What Someone Can Do Is Not What They Owe
When a parent’s health breaks, families usually settle the hardest question without ever asking it. The person who is closest, most organized, or best on the phone with insurers starts doing the work, and the work becomes theirs. Nobody votes. The arrangement simply hardens around whoever proved able first. Capacity determines what a person can do. It does not, by itself, determine what that person owes.
Capacity is also not one thing. It helps to separate four different kinds of it: what a person can do or learn, what they actually accomplish under real conditions, what they can keep doing without unacceptable costs, and what is left after their other legitimate obligations and rest. A person who keeps stretching to cover every gap is showing functional capacity, not spare capacity. Repeated overextension is evidence of strain, not of room.
Willingness is a separate question from ability. The critical distinction is between being unable to contribute in a particular way and declining responsibility altogether. A sibling who has never managed a medication list can learn one, the same way the sibling who manages it now learned by doing. Capacity is evidence of ability, not a debt owed to others. High capacity should create possibility, not servitude.
Who Actually Does the Care
The scale is larger than most families realize. The Caregiving in the U.S. 2025 survey from AARP and the National Alliance for Caregiving counted 63 million Americans, nearly one adult in four, giving ongoing care to an adult or to a child with a complex condition, 20 million more than in 2015. Three in five are women. Nearly one in four gives 40 or more hours a week, and over half handle medical and nursing tasks such as injections, wound care, and medication management, mostly without formal training.

Care for older parents has grown heavier, not just more common. A companion report on caregivers of adults 50 and older found the average caregiver had been at it for five years and was spending 26 hours a week, up from 3.9 years and 22 hours in 2020. Federal time-use data add a useful correction: on the days they give care, women and men spend about the same number of hours. The gap is in who becomes the steady, primary caregiver, which is where sibling research looks.
The work has a price even when nobody is paid. AARP’s Valuing the Invaluable 2026 put the unpaid care given to adults in 2024 at 49.5 billion hours, worth about $1.01 trillion, more than all federal, state, and local Medicaid spending that year.
Daughters, Sons, and the Sibling Game
The most cited study of how siblings split parent care is Angelina Grigoryeva’s 2017 analysis in the American Sociological Review. Daughters gave more care than sons, and the split depended on the whole sibling group: sons gave relatively less care when they had sisters, and daughters gave relatively more when they had brothers. A brother’s care tends to shrink when a sister is available. A sister’s tends to grow when a brother is in the picture.
Economist Timothy Hunt asked why. His model treats parent care as a game played over years, in which each adult child decides where to live, how much to work, and how much care to give while watching the others. Daughters provide roughly three times the care hours of sons in his data, and part of that gap comes from strategy: sons expect a sister to step in, so they hold back, and she compensates. Differences in earnings explained far less of the gap than differences in preferences.
None of this requires bad intent. A brother can sincerely believe his sister is better at it, and the arrangement can still be inequitable. The cost often shows up as sibling tension, which research links to which child takes on care and whether siblings believe a parent favors one of them.
A Cost Is Not an Incapacity
The most common reason given for not sharing care is money: I would lose pay. That is a real cost. It is not, by itself, evidence of incapacity. The ability to rearrange your work is not the same as the ability to afford losing that work, and the sibling with a flexible schedule is usually paying in a currency that does not show up on a pay stub: declined clients, work made up at night, weekends that disappear, time taken from their own children.
Families tend to keep two different sets of books. One person’s lost wages are treated as a cost that must be avoided. The other person’s lost opportunities, evenings, and weekends are treated as the cost of making everything work. Both have paid a price. The fact that the sacrifices appear in different places does not make one of them nonexistent.
A simple illustration shows why the higher earner’s time is not cheaper. If one sibling’s work is worth $75 an hour and the other’s $30, ten hours of care displaces $750 of potential earnings for the first and $300 for the second. The numbers are invented for the example, and gross billings are not net profit, but the direction holds. The research on real caregivers points the same way: women who give care cut their paid hours and earn less, and Social Security earnings records show caregiving itself pulling women’s earnings and employment down. Caregivers also spend an average of $11,508 a year of their own money.
Real hardship matters. A dollar lost can hurt one household more than another, and a sibling living close to the edge is not the same as a sibling who simply prefers not to give up income. But greater earning power, professional independence, or financial success does not create an automatic obligation to subsidize another capable adult’s share of family responsibilities.
Hypothetical rates of $75 and $30 an hour, chosen for illustration only. Not survey data; gross earnings are not net profit.
Leave Law as It Actually Is
The federal Family and Medical Leave Act gives eligible employees up to 12 workweeks of job-protected, unpaid leave in a 12-month period to care for a spouse, child, or parent with a serious health condition, with group health coverage kept in place. Eligibility is narrower than people assume: the employer must be covered, and the employee must have worked there at least 12 months and 1,250 hours, at a site with 50 employees within 75 miles. The law passed in 1993 after two presidential vetoes.

