• Home
• About
• Login
• Register
• Search
• Current
• Archives
• Announcements
• Pre-2013 Index

Reforming the Deprivation of Liberty Safeguards (DOLS): What Is It Exactly that We Want?

Peter Bartlett [1]

Cite as: Bartlett, P., "Reforming the Deprivation of Liberty Safeguards (DOLS): What Is It Exactly that We Want?", (2014) 20(3) Web JCLI.

Abstract

The Deprivation of Liberty Safeguards (DOLS) are the rules by which people who lack mental capacity can be deprived of liberty, most commonly in care homes or hospitals but also in supported accommodation and in the community more broadly. The present DOLS are not viewed as a success, and the Law Commission has a mandate reconsider a the legal framework for them. This paper is to start a debate as to what those revisions might look like.

The paper argues that the current focus on deprivation of liberty as contained in Article 5 of the ECHR is insufficient, both because the law is moving beyond this, and also because it has resonance with neither the professional carers who are likely to be the primary administrators of the new system nor the people perceived to lack capacity and their families. Our starting point should instead be to ask what will have a practical and beneficial effect on the rights and experiences of the individuals concerned.

The paper further considers the advantages and disadvantages of a system restricted to people lacking mental capacity. It makes tentative proposals as to administrative systems which might be used as a basis for discussion, and proposes that enhanced use of advocates may be a useful approach.

1. Introduction

In 2005, the European Court of Human Rights held in HL v the United Kingdom [2] that a man who lacked capacity to decide whether to be admitted to hospital and who was admitted as an informal patient to a psychiatric hospital had been deprived of his liberty in violation of Article 5 of the European Convention for the Protection of Human Rights and Fundamental Freedoms (ECHR). The English and Welsh regulatory system was held to be insufficiently robust to meet the requirements of that article to justify such detention, and the United Kingdom was therefore found in breach of its convention obligations.

In response, the government passed the so-called 'deprivation of liberty safeguards' (or 'DOLS') to regulate the deprivation of liberty of adults lacking capacity in hospitals, care homes, and other settings. [3] While these were amendments to the Mental Capacity Act 2005 (hereinafter the MCA), they were passed as part of the Mental Health Act 2007. The broad structure of these amendments was not complex. Adults lacking capacity could only be deprived of liberty other than in care homes or hospitals pursuant to a court order. Adults lacking capacity and accommodated in care homes or hospitals could be deprived of liberty pursuant to an administrative system: two independent assessors (one a doctor with psychiatric expertise, the other a social worker or similar professional) would need to certify that the individual was indeed deprived of liberty and lacked capacity, and that detention was in the individual's best interests (defined consistently with the remainder of the MCA). An attempt was made to draw a line between DOLS and Mental Health Act detention for people admitted to psychiatric facilities, and a few and broadly sensible, generally minor, limits were introduced (eg., the DOLS could not be used to hold a person for the purpose of providing treatment that the individual had refused, and could therefore not be given). An individual, generally a family member, (a 'relevant person's representative' or RPR) was to be identified to keep an eye on things, and an independent mental capacity advocate (IMCA) was to be available to assist the person lacking capacity and the RPR.

The Deprivation of Liberty Safeguards ('DOLS') have always been subject to criticism. While the previous paragraph indicates their overall direction, the detail is excessive and the drafting often borders on the incomprehensible, even to lawyers experienced in statutory interpretation. The relevant schedules take as much space in the statute book as the main body of the MCA itself. The forms required for each admission are voluminous, running to 172 pages when downloaded as a block. The form certifying that the person deprived of liberty (following the main body of the MCA, hereinafter referred to as 'P') [4] is over the age of eighteen years itself runs to three pages. The DOLS have made little impact in practice: on 31 March 2012, only 1667 people were subject to the regime - 1421 in care homes and 246 in hospitals. [5] The Report of the MCA Post-Legislative Scrutiny Committee of the House of Lords in 2014 recommended that they be replaced entirely with a new system, following full consultation and full parliamentary scrutiny. [6]

Shortly after the House of Lords made its report, the Supreme Court gave its judgment in P v Cheshire West and Chester Council, [7] which interpreted the scope of deprivation of liberty more broadly than previous domestic jurisprudence had. While the case is compelling in its logic and interpretation of the relevant jurisprudence, even the leading judge for the majority has acknowledged that the practical consequences are 'alarming': 'It means that a great many elderly and mentally disabled people, wherever they are living, must have the benefit of safeguards and reviews, to ensure that their living arrangements are indeed in their best interests.' [8] The result of the case was a vast increase in the number of DOLS applications: in the first quarter following the judgment, 21,563 DOLS applications were made under the administrative process (ie., excluding court applications relating to people outside care homes and hospitals), as compared to 12,414 for the entire previous year. The quarter further showed an increasing number of applications: 4506 were made in April, 7477 in May, and 9583 in June, [9] suggesting that the first quarter results may under-represent the effect of the judgment.

