6.3 Applying the Post-AGNI Deprivation of Liberty Test – Practice Guidance

1. What Changed Following the 2026 Supreme Court Judgment

In A Reference by the Attorney General for Northern Ireland of a Devolution Issue under Paragraph 34 of Schedule 10 to the Northern Ireland Act 1998 [2026] UKSC 16, the Supreme Court changed the way in which a deprivation of liberty is identified for the purposes of Article 5 of the European Convention on Human Rights (ECHR).

The previous approach, commonly known as the ‘acid test’, looked at whether the person was subject to continuous supervision and control and was not free to leave. That test should no longer be used to determine whether a person is deprived of their liberty.

The ‘acid test’ has been replaced by a multifactorial assessment which requires consideration of the person’s specific and concrete situation, the restrictions which operate in practice, their degree and intensity, their effect on the person and the wider context in which they occur. No single factor is determinative.

The assessment should remain practical and realistic. It should focus on what the person’s life actually looks like, rather than simply whether particular restrictions appear in a care and support plan.

2. The Three Elements of Deprivation of Liberty

Under Article 5 of the European Convention Rights (‘ECHR’), a person is deprived of their liberty if three elements are satisfied:

1. an objective element: the person is confined in a restricted place for more than a negligible period of time;
2. a subjective element: that the person has not given valid consent to that confinement;

3. imputability to the state: the confinement is directly or indirectly attributable to the state.

These elements should be considered separately.

3. Objective Confinement: Constructing the Multifactorial Assessment

The first question is whether the person is confined in a restricted place for more than a negligible period of time. The distinction between a restriction on liberty and a deprivation of liberty is one of degree and intensity.

A useful way to approach the objective element is as a series of linked lines of enquiry. These are not separate legal tests and should not be scored. They provide a structure for gathering and analysing the evidence needed for the overall professional judgment.

3.1 Start with the person’s ordinary day

Begin with the person’s everyday experience rather than with a legal conclusion. Build a picture of what an ordinary day actually looks like.

• where the person can go independently within the setting;
• whether they can access outside space and the wider community;
• what supervision or observation actually takes place;
• whether arrangements differ during the day, evening or night;
• what control the person exercises over their daily routine;
• whether they can spend meaningful time alone;
• how contact with family, friends and others takes place;
• what happens when the person makes an unexpected or spontaneous choice.

The aim is to understand the person’s real life, not simply reproduce the contents of a care and support plan.

3.2 Identify the restrictions which actually operate

Do not assume that a measure exists simply because it appears in a care and support plan. Equally, do not assume that a restriction is absent because it has not been formally recorded.

Relevant measures may include locked or controlled doors, restricted access to internal areas, supervision or observation, staff accompaniment, physical intervention or restraint, bedrails or other equipment, monitoring technology, restrictions on community access or contact, and medication which may materially affect presentation or the expression of wishes.

For each material restriction, establish:

• what actually happens;
• when it applies;
• who implements it;
• whether it is current, intermittent, historical or proposed;
• how frequently it operates;
• how long it operates for;
• its practical effect upon the person.

3.3 Understand each material restriction

Nature

Consider what the restriction actually controls. Does it prevent movement, require another person’s permission, prevent the person acting on a choice, or control when and how they leave? Is the control environmental, physical, technological or exercised by staff? Does it operate continuously or only when the person seeks a particular freedom?

The label attached to a measure is less important than its practical operation. Staff accompaniment, for example, may enable one person to access the community. For another, it may mean that every attempt to go out independently is prevented.

Frequency and duration

Consider how long the restriction operates when it occurs, how often it occurs, and how long the overall pattern has existed. A brief restriction once a month has a different character from the same restriction operating repeatedly throughout every day.

Relevant considerations include whether the arrangement is temporary, newly introduced, intermittent, reducing, escalating, long standing or expected to continue.

Manner of implementation

How a restriction is implemented can be as important as the restriction itself. Establish what happens where the person’s wishes conflict with the arrangement. This may range from support, negotiation and redirection through to refusal, physical prevention or restraint.

Practical effect

Consider what the restriction actually means for this person. What does it prevent them doing? What does it require them to do? How much opportunity do they have to act spontaneously? How often does a restriction interfere with something they want to do? Does it produce frustration, distress or conflict? Does the person repeatedly seek the opposite choice?

