Prevalence of deprivation of liberty: a survey of in-patient services

Aims and method A trust-wide survey of all in-patients to estimate the prevalence of likely deprivation of liberty and to investigate how two different approaches to measuring deprivation might affect the number identified.

Results A notable difference in the results was detected when comparing the two methods. One survey method identified deprivation of liberty factors in 84% of informal incapacitous patients, whereas a different approach that weighed up the factors in accordance with the UK government’s interpretation of the case law detected only 11% as potential cases.

Clinical implications To be usefully implemented, the deprivation of liberty safeguards require an accurate method of assessment in order to target those the process is designed to protect. This survey demonstrates the difficulties in defining deprivation of liberty and highlights the issue of how it is best measured.

The deprivation of liberty safeguards, introduced in April 2009 in England and Wales, are an addition to the Mental Capacity Act 2005. They provide additional legal protection for those over 18 years old who lack capacity to consent to their care or treatment (including residence in a hospital or care home) and who are at risk of being 'deprived of their liberty'. The right to liberty was included in Article 5 of the European Convention of Human Rights. The UK signed the Convention in 1951, and incorporated it into domestic law in the Human Rights Act 1998. The European Court of Human Rights stated in the case of Guzzardi v. Italy that the aim of Article 5 is that 'no one should be dispossessed of liberty in an arbitrary fashion' and that this should be assessed 'using a whole range of criteria such as the type, duration, effects, and manner of implementation of the measure in question'. 1 The new safeguards do not apply to those detained under the Mental Health Act 1983, which has a separate set of standards to protect the detained person's rights under Article 5.
The deprivation of liberty safeguards were introduced following the well-known case of HL v. the United Kingdom (the Bournewood judgment). 2 HL did not have capacity to consent to his admission to hospital, and as he made no attempt to leave, he was treated under the common law doctrine of necessity. The High Court found that this did not amount to an illegal detention in hospital, but the Court of Appeal overturned this decision, as the hospital would prevent HL from leaving if he attempted to do so. The House of Lords subsequently found against HL, and the case was taken to the European Court of Human Rights. They analysed the case using the precedent set by Guzzardi v. Italy and intervening cases, finding that as HL would be prevented from leaving hospital, his carers were prevented from visiting him, and as the hospital 'exercised complete and effective control of his care and movements' he was unlawfully detained. The safeguards were therefore introduced to allow a hospital or care home to make an application so that a necessary deprivation of liberty may be made lawful. The detained person has the right to appeal against the decision via the newly created Court of Protection.
In addition to the Bournewood case, the Code of Practice for deprivation of liberty safeguards cites several other cases as contributing to the legal definition of deprivation of liberty: LLBC v. TG, 3 Nielsen v. Denmark,4 HM v. Switzerland, 5 DE and JE v. Surrey County Council, 6 and Storck v. Germany. 7 The Code of Practice states 'there is no simple definition of deprivation of liberty'. 8 This raises questions about how to decide who will require a deprivation of liberty safeguards assessment. The Code of Practice states that the distinction between deprivation of liberty (which requires authorisation to be made legal) and restriction of liberty (which does not) is one of 'degree or intensity and not one of nature or substance'. 8 Factors that have been identified in case law to be relevant to this distinction are summarised in Table 1.
It is important to note that this distinction is based on case law that has developed in an ad hoc manner, and as such it can be complex and confusing. For example, it Aims and method A trust-wide survey of all in-patients to estimate the prevalence of likely deprivation of liberty and to investigate how two different approaches to measuring deprivation might affect the number identified.
Results A notable difference in the results was detected when comparing the two methods. One survey method identified deprivation of liberty factors in 84% of informal incapacitous patients, whereas a different approach that weighed up the factors in accordance with the UK government's interpretation of the case law detected only 11% as potential cases.
Clinical implications To be usefully implemented, the deprivation of liberty safeguards require an accurate method of assessment in order to target those the process is designed to protect. This survey demonstrates the difficulties in defining deprivation of liberty and highlights the issue of how it is best measured.
Declaration of interest None.