The government has responded by asking the Law Commission to consider the legal framework of the current safeguards regarding deprivation of liberty (both for people in care homes and hospitals and for people deprived of liberty elsewhere - most notably in supported accommodation, but also potentially in private homes and similar accommodation). A consultation paper is expected to be issued from the Law Commission in the summer of 2015. This paper is intended to start a conversation as to what a reformed system could or should look like. While it indicates some possible directions and approaches, deliberately no firm new system is proposed. That is in part because one of the key claims in the paper is that consultation really does matter in this area; and in part because we are at the beginning of a process of reform, not the end, and posing broader questions is therefore likely to be more use than advocating for a position.

2. Framing Questions

2.1 The Problem of HL and Article 5

What is it, exactly, that we want? That is a question that does not seem to have been adequately addressed in the initial DOLS consultations: those were all about developing a response to HL, a specific legal case. No doubt buttressed by the technical legal wording of the resulting legislation, the result is that care providers, insofar as they think about DOLS at all, seem to perceive it as a technical legal problem, with a technical legal solution - not something that can actually benefit P in any tangible way. The legal discussions concerning deprivation of liberty do not assist this: whether you can be deprived of liberty if you are asleep throughout the alleged deprivation, [10] or whether close accommodation of a person with dementia is or is not like confining a wife until she is prepared to restore conjugal relations with her husband [11] may be fascinating for legal academics; they do not chime with the experience of the care providers who are the current front line administrators of the DOLS, and are likely to administer any post-DOLS system. Further, as the House of Lords Committee noted, the term itself is alienating: to service providers, 'depriving of liberty' sounds too much like offering bad care, [12] and service providers want little to do with it as a result. Without developing a much more positive shared understanding among all stakeholders (including a wide array of professional carers, family carers, and of course service users) it is difficult to see how a proposed system will obtain the required buy-in to make it work. We can force people to fill in forms, but it is very difficult to force people to fill in forms carefully, unless they see a potential benefit.

To focus on article 5 and HL specifically in determining the way forward is problematic on its own terms. The Strasbourg jurisprudence is moving well beyond Article 5 in its engagement with the rights of people perceived as lacking capacity. Recent litigation on Articles 3 and 8 may be suggesting that Strasbourg is becoming more interventionist regarding the provision of medical treatment without consent. [13] It is not yet entirely clear how jurisprudence will affect people lacking capacity specifically, but it seems likely that Strasbourg will be exploring these matters in the foreseeable future. We are further starting to see Strasbourg taking an interest in conditions of care. [14] Article 8 (right to privacy and family life) has long been an issue regarding rights to information and rights to correspond freely with those outside the institution. [15] It is no longer obvious that the sole ECHR issue is detention. As issues arise outside article 5, the English jurisprudence already makes it clear that the DOLS system as it stands cannot adequately respond. [16]

In all of this, we may be seeing an increased relevance and interplay between the work of the Strasbourg Court and the European Committee for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment (the CPT). Certainly, for example, the requirement in X v Finland that involuntary psychiatric treatment cannot flow directly from involuntary psychiatric admission but requires further procedural and substantive safeguards should not have come as a surprise, since that has been a part of the CPT standards since 1998. [17] The CPT standards are broad in their remit, however, and insofar as they are influencing the Court, it suggests a considerable scope for judicial involvement in the future. In that event, specific legal safeguards may be necessary for a great deal more than deprivation of liberty. The ECHR is a moving target, and it is moving at a considerable pace. Unless we intend to re-visit DOLS-like processes after each new Strasbourg case - surely an undesirable approach - we need to think about a process that can take account of the law not merely as it has been, but as it almost certainly will be. That in turn means expanding the analysis well beyond HL and Article 5.

Even an expanded understanding of the ECHR issues is insufficient as a starting point for analysis, in that it looks up the judicial chain to find a shopping list of matters of relevance. For human rights to be meaningful, they are not about technical compliance with jurisprudence; they are about changing life on the ground. An approach that starts with the ECHR diverts attention from the practical question of what sort of a system will benefit P, either in terms of enhancing P's rights or enhancing P's quality of life, as P actually experiences them. Those are surely the core objectives for the post-DOLS system that we want.

HL and Article 5 have become the tail that is wagging the dog. Let us change the conceptual starting point from 'how do we comply with Strasbourg' to 'clearly, there are times when care becomes sufficiently intrusive that safeguards are required; how do we define those situations, and what should the safeguards look like?'. The answer to this will depend in part on the context of the care: it is not obvious that the same logic will apply to accident and emergency wards as to people with dementia in care homes. It is also something on which broad consultation is required, both because (as noted above) it is a necessary prerequisite to getting stakeholders to engage with the system, and also because they may well have highly perceptive insights.

