By Jonathan Cassady, National Sales Director, Medicare Supplement Plans‍

A data-grounded briefing for the executives who own Medicare and Medicare Advantage P&L. The numbers behind dementia spend, and the evidence-based lever that turns the most expensive part of it into the most manageable.

Dementia is one of the most expensive conditions in American healthcare. It is also one of the few where the largest costs are the most preventable. For a Government Programs leader, a Medicare product owner, or a Chief Actuary staring at a compressing margin, that combination is not a talking point. It is a strategy.

The margin math has changed, and not in your favor

Every executive with Medicare exposure feels the squeeze, and the data confirms it is structural, not cyclical. PwC projects a 2026 medical cost trend of 8.5 percent in the group market, the third consecutive year at that elevated level¹, while Centers for Medicare and Medicaid Services (CMS) projects national health expenditures to grow 5.6 percent annually through 2032, with Medicare growing 7.4 percent per year².

For Medicare Advantage, the pressure is sharper: after a 2026 effective rate increase of 5.06 percent, CMS finalized a 2027 increase of just 2.48 percent before risk-score trend, and the V28 risk model reached full phase-in on January 1, 2026. Based on a 2026 KFF Report, looking at 2024 health insurers’ financial performance, MA gross margins fell to $1,655 per enrollee in 2024, down 17 percent year over year.³

Traditional Medicare tells a parallel story. In 2025, MedPAC reported that 79 percent of Part A and Part B beneficiaries now sit in Medicare Advantage or an accountable care model, and that the top 5 percent of fee-for-service beneficiaries drive 46 percent of spending.⁴ Medicare Supplement carriers are not insulated either: Plan G rate filing increases for 2026 range from roughly 12 percent to more than 26 percent as claims costs rise.⁵

Whether you carry risk through an MA bid, an ACO benchmark, or a Medigap loss ratio, the same question now defines financial performance: where is the avoidable spend, and who can intercept it before it lands?

Where dementia fits: a small population with outsized spend

Start with prevalence and concentration. An estimated 7.4 million Americans age 65 and older are living with Alzheimer's dementia in 2026, about 1 in 9 people in that age group, and roughly 11 percent of Medicare beneficiaries carry a dementia diagnosis.⁶ They are a modest share of the membership but consume a share of total spend far out of proportion to their numbers.

However, it is the multiplier that should hold any health plan or ACO leader’s attention. The Alzheimer’s Association data demonstrates that Medicare pays roughly three times more per year for a beneficiary with dementia than for one without, and Medicaid pays more than 22 times more. Total payments for the health care, long-term care, and hospice of people with dementia have climbed on a steep, predictable curve, from $321 billion in 2022 to $409 billion in 2026, roughly two-thirds of it borne by Medicare and Medicaid, with the trajectory pointing toward nearly $1 trillion by 2050.^6, 7 

The practical translation for a plan or ACO is simple: a single-digit percentage of your panel quietly drives a large, growing, and concentrated slice of total cost, precisely the concentration that rewards a targeted intervention.

Where the cost actually comes from

The data points to a key insight that many strategies overlook: spending is rarely driven by dementia itself. Instead, costs are concentrated in acute care, emergency visits, hospitalizations, and readmissions, often triggered by unrelated conditions.

People with dementia have about twice as many hospital stays per year as other older adults and carry heavier burdens of heart disease, diabetes, and kidney disease. Their 30-day readmission rate runs materially higher, roughly 21.5 percent versus 14.7 percent for the general geriatric population in one large cohort.⁸ Much of this is preventable: the most common potentially preventable hospitalization diagnoses for patients with Alzheimer's and related dementias are sepsis, injuries such as falls, urinary tract infections, and heart failure.⁹

Read that list again, because it is the whole argument. A urinary tract infection. A fall. Poorly managed heart failure. In a cognitively healthy patient, these issues can be more proactively managed in primary care with traditional care management programs. In a patient with dementia, they become crises, because the early warning sign went unnoticed and unreported until it arrived by ambulance. Cognitive impairment removes the patient's own ability to detect and report a change in condition.

