Safeguarding Older and Disabled People in Taiwan: Prevention, Protection and System Accountability

An older person receiving support at home begins losing weight. Her daughter says she simply refuses meals, while a home-care worker notices there is little food in the refrigerator and that the woman becomes quiet whenever money is discussed. Elsewhere, a disabled adult tells a community worker that a relative prevents him leaving home alone. In a residential service, unexplained bruising appears after several difficult transfers. Each situation may have a different explanation, but each requires somebody to recognise that ordinary care concerns can also contain questions about protection, rights and possible abuse or neglect.

In Taiwan, safeguarding does not sit within one single long-term care procedure. The protection architecture explored within the Taiwan Ageing, Long-Term Care & Community Support Knowledge Hub spans the Long-Term Care Services Act, Senior Citizens Welfare Act, People with Disabilities Rights Protection Act, Domestic Violence Prevention Act and the work of national and local protection services. Different legal routes can therefore become relevant according to the person, setting and nature of the concern.

This matters increasingly as Taiwan combines rapid population ageing with greater reliance on home and community support. Long-Term Care 3.0, approved at the end of 2025 and implemented from 2026, strengthens the ambition for integrated, continuous support extending from families and homes through community and residential care. That direction can increase independence, but it also makes safeguarding more distributed. Risk may be visible to a home-care worker, hospital, social worker, police officer, community service, neighbour or family member rather than to one organisation with complete oversight.

The strategic challenge is therefore not simply whether Taiwan has reporting laws. It is whether concerns are recognised early, reported appropriately, investigated proportionately and converted into protection without unnecessarily removing the person’s voice, relationships or independence.

Taiwan’s protection framework crosses several legal systems

The Long-Term Care Services Act contains explicit protections for people receiving long-term care. Long-term care institutions and their personnel are required to provide proper care and protection and must not abandon, physically or mentally abuse, discriminate against, harm or improperly restrict the physical freedom of service users, or otherwise infringe their interests. The Act also addresses privacy, contracts and other rights of people receiving long-term care.

That framework is important, but it is not the whole safeguarding system.

The Senior Citizens Welfare Act provides a protection route for older people experiencing circumstances including abuse, neglect and abandonment. It places reporting responsibilities on specified professionals and other personnel carrying out elderly welfare work and requires municipal and county or city governments to respond to suspected cases. It also envisages local protection systems involving social administration, health, police and other relevant organisations.

For disabled people, the People with Disabilities Rights Protection Act prohibits acts including abandonment, physical or psychological abuse, restriction of freedom and leaving a person who cannot protect themselves in circumstances where they may be exposed to danger. It similarly establishes reporting and local-government response duties.

Domestic violence legislation can intersect with both frameworks where harm occurs within family or intimate relationships.

The result is a safeguarding architecture based on overlapping responsibilities rather than one universal process. Operationally, workers do not need to turn themselves into legal specialists before acting on a concern. They do, however, need to understand that the route may depend on the circumstances and that uncertainty about which framework ultimately applies should not become a reason for delay.

This is where strong incident response, protection and escalation becomes essential. Recognition, immediate safety and appropriate referral need to precede organisational debate about ownership.

Mandatory reporting turns recognition into a professional responsibility

Taiwan’s protection legislation gives reporting a particularly important role. Under the Senior Citizens Welfare Act, specified professionals and personnel who become aware through their work of suspected circumstances requiring older-person protection must report to the relevant municipal or county or city government. The People with Disabilities Rights Protection Act similarly requires specified personnel to report suspected prohibited treatment of disabled people, with the legislation establishing a 24-hour reporting expectation.

The Ministry of Health and Welfare’s protection arrangements also bring relevant statutory reporting routes together through the national social-safety infrastructure. Professionals encountering suspected domestic violence, abuse, neglect, abandonment or prohibited treatment of disabled people are expected to make the relevant notification promptly and protect the confidentiality and privacy of the person concerned.

The existence of a reporting duty solves only part of the operational problem. Workers first have to recognise what they are seeing.

Some concerns are obvious. Physical assault, sexual violence or deliberate abandonment can present an immediate protection issue. Others are much less clear. A person may be losing money gradually. Medication may repeatedly be unavailable. A relative may answer every question on the person’s behalf. Restriction may be described as necessary for safety. Neglect can develop because an exhausted family caregiver is no longer able to meet increasing needs rather than because harm was intentionally planned.

