End-of-Life and Palliative Care for Older People in Hong Kong: Choice, Dignity and Coordination

For an older person approaching the end of life, the most important question is not simply where treatment is available. It is whether care remains aligned with what matters to them as illness advances. A person may want relief from pain, fewer burdensome hospital transfers, time with family and the reassurance that future treatment decisions will reflect preferences expressed while they were still able to make them clearly. Achieving that requires healthcare, long-term care and family support to work together before a crisis forces decisions.

This makes end-of-life care an increasingly important part of the Hong Kong Ageing, Long-Term Care & Community Support Knowledge Hub. Hong Kong already provides Hospital Authority palliative care across all seven hospital clusters through inpatient, outpatient, consultative, home, day, psychosocial and bereavement services. At the same time, population ageing means more people will approach the end of life while living with frailty, dementia and multiple chronic conditions rather than one clearly defined terminal diagnosis.

The policy environment has also changed materially. The Advance Decision on Life-sustaining Treatment Ordinance came into operation on 31 July 2026, creating a statutory framework for advance medical directives and do-not-attempt cardiopulmonary resuscitation orders. That reform strengthens the legal infrastructure around patient autonomy, but legislation alone cannot deliver good end-of-life care. The harder work lies in communication, clinical judgement, community capacity, family understanding and ensuring that preferences remain visible across hospital, home and residential settings.

Palliative care is broader than the final days of life

Palliative care is sometimes understood as care provided only when active treatment has stopped. That is too narrow.

The Hospital Authority describes palliative care as holistic care for people with life-threatening and life-limiting conditions and their families. In practice, it can include symptom control, psychological support, communication about goals of care, family support and planning for later stages of illness.

This means palliative care can begin alongside other medical treatment rather than waiting until death is imminent.

An older person with advanced heart failure may still receive active clinical management while also needing help with breathlessness, fatigue, anxiety and decisions about future hospital treatment. Someone with metastatic cancer may receive treatment intended to slow disease progression while also benefiting from specialist pain control and family support.

The distinction matters because late referral can compress complex conversations into the final days of life.

Earlier palliative involvement creates more time to understand what the person values, anticipate likely deterioration and prepare family members for what may happen next.

Hong Kong’s palliative system is organised mainly through the Hospital Authority

Public palliative care in Hong Kong is primarily organised through the Hospital Authority, with services available across its seven hospital clusters.

The model includes several forms of care rather than one single service setting. Depending on clinical need and local arrangements, people may receive inpatient palliative care, specialist consultation while admitted under another clinical team, outpatient review, home-based support, day services, psychosocial care and bereavement support.

This range is important because end-of-life needs change over time.

A person may initially attend outpatient follow-up while still relatively independent. Later they may need home visits because travel becomes difficult. Acute symptom deterioration may require inpatient care. Family members may continue needing bereavement support after the person dies.

The strongest pathway therefore connects these modes rather than treating them as unrelated episodes.

For an international reader, the structural point is important: Hong Kong does not operate a separate long-term care insurance-funded hospice system equivalent to models found in some other jurisdictions. Its public palliative architecture sits substantially within the Hospital Authority and needs to connect outward to community, residential and family care.

Ageing is changing the profile of people who need palliative care

Palliative services historically developed strongly around cancer, where disease trajectories can sometimes make the transition towards end-of-life care more visible. Population ageing broadens that challenge.

Older people increasingly approach the end of life with heart failure, chronic respiratory disease, advanced neurological conditions, dementia, renal disease and severe frailty, often in combination.

These conditions may decline less predictably.

A person with cancer may experience an identifiable progression despite individual variation. A frail older person with several chronic diseases may deteriorate through repeated infections, falls, hospital admissions and partial recoveries over a much longer period.

This uncertainty can delay palliative thinking because nobody is certain whether the current deterioration represents the final phase.

The better question is not always, “Is this person dying now?” It may be, “Would this person benefit from palliative principles now?”

That shift can improve symptom management and advance care planning without requiring professionals to predict an exact prognosis.

Advance care planning is the process; an advance medical directive is one possible legal instrument

Advance care planning and advance medical directives are related but not interchangeable.

Advance care planning is the broader process through which a person discusses values, preferences, goals and expectations for future care. It may include where the person would prefer to receive care, what outcomes they consider acceptable and which treatments they would or would not want in particular circumstances.

An advance medical directive is more specific.

Under Hong Kong’s current legal framework, an adult who is mentally capable of making decisions about life-sustaining treatment can make an advance medical directive stating which life-sustaining treatment they wish to refuse if specified clinical conditions arise later and they are no longer mentally capable of making that decision.

