How to Size Up What Kind of Help Someone Actually Needs

Where to start when you do not know what you are looking at

The place to start is tasks, not a diagnosis, because tasks are what money buys. Two categories cover almost everything: the basic self-care activities a person performs on their own body — bathing, dressing, eating, toileting, transferring, continence — and the instrumental ones that keep a household running, such as meal preparation, managing money, managing medications, shopping, and housework 1Ref 1Cleveland Clinic (2023).Activities of Daily Living (ADLs and IADLs).The two task categories an assessment walks — basic activities of daily living (bathing, dressing, eating, toileting, transferring, continence) versus instrumental activities of daily living (meal preparation, managing money, managing medications, shopping, housework) — and their use in assessing how much help a person needs.. Each list is worth walking separately, and each task is worth marking in three states rather than two.

Most families use two states, can or cannot, and the middle one is where the entire truth lives. Does it badly covers the man who bathes once a fortnight because the tub frightens him — he can, and he doesn't. Someone else does it now covers the daughter who has been paying the bills online since March without ever deciding to. From the outside, both of those look like independence.

A task a family member has already absorbed is not a task the person can still do. It is a task that failed quietly, and was paid for out of someone's evenings.

Why asking "are you managing?" never works

Because the answer is nearly always yes, and it is usually not a lie. Self-report fails here for reasons that have little to do with honesty: the same capacity that plans a meal is the capacity that keeps an accurate account of the week, so the tasks most worth asking about are the ones the person is least equipped to report on. Everyone narrates their own competence generously. Asking produces a summary. What the situation needs is evidence.

There is a second failure built into the visit itself. A parent who knows an assessment is happening will rally for it, and rallying is a real capacity — it is simply not a sustainable one. A person can hold themselves together beautifully for four hours on Saturday and be unable to hold themselves together at all on Tuesday at four in the afternoon. The performance is not deception. It is effort, and the cost of it arrives after the visitor leaves.

What gets past both problems is unremarkable: more time, at ordinary hours, doing nothing in particular. Late afternoon and evening are more informative than late morning. A weekday is more informative than a holiday.

Reading the rooms

A house keeps a more honest record than a conversation does. The kitchen, the medicine cabinet, the bathroom, the mail, and the car each answer a question that cannot be asked directly, and none of them is proof on its own — a single expired yogurt is a yogurt. What matters is convergence: three rooms telling the same story about the same capacity.

• The refrigerator. Not whether it is full, but whether it is coherent. Food bought and never opened is a shopping trip that happened and a meal that didn't. An empty fridge in a house with money is something else entirely.
• The medicine cabinet. Bottles from two pharmacies, refills that arrived and were never opened, a weekly pillbox with Tuesday still loaded on Friday. Medication is the instrumental task with the shortest fuse.
• The bathroom. A towel bar being used as a grab bar, a bath mat that has migrated, soap that is dry in a place someone claims to shower in daily.
• The mail. Unopened envelopes sort into two piles that mean different things: bills nobody paid, and solicitations somebody answered.
• The car. New scrapes on one consistent corner, and a driver who has quietly stopped making left turns.

These are functional decline indicators, and their value is that they are dated. A pillbox is a record of last week. A conversation is a record of nothing.

Who can do a formal assessment, and what it produces

A family inventory answers what is hard. A formal assessment answers what to do about it, and it opens doors a family list cannot. Area Agencies on Aging coordinate and provide local services — home-delivered meals, homemaker and personal care help, caregiver support — and they are the ordinary front door to a publicly funded assessment and to subsidized services once it is done 2Ref 2Administration for Community Living (2025).Area Agencies on Aging.That Area Agencies on Aging coordinate and provide local services — home-delivered meals, homemaker and personal care help, and caregiver support — that help older adults remain at home, making them the usual public entry point for arranging and subsidizing home-based services..

Other people may already be positioned to assess and simply have not been asked: the clinician who sees the person regularly, a discharge planner if a hospital stay is in the picture, a privately hired care manager. Asking a doctor about function specifically — not "how is she doing" but "she has stopped cooking, is that neurological" — changes the appointment.

What a good assessment produces is not a verdict. It is a level of care determination, a service mix, and a number: how many care hours the plan actually requires, at which times of day. That number is the thing families most need and most often leave without, because it is the number that turns a diagnosis into a budget.