ORIGINAL PAPERS
Prevalence of deprivation of liberty: a survey of in-patient services { remains unclear whether one 'deprivation' factor alone is sufficient for an in-patient to require deprivation of liberty safeguards protection, and it may be that the legal concept of deprivation of liberty evolves over time. It is also unclear how common deprivation of liberty is in the UK. The only surveys to date have been within the Department of Health, which estimate that there will be 21 000 deprivation of liberty safeguards assessments in the first year -a fourfold increase compared with their previous estimate. 10 It was in this context that we undertook a survey of possible deprivation of liberty among in-patients in a mental health trust. We had two aims: to estimate the prevalence of deprivation of liberty and to assess whether differing approaches to its assessment have any impact on this prevalence.

Method
The project was approved by and registered with the West Yorkshire Research and Development Consortium, and the data was collected by four of the authors (J.R., E.M., T.S. and J.B.).
The survey included all in-patients on a particular day across Leeds Partnerships NHS Foundation Trust who were in hospitals and units where the deprivation of liberty safeguards will apply. The Trust provides the majority of adult mental health and learning disability in-patient beds for a city population of 725 000. In total, 400 people were surveyed. This included 246 working age adult in-patients, including specialist eating disorder, perinatal, liaison, assertive outreach, continuing treatment and recovery, and forensic patients on 17 wards and community units; 131 old age patients on 4 wards and 3 community units, and 23 people with learning disability on 2 wards and 3 community units. The survey took place before the deprivation of liberty safeguards were implemented.
We first contacted the manager of each ward and unit and asked for the names of those current in-patients who were informally admitted. Individuals who were detained under the Mental Health Act were excluded from the study. We then contacted a senior psychiatrist (either consultant or higher trainee) or the ward manager. This person was asked to identify which of the informal patients did not have capacity to consent to their admission and care. They were all asked the same question, which we had written in advance: 'Does the person have a disturbance or disability of the mind which affects their consent to admission or any aspect of their treatment?' Those who had capacity were then excluded. We were therefore left with a final sample of 55 informal incapacitous individuals who were the focus of our study. This was because the deprivation of liberty safeguards apply only to this informal incapacitous group.
We identified significant factors from the cases cited by the Code of Practice that may indicate deprivation or restriction of liberty, and which were most likely to occur in an in-patient setting. These were formed into survey  Benign force (i.e. not being used to overcome resistance) used to take a confused patient to hospital Subterfuge being used to ensure the patient's cooperation in coming to the hospital The journey to the hospital is exceptionally long or otherwise onerous for the patient The decision to admit is opposed by relatives or carers who live with the patient Force or a locked door is used to prevent the patient from leaving the hospital in a situation where the patient is making a persistent and/or purposeful attempt to leave Sedation being used to prevent the patient from making an attempt to leave The patient being treated in a locked ward The design of door handles or key pads makes it difficult for confused patients to leave Staff bringing a wandering patient back to the ward Force being used in a non-emergency situation to ensure that a resisting patient receives necessary treatment for their mental disorder The use of benign force to feed, dress or provide medical treatment The use of restraint, medication or seclusion in an emergency situation in order to respond to a patient's disturbed, threatening or self-harming behaviour Attempting to persuade a confused patient to return, using benign force if necessary The hospital is denying a request by relatives for the patient to be discharged into their care A decision by the hospital to deny or severely restrict access to the patient by relatives or carers

Placing reasonable limitations on the visits of carers or relatives
The patient is denied freedom of association within the hospital, or otherwise being subject to a care regime that severely restricts autonomy The patient's access to the community is denied or severely restricted due to concerns about public safety A temporary refusal to let the patient leave in the absence of an escort, who is required to support the patient, not protect the public a. Adapted from Jones. 9 questions (see Appendix). These questions were put to a member of the nursing team, who knew the patient well, in an individual interview. As well as answering 'yes' or 'no' to each factor, the members of staff were given the opportunity to discuss the issues involved where there was any uncertainty. More details were requested if a potential deprivation of liberty factor was identified to ensure the correct decision was reached.