The person’s response is relevant to understanding the effect and intensity of the arrangements. It should not be reduced to a formula in which contentment means no deprivation or objection automatically means deprivation.

3.4 Establish what actually happens around leaving

Freedom to leave remains relevant, but it is no longer a standalone test. Rather than asking only whether the person is free to leave, establish the factual position.

• can the person operate the external door or ask for it to be opened?
• what happens when they ask?
• can they leave independently?
• if they require support, is that support routinely available?
• have they actually asked or attempted to leave, and what happened?
• have they asked to go home or live elsewhere?
• what would happen if they persisted?
• what would happen if they wanted to leave permanently?
• does the response vary according to time, staffing or circumstances?

A statement that the person would be stopped if they tried to leave remains relevant evidence. It is not, by itself, a complete analysis. It should be considered alongside the person’s actual everyday opportunities, community access, retained freedoms and the wider restrictive regime.

3.5 Identify the liberty the person positively retains

The assessment should not become an exercise in collecting restrictions. It is equally important to establish what the person can and does actually do. For example, can they:

• move freely through ordinary areas of the setting;
• access outside space independently;
• spend meaningful private time without observation;
• choose their routine, activities or usual times for getting up and going to bed;
• initiate outings;
• leave the setting independently where this occurs;
• maintain contact with family and friends;
• communicate privately;
• make spontaneous choices which are respected.

Retained liberty should be based on positive evidence. The absence of a recorded restriction is not evidence of freedom. Statements such as “the person has never tried to leave”, “does not ask to go out” or “appears settled” do not, without more, establish retained liberty.

3.6 Understand the person’s own experience

Where possible, ask the person directly about their circumstances. Explore how they feel about living or staying there, whether they regard it as home, whether there is somewhere else they would rather be, whether they can go out when they want, and whether there is anything staff prevent them from doing. Record the person’s actual words where possible.

Where verbal communication is limited, observe behaviour and use information from people who know the person well. Potentially relevant behaviour can include approaching exits, waiting near doors, packing belongings, asking relatives to collect them, referring repeatedly to another home, resisting particular interventions, distress when prevented from doing something, willingly returning after outings, or initiating activities and contact.

Behaviour should be interpreted rather than merely listed. Consider what the behaviour may be communicating, whether there is another reasonable explanation, what other evidence adds, and whether the pattern has changed over time.

3.7 Distinguish intrinsic limitation from external restriction

Where the person cannot leave or move independently, establish why. Consider whether they can walk or mobilise independently, use mobility equipment, request assistance, indicate where they wish to go, or physically move with appropriate support.

Assistance may enable liberty rather than restrict it. Requiring help to transfer into a wheelchair, for example, does not itself amount to confinement. The analysis may be different where the person requests the assistance needed to leave and it is withheld in order to prevent them doing so.

Significant physical or cognitive impairment should not automatically be treated as meaning that the person has no physical liberty capable of being confined.

3.8 Consider purpose and relative normality carefully

The reason for a restriction can provide relevant context, but it should not replace analysis of what the arrangement actually does. A locked external door may exist to prevent a serious road risk, for example, while still having the practical effect of preventing the person leaving independently.

Relative normality may also form part of the overall assessment. Consider opportunities for ordinary community life, privacy, choice, contact with other people, freedom of movement, intensity of supervision, and how far the arrangements resemble ordinary support rather than confinement.

Avoid broad conclusions such as “this is normal in a dementia care home”. Return to the individual person’s circumstances.

3.9 Consider medication, fluctuation and the longitudinal picture

An assessment visit is only a snapshot. Consider whether the person’s presentation during the visit is representative of their usual circumstances.

• whether presentation varies at different times of day;
• whether objection or exit-seeking is time-specific;
• whether communication fluctuates;
• whether recent illness or delirium has affected presentation;
• whether restrictions have recently increased or reduced;
• whether medication may materially affect alertness, behaviour or the expression of wishes;
• whether records describe the current position or an earlier period.

Medication does not automatically amount to a restriction and its effects on the person should not be assumed. Where it may be relevant, establish what evidence exists about its actual effect. If the evidence is uncertain, record the uncertainty rather than filling the gap with an assumption.

