The Last Mile of Care Coordination Is the Family

In brief. Care management programs are built around defined windows, counted contacts and encouragement to follow the evidence-based guidelines of clinical communities. However, care plans and follow-through phone calls cannot guarantee execution. We often find that it becomes the job of the family at home to help with compliance. This article explores where traditional care management models end, the gap that they create and the opportunity to support the family at home. Research suggests that doing so makes good business sense in terms of reducing costs and improving quality outcomes. It closes with five principles that care management programs addressing the last mile must consider.

A Typical Scenario

A patient is discharged. A post-discharge care plan, a list of medications and “available resources” with phone numbers are given to the family when they take their loved one home. The family can understand every instruction in a care plan and still struggle to carry it out. Consider a recommendation for help at home. The loved one is uncomfortable having someone unfamiliar in the house. The patient and family go home with discharge papers and a list of resources with phone numbers. The caregiver is trying to find an available service, understand the cost, and fit the arrangements around work. A clinically reasonable recommendation now depends on several things that the care navigator did not address or consider in the plan.

A care navigator can support discharge and follow-up. Much of the daily work still happens between those contacts, in the kitchen on day three when the medication list does not match the bottles, or when help at home falls through. Family members, neighbors, and others close to the patient may be the ones who see whether the plan works. A model that attempts to coordinate care without them leaves a critical part of the last mile unmanaged. Even with CMS regulations to include the caregivers or support persons in planning for post-discharge^[1], the partnership does not become real until the proper support infrastructure at home is made available.

The gaps in the current care management system

Health plans, providers, and employers have invested in care management for years, including for members with cancer and other chronic conditions. The evidence on benefits has been sobering. The Congressional Budget Office's review of Medicare's disease management and care coordination demonstrations found that most did not reduce spending, and that even the better performers did not save enough to cover their fees.^[2] A randomized trial of an intensive program for high-need patients found no difference in 180-day readmissions (62.3% versus 61.7%).^[3] Medicare's Oncology Care Model reduced episode payments modestly, but not by enough to cover the added payments.^[4] The programs most likely to work were those whose care managers had substantial in-person contact with patients and direct work with physicians.^[2] These results raise a question about whether these programs adequately address what must happen at home.

Today, care management programs are commonly built around a defined period of patient care: an assessment, a plan, scheduled contact, and referrals. CMS billing rules show the pattern: transitional care management runs on a 30-day window, and chronic care management is counted in minutes of staff time per month.^[5]^[6] Both name caregivers, but the structure is time-boxed and counted in contacts, while the household's work is continuous.

What programs are built to do	The gaps created
Set goals Build a patient-centered plan with measurable goals	Who executes the tasks while the patient is recovering Feasibility for home-care, Caregiver capacity, and ability to navigate through adversity with the patient
Stay in contact Contact within set windows, then periodic outreach	Problems arise between contacts Knowledge of what is critical, urgent or normal
Carry out the plan Issue the plan, share a copy, rely on report	The person doing the work, often without formal training^[7] Did actions actually take place; Billing requirements say little about verifying that agreed actions happened
Watch capacity List a caregiver assessment as a possible plan element	CMS billing requirements do not require consistent monitoring of caregiver capacity, including as strain increases
Resolve barriers Refer to a service	Work may end at the referral, while the family meets the wait, the cost, or the denial

Each gap in the table has the same feature: the information or the action sits with the family. A program can close some of them only with the caregiver as a partner. Services offered and navigator activity tell us what the program did. True care coordination needs to show whether agreed actions happened and what followed for the patient and caregiver. We cannot simply put the burden on the family without providing the support and infrastructure in place to facilitate the kind of coordination we all seek.

A caregiver-inclusive program should show how it supports the patient's goals, safety, and quality of life while keeping the demands on the caregiver sustainable. Partnerships rely on common goals, foundational understanding of roles/responsibilities, training, data and infrastructure. We have not consistently provided ample support in any of those areas to our community of caregivers and the loved ones they care for.

It makes business sense to support our caregivers

The industry is moving towards a more fixed fee payment system, and 55% of eligible Medicare beneficiaries are now enrolled in Medicare Advantage.^[8] As such, there is significant motivation to keep patients in the lowest cost setting available where recovery is most plausible. In many cases, this is the home. Research suggests that caregiver inclusion may change cost-relevant outcomes. A 2017 meta-analysis of 15 trials of discharge planning interventions found that integrating caregivers was associated with about 25% lower relative risk of readmission at 90 days.^[9] A separate conference abstract reporting a meta-analysis of 54 randomized trials found that transitional care interventions that engaged caregivers showed a larger reduction in all-cause readmissions at two or more months than those that did not.^[10] The interventions in these studies were multifaceted, and neither analysis can show that caregiver engagement alone caused the difference, but the findings merit further testing. Under fixed-fee arrangements, health systems and payers carry the cost of a readmission, so they are the ones who capture the savings when one is avoided.