Training therefore needs to move beyond memorising categories of abuse. Workers need confidence in recognising patterns, recording observations accurately, distinguishing fact from interpretation and escalating uncertainty.

The safeguarding competence of the workforce becomes especially important in home and community care because frontline workers may be the only professionals seeing the person regularly in their normal environment.

The 113 Protection Hotline creates a national access point

Taiwan’s 113 Protection Hotline provides a national, 24-hour route for protection-related reporting, consultation and referral. Its scope includes domestic violence and protection concerns involving older and disabled people as well as other forms of interpersonal harm. The service can be reached by telephone, with additional channels supporting people who may find ordinary voice calls difficult, and interpretation support is available in several languages.

This matters for long-term care because safeguarding does not arise only during office hours or within formal services. A neighbour may hear violence at night. A migrant caregiver may witness mistreatment within a household. A family member may become frightened about what is happening to an older relative during a weekend. A person may want advice before understanding whether their experience meets a particular legal definition.

A national access route reduces one barrier: knowing exactly which administrative department to contact.

It does not remove the need for local response. Protection work ultimately requires assessment of the person’s circumstances, immediate safety, family and household dynamics, health needs and available alternatives. Depending on the situation, social workers, municipal or county/city services, police, health professionals and other organisations may need to become involved.

This makes multi-agency protection more than an organisational preference. Different actors hold different powers, evidence and relationships with the person. Effective safeguarding depends on those contributions being coordinated rather than simply referred from one organisation to another.

Scenario: possible neglect is hidden inside caregiver exhaustion

An 86-year-old man with reduced mobility and cognitive impairment lives with his daughter. She has gradually reduced her employment to care for him and manages most of his daily needs. Home-care workers visit several times each week.

Over a month, workers notice that he is sometimes wearing soiled clothing, food is becoming less available and prescribed medication is not always organised. His daughter appears exhausted and becomes defensive when asked about the changes. One worker worries about neglect.

The safeguarding response should not depend on deciding whether the daughter is a “bad carer”. The immediate questions concern the older man’s wellbeing, whether essential needs are being met and whether there is evidence of harm or serious risk. Observations need to be recorded accurately and escalated through the appropriate protection and service channels.

At the same time, the wider assessment may identify that the daughter has reached the limits of what she can sustain. Long-term care reassessment, additional home support, day care, respite or other family-caregiver support may form part of the response alongside any formal protection action required.

This distinction is important. Abuse and neglect must not be minimised because a family is under pressure. Equally, safeguarding should be capable of recognising when preventable harm is emerging from unsupported caregiving rather than treating every case solely as intentional maltreatment.

For Taiwan, where family care remains central to the long-term care landscape, prevention increasingly depends on identifying caregiver strain before a household reaches this point.

Prevention begins before a report is made

A protection system focused only on responding after abuse occurs will always operate too late for some people. Long-Term Care 3.0 creates an opportunity to connect safeguarding more deliberately with prevention.

Home care, community services, day care, respite, caregiver support and health services can all create protective contact. Their value is not that they monitor families continuously. It is that they reduce isolation, create relationships outside the household and provide opportunities for changes to become visible.

This makes prevention and early intervention relevant to safeguarding as well as health and functional decline.

Risk can increase when several pressures converge:

• a person becomes increasingly dependent on one caregiver;
• family stress, financial pressure or caregiver exhaustion increases;
• the person has limited communication or cognitive impairment;
• social contact outside the household decreases;
• formal services become inconsistent or are refused; and
• nobody has a sufficiently complete view of the changing situation.

None of these factors proves abuse. Their significance lies in the vulnerability created when support becomes concentrated and invisible.

Preventive safeguarding therefore includes accessible services, caregiver support, continuity of professional contact and opportunities for people to speak privately about their lives. It also requires workers to notice when apparently ordinary care changes are creating a different level of risk.

Protection must not turn automatically into restriction

Safeguarding creates a persistent tension between protection and autonomy. Older and disabled people may choose relationships, routines and levels of risk that professionals or relatives would not choose for them. A person with mobility difficulties may still value going outside alone. Someone with early dementia may want to continue shopping independently. A disabled adult may prefer privacy from family involvement.

Concern for safety does not by itself justify removing ordinary freedoms.

This is particularly important because Taiwan’s Long-Term Care Services Act expressly protects service users against inappropriate restriction of physical freedom, while disability legislation also prohibits improper restriction. Rights therefore sit inside the safeguarding framework rather than outside it.

