Restrictive Practice in Hong Kong Care Services: Governance, Human Rights and Proportionate Support

A restraint may be visible: a belt, vest or other device that limits a resident’s movement. Restrictive practice can also be much less obvious. A resident may be repeatedly redirected away from a door, discouraged from walking independently, prevented from keeping personal possessions, supervised continuously, or expected to follow routines created mainly around staffing convenience. None of these situations should automatically be labelled abusive or inappropriate. Each requires a clear explanation of why restriction is necessary, what alternatives exist and whether the intervention remains proportionate.

This is an increasingly important governance question across Hong Kong’s residential, disability and older-person services. Residential care homes for the elderly and residential care homes for persons with disabilities operate within statutory licensing frameworks that set detailed expectations around restraint, residents’ safety, dignity and care. These controls sit alongside wider principles of participation, privacy and individual need explored through the Hong Kong Ageing, Long-Term Care & Community Support Knowledge Hub.

The core issue is not whether providers should sometimes intervene to prevent serious harm. They should. The governance challenge is ensuring that a necessary intervention does not become the default response simply because it reduces uncertainty for the organisation.

Hong Kong’s current residential-care requirements provide an important starting point. Physical restraint should be avoided wherever possible, alternatives considered first, the least restrictive option used, duration minimised, consent obtained and the person observed and re-assessed. These are not merely technical rules about equipment. They establish a wider principle: restrictions need justification, monitoring and an exit route.

Restrictive practice is broader than physical restraint

Physical restraint is the most regulated and visible form of restriction, but care services can limit people in many other ways.

Environmental controls can restrict movement. Staff practices can restrict access to activities. Technology can create surveillance. Organisational routines can narrow choice. Risk-averse care planning can result in people being permanently accompanied even where their capability increases.

The relevant question is therefore not only “Is a restraint being used?” but “Is the person’s liberty, movement, privacy, choice or ordinary participation being limited, and if so, why?”

Examples can include:

• physical devices that limit movement;
• furniture or environmental arrangements that prevent access;
• constant observation or unnecessary staff accompaniment;
• restrictions on community access, possessions, relationships or communication;
• technology used primarily to monitor rather than enable; and
• routines that people cannot reasonably influence despite no individual safety justification.

Different restrictions carry different legal, clinical and operational implications. They should not all be treated as equivalent. What connects them is the possibility that organisational control is being exercised over an individual’s ordinary life.

This is why least restrictive and just-enough support provides a useful wider quality principle.

Hong Kong’s restraint rules establish a high threshold

The Codes of Practice governing Hong Kong residential care provide detailed requirements for physical restraint. The principle is clear: homes should adopt measures to avoid restraint wherever possible, and physical restraint should be considered only where less restrictive alternatives have proved ineffective or in emergencies where safety, health or wellbeing is jeopardised.

The intervention must then be individualised. The type, size and material of restraint need to suit the resident. Duration and degree of restriction should reflect the principle of least restraint. Required consent arrangements apply, and the person’s condition must be observed while restraint is in use.

Re-assessment is equally important. A restriction that was justified at one point does not become permanently justified simply because it has already been authorised.

The current frameworks specifically anticipate alternatives such as medical review, medication review, input from physiotherapists, occupational therapists, psychologists or social workers, environmental modification, improved seating, leisure activity, additional engagement, changes to daily-care routines and family involvement.

This matters because it shifts the decision from “Which restraint should we use?” to “Why is the behaviour or risk occurring, and what else could address it?”

The adverse effects of restriction form part of the risk calculation

Restrictive practice is sometimes justified by examining only what it prevents. A belt may prevent a resident attempting to stand without support. Continuous observation may reduce the possibility of wandering. Restricting independent movement may reduce some falls.

But every intervention can create other consequences.

Hong Kong’s residential-care guidance explicitly recognises potential physical, psychological and social consequences of restraint. Reduced movement can contribute to physical deterioration. People may experience anger, fear, shame, helplessness or distress. Prolonged restriction can reduce social interaction and may ultimately increase frailty.

Those consequences should form part of the decision itself rather than being treated as secondary effects.

A restriction that reduces one immediate hazard while substantially increasing dependency, distress or physical decline may not represent the best overall care decision.

This connects directly with positive risk-taking for older people. Safety needs to be understood in the context of the whole person rather than as the absence of a particular incident.

