreement with a diagnosis) fails this domain, even if their factual recall was perfect.

Critically, appreciation failures are distinguished from mere disagreement: a patient who understands the diagnosis, appreciates its personal relevance, and reasons through the options — but reaches a conclusion the clinical team disagrees with — has NOT failed appreciation or reasoning. Capacity protects the right to make choices others consider unwise.

Reasoning — evaluating the process, not grading the outcome

Reasoning capacity is tested by asking the patient to compare the disclosed options and explain how they arrived at a preference: "walk me through how you're thinking about this." The clinician evaluates:

• Comparative reasoning — can the patient weigh option A against option B along relevant dimensions (survival odds, quality of life, burden of treatment)? • Consequential reasoning — can the patient trace forward what is likely to happen under each option? • Logical consistency — does the stated reasoning connect coherently to the stated conclusion, even if the values underlying it differ from the clinician's own?

The single most important principle in this domain: reasoning capacity is about the PROCESS, not the OUTCOME. A cognitively intact patient who reasons through the risks and benefits and then declines a life-saving treatment for religious or personal-values reasons retains capacity. An impaired patient who reaches a "reasonable" outcome by confabulation, guessing, or simply repeating what was last said to them does NOT demonstrate reasoning capacity — the surface agreement with medical advice conceals the absence of a genuine reasoning process, and probing for the "why" behind a decision is essential to detect this.

A common assessment error is treating disagreement with the clinical recommendation as evidence of incapacity. The legal and ethical standard explicitly protects a capacitated patient's right to make choices that the care team considers unwise, so long as the underlying reasoning process is intact.

Executive dysfunction as the hidden driver

Appreciation and reasoning depend heavily on executive function — the prefrontal-mediated capacity for abstraction, mental flexibility, working memory manipulation, and future-oriented planning. In vascular dementia and frontotemporal dementia in particular, executive dysfunction can be prominent even when episodic memory (and therefore MMSE-style scores) remain relatively preserved — meaning a patient can pass simple memory screens while still lacking the reasoning capacity for a complex decision.

This is why capacity assessment cannot rely on global cognitive screening tools (MMSE, MoCA) alone. These instruments correlate imperfectly with functional decisional capacity: they are useful as a general cognitive snapshot and a trigger for further assessment, but a specific, decision-focused capacity interview (such as the MacCAT-T) is required to actually test appreciation and reasoning for the decision at hand.

Expressing a Choice — Stability Over Repeated Inquiry

The final domain is the simplest to state and, in advanced dementia, one of the most revealing: can the patient communicate a decision at all, and does that decision remain reasonably stable when asked again — in the absence of any new information that would rationally justify a change of mind? Chaotic, unanchored reversal of a decision from one hour to the next is a hallmark of impaired capacity, distinct from a patient who thoughtfully changes their mind after learning something new.

• Minutes to days: Stability check interval (repeat the question, vary phrasing)
• If reasoned: Reversal ≠ incapacity (new information can justify a change)
• Still valid: Aphasia / non-verbal choice (gesture, writing, assistive device accepted)
• High in dementia: "Yes-saying" bias risk (acquiescence to the last question asked)

Communicating a choice — form matters less than clarity

Expressing a choice does not require verbal fluency. A patient with expressive aphasia following a stroke-related dementia, or advanced Parkinson's-related dementia with dysarthria, can validly express a choice through writing, pointing, an assistive communication device, or a clear and consistent gesture (e.g. a head shake). What matters is that the choice is communicated clearly enough to be acted upon, not the modality.

What defeats this domain is not physical inability to communicate but psychological instability of the underlying decision itself: a patient who cannot settle on an answer, who gives contradictory responses within the same conversation, or who simply parrots back whatever was most recently suggested to them ("yes-saying" or acquiescence bias, common in moderate-to-severe dementia) has not demonstrated a genuine, expressible choice — even if, in the moment, they produce a verbal "yes."

The repeated-ask stability check

Best practice is to ask about the decision more than once, using different phrasing, at different points in the same interview and — for high-stakes, non-emergent decisions — on a separate occasion (hours or days apart). This is the single most practical bedside technique for the "expressing a choice" domain:

1. Ask the question in its natural form: "what would you like to do?" 2. Later in the same conversation, rephrase and ask again: "just to make sure I understand — which option do you prefer?" 3. If time allows, revisit on a different day or after a period of reduced acute stress (post-operative delirium, severe pain, or acute illness all transiently impair capacity and should be treated before a final determination) 4. Compare responses: a stable answer across all three checks (absent any new disclosed information) strongly supports intact choice-expression; a different answer each time — without any rational account for the change — is a red flag

A change of mind is NOT evidence of incapacity if the patient can explain it by reference to new information, reflection, or a shift in personal values. It is the absence of any coherent account for the instability — pure oscillation — that undermines this domain.