The practical challenge is to distinguish protective intervention from unnecessary control. Staff need to understand the specific risk, the person’s wishes, available alternatives and whether a less restrictive response could manage the concern.

Organisations exploring similar decisions can use the Positive Risk-Taking Planner as a transferable way of structuring risk-benefit thinking. It is not a statement of Taiwanese law or a substitute for the applicable legal process, but it can help teams avoid reducing complex decisions to a simple choice between complete freedom and complete protection.

The wider principle of safeguarding, consent and human rights is particularly relevant as Taiwan expands ageing-in-place services. Community living cannot fulfil its purpose if people remain at home only by being subjected to excessive restrictions within it.

Scenario: a safety response begins to remove an older woman’s independence

A woman with early-stage dementia lives alone and attends a community service several times each week. She enjoys walking to a nearby market, where stallholders know her. After she becomes briefly disoriented on one occasion, her son asks the service to prevent her going out without him.

A purely defensive response would treat the incident as evidence that independent walking must stop. A more proportionate approach examines what actually happened. Staff explore the route, time of day, her familiarity with the area, whether cognition has changed, how she normally responds when uncertain and what practical support might reduce risk.

Her preferences remain central. She understands that she became lost and strongly wants to retain the routine because shopping is one of the few activities she still undertakes independently.

The resulting plan might include a more familiar route, agreed check-in arrangements, carrying accessible contact information or appropriately chosen location-support technology if she understands and accepts its use. Her cognitive and health needs are also reviewed rather than assuming that one event automatically represents irreversible deterioration.

If disorientation becomes more frequent, the balance can be reconsidered.

The safeguarding achievement is not that risk has disappeared. It is that the response protects the woman from foreseeable harm while avoiding an immediate loss of autonomy. Good safeguarding can therefore preserve ordinary life rather than progressively narrowing it.

Disabled people require protection that recognises dependency and rights together

Safeguarding disabled people presents some overlapping but distinct issues. A person may depend on relatives or paid workers for communication, mobility, personal care, financial management or access to the community. That dependence can create power imbalances even where support relationships are otherwise positive.

The People with Disabilities Rights Protection Act establishes explicit protection against abandonment, physical and psychological abuse, restriction of freedom and other prohibited treatment. It also requires reporting of suspected cases by specified professionals and personnel.

Operationally, protection needs to remain accessible to people with different communication needs. A reporting system is of limited value if the person cannot communicate privately with those operating it. Workers may need to use alternative communication methods, interpreters or other appropriate support while avoiding the assumption that a family member should always speak for the person.

Safeguarding also needs to recognise financial and social control. A person can be physically safe while being prevented from using their own money, maintaining relationships or participating in community life.

This is why disability safeguarding and human rights should remain connected. Protection is not simply the prevention of injury. It also concerns dignity, participation, freedom from exploitation and the ability to exercise ordinary choices.

Home-based care creates both visibility and blind spots

Taiwan’s long-term care policy increasingly supports people in their own homes and communities. This can reduce institutional dependency and maintain familiar relationships, but safeguarding arrangements need to reflect the privacy of the home.

A home-care worker may see things that would otherwise remain invisible: unsafe living conditions, unexplained injuries, lack of food, controlling behaviour, financial pressure or a family caregiver who is no longer coping. Continuity is particularly valuable because a familiar worker can notice changes that a succession of different workers might miss.

At the same time, workers are visitors in somebody else’s household. They may receive conflicting accounts from relatives and the person receiving care. They may worry about damaging relationships if they raise a concern. Migrant family caregivers can occupy an especially complex position because they may witness family dynamics while also depending on the household for their own employment and accommodation.

Provider governance needs to give workers a safe escalation route. A frontline employee should not be expected to investigate suspected abuse independently or confront a possible perpetrator where doing so could increase danger.

The role of the worker is often to observe, listen, record, respond to immediate danger and report through the appropriate route. Responsibility then needs to move to people with the authority and expertise to assess the concern.

Organisations can use the Governance Maturity Assessment to test similar questions about escalation, accountability and organisational oversight. Used internationally, the framework should support governance thinking rather than be interpreted as a Taiwan-specific safeguarding standard.

Residential services need controls against closed cultures

Residential care creates a different safeguarding environment. Staff are continuously present and formal management structures can make concerns easier to escalate. Yet residents may also depend extensively on the same organisation for accommodation, personal care, food, medication, social contact and access to the outside world.