The most expensive part of dementia is not the diagnosis. It is the avoidable acute event that cognitive impairment allowed to escalate.

The lever the data keeps surfacing: the activated family caregiver

If the costs are concentrated in avoidable utilization, the strategy must be to prevent the avoidable. Yet most traditional care management programs are not designed around the realities of this population. They depend heavily on patient self-management, episodic clinical contact, and outreach that occurs after a risk has already surfaced. For a person living with cognitive impairment, those assumptions often break down.

The result is a critical blindspot. The individual most likely to notice the earliest signs of deterioration is the family caregiver; however, they are rarely treated as an activated member of the care team. And the evidence points, with unusual consistency, to this underused resource as one of the most actionable levers available to value-based care.

In 2025, nearly 13 million Americans provided roughly 19 billion hours of unpaid dementia care, an effort valued at $446 billion.⁶ This person already sits at the bedside and observes the subtle change in appetite, gait, mood, or continence that precedes a crisis. What they typically lack is the training to interpret those changes, the confidence to respond, and a direct pathway to a clinical team before the emergency department becomes the default option.

This is the gap traditional care management has largely left unaddressed. Care coordination alone is not enough. Dementia requires a model that extends beyond the clinic, equips the family caregiver to recognize and communicate meaningful change, and creates a reliable connection between what is happening in the home and the clinicians responsible for the patient’s care.

Policymakers have reached the same conclusion. In July 2024, CMS launched the GUIDE Model, an eight-year demonstration built around care coordination, caregiver training, a 24/7 support line, and respite, with the goals of improving quality of life, reducing hospital and ED utilization and delaying nursing-home placement.¹⁰ The federal government is now paying, on aper-beneficiary basis, for exactly the caregiver-centered model the utilization data supports.

What caregiver activation actually delivers

This is where Ceresti Health has built its model and its evidence. Ceresti does not simply support caregivers. It activates them. Through personalized education delivered via acellular-enabled tablet and proactive one-on-one coaching by a dedicated caregiver coach, Ceresti equips the family caregiver with the skills and confidence to better manage this condition, and the means to detect changes in condition early, avoiding hospitalizations. Enrollment is managed directly by Ceresti and the model operates in all 50 states. As one of the few national CMS GUIDE Participants, Ceresti also keeps members within an ACO's existing network.

In a 2024 analysis by the Validation Institute, patients whose caregivers enrolled in Ceresti's program saw total medical costs fall by 42 percent relative to a propensity-matched comparison group, a difference of $569 per member per month, with inpatient costs alone falling 65 percent, both statistically significant.¹¹

PacificSource Health Plans and Ceresti published additional outcomes in 2026, whereby 444 enrolled Medicare Advantage and D-SNP members over 12 months had reduced medical costs by 42 percent versus a matched comparison group, a statistically significant saving of $523 per member per month (p = 0.0165).¹² Ceresti reports comparable results across other health-plan populations, including reductions in avoidable hospitalizations of more than 50 percent and savings of $6,828 per enrolled patient per year.

The savings are driven almost entirely by avoided utilization. This is not a wellness program hoping for a halo effect. It is a targeted intervention aimed precisely at where the money leaks.

