Self-Directed Medicaid Waivers vs. Agency-Directed Home Care

Choosing between agency-directed and self-directed Medicaid home care determines something very specific: whether a family member can be paid to provide care, who carries the administrative weight of that arrangement, and whether care keeps running as the paid labor market for home care tightens. Self-direction is now the majority model in several states, not a workaround, and understanding how it actually functions, alongside what agency care still does well, is the only way for a family to match their situation to the program that fits it.

Why the agency-vs-self-direction choice matters now

A study by Kuang et al., published in Health Affairs Scholar, found that more than half of dually eligible older adults receiving personal care now use self-direction. That figure settles a question that used to be theoretical: self-direction has moved from alternative arrangement to the dominant model in a meaningful share of states, and any family starting this process today is choosing among two mainstream paths, not a standard option and a niche one.

The timing matters because the paid caregiving workforce that agency care depends on is under real strain. At the same time, Medicaid itself is absorbing the largest projected funding reduction in the program's history through the One Big Beautiful Bill Act, and roughly two-thirds of home care spending runs through Medicaid. Both models face rate pressure from that, though not in the same way or to the same degree.

New York's experience with CDPAP shows how fast these structural shifts can become personal crises. The state's 2025 transition to a single statewide fiscal intermediary triggered immediate litigation and retroactive pay disputes affecting workers and families who had no part in the policy decision behind it. Which fiscal intermediary processes payroll is a choice that looks administrative on paper, but it can delay a caregiver's pay.

What agency-directed care means for the family receiving it

Agency-directed care puts a licensed home care agency in the employer role. The agency recruits, hires, trains, schedules, and pays the caregiver, and the family receives a worker the agency has placed, working the agency's schedule.

That employer status is also where the model's real value sits. The agency runs payroll, handles employment taxes, carries workers' compensation, and manages Electronic Visit Verification compliance through its own systems, so the family carries none of that administrative load. Training is agency-administered and meets state requirements, which gives families a baseline assurance that the worker in their home has met credentialed standards, without the family having to design or verify any of that training themselves.

In most agency arrangements, family members cannot be paid caregivers, because the agency hires from its own worker pool rather than contracting with people the family already knows. The family cannot select a specific individual, cannot set the schedule on their own terms, and has no input into what the agency pays the worker providing the care.

Framed this way, agency care is a coordinated service model built for households that need professional backup more than they need control over who walks through the door.

What self-direction means and the federal framework behind it

Self-direction is a federally defined Medicaid option, grounded in regulation. It gives the person receiving care, or their representative, legal employer authority over their own caregiver, and in programs that also grant budget authority, control over how a set pool of Medicaid dollars gets allocated across approved services.

Federal regulation at 42 CFR 441.450, the 1915(j) option, defines self-direction as the participant's opportunity to exercise choice and control over budget, planning, and the purchase of services, down to the amount, duration, scope, provider, and location of care. Four federal authorities can carry self-directed services: 1915(c) HCBS waivers, the 1915(j) Self-Directed Personal Assistance Services state plan option, the 1915(k) Community First Choice option, and the 1915(i) HCBS state plan option. States choose which of these authorities to adopt, and that choice is why the program a reader encounters in one state looks nothing like the program next door, even though both rest on the same federal foundation.

Employer authority and budget authority are distinct, and many programs combine them. Employer authority means the participant recruits, hires, trains, supervises, and can dismiss the caregiver they choose. Budget authority means the participant directs a fixed pool of Medicaid dollars across approved services. One common misconception deserves direct correction: self-direction does not mean the family runs its own payroll or files its own employment taxes. A Financial Management Service, typically shortened to FMS, handles payroll processing, tax withholding, and workers' compensation on the participant's behalf. The participant is the employer in a legal and practical sense, but not the payroll administrator, which is the detail that tends to be missing from how families imagine this model before they look into it.

Program names multiply the confusion: CDPAP in New York, CDASS in Colorado, IRIS in Wisconsin, CDCS and CFSS in Minnesota. Each is a state's branding of the same underlying federal model. MACPAC's report to Congress found self-direction available in more than half of all Section 1915(c) waivers nationally, and every state now offers at least one consumer-direction option for long-term services and supports. Navigating the federal pathways, working out which of the four authorities applies and which program name attaches to it in a given state, is exactly the kind of bureaucratic tangle that platforms built to handle Medicaid self-direction enrollment end to end, Brevy among them, exist to resolve, moving a family from understanding the options on paper to actually being enrolled in one.

The family member payment question

Whether a specific relative can be paid to provide care is a function of the state rules layered on top of self-direction as a model, and those rules vary enough that the answer for a family in one state can be the exact opposite for a similar family just across the border.

Under the 1915(j) option, a state may choose to allow payment of legally liable relatives, a category that can include a spouse, the parent or guardian of a minor child, or a legally assigned caretaker relative. States are not required to allow this, and many draw the line differently. Agency-directed care, by contrast, excludes family payment in most states as a structural matter rather than an oversight: the agency is the employer, so a relative cannot simultaneously be hired through the agency's pool and function as the qualifying family caregiver.

NASHP's analysis of how Connecticut, Florida, and Virginia built their consumer-directed programs makes the variation concrete. Oregon went further and created a new 1915(c) waiver specifically to let parents of minor children be paid for attendant care, an innovation with no equivalent anywhere in the agency model.