This concentration of power creates its own risks.

Taiwan’s Long-Term Care Services Act places direct obligations on long-term care institutions and personnel to provide proper care and protection and prohibits abusive, discriminatory, harmful and improperly restrictive treatment. Institutional evaluation and local competent-authority oversight provide additional layers of assurance.

Safeguarding, however, cannot rely only on formal inspection. Harm can become normalised within everyday routines: rough handling described as efficiency, unnecessary restriction justified as safety, distress dismissed as difficult behaviour or residents discouraged from raising complaints.

The strongest protective environments therefore combine external scrutiny with internal openness. Residents and families need credible ways to raise concerns. Staff need to be able to report poor practice by colleagues or managers. Leaders need to examine patterns rather than wait for one incontrovertible serious incident.

A culture in which every concern is treated as disloyalty can conceal risk even when written safeguarding procedures appear strong.

Scenario: bruising after transfers reveals a practice problem

A resident with significant physical disability develops repeated bruising around her arms. Staff initially attribute this to fragile skin and the assistance required during transfers. No individual incident appears severe enough to explain the pattern.

A supervisor reviewing several records notices that the bruising tends to follow particular shifts. The response does not begin by assuming deliberate abuse, but neither is the pattern dismissed.

The resident is spoken with in a way that supports her communication. Relevant health causes are considered. Transfer practices are observed and equipment, staffing and training are reviewed. The investigation identifies that workers on certain shifts have been using an inconsistent manual technique when staffing is pressured. The resident has sometimes expressed discomfort but believed it was simply part of receiving care.

Immediate practice is corrected and the appropriate protection and management processes are followed according to the findings. Other residents who receive similar assistance are also considered because the risk may not be confined to one person.

The governance response goes further than retraining the workers involved. Managers examine why supervision had not identified the variation, whether staffing pressure was contributing and whether residents were being given accessible opportunities to describe their experience.

This is where safeguarding assurance becomes meaningful. The purpose is not simply to prove that a case was processed. It is to establish whether the conditions that allowed harm to occur have changed.

Financial abuse can sit outside traditional care indicators

Older and disabled people can become vulnerable to financial exploitation when another person manages money, shopping, banking or access to benefits and personal resources. Dependence may increase when cognitive impairment, mobility difficulties or digital exclusion make independent financial management harder.

Financial concerns can also be difficult for care services to interpret. Workers may notice that food is scarce despite apparently adequate household resources, that bills remain unpaid, or that a relative repeatedly asks questions about the person’s money. None of these observations alone establishes exploitation.

The worker nevertheless needs a route for recording and escalating concerns without attempting to conduct a financial investigation personally.

The human impact extends beyond money. Financial control can restrict transport, social participation, access to health care or the ability to purchase ordinary personal items. It can therefore increase dependency on the very person exercising control.

Prevention includes accessible information, opportunities for private conversation and professional curiosity when a person’s living conditions appear inconsistent with what is otherwise known about their circumstances.

Information sharing must support protection without erasing privacy

Safeguarding requires information to move between organisations, yet protection systems also hold highly sensitive personal information. Taiwan’s statutory frameworks explicitly recognise confidentiality, including protection of reporter information in relevant reporting processes and privacy protections for people receiving long-term care.

The operational balance is important. Workers should not withhold necessary information simply because it is sensitive, but nor should a safeguarding concern become permission for unrestricted circulation of personal details.

Information should serve a defined protective purpose. Relevant professionals need enough detail to understand the concern, assess immediate risk and decide what response is required. Records should distinguish observed facts, what the person or another individual has said, and professional interpretation.

This becomes increasingly important as Taiwan develops more connected health and long-term care systems under LTC 3.0. Better digital integration can help relevant information follow a person between services, but it also increases the number of systems and professionals potentially able to access sensitive material.

The principle behind safeguarding information sharing is therefore one of purposeful connection rather than unrestricted visibility.

Organisations considering digital coordination can use the Digital Transformation Readiness Assessment to examine governance, data practices and workforce readiness around technological change. The framework does not define Taiwanese information law, but it can help expose whether digital infrastructure is being developed alongside appropriate organisational controls.

Technology can protect, monitor or control

Technology increasingly complicates safeguarding because the same device can support independence or become intrusive depending on how it is used.

Location technology may help a person with cognitive impairment continue walking independently. Sensors may identify falls or unusual inactivity. Remote contact can extend support to people living at a distance from specialist services.