The regulations are broader than the popular summary. Caring for a parent includes arranging a change in care, such as a move to a nursing home, and the employee does not have to be the only family member available. Leave can be taken in pieces, including when care is shared with a sibling, and employers must count it in small increments. That makes FMLA a tool for sharing care, not only for one person taking it all. It does not cover parents-in-law, and it pays nothing unless the employee substitutes accrued paid leave.
A self-employed person has no employer to grant FMLA protection, so a flexible schedule is not a protected leave. Some states fill part of the gap. As of May 2026, paid family and medical leave programs were paying benefits in 12 states and the District of Columbia, and some let self-employed people opt in.
When Conscience Becomes the Trap
An unfair arrangement is exhausting. It can also be morally painful in a specific way. Moral distress in family caregivers describes internal conflict, powerlessness, and self-directed negative emotion when a person cannot act on their values because of constraints around them. A caregiver may be placed where protecting one set of commitments requires violating another: a parent’s safety against a child’s evenings, a job against a hospital’s timeline. A person can make a responsible decision and still feel moral pain, because no choice satisfies every legitimate duty. Researchers have begun linking work and family conflict to moral injury as well, though nothing here is a diagnosis.
Call the pattern the conscience-capacity trap: a person’s skill plus strong moral concern makes them the default responder, while the consequences of stepping back deny them the freedom to decline that others enjoy. The risk is not merely that the caregiver works too hard. It is watching equally capable people exercise freedoms the caregiver cannot exercise without risking harm to someone vulnerable.
The research on choice is blunt. In a national survey, 44 percent of caregivers said they had no choice in taking on the role, and lack of choice predicted more emotional stress and physical strain. Caregiving itself is not uniformly deadly; large matched studies find caregivers live as long or longer than similar noncaregivers. The danger is strain. In the Caregiver Health Effects Study, older caregivers who reported strain had mortality risks 63 percent higher than noncaregivers, and caregiver burnout has its own measurable signs. If the source of distress is an inequitable arrangement, increasing the caregiver’s ability to endure it does not solve the problem.
Presence Is Not a Plan
“You’re here every day, so she has family support at home.”
My presence is not consent to provide care. Federal hospital rules require a discharge planning evaluation of the patient’s needs and of whether the needed services are available. Please document my actual availability in the record and plan for the care she needs without assuming I will provide it unpaid.
Hospitals that take Medicare must meet a federal condition of participation on discharge planning, 42 CFR 482.43. It requires a discharge planning process that treats the patient and caregivers as partners, an evaluation of likely post-hospital needs, and a determination of whether the needed services are available and reachable. The rule does not require a hospital to keep a patient indefinitely because family cannot help, and it does not make every preferred service funded. It does require a real evaluation rather than an assumption.

There is a telling gap in it. When CMS rewrote the rule in its 2019 final rule, it had proposed requiring hospitals to consider caregivers’ capability and availability, and it dropped that requirement as burdensome. Most states have since passed some version of the CARE Act, which requires hospitals to let a patient name a family caregiver, notify that person before discharge, and teach the tasks they will face. Inclusion matters: a meta-analysis found that bringing caregivers into discharge planning cut readmissions by about a quarter. But being named is not agreeing. Maine’s version says it outright.
The presence of a capable relative does not establish a safe, sustainable, or consented-to plan. A family member’s ability to perform care is not evidence of their agreement to provide it. When an institution treats demonstrated involvement as an available resource without checking willingness or limits, that is institutional capacity appropriation, and it is how the institution’s discharge becomes the family’s emergency. The person at the bedside can ask for the evaluation in writing, have their actual availability put in the record, and, for a Medicare patient who believes discharge is premature, use the fast appeal in which the hospital carries the burden of proof.
What the Facility Is Paid to Do
Discharge to rehab does not hand the job of coordinating care to a relative, but the word rehab hides two different kinds of facility. An inpatient rehabilitation facility is a hospital, with intensive therapy and physician supervision, and it falls under the hospital rules. A skilled nursing facility is governed by the long-term care requirements, and Medicare covers it only after a qualifying inpatient stay.

