The study enrolled participants through the hospital’s Dementia Center, a facility certified under Taiwan’s Disease-Specific Care program run by the Joint Commission of Taiwan. Rather than relying on a single survey, the care team conducted face-to-face interviews roughly every six months, beginning at the time of diagnosis. This repeated-measure design matters enormously for dementia research, because the disease is progressive and care needs can change dramatically from one year to the next. The researchers then analyzed the longitudinal data using generalized estimating equations, a statistical technique that accounts for the fact that repeated observations of the same person are correlated, allowing them to identify which factors reliably predicted service use over time rather than at a single snapshot.

The headline result is a significant but sobering rise in utilization. At baseline, only 34 percent of patients used any long-term care resources; three years later, that figure had climbed to 44 percent, a statistically significant increase. In other words, even after three years of progressive illness and repeated contact with a dedicated dementia care team, more than half of patients were still not using any formal services. The most commonly used resources were personal and professional care services, which rose modestly from 18 percent at baseline to 20 percent at year three, followed by assistive devices and home modification, which increased from 14 percent to 19 percent. These categories reflect practical, hands-on support: help with bathing, dressing, and mobility, and equipment or structural changes that make homes safer for people whose balance, judgment, and daily functioning are deteriorating.

One of the most encouraging trends involved community aging care centers, where attendance nearly doubled, from 6.8 percent of patients at baseline to 13 percent three years later. These centers provide daytime supervision, social engagement, and structured activities, which are increasingly recognized as important for slowing functional decline and relieving family caregivers. The near-doubling suggests that when families are repeatedly introduced to community-based options through a dementia care program, adoption grows. Still, at 13 percent, uptake remains low relative to the scale of the dementia population, hinting at persistent barriers such as transportation, stigma, scheduling, and simple unfamiliarity with what the centers offer.

The statistical analysis uncovered a striking pattern in who used services most. Older patients were more likely to use long-term care, with each additional year of age raising the odds by about 4 percent. Poorer activity of daily living function, measured on a standard scale of independence in tasks like eating, bathing, and toileting, was one of the strongest predictors, with each point of worsening function increasing the odds of service use by 8 percent. This makes clinical sense: formal care is typically triggered by visible functional decline. But the caregiver-side factors were equally revealing. Patients whose caregivers had higher education used more services, and patients cared for by a solo informal caregiver were nearly twice as likely to use them, with an odds ratio of 1.98. Higher scores on the Zarit Burden Interview, a validated measure of caregiver strain, also predicted greater use, with each additional point of burden nudging the odds upward.

That last cluster of findings tells a coherent story about how care actually gets arranged. Long-term care systems do not deliver services to patients directly; they deliver them through caregivers who must recognize the need, navigate eligibility assessments, and coordinate schedules. A solo caregiver carrying the full weight of care has no alternative labor to fall back on, so formal services become a necessity rather than a supplement. An educated caregiver is better positioned to understand what is available and how to request it. And measurable burnout, captured by the Zarit score, appears to push families over the threshold into seeking outside help. The system, in effect, responds to caregiver crisis rather than anticipating patient need.

Two groups stood out as using fewer services, and both deserve attention. Male patients were significantly less likely to use long-term care resources than female patients, with an odds ratio of 0.71. The authors and related literature suggest this may reflect differences in how men with dementia are cared for, including a greater tendency for families to absorb their care informally, as well as gendered patterns in help-seeking behavior. Meanwhile, patients who hired a foreign caregiver were less likely to use formal services, with an odds ratio of 0.67 that sat just at the edge of statistical significance. Taiwan, like several of its neighbors, relies heavily on live-in migrant domestic workers to provide around-the-clock home care, and families who employ one often view this as a complete solution, substituting private labor for the public or subsidized service system.

That substitution effect carries hidden costs. A live-in caregiver, however devoted, is typically not trained in dementia-specific techniques for managing behavioral symptoms, may lack access to respite, and can themselves experience isolation and exhaustion. When families with hired help opt out of community centers, professional home services, and assistive technology programs, they forgo supports that could improve quality of life for both the patient and the worker. The study’s data suggest that Taiwan’s long-term care system and its migrant caregiver program operate in parallel silos rather than as an integrated continuum, and that families are left to choose between them rather than combine them.

The authors conclude that demand for long-term care among people with dementia is high and that tailored support strategies are essential for both patients and caregivers to ensure adequate resource provision. Their most actionable recommendation concerns education: programs targeting caregivers, particularly those with lower educational levels, may help improve resource utilization. Because caregiver education independently predicted service use, the gap between the most and least supported families is not just a matter of money or availability but of information and navigation. A caregiver who does not know that home modification subsidies exist, or who cannot complete the paperwork, will not appear in any utilization statistic, no matter how great the need.

The broader implications extend well beyond Taiwan. Most countries face rapidly aging populations and rising dementia prevalence, and most have built long-term care systems that families underuse, sometimes because services are scarce, sometimes because they are unaffordable, and sometimes, as this study shows, because the burden of coordination falls on exhausted relatives. The Taiwanese cohort demonstrates that even within a structured, hospital-based dementia care program with semiannual contact, formal service uptake remains the exception rather than the rule. Closing that gap will require proactive outreach, simplified access pathways, caregiver education designed for people with limited schooling, and integration of live-in caregiver arrangements with community and professional services. As the authors’ three-year data make plain, a care system is only as effective as the families who can actually reach it, and right now, too many are being left on the outside of a door they cannot find.

Subject of Research: Long-term care resource utilization among people with dementia and their caregivers in Taiwan

Article Title: Utilization of long-term care resources for people with dementia in Taiwan: a three-year cohort study

Article References: Tung, Y.-C., Chang, M.-C., Huang, H.-K., Wang, W.-F., & Jhang, K.-M. (2026). Utilization of long-term care resources for people with dementia in Taiwan: a three-year cohort study. BMC Public Health. https://doi.org/10.1186/s12889-026-29758-z

Image Credits: AI Generated

DOI: 10.1186/s12889-026-29758-z

Keywords: dementia, long-term care, caregivers, Taiwan, cohort study, caregiver burden, community aging care centers, assistive devices, activities of daily living, migrant caregivers, BMC Public Health, geriatric care