Caregiver Burden as a Hidden Vital Sign in Dementia Care: Why Routine Measurement Belongs in Geriatric Practice

Dementia Care Is Dyadic Care

Dementia care is rarely delivered to one person in isolation. The patient may be the identified recipient of medical care, but medication administration, supervision, transportation, behavioral management, personal care, financial coordination, and communication with clinicians often depend on an unpaid caregiver. In 2025, nearly 13 million Americans provided more than 19 billion hours of unpaid dementia care, valued at more than $446 billion.[1]

The caregiver’s capacity is therefore part of the patient’s clinical environment. A care plan that is pharmacologically sound but operationally impossible for the caregiver is not workable. The same is true when behavioral recommendations assume uninterrupted supervision, when follow-up depends on transportation that is no longer available, or when a caregiver’s own illness makes essential tasks unreliable.[6,7,15]

This does not mean that clinicians should treat caregivers as extensions of the patient. Caregivers are distinct people with their own health, privacy, preferences, and limits. It means that dementia care should assess the dyad: the person living with dementia and the person or network carrying much of the day-to-day care.[15,16]

What the “Hidden Vital Sign” Analogy Means

Caregiver burden is a multidimensional, caregiver-reported construct that can include emotional, physical, social, personal, and financial strain.[11,12] Calling it a hidden vital sign is useful only if the analogy is kept within bounds.

A physiologic vital sign has standardized units, expected ranges, and well-defined emergency thresholds. Caregiver-burden instruments do not. Scores vary by tool, language, culture, caregiving arrangement, disease stage, and the specific domains being measured. A positive score is not a diagnosis, evidence of poor caregiving, or proof that a caregiver cannot continue in the role.[9-12]

The analogy is still clinically valuable because burden can be:

• elicited quickly;
• recorded in a consistent format;
• trended over time;
• interpreted alongside function, behavior, safety, and social support; and
• used to trigger a more complete assessment or change in the care plan.[2,6-8,15]

In a longitudinal cohort of caregivers enrolled in a comprehensive dementia-care program, a direct question about regularly feeling completely overwhelmed identified caregivers with greater distress related to the patient’s behavioral symptoms, more depressive symptoms, and higher overall burden. These differences persisted for several outcomes at 1 year after adjustment for baseline scores. The one-question signal did not predict long-term nursing home placement or mortality in that cohort.[2] Its best use is rapid triage, not a substitute for comprehensive assessment.

Why Routine Measurement Adds Clinical Information

It detects risk that a patient-only assessment misses

A patient may appear medically stable while the caregiving arrangement is deteriorating. Sleep disruption, wandering, resistance to care, incontinence, medication complexity, financial strain, and loss of backup help can make an unchanged diagnosis substantially harder to manage. A caregiver may also minimize difficulty during a patient-centered visit, particularly when the person with dementia is present.

A standardized question or brief instrument creates a legitimate opening for disclosure. It also reduces dependence on whether the clinician happens to ask the right question during a crowded encounter.[2,4,5]

It supports longitudinal interpretation

A single score is less informative than a trajectory. A rising burden score after hospitalization, functional decline, new nighttime behavior, caregiver illness, or loss of respite may signal a narrowing margin of safety even before a crisis occurs.[2,8,15]

The trend should be interpreted with the item pattern. Two caregivers can have the same total score for different reasons. One may be exhausted by nighttime supervision; another may be distressed by behavioral symptoms, financial pressure, or conflict among family members. The response should address the dominant drivers rather than the total alone.[11,12,15]

It helps test whether the plan is feasible

Caregiver assessment can expose a mismatch between the prescribed plan and the resources available to implement it. This may lead to simplification of medication schedules, more realistic behavioral strategies, home-health or social-work referrals, adult day services, respite, transportation support, or a discussion about a higher level of care.[6,7,15]

It identifies a documented implementation gap

In a national survey of 106 primary care physicians, including 62 geriatricians and 44 general internists, only 10.5% reported using a standardized formal caregiver assessment during the prior year. Lack of time was the most commonly reported barrier, while better referral options and easier referral mechanisms were leading facilitators.[4] In a review of 211 records for veterans with dementia, most records identified a caregiver, but only 28% assessed caregiver well-being and 41% assessed caregiver needs.[5]

These studies do not prove that a scored instrument improves outcomes. They show that caregiver status is commonly visible while caregiver strain remains incompletely assessed.