For certified nursing facilities, 42 CFR 483.21 requires a baseline care plan within 48 hours of admission and a comprehensive, person-centered plan built by an interdisciplinary team, reviewed after each assessment. The same section contains the clause the hospital rule lacks: discharge planning must consider caregiver availability and capacity. A nursing facility also may not require a relative to guarantee payment as a condition of admission or continued stay.
A health care power of attorney gives an agent authority to make medical decisions when the patient cannot. It does not make that agent the facility’s scheduler, transport coordinator, or case manager. The useful move is to ask the facility to name a staff owner for each piece of coordination, using a short set of questions, and to put the answers in writing. What responsibilities is the facility being paid to perform, and why are they being shifted to an unpaid family member?
Divide Care by Domain, Not by Helping
Most sibling arrangements fail in the same way. One person carries the plan, and the others help. Helping sounds generous, but it leaves the planner doing the hardest part: noticing what is needed, deciding who should do it, reminding them, and checking that it happened. Sociologist Allison Daminger called that work cognitive labor: anticipating, identifying options, deciding, and monitoring. It is the work that never shows on a chore chart.

The fix is ownership of whole domains. Instead of one sibling asking another to pick up a prescription, one sibling owns medications entirely: the list, the refills, the pharmacy, the questions for the doctor. Another owns insurance and bills. Another owns transportation, or the house, or the weekly calls. Ownership means the owner notices, plans, and follows through without being managed.
Getting there usually takes a family meeting that ends with written ownership of tasks, not a general promise to pitch in. Before it, ten audit questions help separate what is essential from what is preferred, what has actually been agreed from what has merely been assumed, and what would break if the person doing invisible coordination stopped tomorrow.
Paid Help and Public Programs
Some gaps should be filled by professionals, not relatives. Medicaid home and community-based services pay for in-home help, but most are optional for states and capped through waivers, and about 650,000 people were on waiting or interest lists in 2026. PACE wraps all Medicare and Medicaid care, plus transportation and adult day services, into one team for people 55 and older who qualify for nursing home care but can live safely at home.

Every region has an Area Agency on Aging, reachable through the federal Eldercare Locator. The National Family Caregiver Support Program funds counseling, training, and respite through that network, and the Lifespan Respite Care Program funds state respite systems. Medicare’s GUIDE model covers dementia care navigation and up to $2,500 a year of respite for participating patients. Veterans with serious service-connected conditions may qualify for the VA caregiver program, which pays a stipend to the primary caregiver.
Two lesser-known pieces involve money. Some states let families be paid through Medicaid self-direction, and the 2025 survey counted 11.2 million paid family caregivers. A parent who lives with an adult child and cannot care for themselves can make adult day care eligible for the federal dependent care credit. And Medicare now pays clinicians to train caregivers, which is worth asking for by name.
Why the Most Capable Person Becomes the Default
Families have long had a person who keeps everyone connected. Sociologist Carolyn Rosenthal named the role kinkeeping in 1985, and later research keeps finding that one person, usually a woman, holds it. The work is mostly cognitive: keeping track, anticipating, remembering. Recent studies tie an unequal share of that invisible planning work to depression, burnout, and exhaustion that spills into paid work.
The default person is rarely chosen. It is whoever answered the first call, learned the insurance rules, and became fluent in the system. That success conceals the costs and risks that would otherwise make the system’s gaps visible. Competence makes the care possible. Conscience makes withdrawal difficult. Success makes everyone else’s nonparticipation sustainable. The same pattern runs at three levels: the parent who wants to stay home, the sibling who prioritizes their own household, and the institution that sees an involved relative and calls it support.
Rosalynn Carter spent decades arguing that caregiving is something nearly everyone will meet, as giver or receiver. That is the case for treating it as shared work with clear owners. Both siblings’ time has value, both have families and livelihoods worth protecting, and neither person’s circumstances entitle them to consume the other’s freedom. Family participation is not institutional staffing. Advocacy is not case management. And a relative’s competence does not relieve a healthcare facility of its professional obligations.