Scenario: restraint after repeated attempts to stand

An older resident with cognitive impairment repeatedly tries to stand from a wheelchair despite poor balance. Staff are concerned about falls and consider using a physical restraint while the resident is seated.

A narrow approach would identify the fall risk, obtain the necessary authorisation and begin restraint.

A stronger assessment asks why the resident keeps standing. Staff review pain, seating comfort, toileting needs, medication, environment, communication and daily routine. They discover that the person becomes restless when seated for long periods and frequently attempts to reach the toilet without assistance.

Occupational and physiotherapy input leads to changes in seating and mobility support. The daily routine is adjusted to provide more frequent supported movement, while staff improve recognition of the resident’s toileting cues.

The intervention does not guarantee that physical restraint will never be necessary. It changes the sequence of decision-making. Restriction becomes the last option after the underlying need and less restrictive alternatives have been considered.

Governance evidence should record that reasoning. A completed consent form alone demonstrates that a process occurred. It does not demonstrate whether the organisation genuinely sought to avoid restriction.

Consent is necessary but does not remove the need for proportionality

Consent requirements create an important safeguard within Hong Kong residential care. Depending on the circumstances and the person’s ability to understand, relevant medical and resident or family consent is required before physical restraint is applied, outside emergency circumstances.

Consent, however, should not become a substitute for professional judgement.

A family may strongly request restraint because they fear a fall. A resident may acquiesce without fully understanding the consequences. Staff may assume that once consent is signed, continued use is automatically acceptable.

The organisation still needs to demonstrate that restraint remains necessary, that alternatives have been considered and that use has not expanded beyond the original purpose.

This is particularly important where family anxiety drives restrictive decisions. Family views are valuable, but choice and control should remain connected to the person receiving care rather than being replaced automatically by the preferences of relatives.

Restrictive drift often happens gradually

Many problematic restrictions are not introduced through one conscious decision. They accumulate.

A temporary supervision arrangement after a fall continues for months. A resident who once became lost is no longer encouraged to use community facilities independently. Staff begin locking away an item after one incident and never reconsider whether access can safely resume. A person is consistently redirected from leaving a room because it makes staffing easier.

Each individual step may originally have had a reasonable explanation.

The problem is the absence of review.

Restrictive drift occurs when temporary risk controls become routine and routine becomes organisational culture. Staff joining later may not know why the restriction began at all. They simply inherit it as “how we support this person”.

Governance therefore needs to make duration visible. Restrictions should have review points, and the review should ask whether the intervention can be reduced or removed, not simply whether it should continue unchanged.

Scenario: permanent supervision develops from a temporary precaution

A man living in a residential care home for persons with disabilities experiences a seizure while returning from a nearby community activity. Following medical assessment, staff accompany him closely whenever he leaves the home.

The arrangement is initially sensible. Several months later, his condition is more stable and medical treatment has been reviewed, but continuous accompaniment remains in place. Staff regard it as safer and his family supports the arrangement.

The resident increasingly objects because he previously travelled locally with limited support.

A formal review examines his current health, seizure pattern, ability to seek help, familiarity with the area, communication and what support would be available if something went wrong. The service considers graded alternatives: accompanied travel for unfamiliar journeys, agreed check-in points, carrying emergency information and gradually restoring independent local activity where appropriate.

The original precaution is therefore neither dismissed nor treated as permanent. It is re-evaluated using current evidence.

This illustrates why care planning and review are essential controls against restrictive drift. A risk plan should evolve with the person.

Restriction should never compensate for inadequate staffing

A particularly important governance boundary concerns workforce capacity.

Restrictive practice can become attractive when staffing is stretched. Keeping residents seated reduces demand for mobility support. Limiting activities simplifies rotas. Preventing spontaneous community access makes deployment more predictable. Continuous environmental restrictions can appear easier than providing individual support.

These pressures are operationally understandable, but they create serious risk if organisational constraints become the reason for limiting people.

The principle of least restriction requires services to distinguish between an intervention needed because of the person’s actual risk and an intervention introduced primarily because the organisation cannot provide sufficient support.

This is where safe staffing and deployment becomes a human-rights issue as well as a workforce issue.

Leaders should be alert when restrictive practice increases during vacancy periods, high staff turnover or use of unfamiliar workers. That does not prove causation, but it deserves analysis.

The Predictive Workforce Risk Module can help organisations test whether workforce instability coincides with changing service patterns. It is not a Hong Kong regulatory tool, but it can support analysis of the organisational conditions surrounding restrictive practice.