What this means for your P&L

The strategic implication differs by line of business, but the direction is the same across all of them.‍

• ‍Medicare Advantage and D-SNP plans. With rates decelerating, V28 fully phased in, and margins down 17 percent year over year reducing avoidable inpatient utilization in your dementia cohort is one of the few levers that improves MLR without touching benefits or network. The PacificSource result, $523 PMPM on an MA and D-SNP population, maps directly to the economic forces and affordability challenges plans are currently facing.
• ACOs and value-based organizations. In a model where shared savings turn on beating a benchmark, a claims-identified dementia panel is a high-yield target, and under GUIDE the caregiver program can be substantially funded by CMS while keeping patients attributed to your network.
• Medicare Supplement carriers. On comprehensive Medigap plans (Plan F and G), the carrier absorbs nearly all of a member's Part A/B cost-sharing, so every avoidable hospitalization flows straight to claims and the loss ratio. As carriers push through double-digit rate increases (steepest on closed Plan F blocks), caregiver activation attacks those admissions at the source. And in "switching" states, with birthday rules or guaranteed-issue windows, where higher-risk members can join the block without underwriting, the savings GUIDE generates help offset that adverse selection and supports rate stability.
  ‍

The bottom line

The data tells a hopeful story for the leaders who manage these business units. Costs are rising, and rates are tightening, but the most expensive part of dementia may be the most preventable part, if you build your strategy around the right person. That person is not a new care manager or a new device. It is the family member already in the room, activated to catch the change in condition and connected to a team that can act.

Manage the avoidable utilization, and you manage the cost. The lever is the activated family caregiver.

Take the next step

If you own a Medicare or Medicare Advantage P&L, the question is no longer whether dementia is driving avoidable cost. The data settles that. The question is whether you are using the one strategy proven to intercept it. By simply using your age-band information, Ceresti can model the opportunity in your own population, giving you a directional estimate of the PMPM impact of activating the family caregiver. Start the conversation at ceresti.com.

‍

Sources and references

¹ PwC Health Research Institute (2025). Medical Cost Trend: Behind the Numbers2026. https://www.pwc.com/us/medicalcosttrend

² Centersf or Medicare and Medicaid Services (2024). National Health Expenditure Projections 2023-2032. https://www.cms.gov/newsroom/press-releases/cms-releases-2023-2032-national-health-expenditure-projections

³ KFF Health News (2026). Health Insurer Financial Performance in 2024. https://www.kff.org/medicare/health-insurer-financial-performance/

⁴ MedPAC(2025). Report to the Congress and July 2025 Data Book. https://www.medpac.gov/wp-content/uploads/2025/07/July2025_MedPAC_DataBook_SEC.pdf

⁵ https://kffhealthnews.org/medicare/medigap-medicare-advantage-premiums-rate-increase-few-alternatives/

⁶ Alzheimer's Association (2026). 2026 Alzheimer's Disease Facts and Figures. https://www.alz.org/alzheimers-dementia/facts-figures

⁷ Alzheimer's Association (2022). 2022 Alzheimer's Disease Facts and Figures. Alzheimer's & Dementia. https://alz-journals.onlinelibrary.wiley.com/doi/10.1002/alz.12638

⁸ BMC Health Services Research (2024). Thirty-day readmission among patients with Alzheimer's disease and related dementias. https://link.springer.com/article/10.1186/s12913-026-14601-3

⁹ HCUP National Inpatient Sample analysis (2024). Preventable Hospitalizations in Adults with Alzheimer's Disease and Related
Dementias. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11691352/

¹⁰ CMS Innovation Center (2024). Guiding an Improved Dementia Experience (GUIDE)
Model. https://www.cms.gov/priorities/innovation/innovation-models/guide

¹¹ Validation Institute (2024). Ceresti Health Savings Validation Report. https://[...]Ceresti%20Health%20Savings%20Validation%20Report.pdf

¹² Ceresti Health (2026). PacificSource and Ceresti Health Demonstrate Significant Cost Savings in Medicare Advantage Members Living with Dementia. https://www.ceresti.com/in-the-news/

‍
Note on figures: All statistics are drawn from publicly available reports and cited by source and year. Ceresti's headline outcome figures (for example, more than 50 percent reduction in avoidable hospitalizations and approximately $6,828 in savings per enrolled patient per year) are reported by Ceresti Health; the 42 percent total medical cost reduction is independently validated by the Validation Institute (2024) and separately demonstrated with PacificSource (2026).

‍