Current Clinical and Policy Context

Routine burden measurement is a pragmatic practice recommendation, not a universal United States screening mandate. Current Medicare policy nevertheless creates a clear clinical home for caregiver assessment.

Medicare’s cognitive assessment and care-planning service, commonly reported with CPT code 99483 when all requirements are met, includes identifying caregivers and assessing caregiver knowledge, needs, social supports, and willingness to take on caregiving tasks. It also requires a written care plan and referrals to community resources as needed.[6] The service does not require every practice to use a specific caregiver-burden scale, and clinicians should not imply that a numerical burden score is itself a billing requirement.

The voluntary CMS Guiding an Improved Dementia Experience (GUIDE) Model goes further operationally. GUIDE provides longitudinal dementia care, including care navigation, 24/7 support, caregiver training, community connections, and respite for qualifying caregivers.[7] CMS performance measures monitor caregiver burden over time, and GUIDE participants submit caregiver-assessment information that includes burden level.[8] These features demonstrate that caregiver burden can be incorporated into care delivery, measurement, and resource allocation.

International guidance also recognizes the caregiver as a legitimate focus of assessment and support. The National Institute for Health and Care Excellence recommends advising carers of people with dementia about caregiver assessment, respite assessment, and other support.[16] This supports dyadic care, but it should not be converted into a claim that all health systems mandate a particular screening instrument or interval.

Choosing a Measurement Approach

The best tool is one that the practice can administer consistently, interpret correctly, and connect to an action pathway. Using the same validated version over time is generally more useful than switching instruments. Practices should confirm permissions, approved translations, scoring instructions, and electronic-use terms before deployment.[9-12]

Approach	Practical role	Important cautions
Direct one-question triage	Ask whether the caregiver regularly feels completely overwhelmed; useful when time is extremely limited.[2]	A positive response requires deeper assessment. A negative response does not exclude domain-specific strain.
Zarit Burden Interview, 4-item or 12-item version	Brief burden screening or broader short-form assessment; derived from the 22-item instrument and studied in caregivers of cognitively impaired older adults.[9]	Use the exact validated version and authorized translation. Confirm current distribution and permission requirements.[10] Do not apply a cutoff validated for a different version or population.
Modified Caregiver Strain Index	Thirteen items covering financial, physical, psychological, social, and personal strain; score range 0-26.[11,12]	Higher scores indicate greater strain, but scores are not formally categorized as low, moderate, or high. Professional judgment and item-level review are required.[12]

The one-question approach is best viewed as triage. A short instrument may be preferable for baseline and follow-up because it provides a reproducible score and a broader view of burden. A fuller assessment is warranted when the screen is positive, the score is rising, or the caregiver expresses concern that is not captured by the selected tool.

A Practical Geriatric Workflow

1. Identify the caregiving structure

Document the primary caregiver, backup caregivers, relationship to the patient, contact preferences, and the tasks each person actually performs. Do not assume that the person accompanying the patient is the only caregiver or that a legally authorized representative provides day-to-day care.

2. Create an opportunity for private disclosure

When feasible, offer the caregiver a brief private conversation, paper form, portal questionnaire, or previsit call. Caregivers may avoid discussing exhaustion, resentment, conflict, fear, finances, or thoughts of placement in front of the person with dementia.

Explain the purpose directly: the practice asks because caregiver capacity affects whether the care plan is safe and workable, not because it is grading the caregiver.[15]

3. Use a consistent brief measure

Select one approach for routine use. A practice might use one-question triage at every comprehensive dementia visit and a validated multi-item tool at baseline and selected follow-up visits. Another practice may use the same short instrument throughout. The workflow matters more than choosing a single universally superior tool.