This is an important distinction because good advance care planning should not be reduced to completing a legal form.

The conversation matters.

A person may not yet be ready to make a formal directive but may still want to discuss their priorities with family and clinicians. Those discussions can shape future decision-making even where no specific refusal of treatment has been documented.

The wider principles of end-of-life care and advance care planning therefore sit at the centre of person-centred later-life care.

The Advance Decision on Life-sustaining Treatment Ordinance changes the legal landscape

The Advance Decision on Life-sustaining Treatment Ordinance, which came into operation on 31 July 2026, provides Hong Kong with a statutory framework for advance medical directives and do-not-attempt cardiopulmonary resuscitation orders.

This is a significant development because the Hospital Authority had already supported advance directives and advance care planning under common-law arrangements for many years, but the new legislation gives clearer statutory recognition to advance decisions and establishes specific safeguards and operating requirements.

An advance medical directive allows an adult who is mentally capable of deciding on life-sustaining treatment to indicate in advance which treatment they wish to refuse if they later become mentally incapable and the specified clinical preconditions in the directive are met.

The framework is designed around two important principles: careful decision-making when the directive is created and ease of revocation while the person remains mentally capable.

That balance matters.

Advance decisions should not be made casually or because somebody feels pressure from family or services. Equally, a person who later changes their mind should not be trapped by a previous decision.

The new law strengthens autonomy but does not legalise withdrawal of basic or palliative care

A clear boundary is essential in discussing the new legislation.

An advance medical directive concerns refusal of specified life-sustaining treatment under defined circumstances. It does not authorise euthanasia, nor does it mean that a person can direct services to withhold basic care or palliative care.

This distinction is important for public confidence.

Choosing not to receive a particular life-sustaining intervention does not mean choosing abandonment.

A person who refuses cardiopulmonary resuscitation or another specified treatment may still require intensive symptom relief, nursing, comfort care, communication and support for their family.

In many cases, the need for compassionate professional care becomes greater rather than smaller as burdensome interventions are limited.

End-of-life autonomy should therefore be understood as a choice about the goals and proportionality of treatment, not a withdrawal of care itself.

DNACPR orders create an operational bridge between patient wishes and emergency response

Do-not-attempt cardiopulmonary resuscitation orders are particularly important because cardiopulmonary arrest creates an emergency in which rescuers need to act quickly.

An advance medical directive may record that a person wishes to refuse cardiopulmonary resuscitation under specified circumstances, but emergency responders may not be in a position to determine instantly whether those conditions have been met.

The statutory DNACPR framework addresses that problem by providing a recognisable clinical order that can operate in applicable circumstances.

This becomes especially important outside hospital.

Consider an 88-year-old man with end-stage chronic respiratory disease who has discussed goals of care extensively with his doctors and family. He has an applicable advance medical directive refusing cardiopulmonary resuscitation and an associated DNACPR order.

He chooses to remain at home with palliative support.

If he later suffers cardiopulmonary arrest, the practical effectiveness of his choice depends on the relevant documentation being valid, available and recognised by those responding.

The legal framework therefore creates an operational responsibility as well as a right. Documentation needs to be accessible, healthcare professionals and rescuers need appropriate training, and families need to understand what the order means.

Documentation becomes critical when decisions need to survive movement between settings

End-of-life wishes can be discussed carefully in one setting and then become difficult to identify when the person moves elsewhere.

An older person may make an advance decision while under specialist care, later return home, then be transferred unexpectedly to an emergency department. Another may move from hospital into an RCHE.

If the relevant information does not follow the person, the decision can become practically invisible.

Hong Kong’s implementation arrangements therefore increasingly connect advance decision documentation with eHealth. The first phase allows appropriate electronic storage and viewing of qualifying advance medical directives, alongside paper documentation.

This creates potential for stronger continuity, but digital availability does not remove the need for operational discipline.

Professionals still need to know that a directive exists, understand when it applies and confirm the relevant clinical conditions. Families need clear information about documentation kept at home. Residential staff need to know what to do if a resident deteriorates.

The wider digital records and information governance agenda is therefore directly relevant. End-of-life information is particularly sensitive because errors can affect irreversible treatment decisions.

Advance care planning should begin before decision-making capacity is lost

Timing matters because some conditions gradually reduce the person’s ability to make complex decisions.

Dementia is the clearest example.

A person newly diagnosed with dementia may still understand their illness, communicate priorities and participate fully in planning. Waiting until advanced cognitive impairment develops can remove much of that opportunity.

Frailty creates a different timing problem. A person may remain mentally capable but experience repeated hospitalisations before anyone initiates a serious discussion about future treatment preferences.