An ADL assessment done well should be legible to a stranger. If an agency cannot read the output and know which worker to send, it is not finished.

When memory is the problem, the question changes

Task inventories under-read dementia, because the risk is not what the person cannot do — it is what they might do while nobody is there. For someone living with dementia, in-home dementia care spans companion services for supervision and socialization, personal care covering bathing, dressing, toileting, and eating, homemaker services for housekeeping, shopping, and meals, and skilled care delivered by a licensed professional 3Ref 3Alzheimer's Association (2025).In-Home Care.That in-home care for a person with dementia spans companion services (supervision and socialization), personal care (bathing, dressing, toileting, eating), homemaker services (housekeeping, shopping, meals), and skilled care delivered by a licensed professional.. The question moves from capability to coverage.

That reframing changes the arithmetic completely. Someone who scores well on every task can still need a person in the house from four in the afternoon until bedtime, because the dementia supervision level — not the task list — is what the hours are actually for. Capability is measured once. Supervision is measured in hours, and hours are what get billed.

It also changes who the right worker is. A dementia trained caregiver is not simply a caregiver who has met someone with dementia. The relevant skill is what happens at the moment of refusal: what a person does when the shower is declined for the third day, whether they escalate or come back in twenty minutes. That skill is worth naming out loud when arranging care, because agencies match on what they are asked for.

The person already doing the job

Every assessment of an older adult is quietly an assessment of two people. Around 53 million U.S. adults were unpaid family caregivers in 2020, providing an average of roughly 24 hours of care each week, with a substantial share caring for someone with Alzheimer's or another dementia and reporting real financial strain 4Ref 4AARP and National Alliance for Caregiving (2020).Caregiving in the U.S. 2020.That roughly 53 million U.S. adults were unpaid family caregivers in 2020, providing an average of about 24 hours of care per week, with a substantial share caring for someone with Alzheimer's or dementia and reporting financial strain.. Twenty-four hours a week is not a rounding error. It is a part-time job with no schedule and no end date.

Unpaid family caregivers provided an average of about 24 hours of care per week in 2020 4Ref 4AARP and National Alliance for Caregiving (2020).Caregiving in the U.S. 2020.That roughly 53 million U.S. adults were unpaid family caregivers in 2020, providing an average of about 24 hours of care per week, with a substantial share caring for someone with Alzheimer's or dementia and reporting financial strain..

An assessment that does not count those hours produces a plan that works only because someone invisible is holding it up. The questions worth putting on the same page as the parent's task list are plain ones. How many nights has the caregiver slept through. What have they stopped doing. What happens the week they get the flu.

The answers often change what the plan should be. Not every finding routes to an aide in the house: adult day services are professionally delivered, community-based therapeutic, social, and health-related services designed to help people keep living in the community 5Ref 5National Adult Day Services Association (2025).About NADSA.That adult day services are professionally delivered, community-based therapeutic, social, and health-related services that help people remain living in the community — a daytime option that can come out of an assessment instead of, or alongside, in-home hours., and they hand a working caregiver back a workday. Sometimes the right output of an assessment is not more care for the parent. It is relief for the person already providing it.

Deciding with them rather than for them

An assessment done to someone produces a plan they will dismantle. AHRQ's SHARE Approach is a five-step model for shared decision making — seek the person's participation, help them compare options, assess their values and preferences, reach a decision together, and evaluate that decision — working explicitly through benefits, harms, and what matters most to them 6Ref 6Agency for Healthcare Research and Quality (2020).The SHARE Approach.The five named steps of AHRQ's SHARE Approach to shared decision making — seek the patient's participation, help them compare options, assess values and preferences, reach a decision together, and evaluate the decision — and its explicit exploration of benefits, harms, and what matters most to the patient.. It was built for clinical choices. It fits this one precisely.

The step families skip is the third. Values are not a formality here; they are the whole variable. A person who would trade six months of life for the right to keep sleeping in their own bed is not being irrational, and a plan built without knowing that about them is a plan built for a different person.

The step families skip second is the fifth. Evaluate the decision. Care needs move, which means an assessment is a repeated measure, not an event — the same list, dated, three months later. What that produces is a trend line, and a trend line settles arguments a snapshot cannot: whether this was a bad winter or a direction, whether the help that was sufficient in March is still sufficient now.

Getting this wrong on the first pass is normal and recoverable. Almost nobody's first care plan is their final one, and the plan is supposed to be revised.