2.2 A Capacity-Based System?

The Law Commission's terms of reference expects the new procedures to remain under the MCA. [18] This approach warrants some consideration.

The advantage of this approach is administrative coherence. The MCA contains a good set of principles. These include a requirement that any intervention be the least restrictive alternative, and that a person is not to be considered incapable of making a decision merely because his or her decision is unwise (thus protecting the right of capable people to make their own choices). It contains a nuanced best interests test which, if properly applied, allows considerable weight to be placed on P's wishes, beliefs, values and feelings (now and in the past) and requires consultation with family carers, while keeping objective factors in the picture. It already covers restraint and, in its jurisprudence, a variety of other issues such as access to family and conditions of care that might well serve as triggers for the application of particular safeguards in a new post-DOLS system. In the view of the House of Lords committee, the MCA contained an ethos of empowerment of P; assuming that is what is wanted for the new system, there would be advantages in not reinventing the wheel.

Further, if the system is to work as intended, it needs to be remembered that those implementing it are unlikely to be legal professionals. The MCA, at least until the DOLS were introduced, was generally praised as a particularly clearly written piece of legislation, that people other than lawyers could readily understand. These people are expected to get to know and implement the MCA, quite apart from whatever replaces DOLS; there would thus be a logic to linking the new administrative structure to keep things simple for them, since complexity and multiplicity of design is likely to breed both error and a reluctance to use the system.

There are two sets of difficulties with using the MCA as a home for the new processes, however:

Problems relating to MCA Implementation

The first is that, as the House of Lords Committee found, implementation of the MCA has been at best mixed. This finding is supported not merely by the evidence submitted to the Committee, but by the academic literature. Capacity assessments appear not to be routinely undertaken. [19] When they are undertaken, there is some dispute as to whether the persons performing the assessments are suitably qualified or knowledgeable about the criteria of the MCA, or suitably matched to the disability of the person assessed. Information upon which the capacity assessment is based is not necessarily presented to P in a clear and appropriate way. [20] While there is evidence in some of the literature that the decision-specific nature of capacity assessment is increasingly understood, [21] it is not yet universal. [22]

These difficulties might be addressed through administrative change in the implementation of the MCA, but other difficulties identified in the literature are more difficult to address through administrative means. Interrater reliability - the likelihood that different assessors will reach the same conclusion - is problematic regarding capacity. Olumoroti conducted a vignette study based on a woman refusing a liver transplant. The responses from 284 psychiatrists and psychotherapists showed considerable variation in outcome: at the extremes, 80 per cent of the psychotherapists considered that the woman in the vignette lacked capacity; 36 per cent of old age psychiatrists reached that conclusion. [23] In a study by Raymont, 40 patients on general medical wards were assessed for treatment capacity by a psychologist and four psychiatrists, all of whom had experience in assessing capacity. While interrater reliability was rated as strong - mean kappa = .76 - this still indicates significant variation between assessors. Also of note is that in the seven cases where all five of these assessors agreed that P lacked capacity, 71.4 per cent of the clinical teams in charge of P's treatment thought that P had capacity, and 75 per cent of P's close relatives thought that P had capacity. [24] These are presumably the people who knew P best, so the uniform decision by the experts that P lacked capacity in these cases should not necessarily be viewed uncritically: in the language of social science, the problem here is not just reliability but also validity of the capacity assessments. Insofar as these findings suggest that a finding of incapacity is to a marked degree determined by who is assessing P, this suggests an unacceptable level of arbitrariness in using capacity as a threshold for the determination of human rights safeguards.

The tendency to conflate the determination of capacity with the outcome of P's decision, labelled by Williams and colleagues as the 'concertina effect' [25] also seems pervasive. People who agree with care providing authorities tend to be taken to have capacity. [26] Willner's study found that two thirds of the 71 hospital staff surveyed agreed that 'in assessing capacity, the outcome of the decision needs to be considered. [27] Concerns about risk appear to be of particular importance. This is acknowledged in the jurisprudence:

[I]n cases of vulnerable adults, there is a risk that all the professionals involved with treating and helping that person - including, of course, a judge in the Court of Protection - may feel drawn towards an outcome that is more protective of the adult and thus, in certain circumstances, fail to carry out an assessment of capacity that is detached and objective. [28]

Banner argues that the influence of the outcome of the decision is intrinsic to the assessment process: capacity determination is a normative judgment by the assessor, and therefore the reasonableness of the decision is unavoidably conjoined to the capacity determination. [29] The apparent pervasiveness of the concertina effect gives this claim some credibility; but it equally challenges the notion that the use of capacity as a concept protects autonomous decision-making.