Consider an 84-year-old woman with heart failure, chronic kidney disease and increasing frailty who has been admitted three times in one year. Each admission ends with successful treatment and discharge, but each recovery leaves her weaker.

Her daughter is repeatedly asked urgent questions during hospital episodes but does not know what her mother would want if invasive treatment eventually offered little realistic benefit.

A planned conversation while the woman is stable could clarify her values before another crisis occurs.

The purpose is not to persuade her to refuse treatment. It is to ensure that future choices reflect her own priorities rather than assumptions made under pressure.

Families need to understand that advance planning does not transfer decision-making ownership to them

Family involvement is deeply important in Hong Kong’s older people’s care, but advance care planning should remain centred on the person wherever they are mentally capable of participating.

Relatives may hold strong views about what treatment should continue. They may fear that agreeing to a palliative approach means giving up. Others may favour less intervention because they have seen the person suffer through repeated hospital admissions.

Neither perspective automatically represents the older person’s wishes.

Good conversations therefore need to distinguish support from substitution.

Families can help the person understand options, provide context and later explain previously expressed preferences. But where the person is capable of making the decision, the decision remains theirs.

This reflects wider choice and control principles. End-of-life care is one of the areas where person-centred language must translate into genuine decision-making authority.

Home can be an appropriate place for end-of-life care, but preference alone is not enough

Some older people wish to remain at home as they approach death. Hong Kong has been strengthening support for people who choose to spend their final phase in their place of residence, and Hospital Authority palliative home-care services form part of that pathway.

Home death should nevertheless not be romanticised.

Supporting someone with advanced illness at home can require symptom management, medication, nursing input, equipment, personal care and substantial family involvement.

The physical environment also matters. A small flat may be difficult to manage with mobility equipment. Family members may lack confidence responding to breathlessness or pain. Overnight support can be particularly demanding.

The right question is therefore not simply whether a person prefers home.

It is whether the system can make that preference reasonably achievable without transferring an unsafe or unsustainable burden onto relatives.

A person may ultimately prefer hospital or hospice care once they understand the practical implications. Choice should remain informed and revisable.

Palliative home care requires rapid routes back into clinical support

End-of-life symptoms can change quickly.

Pain may escalate. Breathlessness can become frightening. Delirium may appear. A family member who previously felt confident can suddenly become overwhelmed.

Home-based palliative care therefore needs more than scheduled visits.

Families need to know who to contact when symptoms change, what can be managed at home and when urgent clinical assessment is required.

This does not mean every deterioration should trigger hospital transfer. It means that staying at home needs an escalation infrastructure around it.

Consider an older man with advanced cancer receiving palliative support at home. His pain has been reasonably controlled but worsens significantly overnight. His wife becomes frightened and considers calling emergency services because she does not know whether the change is expected.

A strong pathway gives her access to appropriate advice and clear escalation. The outcome may still be hospital admission if symptoms cannot be controlled safely at home, but the decision is made around clinical need rather than uncertainty alone.

The wider principles within end-of-life and palliative support at home are therefore highly relevant to Hong Kong’s ageing-in-place strategy.

Symptom control is the practical foundation of dignity

Discussions about autonomy and place of care are important, but they mean little if the person remains in uncontrolled pain or distress.

Palliative care needs to address symptoms such as pain, breathlessness, nausea, fatigue, anxiety, constipation, agitation and other problems associated with advanced illness.

The goal is not simply comfort in an abstract sense.

Good symptom control can allow someone to speak with family, sleep, eat, remain at home longer or participate in decisions that would otherwise be overwhelmed by physical distress.

It also requires repeated review because symptoms and medication needs can change rapidly.

For older people with frailty and multimorbidity, treatment can be particularly complex. Medication that reduces one symptom may increase sedation or falls risk. Renal or hepatic impairment may affect prescribing. Cognitive impairment may make self-reporting more difficult.

Palliative medicine therefore relies heavily on clinical judgement and observation rather than a standardised one-size-fits-all protocol.

Dignity depends on ordinary care as much as specialist medicine

End-of-life care can become highly medicalised because decisions about treatment are consequential. Yet much of dignity is delivered through ordinary care.

Being washed without unnecessary exposure, having a dry mouth relieved, receiving food or fluids in a way that is appropriate and safe, being repositioned comfortably and having family nearby can matter as much as sophisticated medical intervention.

This makes coordination between specialist teams and everyday carers essential.

A palliative consultant may establish the clinical plan, but nurses, care workers, family members and residential staff often deliver much of the person’s lived experience between professional reviews.

The care plan should therefore make comfort priorities understandable to everyone involved.