Caregiver health deserves attention in its own right. The evaluation should ask what the demands of the plan mean for the person carrying it out, alongside what happens to patient spending. An Urban Institute analysis using 1999 national survey data linked to administrative records found high caregiver stress was a significant predictor of nursing home entry.^[11] A study published in 2025 followed 1,409 older adults and their caregivers after emergency department discharge in Québec, using data collected in 2019–2021. Each point on the Zarit Burden Interview was associated with modestly higher odds of a 30-day emergency department return visit.^[12]

These findings suggest that a caregiver-inclusive program may improve outcomes and lower spending. They also support asking whether a plan that becomes unworkable at home, or a caregiver who reaches their limit, is a risk the program can recognize and address. What would a care management program need to do differently to make the plan workable at home?

Five principles for care management that reaches the last mile

Care management programs that address the last mile start from the work the household is being asked to do. Five principles follow from the gaps above. For each one, a program should be able to show the last time it changed a family's plan.

Plan with the household, not only for the patient. A plan built on clinical goals alone may not be acceptable to the patient or workable at home. Evidence on personalized care planning for adults with long-term conditions suggests small improvements in some outcomes, although its relevance to this caregiver-inclusive setting remains indirect.^[13] Programs should establish what matters to the patient, whether the proposed help is acceptable, and who in the household has agreed to do what. That includes other family members, neighbors, and paid help, with the patient's consent. The plan should also account for patients who have no available family caregiver.
Stay present between contacts. Scheduled outreach and counted contacts leave long stretches when only the household is watching. Programs should give the people at home a way to recognize what is normal, what is urgent, and what is critical, and a clear path to someone who will respond.
Size the work to the caregiver, and confirm it happened. More than half of family caregivers of adults perform medical or nursing tasks, yet just over a fifth report receiving training for them.^[7] Programs should assess the caregiver's willingness, skills, time, and access to help before assigning tasks, provide training where it is needed, and verify that agreed actions took place. Services offered and navigator activity show what the program did. They do not show whether the plan was carried out.
Watch caregiver capacity, and let the answer change the plan. Caregiver strain is associated with nursing home entry and with return visits to the emergency department.^[11]^[12] Monitoring solely to produce a score does not help the family. Programs should respond when the caregiver needs support or the plan needs to change.
Own the barrier until it is resolved. When a service is unavailable, unaffordable, or not covered the way the family assumed, a program may record the problem without resolving it. Resolution often falls to family members working with the clinical team on the patient's behalf. Programs should name who owns each barrier and follow it through to an outcome, and should be clear about the barriers, such as service shortages and coverage rules, that no caregiver can close.

A program that applies these principles considers what it asks of a family before judging the family's performance. Treating family members as partners does not make them responsible for filling gaps the care system has left unresolved.

Return to the family with the discharge papers and the list of phone numbers. What happens next can affect the patient, the people supporting them, and the organization responsible for the cost of care. Covering the last mile requires treating the family as part of the care team and being honest about what the plan asks of them and what it costs them. Before funding or buying a program, ask it to show what changes when that plan becomes hard to carry out.

Download the one-page version of the five principles (PDF)

Sources

[1] CMS. Hospital and home health discharge planning final rule. Federal Register, September 30, 2019; 84 FR 51836. Hospital discharge planning provisions at 42 CFR 482.43.Back to text
[2] Congressional Budget Office. Lessons from Medicare's Demonstration Projects on Disease Management, Care Coordination, and Value-Based Payment. January 2012. https://www.cbo.gov/publication/42860Back to text: 1, 2
[3] Finkelstein A, Zhou A, Taubman S, Doyle J. Health care hotspotting: a randomized, controlled trial. New England Journal of Medicine 2020;382:152-162.Back to text
[4] Abt Associates. New study from Abt finds Oncology Care Model reduces Medicare payments. https://www.abtassociates.com/who-we-are/news/news-releases/new-study-from-abt-finds-oncology-care-model-reduces-medicareBack to text
[5] CMS. Transitional Care Management Services. MLN908628, August 2025.Back to text
[6] CMS. Chronic Care Management Services. MLN909188, June 2025.Back to text
[7] AARP and National Alliance for Caregiving. Caregiving in the US 2025. Executive summary, July 2025.Back to text: 1, 2
[8] Freed M, Biniek JF, Damico A, Ochieng N, Neuman T. Medicare Advantage in 2026: Enrollment Update and Key Trends. KFF, June 5, 2026 (updated July 1, 2026). https://www.kff.org/medicare/medicare-advantage-enrollment-update-and-key-trends/Back to text
[9] Rodakowski J, et al. Caregiver integration during discharge planning for older adults to reduce resource use: a meta-analysis. Journal of the American Geriatrics Society 2017.Back to text
[10] Levoy K, et al. The moderating effect of caregiver engagement in transitional care intervention outcomes: a meta-analysis. Innovation in Aging 2021;5(Suppl 1):356. Conference abstract.Back to text
[11] Spillman BC, Long SK. Does high caregiver stress lead to nursing home entry? Urban Institute for ASPE, January 2007.Back to text: 1, 2
[12] Germain N, et al. Caregiver burden and 30-day emergency department revisits. JAMA Network Open 2025;8(9):e2531166.Back to text: 1, 2
[13] Coulter A, et al. Personalised care planning for adults with chronic or long-term health conditions. Cochrane Database of Systematic Reviews 2015;(3):CD010523. The review cited by Mate K, et al., Evidence for the 4Ms: interactions and outcomes across the care continuum. Journal of Aging and Health 2021.Back to text