The use of the 'best interests' test is similarly questionable. The House of Lords Committee was critical that family carers were often side-lined in the determination of best interests, notwithstanding the express legislative requirement that they be consulted. [30] While that finding is consistent with the evidence submitted to the Committee, the academic literature is oddly silent on these matters: the studies tend to involve interviews of professionals, rather than service users and their families. [31] Williams' study, which included some interviews with carers, is not encouraging:

[F]rom the small amount of information we had, we felt that carers' views definitely warrant further investigation. Five of the telephone interviews were about the point of view of a carer, and in all five cases, the family member had been effectively disempowered. The only reason some of these cases were successful were that they had been reported by people who had professional expertise to bring to their role as a family carer. The themes that started to emerge were bewilderment, stress and indeed anger. [32]

The academic data relating to the involvement of P in the best interests process is similarly problematic. It does seem that real attempts are made in at least many care sectors to elicit the views of P. [33] As Williams notes, it is much more difficult to determine the dynamics of the decision-making process, and whether P's involvement is actually to take P's views into consideration, or whether it is instead, for example, a mechanism to encourage P to agree with a decision already largely reached. There is also little research as to whether P's involvement is sought in a fashion appropriate to P's circumstances.

All of this suggests that the MCA is not currently a system working smoothly and as intended. Buckling a new DOLS-like structure on to it will need to take that into consideration, and serves as a reminder that the success of a new system would not be independent from a proper implementation strategy for the broader MCA itself. To this end, the House of Lords Committee recommended the establishment of an oversight body to 'own' the implementation of the MCA, and the government has responded by proposing an advisory board with representation from an array of stakeholders from within and outside government. [34] Further details have yet to be published, but the success of this approach is likely to be closely linked to the success of any new post-DOLS system based in the MCA.

While the implementation of the MCA as a whole would be relevant to the new system, it might be argued that the new DOLS-like system would be an added check to ensure that, whatever is happening with the MCA more broadly, it would be implemented properly for the matters identified by the new system. It is easy to imagine a relatively rudimentary system going a considerable distance in this regard. Certainly, it is in principle easy for a routine administrative system to check that the appropriate paperwork is in place: in the context of psychiatric detention, Mental Health Act managers have done this for years, and there is no reason that a similar system could not be introduced for post-DOLS. This can address some of the concerns above, such as ensuring that capacity assessments are done, and that the relevant procedural steps of best interests test are followed. Further, the fact that decisions will be routinely scrutinised may well increase the care taken in reaching and documenting those decisions by the original decision-maker(s), as no doubt occurs in the mental health context.

Such administrative review cannot solve all the problems identified above, however. The capacity assessors in Raymont's study were selected because of their knowledge and experience; no administrative process will address the difficulties above regarding their assessments. Similarly, the pressures on professionals relating to risk will not go away, and it is difficult to see how a relatively rudimentary system of administration can effectively identify and challenge decisions motivated by risk aversion: defensive practice is unlikely to be stated expressly on the forms.

Problems of Capacity More Broadly

The second problem of using the MCA as the legislative home of the new procedures is not restricted to the implementation of the MCA. Increasingly, the appropriateness of capacity as a concept is itself being called into question, at least when, as here, it is determined in whole or in part with reference to disability. These discussions have been pulled into focus by the negotiation and coming into force of the United Nations Convention on the Rights of Persons with Disabilities (CRPD), [35] a convention which the United Kingdom has both signed and ratified. Article 12 of the CRPD enjoins member states to move away from binary conceptions of capacity towards the increased use of supported decision-making. The idea is that one never entirely loses legal authority to make decisions; instead, supports for decision-making become ever stronger, as functional ability decreases.

This is an attractive notion, and may intersect nicely with our instincts about who needs support in a DOLS-like system. At the moment, there is in law a bright line dividing capacity from incapacity. Those with capacity are free to make their own choices, but need be offered no support in doing so; those without capacity are brought within the purview of a considerable legal edifice that contains both restrictions on freedoms and safeguards to autonomy. It is however difficult to see that this provides an appropriate criterion for provision of safeguards in, for example, care home admissions. These may well be complex decisions, based on personal factors specific to P, what care the local authority is prepared to provide (either at P's home, in supported accommodation, or in a care home), what care home is appropriate for P's needs and wishes, and, if P is paying, what P can afford. It is not credible to think that many people in the population, let alone those who just pass the threshold of capacity, can navigate these issues without support, and it is doubtful whether they will know where to look for that support. The Care Act 2014 goes some distance to assisting in this regard by requiring advocates to be provided to people who would 'experience substantial difficulty' in understanding, retaining, using or weighing information or communicating a decision - the capacity criteria in the MCA, but only where there is no family member or other non-professional carer to support the individual. Given the technical nature of the information in question, it is difficult to see that this creates a complete solution. Further, people may well understand themselves in positions of considerable power imbalance when dealing with professional carers and local authorities, both because those actors may already be providing significant care upon which P is already relying, and because P realises that he or she will need to rely on these actors in the future and will therefore be reliant on their good will. It may well be the case that there are significant numbers of marginally capable people in care homes (or in the wrong care home) who do not want to be there and through lack of support have no effective way to change that situation. In ethical terms, it is not clear that they are distinguishable from people in similar situations who lack capacity. If that is correct, it is hard to see the argument for depriving them of similar safeguards. Care home admissions are merely an example here: comparable arguments are likely to apply in other contexts where care is provided that is sufficiently intrusive to warrant safeguards being provided.