Organisations examining how risk, preference and quality of life interact can use the Positive Risk-Taking Planner to structure similar discussions about autonomy and proportionate safeguards. It is not a Hong Kong end-of-life legal or clinical tool, but the underlying principle is relevant: protecting somebody from every conceivable risk should not automatically override the way they wish to live during their remaining time.

Residential care homes are becoming an increasingly important end-of-life setting

For many frail older people, a Residential Care Home for the Elderly becomes their established home rather than a temporary service location. As residents become older and more medically complex, RCHEs increasingly encounter deterioration, advanced dementia and end-of-life decisions as part of ordinary service delivery.

This creates a strategic question about where palliative expertise should sit.

Residential homes cannot simply be expected to reproduce hospital-level clinical capability. At the same time, automatically transferring every dying resident to acute care can disrupt continuity and conflict with previously expressed preferences.

The stronger model brings clinical expertise into the residential setting where this is safe and appropriate.

Hospital Authority outreach, community nursing and geriatric services can support RCHEs alongside the person’s wider medical care. The practical value comes from enabling residential workers to understand what deterioration means, what symptoms can be managed within the home and when hospital assessment remains necessary.

The distinction between “hospital” and “care home” therefore becomes less useful near the end of life than the question of capability. The appropriate setting is the one able to meet the person’s needs safely and consistently with their wishes.

Avoiding an unnecessary transfer is different from avoiding hospital care

Policy interest in reducing avoidable hospital use can become dangerous if interpreted as a target to keep older people out of hospital regardless of circumstance.

Palliative care requires a more discriminating approach.

An acute fracture, uncontrolled haemorrhage or severe symptom crisis may require hospital treatment. Another resident with a known advanced condition may be able to remain in their RCHE when deterioration is expected, symptoms can be controlled and an agreed plan is already in place.

Consider an 91-year-old woman with advanced heart failure and severe frailty who has lived in the same RCHE for four years. She has discussed future care with clinicians and family and has clear preferences about avoiding burdensome intervention where recovery is unlikely.

Over several days she becomes weaker, eats less and spends increasing time asleep. Staff recognise the change as consistent with her expected deterioration, obtain appropriate clinical advice and implement the agreed comfort plan. Her daughter is informed and visits frequently.

The quality of this response lies not in the absence of hospital transfer itself. It lies in the fact that the decision is clinically supported, consistent with the woman’s wishes and accompanied by active care.

If she were instead experiencing a reversible acute problem that could not be managed safely in the home, transfer could remain entirely appropriate.

The wider principles of decision-making and escalation are therefore central to good palliative care. A pathway should support proportionate clinical judgement rather than replace it with a blanket rule.

Advance decisions are only useful when frontline staff understand what they mean

The new statutory framework creates clearer legal recognition of advance medical directives and DNACPR orders, but implementation reaches far beyond doctors who create or verify the documents.

Nurses, ambulance personnel, residential-care workers and other staff may encounter the person during deterioration. They need enough understanding to know that relevant documentation exists, when escalation is required and what they are not authorised to decide independently.

This creates a workforce-development requirement across settings.

Training should distinguish between an advance medical directive, a DNACPR order, an advance care plan and ordinary family preferences. These documents and conversations have different functions.

Staff also need to understand that a DNACPR order applies specifically to cardiopulmonary resuscitation. It should not be interpreted as an instruction to withhold all treatment.

A resident with a DNACPR order may still require antibiotics, oxygen, pain relief, treatment for a reversible infection or hospital assessment depending on the person’s condition and agreed goals of care.

This distinction protects against a serious operational risk: allowing one decision about resuscitation to become an unintended ceiling on all future care.

Dementia makes advance planning both more difficult and more important

Dementia creates a particular challenge because the person’s ability to make complex decisions may decline over time while their need for end-of-life planning increases.

Earlier conversations therefore matter.

A person in the mild stages of dementia may still be entirely capable of understanding options and making decisions about future treatment. Waiting until advanced cognitive impairment develops can leave families and clinicians interpreting what the person might have wanted without a clear record of their views.

Later in the condition, care still needs to remain person-centred even where formal decision-making becomes more difficult.

The person’s previously expressed values, behaviour, family knowledge and clinical interests all become relevant. A man who can no longer articulate a complex treatment preference may still communicate comfort, distress, fear or acceptance through facial expression and behaviour.

The principles within dementia end-of-life care and advance care planning therefore require both prospective planning and careful interpretation of the person’s present experience.

Frailty creates uncertainty about prognosis that should not prevent palliative thinking

Older people living with severe frailty often do not follow a predictable terminal trajectory.