Workforce confidence can reduce unnecessary restriction

Staff often use restrictive interventions because they are worried, not because they intend to reduce someone’s rights.

An inexperienced worker may feel safer preventing a person with dementia from walking. A colleague unfamiliar with autism or intellectual disability may respond to distress by attempting to control behaviour rather than understand its purpose. Staff may interpret every deviation from routine as a hazard.

Competence therefore matters.

Training should address both how approved restraint is used safely and how restrictive interventions can be avoided. Employees need skills in communication, environmental adaptation, de-escalation, mobility support, behavioural understanding and recognising unmet needs.

Supervision then needs to test how those skills operate in practice.

This connects with safeguarding training and competency. A workforce capable of recognising abuse but unable to identify unnecessarily restrictive everyday practice still has an important assurance gap.

Behaviour should trigger understanding before control

Some restrictive interventions arise because behaviour is interpreted mainly as a problem to stop.

A person walks continuously. Another repeatedly tries to leave. Someone removes medical equipment. A resident calls out or becomes distressed during care. Restriction may reduce the visible behaviour while leaving its cause untouched.

Alternative assessment should therefore consider pain, discomfort, fear, boredom, sensory needs, communication difficulties, medication effects, unfamiliar routines and environmental stress.

The objective is not to suggest that every behavioural risk has an easy solution. Some situations remain complex and immediate intervention may be required. The stronger model nevertheless asks what the behaviour communicates and what environmental or support changes might reduce the need for restriction.

For people with intellectual disability, dementia or communication difficulties, this is particularly important because behaviour may be one of the clearest ways of expressing distress.

Scenario: repeated attempts to remove a feeding tube

A resident repeatedly reaches for a feeding tube, creating a significant health risk. Staff begin discussing physical restraint because removal could result in injury and urgent medical intervention.

The risk is genuine. Simply allowing the resident unrestricted access to the tube would not represent positive risk-taking.

Before moving directly to restraint, the team examines why the behaviour has increased. They consider discomfort at the insertion site, positioning, pain, cognition, anxiety, clothing and whether the resident understands the equipment. Medical and nursing advice is sought, and different environmental and engagement approaches are tried.

Some interventions reduce the behaviour but do not eliminate it. During periods of acute agitation, limited restraint may still be assessed as necessary to protect the person.

The difference is that the restraint is linked to a defined risk, used for the minimum necessary period and reviewed alongside continued attempts to address the cause. The organisation does not convert an intermittent clinical problem into permanent restriction.

The case also demonstrates why restrictive-practice governance cannot be reduced to an ideological position that restraint is always wrong. The relevant question is whether intervention is necessary, proportionate, properly authorised, safely delivered and continuously challenged.

Observation protects people only if it leads to action

Hong Kong’s restraint requirements place importance on observation while a person is physically restrained. Staff need to monitor condition, comfort and safety and record relevant observations.

This has practical significance because restraint itself can cause harm.

Observation should therefore be more than evidence that a worker looked at the resident at prescribed intervals. It should identify discomfort, circulatory problems, emotional distress, changes in behaviour or evidence that restraint is no longer necessary.

If observations repeatedly show distress but nothing changes, the monitoring system has become procedural rather than protective.

The same principle applies more widely to restrictive practice. Data should trigger decisions.

Human rights become real through ordinary routines

Human rights in care are sometimes discussed mainly through serious abuse, unlawful detention or major restrictions. Everyday service design matters too.

Privacy when receiving personal care, being able to communicate with family without unnecessary observation, choosing clothing, accessing personal possessions, taking part in relationships and having some influence over daily routines all affect dignity and autonomy.

Hong Kong’s residential-care framework explicitly links care with dignity and privacy. Service Quality Standards similarly emphasise rights and participation.

For providers, this means restrictive-practice governance should not focus exclusively on restraint registers. It should also examine whether organisational routines unintentionally remove ordinary control from people’s lives.

A service can have very low physical-restraint use while still being highly restrictive in culture.

Technology creates a new generation of restrictive practice questions

Hong Kong’s continuing investment in innovation and technology for elderly and rehabilitation care creates substantial opportunities to support independence. Sensors, bed-monitoring systems, fall-detection technologies, location tools and other connected devices can reduce reliance on direct observation.

Used well, technology can therefore reduce restriction.