3.10 Triangulate the evidence

Relevant evidence may come from the person, direct observation, care records, staff, family and friends, advocates and other professionals. Where accounts differ, explore why.

An apparent discrepancy may reflect genuinely conflicting evidence, different periods of time, different circumstances, variations between staff, outdated records or different interpretations of the person’s behaviour. Important uncertainty should remain visible in the assessment rather than being removed simply because a conclusion is required.

3.11 Step back and consider the arrangements cumulatively

Once the evidence has been gathered, step away from the individual restrictions.

Consider together the nature, frequency and duration of restrictions; their manner of implementation; what happens when the person acts contrary to them; their practical effects; the person’s experience; retained liberty; supervision and control; community access; intrinsic limitations; wider context; and changes over time.

Restrictions which appear relatively modest when viewed individually may have a different character when they operate together. Equally, the existence of significant restrictions does not remove the need to consider substantial freedoms which the person genuinely exercises.

The final reasoning should explain what the combined arrangements actually amount to, rather than simply list the factors considered.

3.12 A final sense-check before concluding the objective element

• could another practitioner understand what this person’s ordinary life actually looks like?
• have the arrangements been established in practice rather than mainly from labels in the care and support plan?
• is it clear what happens when the person seeks to act differently?
• have both restrictions and positively evidenced freedoms been considered?
• has the person’s own experience been explored?
• has intrinsic inability been distinguished from externally imposed restriction?
• is the evidence representative over time?
• are important conflicts or gaps visible?
• have the arrangements been considered cumulatively rather than by searching for one decisive feature?

4. Valid Consent: A Separate Enquiry

If the multifactorial assessment identified objective confinement, consider separately whether the person validly consents to that confinement for Article 5 purposes. Valid consent is not the same question as mental capacity under the Mental Capacity Act 2005.

A person may lack capacity to make a decision about residence or care and still have sufficient awareness of their immediate circumstances to register whether they are broadly happy or unhappy with them and to express an attitude towards remaining there.

4.1 Look for positive evidence of genuine acceptance

Depending on the person’s communication, evidence may include verbal statements, reliable non-verbal communication, behaviour, choices, the person’s response when returning to the setting, and consistent expressions of attitude over time.

Family members, advocates, carers and professionals who know the person well may assist in interpreting communication, but their views should support rather than replace evidence of the person’s own attitude.

4.2 Do not confuse compliance or settledness with consent

A person should not be regarded as consenting merely because they comply with staff, accept care, appear settled, do not ask to leave, do not physically resist, or have stopped trying to leave.

There may be many explanations for apparent acceptance, including genuine contentment, acquiescence, communication difficulty, learned compliance, fear or trauma, medication effects, or lack of opportunity to express a different preference.

4.3 Consider objection in words and behaviour

Objection may be expressed through words or behaviour. Relevant evidence can include the person saying they do not want to live there, repeatedly asking to go home, trying to leave, waiting at exits, packing belongings, asking someone to collect them, or becoming distressed when prevented from leaving.

Context remains important. Resistance to personal care, for example, does not automatically establish objection to the overall placement or confinement.

4.4 Consent can change

Consent is not necessarily permanent. A person who genuinely accepts the arrangements today may later object. Equally, a person who initially objects may later become genuinely accepting once they have experienced the placement.

Where the conclusion that there is no deprivation of liberty depends upon valid consent, changes in the person’s wishes, behaviour, communication, medication or circumstances may require the Article 5 position to be reconsidered.

5. Imputability to the State

If objective confinement exists and valid consent is absent, consider whether the confinement is directly or indirectly attributable to the state.

This will commonly be present where arrangements are made, commissioned, funded, approved or knowingly overseen by a local authority, NHS body or other public authority. Private arrangements may require closer consideration.

Only where objective confinement, absence of valid consent and state imputability are all present is there an Article 5 deprivation of liberty.

6. Reaching the Article 5 Conclusion

The reasoning should follow a clear sequence:

1. is the person objectively confined when the arrangements are considered cumulatively?
2. if so, have they given valid consent to that confinement?
3. if valid consent is absent, is the confinement attributable to the state?

The conclusion should identify the evidence given particular weight and acknowledge important evidence pointing the other way. Where there is genuine uncertainty, the uncertainty should be visible and reasoned through rather than hidden.