A person may become seriously unwell, recover partially, remain stable and then deteriorate again several months later. Repeated hospital admissions can make it difficult to identify exactly when the person has entered the final phase of life.

This uncertainty can delay discussions about goals of care.

Yet palliative care does not require certainty about the exact timing of death.

An 87-year-old with advanced frailty, recurrent aspiration pneumonia and increasing dependence may benefit from discussions about future hospital treatment even if they could live for many months.

The objective is not to label the person as dying prematurely. It is to reduce the likelihood that major decisions are being made for the first time during an emergency.

This is particularly important where the balance of benefit and burden from invasive treatment may change as resilience diminishes.

Medication management changes as goals of care change

Older people approaching the end of life frequently take medication for several chronic conditions as well as drugs used specifically for symptom relief.

The role of medication therefore needs to be reconsidered as priorities change.

A treatment that was valuable when the objective was preventing a complication over several years may offer less benefit when life expectancy has shortened substantially. Conversely, medication for pain, breathlessness, nausea, anxiety or other symptoms can become increasingly important.

This does not justify indiscriminate withdrawal.

Medication review remains a clinical process requiring consideration of diagnosis, prognosis, symptom burden, interactions and the person’s preferences.

The practical challenge is ensuring that changes are understood wherever care is delivered.

A family supporting someone at home needs clear instructions. RCHE staff need an accurate medication plan. Community professionals need to know which medicines remain active and which have been discontinued.

Where several prescribers are involved, reconciliation becomes particularly important.

Anticipatory planning can reduce avoidable uncertainty during deterioration

Some deterioration near the end of life is foreseeable even if its timing is not.

A person with advanced illness may be expected to develop increasing pain, breathlessness, agitation, nausea or difficulty swallowing. Planning for those possibilities can make home or residential care more resilient.

Anticipatory planning is not the same as assuming every symptom is inevitable.

New problems should still be assessed appropriately. But where expected symptoms emerge, the care team can respond more quickly if agreed medication, contact routes and professional responsibilities are already understood.

For families, this can reduce the sense that every change is an unexpected emergency.

For staff, it creates clearer boundaries around what can be managed locally and when additional clinical input is required.

Eating and drinking decisions can become emotionally difficult

Reduced appetite and swallowing difficulty are common in advanced illness and severe frailty. For families, these changes can be particularly distressing because food is closely associated with care, affection and survival.

A relative may interpret reduced intake as evidence that the person is being neglected. Staff may feel pressure to encourage eating beyond what is comfortable. Clinical teams may need to discuss whether artificial nutrition or hydration would offer meaningful benefit in the person’s circumstances.

These decisions require sensitive communication.

The person’s condition, prognosis, ability to swallow safely, comfort and previously expressed wishes all matter.

A simplistic choice between “feeding” and “not feeding” can obscure the actual clinical and ethical questions.

Care may instead focus on offering food and drink that the person can manage comfortably, maintaining mouth care and responding to thirst or enjoyment while avoiding interventions that create greater burden than benefit.

The operational requirement is consistency. Families and staff need a shared understanding of the care plan so that different shifts or settings do not produce contradictory approaches.

Cultural expectations influence how end-of-life conversations are approached

End-of-life decision-making does not occur outside culture.

Some families may prefer difficult information to be discussed with relatives before it is discussed directly with the older person. Others may worry that talking openly about death removes hope or causes distress.

Professionals need cultural sensitivity without assuming that every person from the same background holds the same preference.

The strongest approach asks the individual how they want information communicated and who they want involved.

Some people will want detailed clinical explanations and direct participation. Others may prefer their family to play a larger role in communication. Preferences may also change as illness progresses.

Respecting culture therefore means adapting communication around the individual rather than replacing individual autonomy with presumed family norms.

The principles of cultural and identity needs are particularly relevant because sensitive communication can determine whether advance planning feels supportive or imposed.

Language and health literacy affect whether choice is genuinely informed

Advance care planning involves difficult concepts: prognosis, life-sustaining treatment, cardiopulmonary resuscitation, mental capacity and the circumstances in which a future directive would apply.

Legal and medical terminology can make these discussions inaccessible even to people who are otherwise confident navigating healthcare.

Professionals therefore need to translate complex concepts into understandable choices without oversimplifying them.

A person should understand that refusing one form of future treatment does not mean refusing all care. They should know that circumstances specified in an advance medical directive matter. They should also understand that preferences can be revisited while they remain capable of making the relevant decision.

Where communication needs, sensory impairment or language barriers exist, additional support may be required.

Choice without understanding is not meaningful autonomy.

Families need support during the dying phase, not only information about the patient

End-of-life care affects an entire household.