4. Interpret the result in context

Review the total, item pattern, and change from prior measurement. Pair the result with the broader cognitive-care and dyadic assessment.[6,15] Consider:

• neuropsychiatric and behavioral symptoms;
• activities of daily living and instrumental activities of daily living;
• nighttime supervision and sleep disruption;
• medication-management demands;
• falls, wandering, driving, or home-safety concerns;
• caregiver depression, anxiety, substance use, and physical illness;
• family conflict and availability of backup help;
• financial and employment strain;
• language, health literacy, cultural expectations, and access barriers; and
• the caregiver’s stated willingness and limits.

Do not use the score alone to determine capacity, neglect, legal competence, or need for institutional placement.

5. Match the response to the pattern

Clinical pattern	Immediate response	Follow-up
Stable score, no concerning item, care plan remains feasible	Reinforce effective strategies; provide anticipatory guidance and accessible resource information.	Repeat at a planned interval and after major change.
New or rising burden, nonurgent domain-specific strain, or caregiver request for help	Identify the dominant drivers; modify the patient plan; involve social work, nursing, pharmacy, care navigation, or community services; address respite and backup coverage.	Assign an owner and reassessment date. Confirm that referrals were reached and usable.
Acute mental-health or safety concern, fear of losing control, suspected abuse or neglect, inability to provide essential care, or imminent breakdown of the home plan	Conduct immediate risk assessment and use local emergency, crisis, adult-protective, or safeguarding procedures as indicated.[6,15,16]	Document the safety plan, responsible clinician, and rapid follow-up.

6. Close the loop

A screening result without a documented response can become another unacted-on data point. The record should show what was identified, what changed, who owns the next action, and when burden will be reassessed.

When Should Burden Be Reassessed?

No universal evidence-based interval applies to every instrument, caregiver, or dementia stage. “Routine” should mean intentional and repeatable, not necessarily every visit.[2,8-12]

A pragmatic schedule is:

• Baseline: at diagnosis, entry into a dementia-care program, or the first comprehensive care-planning visit.
• Event-triggered: after hospitalization, emergency care, rehabilitation, a major medication or behavioral change, new wandering or nighttime disturbance, functional decline, caregiver illness, loss of backup support, relocation, or a discussion about long-term care.
• Periodic: during longitudinal follow-up, with frequency based on prior burden, disease trajectory, and practice capacity.
• After intervention: after respite, care-plan modification, treatment of a behavioral driver, new home services, or caregiver training, to determine whether the targeted problem changed.

A practice may choose an annual anchor for clinically stable dyads, but that interval is an operational choice rather than a universally validated recommendation. Higher-risk or rapidly changing dyads warrant earlier reassessment.

What Should Happen After a Positive Screen?

The first task is not to lower the score. It is to understand what the score represents.[11,12,15]

Reassess the patient-level drivers

Agitation, psychosis, pain, constipation, urinary symptoms, delirium, sleep-wake disruption, medication adverse effects, and unmet sensory or environmental needs can intensify caregiver strain. The patient’s change may require clinical evaluation rather than caregiver education alone.

Reduce avoidable workload

Medication reconciliation may uncover duplicative dosing times, difficult formulations, or nonessential complexity. Occupational therapy, home-health assessment, assistive devices, continence planning, transportation, meal support, and home-safety modification may reduce task burden.

Provide targeted skills and support

Caregiver education is most useful when linked to a current problem, such as responding to repetitive questioning, bathing resistance, nighttime behavior, or transitions. Support groups, counseling, care navigation, and respite may be appropriate, but referrals should account for language, transportation, cost, technology, and the caregiver’s available time.

The REACH II randomized trial showed that a structured, multicomponent intervention tailored to individual risk profiles improved a composite quality-of-life outcome in several racial and ethnic groups and reduced the proportion meeting a study definition of clinical depression. However, institutional placement did not differ significantly at 6 months.[13] These findings support targeted, multicomponent support rather than a generic handout.