Analysis
Three of the authors (J.R., E.M. and T.S.) analysed the data. Agreement was reached between these authors as to whether a deprivation of liberty was occurring. Due to the lack of clarity about the definition of deprivation of liberty, we analysed the data using two different approaches -each designed to be within the spirit of the Code of Practice, but also to investigate whether differences in outcomes exist when a different approach to assessment is employed. In the absence of a clearer definition of deprivation of liberty, it is likely that most clinicians will adopt an approach to assessment between these two extremes.

Approach 1
This was designed to be highly sensitive to any potential deprivation of liberty. Individuals were recorded as being deprived of their liberty where there was one or more individual factor from the survey that has previously been associated with a deprivation of liberty.

Approach 2
This more specific method was designed to allow the assessor to 'balance' the likelihood of deprivation of liberty against the likelihood of restriction of liberty. Taking all these factors into account, we weighed up whether that person's care most closely matched the overall clinical description of someone who is deprived of their liberty -or merely restricted in their liberty.

Results
All 400 in-patients were included in the survey. The survey and its analysis is summarised in Fig. 1. In the vast majority of units and wards for working age adults, there were no individuals who were informal and incapacitous. All nine learning disability in-patients in the assessment and treatment unit were detained under the Mental Health Act. Table 2 summarises the results for those services that did have informal incapacitous patients (n = 55), and which are therefore the focus of this study.

Approach 1
According to approach 1, 46 in-patients were deprived of their liberty as they had at least one deprivation of liberty safeguards factor, and some were identified as having several. The most common factors related to discharge planning. In 39 cases (85% of the 46 deprivation of liberty cases) the staff team would prevent that person from taking their own discharge, in 15 (33%) from changing the place that they are discharged to, and in 11 cases (24%) a carer's request for discharge would be refused. Restrictions on the ward itself were less prevalent. In 12 cases (26%) there was regular physical restraint, and 6 patients (13%) were never allowed temporary leave. Restriction on making contact with the outside world was present in only two cases (4%both involving restricting the use of the ward telephone, due to continual calls to either family members or the police). In two further cases (4%), visits from family members were restricted on the basis that the treating team did not feel that family members were acting in the best interests of the person. All of these cases were within old age services.
Within the old age setting, ten (21%) informal and incapacitous individuals were identified as being unable to participate in the Care Programme Approach (CPA) owing to their significant communication difficulties. This figure was six or seven in the learning disability setting. Although involvement of family and carers was frequent, occurring in 81% of population of older people and in 100% of the learning disability sample, there were no cases where independent advocacy had been involved.

Approach 2
When the same data were analysed using a system that weighed up the factors that were in keeping with restriction or deprivation of liberty, the prevalence of deprivation of liberty decreased -to only six patients (3% of all inpatients, 6% of informal patients, 11% of informal incapacitous in-patients).
The six patients detected as possible deprivation of liberty were all from old age psychiatry settings; five in community units and one in a ward. Five people scored positively for possible deprivation as a result of force (physical restraint) being used regularly in a non-emergency situation. The frequency and intensity of this restraint, often requiring several staff members and occurring several times weekly, was felt to be too high to be classed as mere restriction and in almost all cases the individual was distressed by this restraint. For the remaining patient, access to the community was prevented because of concerns about safety both to the patient and others. This individual was also nursed on continuous observations. All six patients also scored positively for several restriction factors (between three and five restriction factors for each patient). Our survey did not specifically include questions about whether a patient would meet the criteria for detention under the Mental Health Act. However, at least one of these six deprivation of liberty patients (who had a diagnosis of schizophrenia) was identified by staff as possibly meeting Mental Health Act criteria.