A spouse may be exhausted after months of caregiving. Adult children may disagree about treatment. Relatives living overseas may struggle with whether to return to Hong Kong. Some family members may be managing work and childcare alongside bedside responsibilities.

Clinical updates are necessary but do not address all of these pressures.

Families may need preparation for physical changes during dying, practical guidance about whom to contact and reassurance about symptoms that can appear distressing.

They may also need permission to recognise their own limits.

Consider a 78-year-old woman caring for her husband at home with advanced cancer. She strongly supports his wish to remain at home but has slept very little for several nights because she is frightened he will deteriorate while she is asleep.

The issue is not simply whether she remains willing to care. Her capacity is becoming unsafe.

A stronger response considers additional professional support, respite or an alternative care setting if the home arrangement can no longer be sustained.

This reflects the wider family partnership and carer-support agenda. Respecting the patient’s preferences should not require an exhausted relative to carry responsibility beyond what is realistically manageable.

Disagreement within families requires structured clinical communication

End-of-life decisions can expose longstanding family tensions.

One relative may favour continued active intervention, while another believes the person would want comfort-focused care. Family members living overseas may understand the person’s current condition differently from those providing daily support.

Where the person has already expressed valid preferences, those should provide the central reference point rather than family consensus alone.

Where wishes are unclear, healthcare professionals need to communicate the clinical situation carefully and avoid framing the discussion as a simple choice between “treatment” and “no treatment”.

The relevant question may be which treatments offer realistic benefit and which would impose burden without changing the underlying trajectory.

Good communication cannot remove every disagreement, but it can make clear which decisions are clinical, what the person previously expressed and where family input remains important.

Bereavement support is part of the palliative pathway

Palliative care does not end at the moment of death.

Family members may have spent months or years providing care. The death can bring grief, exhaustion, relief, guilt and practical disruption simultaneously.

Hospital Authority palliative services include bereavement support as part of their broader model.

The intensity of support required will vary. Many families draw primarily on relatives, friends, faith communities and existing networks. Others may need more structured professional support, particularly where the death was difficult, the caregiving period was prolonged or mental-health concerns emerge.

The important system principle is continuity. Families who have developed relationships with palliative teams should not experience death as an abrupt administrative endpoint.

Supporting death outside hospital requires coordination after death as well as before it

When a person dies at home or in an RCHE, families and staff need clarity about the procedures that follow.

Uncertainty at this point can undermine an otherwise well-planned end-of-life pathway.

Hong Kong has been developing arrangements that make dying in place more practicable, including changes that support death in residential settings and revised procedures around after-death arrangements.

The operational significance is considerable.

If a person’s preference is to remain in an RCHE, the home needs to understand the applicable process, staff need confidence in their responsibilities and relatives need accurate information about what will happen after death.

This is one example of why end-of-life policy cannot be separated into clinical and administrative components. A technically lawful option becomes meaningful only when frontline procedures make it workable.

Workforce competence needs to extend well beyond specialist palliative teams

Specialist palliative clinicians cannot deliver every element of end-of-life care personally.

Older people may spend most of their final months with family doctors, hospital clinicians from other specialties, community nurses, homecare workers or residential staff.

These workforces need different levels of competence.

Not everyone needs specialist expertise, but relevant staff should understand:

• how to recognise significant deterioration;
• when specialist palliative input may be helpful;
• the difference between advance care planning, advance medical directives and DNACPR orders;
• how to communicate respectfully with people and families;
• their own professional boundaries; and
• how to escalate symptoms or concerns promptly.

This connects with the wider older people’s workforce and practice competence agenda. End-of-life care becomes sustainable when specialist expertise is supported by wider generalist capability rather than isolated from it.

Staff wellbeing matters because end-of-life work carries emotional weight

Repeated exposure to death and family distress affects workers.

This is particularly relevant in RCHEs, where staff may have known residents for several years. A care worker can experience genuine bereavement when somebody they have supported daily dies.

If organisations treat death only as a clinical event or administrative closure, the workforce dimension can be overlooked.

Reflective supervision, peer support and opportunities to discuss difficult cases can help staff process the emotional and ethical complexity of the work.

Support should not pathologise normal sadness. It should recognise that emotionally demanding care requires organisational structures that allow people to reflect and learn.

Digital coordination can strengthen continuity only if information is trustworthy

The availability of advance decision information through electronic systems can improve continuity across care settings, particularly where a person cannot communicate their preferences during an emergency.

But end-of-life information has unusually high consequences.

Records need to be accurate, current and clearly attributable. Superseded documentation must not create confusion. Access needs appropriate information-governance safeguards.