This argument is strengthened if the objective of the new system is understood to involve safeguarding. While 'abuse' is an emotive term, the sort of thing that a DOLS-like system is likely to be protecting against can be characterised in this way. No Secrets, the government's landmark guidance on safeguarding of adults, provides the following starting point for a definition of abuse: 'Abuse is a violation of an individual's human and civil rights by any other person or persons.' [36] While the Care Act 2014 itself does not define abuse beyond making clear that it includes financial abuse, [37] the guidance under that Act provides a number of examples, including physical abuse, sexual abuse, psychological abuse, exploitation, financial abuse, neglect and acts of omission, discrimination, and institutional abuse. This last is stated to include, by way of example, 'neglect and poor care practice within an institution or specific care setting like a hospital or care home', whether as an individual incident or as part of a course of conduct. [38] This suggests that a post-DOLS system may be at the centre of that safeguarding policy. Both No Secrets and the Care Act Guidance also make it clear that that abuse can arise from neglect or poor practice: [39] express ill-will directed towards P is not a requirement. Within the current safeguarding system, incapacity is not a requirement for an enquiry to be commenced. The Care Act 2014 requires a local authority to launch a safeguarding equiry when it has reasonable cause to suspect that an adult in its area

(a) has needs for care and support (whether or not the authority is meeting any of those needs),

(b) is experiencing, or is at risk of, abuse or neglect, and

(c) as a result of those needs is unable to protect himself or herself against the abuse or neglect or the risk of it. [40]

Capacity is thus not relevant to the question of whether safeguards (in this case, an investigation) ought to be provided in a safeguarding context.

Capacity is relevant in safeguarding under the Care Act 2014 after the enquiry: if capable, P can refuse any care or support offered by the local authority, and if incapable, measures may be imposed consistent with the best interests test of the MCA. This raises an argument in favour of a continued role for capacity, is that it protects the right of people to make apparently inappropriate decisions. As discussed above, however, the bright line separating capacity and incapacity is not quite so bright as the statutory framework would suggest. As discussed above, it would seem that in practice the outcome of a choice is significant in determining whether P has capacity. Further, the continued use of the so-called 'inherent' jurisdiction, allowing intervention in circumstances where P has capacity but is perceived as unable to exercise a 'real' choice, similarly blurs the distinction between capable and incapable decision-making. [41]

It seems now clear that, at least in the view of the CRPD Committee, the United Nations body charged with the oversight of the CRPD, the use of capacity in the MCA (and, by extension, any post-DOLS framework that adopts a similar capacity criterion based on disability) will not be consistent with the Convention. [42] Whatever the merits of that approach - and they are many - there is no consensus on how to implement it in the near future in England, particularly concerning people with profound mental impairments. The perfect may nonetheless be the enemy of the good, and it may be worth serious consideration of moving away from a strict capacity-based framework in which capacity/incapacity is viewed as simply dichotomous and instead to one where supported decision-making has a much greater place.

2.3 Problems of Scale

The scale of the problems to be addressed by the new system will be vast. As of 31 March 2012, there were 13,134 residential care homes with 247,824 beds registered in England, and 4,672 nursing homes with 215,463 beds. [43] We are a country that has long emphasized the desirability of community-based care, and it is therefore likely that those remaining in care homes will often be among those with the highest needs and, correspondingly, the most intrusive care. On that basis, the 21,563 DOLS applications made through the administrative scheme in the first quarter of 2014-15 may yet be a considerable under-representation of the overall number of people deprived of liberty in these settings. The Association of Directors of Adult Social Services (ADASS) estimated near the beginning of that period that care home applications are likely to grow to roughly 94,000 in 2014-15; that estimate may prove closer to accurate. ADASS further estimated that the number of Court of Protection applications relating to people deprived of liberty outside hospitals and care homes would rise from 134 in 2013-14 to 18,633 in 2014-15, for a total of 112,533 overall. [44] As yet, the growth in Court of Protection applications has been more limited than this estimate would suggest: 47 in the quarter before P v Cheshire West was decided, 148 in the quarter immediately following the case, April to June 2014, and 163 between July and September 2014. [45] Whether they will grow significantly in the coming months remains to be seen. [46] A move away from deprivation of liberty and incapacity as the defining criteria to a broader set of entry points for safeguards would expand, not reduce, those numbers. By way of comparison of scale, there were 105,224 admissions to psychiatric facilities in 2012-13, of which 45,202 were detained under the Mental Health Act 1983. [47] The prison population stands at roughly 86,000. [48] However it is eventually defined, the post-DOLS régime is likely to be significantly larger than either of these.