Yet monitoring can also become continuous surveillance. A family may install technology for understandable reasons while giving little consideration to the older or disabled person’s preferences. Data may be shared more widely than expected. A technological alert can encourage an unnecessarily restrictive response if every deviation from routine is treated as danger.

The Long-Term Care Services Act itself recognises the sensitivity of monitoring. It protects service users against unauthorised recording while allowing long-term care institutions to install monitoring equipment where required for service-user safety, subject to the applicable notification requirements.

The broader governance principle is that technology should have a defined purpose, proportionate use and clear responsibility. Safeguarding technology should increase the person’s safety and agency where possible rather than simply making organisational risk easier to manage.

Scenario: digital monitoring becomes more intrusive than the risk requires

An older man lives with mild cognitive impairment and receives home and community support. After he leaves his home late one evening and is brought back safely by a neighbour, his family proposes installing several cameras and continuous location monitoring.

The family’s concern is genuine. The proposed response, however, would create extensive observation of the man’s daily life.

His care team explores the circumstances surrounding the incident. It occurred after he woke from an evening sleep and believed it was morning. There have been no previous episodes. Medication, sleep pattern and cognition are reviewed, and the family discusses less intrusive options with him.

A proportionate plan focuses first on the circumstances that created the confusion, agreed contact arrangements and technology limited to the specific risk rather than continuous surveillance throughout the home. The approach can be revised if the pattern changes.

This is not an argument against assistive technology. It demonstrates why person-centred technology requires the same ethical discipline as other safeguarding interventions.

The relevant measure of success is not how much information the technology can collect. It is whether it helps the person live with greater safety while preserving as much privacy and ordinary life as possible.

Safeguarding data should reveal patterns, not merely count reports

Reporting volumes matter, but they are difficult to interpret in isolation. A rise in safeguarding reports can indicate increasing harm, better recognition, stronger reporting confidence or some combination of these factors. A low reporting rate can reflect safety, but it can also reflect weak awareness or barriers to disclosure.

Governance therefore needs context.

Municipal and county or city authorities can examine patterns in types of concern, settings, recurrence and outcomes. Providers can analyse internal incidents, complaints, injuries, restrictive practices, workforce instability and other information that may reveal emerging risks.

At provider level, useful safeguarding questions include whether concerns are concentrated around particular times, locations or teams; whether the same people experience repeated incidents; whether actions prevent recurrence; and whether workforce or service pressures are appearing alongside increased risk.

The Quality Dashboard Builder can help organisations structure this type of multi-source oversight. Used in an international context, it offers an analytical method rather than a prescribed Taiwanese reporting framework.

The strongest data system does not simply produce more safeguarding indicators. It helps decision-makers ask better questions.

Accountability needs to follow concerns beyond the initial response

A protection system can respond correctly to an individual case yet still fail to learn from recurrence. This is why safeguarding governance needs a feedback loop.

Where a concern involves an individual worker, the response may include investigation, employment action, additional supervision or referral through relevant professional or legal routes. Where the issue arises from organisational conditions, broader action may be necessary: staffing, training, management oversight, environmental design or service processes may need to change.

Where patterns occur across several organisations or localities, responsibility moves again. Municipal, county or city authorities and national agencies may need to consider whether service design, workforce supply, guidance or regulation is contributing to persistent vulnerability.

This layered accountability is particularly important under LTC 3.0 because greater integration means more interfaces. A concern may involve a family caregiver, migrant caregiver, home-care organisation, hospital and local long-term care system at different points.

No single actor should assume that making a referral ends its responsibility for the person. The referring service may still have an important role in maintaining ordinary support while specialist protection processes proceed.

Workforce culture determines whether formal protections become real

Safeguarding legislation can require reporting, but organisations shape whether workers feel able to act.

A care worker may hesitate because the suspected perpetrator is a respected colleague. A migrant caregiver may fear losing employment. A junior employee may believe a manager will dismiss their concern. A worker visiting a family home may worry that raising an issue will cause the family to cancel the service.

These pressures cannot be solved by annual training alone.

Leaders need to create a culture in which uncertainty can be raised without requiring the worker to prove abuse first. Staff should understand immediate safety actions, reporting routes, confidentiality and the limits of their investigative role. Supervision should provide space to discuss concerns that remain ambiguous.

Whistleblowing and reporting arrangements also need credibility. A policy has limited protective value if workers believe using it will expose them to retaliation.