A movement sensor may allow someone to move more freely than constant staff supervision. A location device may support community access. Digital prompts may allow a person to manage tasks with less direct assistance.

However, the same technology can extend organisational control into spaces that were previously private.

Continuous tracking, extensive camera use or automated alerts for normal behaviour can create digital safeguarding and surveillance risks.

The governance questions are similar to physical restriction: is the intervention necessary, what purpose does it serve, is there a less intrusive option, who can see the information and when will use be reviewed?

The Digital Transformation Readiness Assessment can help organisations examine governance, privacy, workforce capability and operational readiness before technology is embedded into care.

Family reassurance can unintentionally increase restriction

Families understandably want relatives to be safe. After a fall, episode of wandering or other frightening event, family requests may become highly risk-averse.

A family may ask for bedrails, restraint, continuous supervision, location monitoring or restrictions on community activity because these controls provide reassurance.

Providers need to listen carefully without assuming that the most restrictive request is automatically the correct response.

They should explain the benefits and potential harms of interventions, the person’s views, available alternatives and the limits of what restriction can achieve.

This can be difficult, especially where families believe that any subsequent adverse event proves the provider should have imposed tighter controls.

Clear communication and documented reasoning help establish that proportionate care is not careless care.

Regulatory enforcement makes restraint governance an active assurance issue

Restrictive-practice governance is not merely theoretical within Hong Kong’s residential-care sector. The Social Welfare Department’s public warning arrangements demonstrate active enforcement where homes do not comply with required standards.

Recent regulatory warnings have addressed improper application of physical restraints, inadequate observation, unsafe fixing of restraint equipment and failure to review whether continued restraint remained necessary when residents’ conditions changed.

This external visibility matters for operators.

It shows that restraint governance involves both individual safety and organisational accountability. Poor practice can result in regulatory attention, public warning records and reputational consequences.

Providers should therefore be able to demonstrate not only that restraint paperwork is present, but that:

• alternatives were genuinely considered;
• the intervention was individually justified;
• consent and assessment requirements were addressed;
• staff used the equipment safely;
• observation occurred and informed decisions;
• re-assessment challenged continued use; and
• the organisation monitors patterns across the service.

This is where regulation and oversight connects directly with frontline practice.

Governing bodies need to see the pattern, not every form

Individual restraint decisions belong close to the person and require appropriate clinical, care and managerial input. Organisational governance operates at another level.

Senior leaders and governing bodies need evidence about whether restrictive practice is increasing, concentrated in particular services or associated with staffing, incidents or particular management arrangements.

Useful indicators might include numbers of residents subject to physical restraint, duration, repeated use, emergency applications, overdue reassessments, restrictive-practice incidents, complaints, injuries and successful reductions.

Numbers require context. A service supporting people with more complex needs may reasonably have a different profile from another service. The purpose is not to create an arbitrary zero-restraint target where intervention remains necessary for immediate safety.

The objective is to identify unexplained variation and restrictive drift.

Organisations can use the Quality Dashboard Builder to bring restrictive-practice data alongside incidents, complaints, workforce information and outcome measures.

Scenario: organisational data reveals a management difference

A provider operating several residential services reviews restraint data for the first time across the whole organisation. Two homes support broadly comparable groups of residents, yet one reports substantially more frequent use of physical restraint.

There is no immediate evidence that staff in the higher-use home are deliberately acting improperly. Individual assessment forms are largely complete.

Senior leaders therefore explore the difference rather than treating forms as proof that practice is equivalent.

They find that the lower-use home has stronger occupational-therapy input, more consistent supervision and a management culture in which staff are expected to trial environmental and behavioural alternatives before escalating to restraint. In the higher-use service, staff are more likely to view restraint as a standard falls-management response.

The organisation uses the comparison for improvement rather than blame. Cross-service practice review is introduced, supervision is strengthened and managers begin tracking attempted alternatives and successful reductions.

Over time, restraint use in the higher-use home falls without an equivalent rise in serious injuries.

The scenario illustrates why quality data and performance metrics can expose cultural differences that are invisible within individual records.

Restrictive-practice reduction needs organisational ownership

Reduction cannot depend entirely on individual frontline workers.

Some alternatives require environmental changes, equipment, specialist input, workforce redesign or investment. Staff cannot independently solve a poorly designed environment or chronic deployment problem.

Senior leadership therefore needs to understand which restrictions are maintained because better alternatives require resources.