The Digital Transformation Readiness Assessment can help organisations examine similar questions about digital governance, cyber resilience, workforce adoption and information quality. It is not a Hong Kong eHealth or palliative-care framework, but the underlying readiness questions are relevant whenever clinical decisions increasingly depend on digital information.

Quality measurement should reflect whether care matched the person’s goals

End-of-life quality cannot be judged by one simple indicator. A death at home is not automatically better than a death in hospital, just as avoiding intensive treatment is not inherently evidence of good care.

The more meaningful question is whether the care received was appropriate to the person’s condition, preferences and changing needs.

Useful quality evidence can therefore include symptom control, timeliness of palliative referral, documentation of advance care planning, avoidable emergency transfers, family experience, continuity between settings and whether agreed treatment preferences were understood and followed.

Place of death can still be relevant, but it needs context.

If a person wanted to remain at home but was transferred repeatedly because community support could not manage symptoms, that may reveal a capacity problem. If another person preferred hospital care because they felt safer there and symptoms were complex, hospital death may be entirely consistent with good person-centred care.

This is why the Quality Dashboard Builder can help organisations structure a wider view of quality, outcomes and service reliability. It is not a Hong Kong palliative-care performance framework, but the principle is relevant: quality indicators need to illuminate whether care remained aligned with the person rather than rewarding one setting or intervention automatically.

Governance should distinguish difficult outcomes from poor care

End-of-life care includes outcomes that are inherently difficult. A person may die shortly after hospital admission despite appropriate treatment. Symptoms may become difficult to control even when professionals respond well. Family members may remain distressed despite careful communication.

Governance therefore needs judgement as well as measurement.

Repeated patterns are more informative than isolated events.

If several residents are transferred from the same RCHE during expected deterioration because staff are uncertain about end-of-life plans, the issue may be workforce competence or clinical access. If advance medical directives repeatedly cannot be located when needed, documentation systems require review. If families consistently report that they did not understand what to expect, communication needs strengthening.

The objective is not to remove uncertainty from dying. It is to identify organisational problems that create avoidable distress, unwanted treatment or fragmented care.

This aligns with the broader principles of learning from incidents and continuous improvement. End-of-life governance is strongest when difficult experiences become intelligence for redesign rather than simply retrospective case closure.

The implementation of the new legal framework will depend on everyday operational reliability

The Advance Decision on Life-sustaining Treatment Ordinance creates a clearer statutory framework, but implementation quality will determine whether that framework changes lived experience.

Several practical conditions need to work together.

• People need understandable information before making advance decisions.
• Healthcare professionals need confidence in the legal and clinical requirements.
• Documents need to be valid, current and available when needed.
• DNACPR orders need to be recognised appropriately across emergency and care settings.
• Staff need to understand what a directive does and does not authorise.
• Families need enough explanation to reduce avoidable conflict during deterioration.

These are implementation issues rather than abstract legal principles.

A strong statutory framework can still fail operationally if a document cannot be found, if frontline staff misunderstand its scope or if the person was never supported to make an informed decision in the first place.

Organisations examining similar implementation questions can use the Governance Maturity Assessment to test whether policy, accountability, workforce competence and assurance connect in practice. It is not a Hong Kong legal compliance tool, but its underlying purpose is relevant: formal policy should be traceable into reliable operational behaviour.

Public understanding will influence whether advance planning becomes normal practice

Legal reform can create rights, but people need to know those rights exist before they can use them.

Advance care planning can be difficult to discuss because it involves illness, decline and death. Some people may avoid the subject until a crisis occurs. Others may misunderstand an advance medical directive as a request for all treatment to stop.

Public education therefore has an important role.

The aim should not be to encourage older people to refuse treatment. It should be to make future decision-making more informed.

People need to understand that advance planning can include preferences for active treatment, comfort, place of care and the circumstances in which particular interventions would or would not be acceptable.

Normalising these discussions could also reduce pressure on families, who otherwise may be asked to interpret the person’s wishes without sufficient prior conversation.

Primary healthcare can help move end-of-life planning upstream

Advance care planning often becomes visible during specialist or hospital care, but primary healthcare has an important role because it offers continuity before the person reaches a crisis point.

Family doctors may know the person’s chronic conditions, previous hospital use and general preferences over a longer period.

For older people with advanced frailty or several progressive conditions, primary healthcare can help identify when goals-of-care discussions would be useful even if specialist palliative referral is not yet required.

This does not mean every older person should be approached with end-of-life discussions simply because of age.

The conversation should remain proportionate to clinical circumstances and individual readiness.

The opportunity lies in avoiding a system in which the first serious discussion occurs only after an emergency admission.