7. Where There Is No Deprivation of Liberty: Wider Safeguards

A negative Article 5 conclusion should not automatically end the assessor’s consideration of the restrictions identified during the assessment. The assessment may have uncovered substantial restrictions, vulnerability, unresolved decision-making issues or matters requiring review even though the arrangements do not meet the Article 5 threshold.

Where appropriate, the Best Interests Assessor (BIA) should consider whether observations or recommendations could strengthen the other legal and practice safeguards around the person. The BIA’s recommendation is not itself a statutory direction. However, it may draw attention to an existing legal duty or statutory principle which remains applicable regardless of the Article 5 outcome.

7.1 Start with why there is no deprivation

The recommendations should reflect the basis of the conclusion.

Where the arrangements do not amount to objective confinement

Consider whether the conclusion depends upon particular freedoms or upon the present degree and intensity of restrictions. If those circumstances materially change, the Article 5 analysis may also change.

• review if supervision, control or restrictions materially increase. (Article 5 ECHR; Care Act 2014, s.1(3)(h) and s.27; MCA 2005, s.1(6));
• ensure material restrictions and the freedoms relied upon in the assessment remain visible within care planning and review. (Care Act 2014, ss.24-25 and s.27);
• consider whether restrictions can be reduced further without undermining the purpose of the care arrangements. (Care Act 2014, s.1(3)(h); MCA 2005, s.1(6)).

Where the conclusion depends upon valid consent

Where objective confinement exists but the person is not deprived of their liberty because they are validly consenting, particular attention should be given to how changes in the person’s attitude will be recognised.

• record and respond to withdrawal of agreement, emerging objection, attempts to leave, requests to live elsewhere or behavioural expressions of objection. (Article 5 ECHR; Care Act 2014, s.1(3)(b) and s.27 where applicable);
• reconsider the Article 5 position if communication, medication, illness or other circumstances materially affect the evidence of consent. (Article 5 ECHR; Care Act 2014, s.27 where applicable).

Where intrinsic limitation was material to the conclusion

Consider whether changes in mobility, physical ability, communication, assistance, environmental restrictions or supervision could change the analysis. The relevant question is whether the factual basis of the assessment continues to hold.

7.2 Mental Capacity Act protections

A finding of no deprivation of liberty does not remove the Mental Capacity Act 2005 from restrictive care. Where the evidence identifies a need, recommendations may include:

• further practicable steps to support the person to make a particular decision. (Mental Capacity Act 2005, s.1(3));
• a decision-specific assessment of capacity where capacity for a particular matter remains unclear. (Mental Capacity Act 2005, ss.2-3);
• clearer recording of a best interests decision. (Mental Capacity Act 2005, s.4);
• fuller consideration of the person’s past and present wishes and feelings, beliefs and values. (Mental Capacity Act 2005, s.4(6));
• consultation with people engaged in caring for the person or interested in their welfare. (Mental Capacity Act 2005, s.4(7));
• reconsideration of whether the purpose can be achieved in a less restrictive way. (Mental Capacity Act 2005, s.1(6));
• review of restraint to ensure that it remains necessary to prevent harm and proportionate to the likelihood and seriousness of that harm. (Mental Capacity Act 2005, s.6);
• clarification that acts of care or treatment are supported by reasonable steps to establish capacity and a reasonable belief that the act is in the person’s best interests. (Mental Capacity Act 2005, s.5).

Recommendations should identify the actual decision or restriction requiring attention rather than simply recommending “an MCA assessment”.

7.3 Care Act protections

Where the person receives care and support within the Care Act framework, a conclusion that Article 5 is not engaged does not remove the local authority’s wider duties. The wellbeing principle in section 1 is particularly important in restrictive care.

• ensure that restrictions upon the person’s rights or freedom of action are kept to the minimum necessary. (Care Act 2014, s.1(3)(h));
• ensure that the person’s views, wishes, feelings and beliefs are properly considered. (Care Act 2014, s.1(3)(b));
• increase the person’s control over day-to-day life and the way care and support is provided. (Care Act 2014, s.1(2)(d));
• support and maintain domestic, family and personal relationships. (Care Act 2014, s.1(2)(g));
• review whether the person’s accommodation remains suitable. (Care Act 2014, s.1(2)(h));
• review the care and support plan where circumstances warrant this. (Care Act 2014, s.27);
• reassess needs where there has been a material change. (Care Act 2014, s.9);
• ensure important matters identified through assessment and review are reflected in care and support planning. (Care Act 2014, ss.24-25).