That in turn raises the question of how safeguards should be structured in administrative terms. On the one hand, the system cannot rely simply on investigation of abuses reactively, as and when they come to light. Whether the system is restricted to people who lack capacity or is expanded to include some people who have capacity, it cannot be assumed that these people are in a position to undertake a meaningful investigation and appraisal of what their options are, and to press for their rights if they are unsatisfied with their present situation. These people presumably may suffer in silence, or risk being ignored if they do attempt to complain. While this is most obvious for those without close family or friends, even those with good social circles may be at risk. As noted above, many of the issues are of a technical nature (such as, for example, assessing care options and standards of care that a local authority may be required by law to provide) and it is not obvious that family carers will know what questions to ask, or what services to press for. By definition, the population at issue in a DOLS-like structure will be vulnerable, and the system needs to take that vulnerability into account.

That would suggest the need for an extensive and intrusive system of safeguards, with routine and close inspections of all those whose care meets whatever standard of intrusiveness is selected - deprivation of liberty, under the present system, or quite possibly some other criteria if the arguments above are accepted. Attractive though that may appear in principle, in practice such a system is open to question. It is not merely that such a system might be extremely expensive at a time of austerity, and thus raise political difficulties. It is also the risk that routine investigations can become rubber-stamping exercises, at least when the investigation is more than a purely mechanical exercise to ensure administrative completeness. When investigators consider applications merely because they are told they must, they may miss problematic and important issues that warrant closer investigation. [49] Many central European countries require court orders for any person detained in psychiatric facilities; it would seem that these routine procedures do not necessarily result in meaningful safeguards for the persons being detained. [50] The risk is that we end up with an expensive and intrusive system that does not actually provide tangible benefits to people on the ground - surely the worst of all possible outcomes.

Medical evidence in the current system is perhaps an example of this. Certainly, clear medical evidence of a 'true' mental disorder is required for any care engaging an ECHR right. [51] While this must be established through 'objective medical expertise', [52] the Strasbourg court has never expressly required that it be provided by a consultant psychiatrist. Certainly, if the justification for the intrusive intervention giving rise to the safeguard is based on a mental disorder, such medical evidence must be present. This is not necessarily the case currently: it would seem that a majority of people with dementia in England, for example, have never been properly diagnosed. [53] It is further fair however to expect that people undergoing sufficiently intrusive care that safeguards may be warranted should be getting appropriate medical advice throughout that care quite apart from the DOLS process: people in care homes or supported housing, for example, should get proper GP and appropriate specialist care as a matter of routine. If that is being done properly and appropriate and up-to-date evidence of the mental disorder is in place from those sources, it is not obvious how much a different specialist medical view will bring to the process, at least in cases where the diagnosis or similar medical matters are not contested. The current DOLS system requires a fresh report from a psychiatric expert for each DOLS application, quite apart from the general medical care received by P. There has been no study of these, so it is not known whether they have an effect on outcomes or whether overly routinized examination leads to overly formulaic responses. If the latter is the case, it might well be the case that the appropriate way forward is to insist on proper general medical care on an ongoing basis rather than part of the specific safeguards, and to focus the medical investigations within the safeguards system on cases where medical issues such as diagnosis are actually contested, and where the investigation by the medical expert can be correspondingly more focussed.

This is arguably true about the whole of the system, not just medical evidence: most resources should be focussed on cases (and aspects of cases) that give rise to real matters of contention, for reasons of practicality and resource management and also to protect standards of decision-making. While such selectivity may be thus desirable, it also has its complications. Unless the selection is based on simple and unambiguous criteria, there is a significant risk of error in its application. Further, the process of selection itself is an administrative process that must be staffed: it has an economic cost. The more complex the criteria, the greater the time that will be necessary to administer them, and the greater the economic cost of the process of selection.