Palliative integration should extend beyond cancer pathways

Hong Kong’s future palliative model will need to serve increasing numbers of people whose end-of-life trajectories arise from non-cancer conditions.

Heart failure, chronic respiratory disease, dementia, neurological illness and advanced frailty can all create substantial symptom and family-support needs.

Referral models should therefore be capable of recognising need rather than relying too heavily on diagnosis.

A person with repeated breathlessness, declining function and several hospital admissions may benefit from palliative input even if prognosis remains uncertain.

The shift is important because ageing populations increasingly die with multiple interacting conditions rather than from one neatly bounded disease pathway.

Capacity outside hospital needs to grow if greater choice of place is to be realistic

Policy ambition around dying in place depends on practical service capacity.

If more people are to spend their final phase at home or in RCHEs, those settings need reliable access to clinical advice, medication, nursing support, equipment and escalation.

Families also need support that reflects the intensity of caring near the end of life.

Without sufficient community capacity, preference can become symbolic rather than actionable.

A person may state clearly that they wish to remain at home, but if symptoms cannot be controlled or the family cannot sustain overnight care, hospital transfer may still become necessary.

The policy implication is that choice requires infrastructure.

It is not enough to record preferred place of care. The system needs to understand what resources make that preference feasible.

Technology can support coordination but should not depersonalise dying

Digital tools can improve access to records, support remote clinical review and help professionals coordinate across settings.

For some people, video consultation may reduce burdensome travel. Electronic advance decision records can make preferences more visible during emergency care. Remote monitoring may help identify symptom deterioration between visits.

But technology should remain subordinate to the human purpose of palliative care.

A dying person should not be surrounded by intrusive monitoring simply because the technology exists. Families should not be expected to manage complex digital systems while under emotional strain.

The strongest use of technology is to simplify coordination, reduce duplication and extend professional reach while preserving privacy and relationships.

International learning lies in connecting autonomy with service capability

Different countries use very different legal and service structures for advance decisions, palliative care and end-of-life treatment. Some have long-established statutory advance-directive frameworks. Others rely more heavily on clinical guidance, common law or family decision-making. Funding and hospice arrangements also vary substantially.

Hong Kong’s current reform therefore cannot be reduced to whether another country has a similar statute.

The transferable lesson lies in the interaction between legal autonomy and operational capability.

A right to refuse future treatment needs reliable documentation and informed clinical practice. A preference to die at home needs community services capable of supporting the person there. A DNACPR decision needs to be understood correctly by the workers who encounter the person during deterioration.

Other systems can adapt that principle without replicating Hong Kong’s institutional model: rights become meaningful only when services can implement them consistently.

The future direction is earlier planning, stronger community capability and better continuity

Hong Kong has already taken significant steps towards a more person-centred end-of-life framework. The legal recognition of advance decisions, established Hospital Authority palliative services, developing eHealth support and greater attention to dying in place create important foundations.

The next challenge is implementation at scale.

More older people with frailty and multiple conditions will need palliative approaches outside traditional cancer pathways. RCHEs will increasingly need clinical support around expected deterioration. Families will require better preparation and respite. Primary healthcare can help begin conversations earlier, while digital systems can support continuity when information is accurate and accessible.

The strongest future model will not be defined by reducing hospital deaths or increasing advance directives as targets in themselves.

It will be defined by whether older people understand their choices, whether preferences remain visible as care settings change and whether services can respond safely when those preferences favour less institutional or less invasive care.

Conclusion

End-of-life and palliative care in Hong Kong is entering an important period of transition. Population ageing is increasing the number of people approaching death with frailty, dementia and multiple chronic conditions, while the Advance Decision on Life-sustaining Treatment Ordinance creates a clearer statutory basis for advance medical directives and DNACPR orders.

The significance of that reform will depend on implementation. Patient autonomy needs understandable conversations, valid documentation, confident professionals and reliable information across hospitals, homes and RCHEs. Greater choice about place of care requires community and residential capacity capable of managing symptoms safely. Families need support without being expected to absorb unlimited responsibility, and specialist palliative expertise needs to connect with the wider workforce supporting older people every day.

The strongest direction for Hong Kong is therefore not simply more end-of-life policy, but better continuity between preference and practice. Earlier advance care planning, stronger symptom management, proportionate use of hospital care, better workforce capability and integrated quality governance can help ensure that dying is not treated as a final emergency in an otherwise fragmented pathway.

Good end-of-life care cannot remove loss or uncertainty. It can, however, reduce avoidable distress and give older people greater confidence that the care they receive will remain aligned with their values, dignity and wishes when they are least able to advocate for themselves.