Discussion
Despite the introduction of deprivation of liberty safeguards in April 2009, there is a lack of clarity about how deprivation of liberty is defined. Although the factors that may lead to a deprivation of liberty are clear, this survey has shown that two different methods of weighing up these factors produced remarkably different results. As well as the confusion over the definition, there is also some dispute as to whether the introduction of the deprivation of liberty safeguards was needed at all. 11 As 'the distinction between a deprivation of, and restriction upon, liberty is merely one of degree or intensity and not one of nature or substance', 8 this judgement is a finely balanced one, and will need to be carried out by professionals with a high level of training and experience in this field. The cost of providing appropriate training and support for those charged with implementing this new legislation will clearly be significant -as will the cost of the assessments themselves. The government estimates that the deprivation of liberty safeguards will cost £13.9 million in the first year. 12 When assessing the potential impact of the deprivation of liberty safeguards, the government stated that it 'maintains the view that the court judgments to date indicate that a particular factor, or combination of factors, do not inevitably constitute a deprivation of liberty'. 12 Using this approach they estimate that 10% of 'the relevant population' will be subject to a deprivation of liberty safeguards assessment, of whom 25% will have an authorisation granted. 12 It is striking that our second system of assessment, which mirrors the approach that the government has followed, found a similar prevalence of deprivation of liberty to the government estimate (11% of informal incapacitous patients -the 'relevant population' in this situation). However, our survey did not extend to registered care homes or medical hospitals, where it is reasonable to assume that the prevalence of both incapacity and deprivation of liberty will vary.
Our first method of analysis has also shown that staff can identify factors associated with deprivation of liberty in 84% of informal incapacitous patients. Although it is unlikely that this high figure represents the true prevalence of deprivation of liberty, it may suggest that the number of deprivation of liberty safeguard assessments will be higher than has been predicted, especially during the first year. This also demonstrates how specific the UK government's interpretation of the case law has been. If this approach is questioned by the courts, the prevalence of deprivation of liberty could be much higher. This is entirely possible, as the 'threshold' for deprivation of liberty varies even within the case law itself. 13 Our study has a number of limitations. Clearly, we have relied upon the judgements of a large number of medical and nursing staff who, despite our best efforts, may not know the patient well or have fully assessed their capacity. These assessments are by their very nature subjective, and even though we were able to reach a consensus in each case, our conclusions are open to challenge. As the decisions were reached by consensus, we were not able to assess interrater reliability. However, this approach may also be seen as a strength of the study, as it closely mirrors the team discussions that most clinicians will have when assessing deprivation of liberty.
So what advice can we give to clinicians who suspect that they are caring for individuals who are being deprived of their liberty? First, the Code of Practice makes it clear that patients must be cared for in the least restrictive way possible. We have identified several frequently occurring deprivation of liberty factors that are easily addressed to reduce the restriction of patients. Deciding as a team how a request for discharge will be managed will avoid any suggestion that the individual is subject to de facto detention; the use of independent mental capacity advocates will open care planning to a degree of independent scrutiny; and in some cases it may be possible to make more arrangements for temporary leave from the ward -perhaps with an escort. We have identified potential deprivation of liberty cases where the patient also appeared to meet the  Approach 2 0 (0) 0 (0) 1 (2) 5 (9) 6 (11) a. Percentages are as a percentage of the total number of informal incapacitous patients and are rounded to the nearest whole number.
criteria for detention under the Mental Health Act. The case of GJ 14 (which was heard following the data collection) suggests that where the purpose of the deprivation is mental health treatment, the Mental Health Act should be used. The Act also confers safeguards such as the supervision of medication for mental disorder and established procedures for appeal. Finally, if a deprivation of liberty is suspected, we have shown that it is relatively easy for a group of professionals to reach a consensus decision as to whether a deprivation of liberty is taking place. A meeting of all those involved in the care of the individual, with appropriate advice from the local deprivation of liberty coordinator if needed, will in the vast majority of cases, be able to allay the anxieties of those involved. Clinicians are understandably anxious. They are being asked to apply a legal concept that has emerged from ad hoc case law into real-life clinical situations. Our study has shown how difficult it is to clearly define deprivation of liberty, and demonstrated that further case law could lead to a dramatic change in deprivation of liberty prevalence. Despite this, we have also identified simple steps that can reduce the prevalence of deprivation of liberty, and shown that professionals are usually able to reach a consensus as to whether a deprivation of liberty is taking place.