The criteria recently published by the Court of Protection intended to determine which of its DOLS cases require oral hearings show some of these difficulties. [54] The Court provides a list of fourteen elements that must be addressed in an application for approval of a deprivation of liberty. It then determines which elements of that list must be contested, which not complied with, and which raising concerns for an oral hearing generally to be required. An oral hearing is further to be required if 'any objection' is made by P, or where the court thinks an oral hearing is necessary or appropriate. The selection process is thus complex, and will take a significant amount of time. As it is to be performed by a judge of the Court of Protection, the economic cost of this selection will be significant. The criteria further contain elements of flexibility and ambiguity: how much of an objection is required from P before an oral hearing will be required, for example, or how big a 'concern' warrants an oral hearing? There are also evidential questions: how will a judge be able to evaluate, based on the papers alone, the seriousness of P's objection? It is not generally going to be possible for the judge doing the selecting to look behind the papers or to test the evidence presented, raising questions of how for example excessive or inappropriate intervention flowing from administrative apathy will be identified: the applicant is unlikely to emphasise this in the papers. This sort of problem is likely to arise in any complex system of selection where evaluation of the sufficiency of evidence is required, but it does serve as a reminder that selection comes with its own difficulties.

Safeguards in the post-DOLS context should not be confused with administrative audits, inspections and other administrative forms that are designed to ensure overall standards of care, and serious consideration will need to be given as to the delineation of roles between the post-DOLS structures and the quality control mechanisms. Unlike mechanisms to ensure standards of care overall, safeguards are to protect each individual. In that sense, there is a comprehensive side to them, and the selectivity in question cannot properly be attained by, for example, ensuring that a random sample of people in places where intrusive care is likely to be practiced are in fact dealt with appropriately. That said, the post-DOLS regime will exist in an administrative universe that also contains a wide array of quality control structures. If in environments such as care homes the post-DOLS structures affect the number of people noted above, they may well come to feel like general quality control mechanisms, simply because they may apply so pervasively to so many people.

While they are different, it is difficult to see that the safeguards can be entirely independent from the quality control mechanisms. Insofar as the safeguards take documents at face value, they will be relying on the quality controls to ensure the quality and probity of the documents; and the quality control mechanisms will presumably be the primary way to ensure that the outcomes of the safeguards are enforced in practice. This is probably desirable: there is no obvious reason to establish parallel quality control procedures as part of the safeguarding process to procedures already in place elsewhere. It does mean however that the efficacy of the safeguarding procedures will be dependent on the efficacy of the quality control mechanisms, and some providers of those quality control mechanisms have been severely criticised in recent years.

Almost certainly, dichotomous decision-making will be unsatisfactory in a post-DOLS system. To use deprivation of liberty itself as an example, it seems unlikely that many of the people currently deprived of liberty in care homes are in a position simply to leave the home on their own. If leaving is possible at all, it is likely to be possible only with considerable support in the community. If leaving is not possible, changes may still be able to be made within the care home that may significantly improve P's experience. This is something the existing DOLS system gets right, at least on paper: the DOLS assessors do not simply approve or reject applications without amendment, but can require that changes to the terms of the accommodation be made that would benefit P. There appears to be no study on how often this occurs, but it does seem a vital element of the process.

3. Ways forward

3.1 Mapping the terrain

Deprivations of liberty of people lacking capacity occur in a wide variety of contexts, and the existing DOLS do not make sense in all of those contexts. As we move on to articulate what we want from a new system, one not necessarily based on deprivation of liberty as the sole triggering criteria for safeguards, this problem is unlikely to go away: administrative systems that are perfectly workable for care home admissions may not be remotely practical for accident and emergency departments of hospitals.

The present DOLS are a good example. It may well be appropriate to restrain to the point of depriving of liberty a person incapacitated by the after-effects of an anaesthetic administered as part of an operation. That person may well have recovered capacity and been discharged home before any DOLS assessors would arrive to implement the present safeguards, however. What exactly is the point of requiring an application for a DOLS authorization in these circumstances? It does not follow from this that no safeguards should apply; merely that the system of a central authority sending out assessors will not be practical, given the time constraints. In the example, it may make considerable sense to require the treatment team to discuss the after-effects of the anaesthetic with the patient prior to the operation, and for the patient to consent to the post-op restraint as he or she consents to the remainder of the operation. That would work for operations planned in advance; it would be more difficult to apply in accident and emergency departments where care is, almost by definition, not planned in advance.