7.4 Advocacy and representation

Where DoLS does not apply, safeguards which depend upon an assessment or authorisation may no longer be available. Consider whether the person requires another form of independent support.

• consider independent advocacy to support participation in assessment, care and support planning or review where the statutory criteria are met. (Care Act 2014, s.67);
• consider independent advocacy during safeguarding processes where the statutory criteria are met. (Care Act 2014, s.68).

Advocacy should not be recommended automatically. The question is whether the person has an effective means of participating in the processes which now provide the safeguards around their care.

7.5 Safeguarding

A finding that the person is not deprived of their liberty does not determine whether restrictive practice is lawful, proportionate or safe. Where the assessment identifies possible abuse or neglect, consider whether the safeguarding duty may be engaged.

• disproportionate or unexplained restraint. (MCA 2005, s.6; Care Act 2014, s.42);
• restrictions repeatedly imposed without appropriate capacity and best interests decision-making. (MCA 2005, ss.1 and 4-6; Care Act 2014, s.42);
• blanket or organisational restrictions which unnecessarily restrict freedom or autonomy. (Care Act 2014, s.1(2)(d), s.1(3)(h) and, where the safeguarding criteria are met, s.42);
• neglect of the person’s relationships or community access. (Care Act 2014, s.1(2)(g) and s.1(3)(h));
• concerns about coercion, ill-treatment or wilful neglect. (MCA 2005, s.44 where applicable. Care Act 2014, s.42).

A negative deprivation of liberty finding should never be used as evidence that restrictive care is necessarily safe or lawful.

7.6 Human rights beyond Article 5

A person who falls outside Article 5 remains protected by the other Convention rights. Article 8 may remain particularly important where arrangements affect privacy, personal autonomy, family and personal relationships, visitors, communication, surveillance, access to the community, daily routines, intimate care, or decisions about where and how the person lives.

• public authorities must act compatibly with Convention rights. (Human Rights Act 1998, s.6);
• restrictions affecting private and family life should have a lawful basis and be necessary and proportionate. (Article 8 ECHR);
• in serious cases involving degrading treatment, ill-treatment or neglect, Article 3 may also require consideration. (Article 3 ECHR).

8. Making Useful, Legally Anchored Recommendations

The purpose is not to produce a standard list for every negative assessment. Recommendations are most useful when they arise directly from the evidence gathered during the assessment.

Where possible, each recommendation should make clear:

1. what was identified during the assessment;
2. what action or consideration is recommended;
3. which legal duty or statutory principle is relevant;
4. who is best placed to respond;
5. what change should prompt further review, where relevant.

Useful final questions for the BIA include:

• what has this assessment identified which somebody still needs to act upon?
• what protection was the DoLS process expected to provide, and if Article 5 does not apply, is that protection available through another route?
• what does my conclusion depend upon remaining true?
• how will anyone know if that changes?
• does the person still have an effective voice in decisions about their care?
• are any restrictions capable of being reduced?
• is there another statutory process which should now do the safeguarding work?

Where there is nothing requiring further action, it is appropriate to record that. The objective is not to keep people unnecessarily within the DoLS system. It is to ensure that falling outside Article 5 does not mean falling outside meaningful protection.

9. Relevant Legislation, Case Law and Guidance

• A Reference by the Attorney General for Northern Ireland of a devolution issue under paragraph 34 of Schedule 10 to the Northern Ireland Act 1998 (June 2026);
• Guzzardi v Italy (1980) 3 EHRR 333;
• Oxfordshire County Council v P [2026] EWCOP 33 (T2);
• Mental Capacity Act 2005, including Schedule A1;
• Care Act 2014;
• Human Rights Act 1998 and Articles 3, 5 and 8 ECHR;
• DHSC interim guidance following the 2026 Supreme Court ruling;
• Mental Capacity Act Code of Practice and Deprivation of Liberty Safeguards Code of Practice, read subject to the 2026 Supreme Court judgment.