The broader lesson is that safeguarding competence has both a technical and cultural dimension. People need to know what to do, and they need confidence that the organisation will support them when they do it.

Community-based care changes where prevention needs to happen

Taiwan’s long-term care reforms are deliberately strengthening support outside institutions. This creates a wider safeguarding environment encompassing homes, neighbourhood services, day-care centres, hospitals, community organisations and informal networks.

That distributed model can be protective because more people may have meaningful contact with somebody who would otherwise be isolated. It can also create ambiguity because responsibility is shared across several relationships.

The stronger opportunity under LTC 3.0 is to make safeguarding part of ordinary integrated care rather than treating it as a specialist process activated only after serious harm.

A day-care service noticing unexplained withdrawal, a hospital identifying repeated injuries, a home-care worker observing financial control and a care manager recognising caregiver exhaustion are all seeing different parts of the same protection landscape.

Integration should make those observations easier to connect while preserving appropriate confidentiality.

For older and disabled people, this also means safeguarding should support independence and community inclusion. The purpose is not to construct a risk-free care system by restricting people’s lives. It is to create enough protection, support and accountability for people to live ordinary lives with greater security.

International learning: protection works best when it is connected to ordinary care

Taiwan’s safeguarding arrangements reflect its own legal and administrative structure. The combination of the Long-Term Care Services Act, Senior Citizens Welfare Act, People with Disabilities Rights Protection Act, Domestic Violence Prevention Act, local government protection responsibilities and the national 113 Protection Hotline cannot simply be transferred to another jurisdiction.

The underlying principles are more portable.

First, safeguarding benefits from a visible route through which professionals and the public can raise concerns without needing to understand every administrative boundary. Second, statutory reporting is only effective when frontline workers can recognise less obvious forms of harm. Third, local protection needs access to social, health and policing responses because abuse rarely fits neatly inside one service domain.

Perhaps most importantly, safeguarding should connect with prevention. Caregiver support, community participation, reliable home care, accessible communication and continuity all reduce conditions in which risk can become hidden.

Other systems can adapt those principles without replicating Taiwan’s legal mechanisms. The transferable lesson lies less in the particular reporting route than in connecting recognition, protection, rights and service improvement.

The next safeguarding challenge is system visibility without excessive control

As Taiwan’s population ages and long-term care expands, safeguarding will become both more important and more complex. More people will receive support in private homes. Dementia and multimorbidity will increase the complexity of some decisions. Families will remain important while household structures continue to change. Migrant caregivers will continue to occupy a significant place in the care economy. Digital technology will make new forms of monitoring possible.

None of these developments automatically increases or reduces safety.

The strategic requirement is to make vulnerability more visible without making people themselves excessively visible. Data can identify patterns without turning every household into a monitored environment. Workers can recognise risk without assuming that disability or cognitive impairment removes autonomy. Families can receive support without safeguarding systems overlooking harm when it does occur.

That balance will require continuous learning across protection services and long-term care. It will also require evidence about what happens after reports are made: whether people become safer, whether support remains sustainable and whether recurring organisational risks actually change.

Conclusion

Taiwan has a substantial legal and operational foundation for protecting older and disabled people. The Long-Term Care Services Act establishes explicit rights for people receiving long-term care; the Senior Citizens Welfare Act and People with Disabilities Rights Protection Act create additional protection and reporting duties; domestic violence arrangements address harm within family relationships; and municipal and county or city systems, supported by the 113 Protection Hotline, provide routes through which concerns can enter the protection system.

The central challenge under Long-Term Care 3.0 is increasingly one of connection. Safeguarding concerns may first appear as a missed meal, unexplained injury, caregiver exhaustion, financial irregularity, restriction of movement or subtle change in behaviour. Recognising those signals requires competent workers, accessible reporting, local protection capacity and governance able to see patterns beyond individual incidents.

Protection must also remain proportionate. Older and disabled people do not become safer simply because other people make more decisions for them. Effective safeguarding protects people from abuse, neglect, exploitation and inappropriate restriction while preserving dignity, privacy, relationships and participation wherever possible.

As Taiwan extends care further into homes and communities, its strongest safeguard will therefore not be one additional procedure. It will be a system in which ordinary care, protection services, families, providers and public authorities can recognise vulnerability early, act when necessary and learn from what the experience reveals. That is how safeguarding becomes part of a mature long-term care system rather than a response reserved for its most serious failures.