Protect against overpromising

The 2025 D-CARE randomized clinical trial compared health system-based dementia care, community-based dementia care, and usual care in 2,176 dyads. The active models did not significantly reduce caregiver strain or patient behavioral symptoms over 18 months compared with usual care, although caregiver self-efficacy was higher.[14] The trial reminds us that comprehensive programs can improve selected capabilities without reliably reducing measured burden.

Screening should therefore be justified by better recognition, care-plan fit, risk detection, and access to support. It should not be sold as a proven stand-alone method for preventing depression, hospitalization, institutionalization, or caregiver collapse.

Operationalizing Measurement Without Overloading the Visit

A workable program assigns tasks across the team.[7,8,15]

• Before the visit: a medical assistant, nurse, portal workflow, or care navigator identifies the caregiver and administers the selected measure.
• During the visit: the clinician reviews positive responses, evaluates patient-level contributors, and determines urgency.
• After the visit: social work, nursing, pharmacy, behavioral health, or navigation staff complete referrals and follow-up.
• In the record: use a dedicated field for instrument, version, date, score, urgent item flags, action plan, owner, and reassessment date.
• At the population level: create a registry or work queue for high or rising burden rather than relying on free-text recall.

A 2026 implementation framework for dementia care navigation emphasizes assessment, person- and family-centered planning, monitoring, coordination, caregiver support, culturally responsive care, and information systems that make the work trackable.[15] These are the same infrastructure elements needed for burden measurement to produce action.

Practices should build the referral map before launching routine screening.[4,15] At minimum, the map should identify local and virtual options for caregiver education, support groups, respite, adult day services, Area Agencies on Aging, transportation, home care, legal or financial assistance, behavioral health, crisis response, and adult-protective services. Availability and eligibility vary by locality.

Equity, Privacy, and Documentation

Caregiver burden is shaped by culture, income, employment, housing, immigration concerns, rurality, family structure, and prior experience with health systems. A score can mislead if the instrument is not understood, the translation is inappropriate, or the response options do not fit the caregiver’s circumstances. Practices should use validated translations when available and provide accessible alternatives for caregivers with sensory, literacy, or technology barriers.[10,15]

Clinicians should also avoid framing caregiving only as a burden. Many caregivers report meaning, reciprocity, commitment, or positive aspects alongside strain. Measuring burden should create room for both realities.

The caregiver is not automatically the clinician’s patient. Obtain permission for communication as required, offer privacy where feasible, document only information relevant to the dementia care plan, and encourage caregivers with health or mental-health needs to seek care from their own clinicians. Immediate safety concerns require action under applicable clinical, legal, and institutional procedures.[6,15,16]

Limitations of the Evidence

The case for routine measurement is clinically plausible and supported by cohort evidence, implementation studies, and the design of current dementia-care programs. Important limitations remain.

First, burden is subjective and context-dependent. Instruments measure overlapping but nonidentical constructs, and scores are not interchangeable. Second, much of the evidence linking burden with patient outcomes is observational. For example, high Zarit burden was associated with nursing home placement in a large prospective cohort, but placement also reflects patient severity, caregiver age, preferences, culture, finances, and service availability.[3] Third, studies do not establish one optimal tool, cutoff, or reassessment interval for all settings. Fourth, caregiver interventions have heterogeneous effects, and a lower score is not guaranteed even when care coordination or self-efficacy improves.[13,14] Finally, screening can create frustration or ethical concern when services are inaccessible.

These limitations argue for careful implementation, not for ignoring the signal. The defensible approach is to measure burden as one component of a broader dyadic assessment, preserve uncertainty, and evaluate whether the local response pathway produces meaningful care changes.