States can also tighten the arrangement financially even where family payment is allowed, through waiver enrollment caps, overall spending caps, or reimbursement rates capped at a percentage of what an agency would be paid for the same work. A family needs its specific state's rules, and confirming which self-direction and agency-directed options actually exist in a given state and under what terms is a discovery step well suited to a service like Brevy, which checks eligibility against those state-level variations before a family commits to one path or the other.

Research on outcomes, costs, and caregiver well-being

The evidence base here runs deeper than most policy debates get, largely thanks to the Cash and Counseling demonstration, one of the most studied experiments in Medicaid long-term care. Across that demonstration and the research that followed it, self-direction comes out ahead on participant satisfaction, on keeping people out of nursing homes, and on caregiver well-being, while total Medicaid costs between the two models run about the same.

The Kuang et al. study adds an important wrinkle: self-direction users in the data had a higher prevalence of chronic disease and higher Medicare costs than agency users. If anything the population using self-direction skews sicker, yet it was associated with significantly fewer Medicare-covered inpatient days in the primary sample (though the effect did not reach statistical significance in the larger sample), a pattern consistent with better day-to-day management of chronic conditions at home.

Cash and Counseling participants under self-direction were more likely to actually receive the paid care they were authorized for, reported higher satisfaction than agency-based counterparts, and had comparable or better health outcomes alongside a lower likelihood of ending up in a nursing home. A systematic review found, with moderate certainty, that self-direction improves caregivers' personal and social well-being, and caregivers also reported fewer unmet needs and better access to care for the people they support, though those two findings carry lower certainty than the well-being result itself.

The fraud and oversight debate

The strongest objection to self-direction is structural, and it deserves to be taken at face value. There is no agency intermediary standing between the Medicaid dollar and the service delivered, so the state itself bears more of the monitoring burden, and Electronic Visit Verification alone confirms that a visit happened, not that the care provided during it met any particular standard.

That critique has real policy weight behind it right now. New York's CDPAP spending grew very rapidly between 2019 and 2023, drawing intensified federal scrutiny, and in January 2026 CMS announced its intent to withhold quarterly Medicaid funds from Minnesota over compliance concerns, though the withholding was blocked once Minnesota requested a hearing. Both episodes show a political and fiscal environment around self-direction that has tightened.

Researchers and disability-rights advocates push back on reading that spending growth as evidence of fraud on its own. KFF's analysis found nearly half of states are already using or planning to use AI in program integrity and monitoring, with North Carolina and Missouri among the states already running AI-enabled tools, which may narrow the oversight gap between the two models over time.

For a family actually choosing between the two, the fraud debate is a policy-level argument playing out above their heads. What they can check directly is whether their state's self-direction program has solid FMS infrastructure, clear EVV requirements, and a case management review process that functions in practice. All three are verifiable before enrollment, and they matter far more to an individual family's experience than where any state lands in the broader argument.

How the current policy environment is shifting the tradeoffs for families

Several forces converging across 2025 and 2026 are making agency-directed care harder to access in practice, which is tilting the relative appeal toward self-direction for a growing number of families. That shift has little to do with self-direction being inherently superior and everything to do with the agency model's dependencies coming under direct strain.

Agency care runs on a paid direct-care workforce that is shrinking. Self-direction, which substitutes a family member for the depleted outside labor pool, does not face that same pressure to nearly the same degree.

Medicaid cuts under the One Big Beautiful Bill Act will reach both models, but agency-directed care carries higher administrative overhead per hour of care delivered, which leaves it more exposed if states respond to federal cuts by compressing reimbursement rates. Eligibility rules are moving too, not just funding. Federal legislation introduced in February 2026, the Combating Deceptive Practices in Assistance Programs Act, would tighten eligibility requirements for Medicaid personal care services, effectively raising the ADL threshold nationally. New York's own ADL requirement change for CDPAP, effective September 1, 2025, shows how fast a state can move the eligibility line on its own, with immediate consequences for who still qualifies. Any family weighing self-direction needs to check their loved one's qualifying need level against both their state's current standard and whatever is still pending in Congress, not just against the rules as they stood a year ago.

The decision framework: which model fits which family situation

Neither model wins in every case. The right fit turns on four things: whether a family has a relative who is both willing and able to take on caregiving, whether that relative is actually permitted to be paid under the state's specific rules, how much administrative capacity the family can realistically take on, and how much continuity of care the recipient needs.

Self-direction tends to fit best under a few recognizable conditions. A trusted family member, an adult child or other close relative, may already be providing the care informally, and the family wants that person paid. And where the care recipient has complex or high chronic-disease needs, a caregiver who already knows their baseline day to day can offer a kind of continuity that rotating agency staff rarely matches.

Agency-directed care tends to fit better under a different set of conditions. No family member may be available, willing, or qualified to take on the employer role. And some families simply want state-certified training and agency-level supervision of the worker without having to build or monitor that training on their own.

Before committing either way, a family has a short, concrete list worth confirming: which self-direction option or options their state offers and under which federal authority, whether the specific relative they have in mind can legally be paid under that state's rules, which FMS providers operate in their area and how enrollment and payroll actually work, and whether any pending state or federal eligibility change, including a new ADL threshold, would affect the care recipient's qualification. Because Medicaid rates and program availability shift by state and by waiver type, confirming eligibility before committing to either path matters, and a free eligibility check through a service like Brevy can clarify which models are genuinely on the table in a given jurisdiction. For most families, the opaque state-by-state variation covered throughout this piece, not the underlying decision between agency care and self-direction, is the actual barrier standing between them and a program they already qualify for.