The governance response may include investment in equipment, environmental improvement, specialist assessment, additional staff development or redesign of routines.

The Governance Maturity Assessment can help organisations examine whether responsibility, challenge and improvement action around restrictive practice reaches the appropriate level of leadership.

Incidents should not automatically drive services towards more restriction

After a serious fall, missing-person incident or other adverse event, the organisational instinct may be to introduce tighter controls.

Some changes will be justified. Others may be reactive.

Incident review should therefore ask what actually contributed to the event before assuming that greater restriction is the correct solution.

A fall may relate to medication, environment, footwear, acute illness or poor transfer technique rather than insufficient restraint. A missing-person incident may reveal weak communication or an unrecognised pattern of distress. A behavioural incident may relate to sensory overload or pain.

Linking restrictive practice with learning from incidents helps prevent an automatic cycle in which every adverse event produces another control.

The relevant test is whether the new restriction addresses the identified cause and whether less restrictive alternatives could achieve the same objective.

Service-user voice should be part of restrictive-practice review

A governance system can be technically strong while failing to capture how restriction feels to the person.

Residents may experience restraint as frightening or humiliating. Continuous supervision may feel intrusive. Restrictions on possessions, communication or movement can reduce identity and dignity even where staff believe they are acting protectively.

People should therefore be involved in reviews wherever possible, using communication approaches suited to their needs.

For someone unable to describe their experience verbally, staff should still observe distress, resistance, withdrawal or changes in behaviour. Family insight may help, but it should not automatically replace the person’s own expressed or observed preferences.

Experience evidence is particularly important because successful restrictive-practice reduction should improve more than a compliance metric. It should increase meaningful freedom, participation or comfort.

International learning: strong regulation should challenge restriction, not institutionalise it

Countries organise restraint law, consent, capacity and adult protection differently. Hong Kong’s residential-care framework therefore cannot simply be translated into another system’s legal terminology.

The transferable principle is nevertheless important.

Regulation can make restrictive practice safer, but the strongest regulation also asks whether restriction is necessary at all. Hong Kong’s requirement to consider alternatives, apply least restraint, minimise duration, observe the person and re-assess continued need reflects that distinction.

Other systems can adapt the underlying principle without replicating the mechanism: restriction should have a clear purpose, credible evidence, proportionality, review and an expectation that it will end when no longer necessary.

That final point matters most. A restrictive-practice system that becomes exceptionally efficient at documenting long-term restriction may still be weak if it rarely succeeds in reducing it.

The future is restrictive-practice intelligence, not simply restraint compliance

Hong Kong’s care system is moving towards greater use of technology, more sophisticated workforce roles and strengthened residential-care oversight. These developments create an opportunity to make restrictive-practice governance more analytical.

Provider systems could increasingly connect restraint records with falls, medication changes, staffing, complaints, behavioural patterns and service-user outcomes. Digital tools could prompt reassessment when authorisations approach review dates or identify services where restrictive practice differs substantially from organisational norms.

Technology should assist human judgement rather than determine whether someone is restricted.

The future opportunity is therefore not merely digitising restraint forms. It is building intelligence that helps organisations identify where restriction is increasing, why it is happening and which alternatives actually work.

That moves governance from documenting intervention towards preventing unnecessary intervention.

Conclusion

Hong Kong’s residential-care framework provides a clear starting point for restrictive-practice governance: restraint should be avoided where possible, alternatives should be attempted, intervention should be least restrictive, consent and assessment matter, observation must protect the person and continuing use requires re-evaluation. Current regulatory enforcement demonstrates that these expectations have practical consequences.

The wider challenge is preventing restriction from becoming normal in forms that are harder to see. Staffing pressure, family anxiety, organisational routine, behavioural uncertainty and digital surveillance can all narrow a person’s freedom even when no formal restraint is recorded.

Strong governance therefore needs to ask two questions simultaneously: is the intervention being used safely, and does it still need to be used at all? The first protects people during restriction. The second protects them from unnecessary restriction.

For Hong Kong’s older people and adults with disabilities, the strongest direction is neither unrestricted risk nor institutional control. It is proportionate support built around dignity, individual need, professional judgement and visible accountability. Restrictive interventions will sometimes remain necessary, but their legitimacy depends on being exceptional, individually justified and continually challenged. The measure of a mature system is therefore not simply how well it records restriction, but how effectively it creates the conditions in which less restriction becomes possible.