References

1. Alzheimer’s Association. Alzheimer’s Disease Facts and Figures. 2026. Accessed August 12, 2026. https://www.alz.org/alzheimers-dementia/facts-figures
2. Reuben DB, Romero T, Evertson LC, Jennings LA. Overwhelmed: a Dementia Caregiver Vital Sign. J Gen Intern Med. 2022;37(10):2469-2474. https://doi.org/10.1007/s11606-021-07054-3 PMID: 34389938.
3. Yaffe K, Fox P, Newcomer R, et al. Patient and caregiver characteristics and nursing home placement in patients with dementia. JAMA. 2002;287(16):2090-2097. https://doi.org/10.1001/jama.287.16.2090 PMID: 11966383.
4. Riffin C, Wolff JL, Pillemer KA. Assessing and Addressing Family Caregivers’ Needs and Risks in Primary Care. J Am Geriatr Soc. 2021;69(2):432-440. https://doi.org/10.1111/jgs.16945 PMID: 33217776.
5. Davis ML, Hendrickson J, Wilson N, et al. Taking Care of the Dyad: Frequency of Caregiver Assessment Among Veterans with Dementia. J Am Geriatr Soc. 2019;67(8):1604-1609. https://doi.org/10.1111/jgs.15882 PMID: 31002403.
6. Centers for Medicare & Medicaid Services. Cognitive Assessment and Care Plan Service (A59036). Accessed August 12, 2026. https://www.cms.gov/medicare-coverage-database/view/article.aspx?articleid=59036
7. Centers for Medicare & Medicaid Services. GUIDE (Guiding an Improved Dementia Experience) Model. Accessed August 12, 2026. https://www.cms.gov/priorities/innovation/innovation-models/guide
8. Centers for Medicare & Medicaid Services. GUIDE Model Frequently Asked Questions. Updated May 22, 2026. Accessed August 12, 2026. https://www.cms.gov/priorities/innovation/guide/faqs
9. Bedard M, Molloy DW, Squire L, Dubois S, Lever JA, O’Donnell M. The Zarit Burden Interview: a new short version and screening version. Gerontologist. 2001;41(5):652-657. https://doi.org/10.1093/geront/41.5.652 PMID: 11574710.
10. Mapi Research Trust. Zarit Burden Interview (ZBI). Updated June 16, 2026. Accessed August 12, 2026. https://eprovide.mapi-trust.org/instruments/zarit-burden-interview
11. Thornton M, Travis SS. Analysis of the reliability of the Modified Caregiver Strain Index. J Gerontol B Psychol Sci Soc Sci. 2003;58(2). https://doi.org/10.1093/geronb/58.2.S127 PMID: 12646602.
12. Onega LL. The Modified Caregiver Strain Index (MCSI). Hartford Institute for Geriatric Nursing. Accessed August 12, 2026. https://hign.org/consultgeri/try-this-series/modified-caregiver-strain-index-mcsi
13. Belle SH, Burgio L, Burns R, et al. Enhancing the quality of life of dementia caregivers from different ethnic or racial groups: a randomized, controlled trial. Ann Intern Med. 2006;145(10):727-738. https://doi.org/10.7326/0003-4819-145-10-200611210-00005 PMID: 17116917.
14. Reuben DB, Gill TM, Stevens A, et al. Health System-Based, Community-Based, or Usual Dementia Care for Persons With Dementia and Caregivers: The D-CARE Randomized Clinical Trial. JAMA. 2025;333(11):950-961. https://doi.org/10.1001/jama.2024.25056 PMID: 39878968. Correction: JAMA. 2025;333(17):1549. https://doi.org/10.1001/jama.2025.4753
15. Daven M, Bass DM, Deaner N, et al. A framework for implementing high-quality dementia care navigation: recommendations from the Alzheimer’s Association Dementia Care Navigation Roundtable. Alzheimers Dement. 2026;22(2). https://doi.org/10.1002/alz.71102 PMID: 41612921.
16. National Institute for Health and Care Excellence. Dementia: assessment, management and support for people living with dementia and their carers. NICE guideline NG97. Published June 20, 2018. Accessed August 12, 2026. https://www.nice.org.uk/guidance/ng97/chapter/
    recommendations