Distinguishing genuine ambivalence from incapacity

High-stakes decisions provoke genuine ambivalence in cognitively intact people too — declining a risky surgery is an emotionally difficult choice that reasonably invites second thoughts. The clinician's task is to distinguish:

• Reflective ambivalence — the patient articulates the tension between competing values ("I'm scared of the surgery but I also don't want to leave my family too soon") and eventually settles, or explicitly states they need more time — this is compatible with intact capacity • Unanchored oscillation — the patient states one answer, is asked again minutes later, and gives the opposite answer with no memory of having just answered differently, no acknowledgment of the change, and no reasoning offered — this is a capacity red flag, especially when paired with impaired understanding or appreciation on the same decision

When oscillation is observed, the correct clinical response is not to force a single answer but to treat reversible contributors (pain, delirium, medication timing) and reassess, and to loop in family/surrogates and, where needed, a formal psychiatric capacity consultation.

Capacity Determination — Integration, Surrogate Activation, and Capacity vs. Competency

The four domain assessments are integrated into a single, decision-specific determination: does this patient, right now, have capacity for this decision? If yes, their choice is honored and documented. If not, the clinical and legal system shifts to a surrogate decision-making pathway — but the assessment is never final for all future decisions, and it is distinct from the formal legal finding of incompetency made by a court.

• Clinical: Capacity assessment (made by treating physician / team)
• Legal: Competency finding (made only by a court of law)
• Spouse → adult child → parent…: Surrogate hierarchy default (varies by jurisdiction)
• On incapacity finding: Advance directive trigger (healthcare proxy activates)

Integrating the four domains into one determination

There is no universal numeric formula that converts four domain scores into a capacity verdict — the MacCAT-T and similar tools structure the clinical picture, but the final judgment is an integrative one, weighted by the sliding-scale principle: a decision with low stakes and easy reversibility (choosing a meal, agreeing to a blood pressure check) requires only modest performance across the domains; a decision with high stakes and irreversibility (major surgery, disposing of the family home, revoking a power of attorney) demands a correspondingly higher and more rigorous demonstration in all four domains, particularly appreciation and reasoning.

A patient may pass understanding and choice-expression cleanly while showing clear impairment in appreciation (denial that the diagnosis applies to them) or reasoning (unable to connect stated risks to a coherent conclusion) — because the domains are not perfectly correlated, all four must be examined individually rather than inferred from a single global impression or diagnosis label.

Capacity (clinical) is not competency (legal)

These two terms are frequently conflated but are legally and procedurally distinct:

• Capacity is a clinical, functional, decision-specific determination made at the bedside by a treating physician (often with psychiatric consultation for complex or contested cases). It can be reassessed at any time, applies only to the decision evaluated, and requires no court involvement for routine clinical decisions.

• Competency is a legal, global status determined by a court, typically in the context of guardianship/conservatorship proceedings. A judicial finding of incompetency is more durable, broader in scope, and strips specific legal rights (e.g., to manage finances, to vote in some jurisdictions, to make medical decisions) until reversed by the court.

In practice, most capacity determinations never reach a courtroom — the vast majority of medical and even many financial capacity questions are resolved clinically, with surrogate decision-makers or existing advance directives stepping in. Court involvement is reserved for disputed cases, cases with no available surrogate, or situations requiring formal guardianship.

Surrogate decision-making and healthcare proxy activation

When a patient is found to lack capacity for a specific decision, authority shifts to a surrogate, following a hierarchy that varies by jurisdiction but typically runs:

1. A previously-designated healthcare proxy / durable power of attorney for healthcare (activates automatically upon a documented incapacity finding — this is the cleanest pathway, which is why advance care planning while capacity is intact is so clinically valuable) 2. A court-appointed legal guardian, if one already exists 3. A statutory default surrogate hierarchy: typically spouse/domestic partner → adult children → parents → adult siblings → other next of kin, though exact order and inclusion of close friends varies by state/country 4. In the absence of any available surrogate and for high-stakes irreversible decisions, court-appointed guardianship may be required

The surrogate is generally instructed to apply substituted judgment — deciding as the patient themselves would have decided, based on known values and any advance directive — rather than simply what the surrogate believes is medically "best." When the patient's own wishes are unknown, the surrogate falls back to a best-interest standard.

The ethical obligation to maximize capacity before concluding it is absent

Because an incapacity finding removes a person's right to direct their own care, clinicians carry a strong ethical (and often legal) obligation to first optimize every modifiable factor before concluding capacity is lacking:

• Treat reversible contributors — delirium, untreated pain, depression, medication side effects, metabolic derangements, and sensory impairment (uncorrected hearing or vision loss) can all masquerade as incapacity and must be addressed first • Simplify and repeat information delivery — plain language, teach-back, written aids, involvement of family for context • Choose optimal timing — assess during the patient's cognitively best window of the day, avoid assessing during acute crisis whenever the decision is not emergent • Reassess rather than assume permanence — capacity in dementia fluctuates, and a patient found to lack capacity for a complex decision today may regain adequate capacity for that same decision, or retain capacity for simpler decisions, at another time

Only after these steps have been taken, and the four domains still show clear, decision-matched impairment, should a determination of lacking capacity — and activation of the surrogate pathway — be finalized.

A capacity determination is a snapshot, not a life sentence. Dementia progresses, but capacity can also fluctuate day to day with delirium, pain control, sleep, and medication timing — every new significant decision, in principle, deserves its own assessment.