If, following the above arguments, we are to move outside deprivation of liberty as the sole defining criterion for the provision of safeguards, mapping the terrain first requires some iteration of when care is sufficiently intrusive to warrant such safeguards. That itself may be a matter that warrants consultation, both because it is necessary to ensure appropriate ownership of the new system among stakeholders, and also because such stakeholders may have particularly informed and astute views as to where abuse occurs. It would therefore be premature to map the terrain here. The comments that follow are instead written with relatively long-term care in mind - care for a period that would allow relatively intensive administrative processes to be used, if we wish them to be. That does not change the overall puzzle, however, since relatively short-term care can be intrusive enough to warrant safeguards. This is true even if the current standard of deprivation of liberty is maintained, since the Strasbourg court has taken seriously that a deprivation of liberty might be as short as thirty minutes. [55]

3.2 Sketch of an Overall Approach

If the objective is to make a difference for P on the ground, a two-stage process is perhaps appropriate:

1. A procedure to ensure that P has benefitted from the provision of appropriate care planning to ensure that the care offered takes into account P's wishes and feelings and is as minimally intrusive as is practicable under the circumstances. This should as far as possible be linked to existing requirements, to limit the amount of new bureaucracy required for carers. By ensuring that this is done properly up front, it may be possible to reduce the number of cases that become problematic. That is beneficial both by reducing the overall administrative burden (more intensive safeguards are more burdensome) and to P by limiting as far as possible the inappropriate or unwelcome intrusion into his or her life.
2. More intensive safeguards in the event that care is particularly intrusive, however that is to be defined, or particularly controversial, or objected to by P.

To better examine the possibilities within this structure, let us consider what it might look like in the context of care home admissions.

Regarding the first stage, it would not in principle be difficult to require care homes to submit copies of key documents to a central authority for approval. The care plan would be the obvious document in question: they are supposed to exist for everyone in care homes anyway, and if that requirement is being followed, the additional burden to the care home of submitting them would be minimal. These would be expected to be drafted with a view to articulating not merely the objective care to be offered, but why the proposal reflects the least intrusive level or type of care is necessary, and what P thinks about it. This will already be good practice under the Care Act 2014; again, it is difficult to see that there would be much additional administrative burden to what will already be expected. Presumably such plans should also already include medical views as to diagnosis and the proposed care; again, making that compulsory seems to add little to the existing requirements.

Should this requirement be limited to a subset of care home admissions? Certainly, not all care home admissions are the same, and many will involve entirely amicable relations between competent individuals and the care home. If the individual is paying for his or her own care, he or she may indeed have practical levers as to the care that is offered. It is perhaps arguable that these people do not need the safeguards. On the other hand, this might be a case where the criteria of selection may create more problems than they solve. If the initial scrutiny process is relatively mechanical - are the required forms completed properly and vaguely convincingly - then the workload need not be oppressive. As an obvious comparator, the Office of the Public Guardian currently scrutinises approximately a quarter of a million LPOAs prior to registration every year [56] - a number roughly corresponding to total care home admissions. The administrative burden of that is certainly not intolerable, and it may well be that introducing some system of selection not merely leaves open the possibility of error, but is also more administratively burdensome.

The degree of scrutiny that can be expected under such a system is limited. If however it resulted in the people arranging care to take care planning seriously, and to take P's views, wishes, values and feelings seriously as they do so, it might provide a significant benefit to P. This is not necessarily the case at the present time. A recent study by the Care Quality Commission of care homes catering to people with dementia found that in 29 per cent there were deficiencies in the assessment of care needs, in 34 per cent variable or poor care regarding how care met the physical, mental health, emotional and social needs of P, and in 33 per cent bad practice in the involvement of P or of family carers. [57] Care homes that have internalised good practice already would see little change; care homes that have not would be pressured to do so. Anecdotally, the routine scrutiny of sectioning papers under the Mental Health Act 1983 does seem to create an environment where those in charge of admission complete the papers with additional care. That would be likely to be a beneficial result, if it is replicated more broadly.

If the plan envisaged care that should trigger a stage 2 process, the initial scrutiny could identify that, and if papers were on their face inadequate, it might identify that; but otherwise it would have to be beyond the role of this initial scrutiny to look behind the papers. Questions of completeness, accuracy or probity of the care plans would need to be dealt with through other mechanisms. This might be periodic and systematic audit by the CQC, or a response to a complaint or whistle-blowing, much as happens now with LPOAs.

This initial scrutiny is also incomplete as a gateway to safeguards. The safeguards will need to require that approval of care be renewed on a periodic basis. The current DOLS allow deprivation of liberty for up to one year. [58] If the new safeguards adopted a similar time frame, intrusive care might well arise during the period of an authorisation. Thus if intrusive restraint on routine occasions were a trigger to safeguards, that might arise unexpectedly during the authorisation. Clearly, some mechanism would be needed for such changes during the life of the authorisation to be identified and pursued.

Should an RPR - a person such as a family member who can keep an eye on things - be routinely appointed for all care home admissions, and perhaps be required to sign off on the care plan and to sign a document acknowledging his or her role? The argument against this is that it does intrude somewhat on P's privacy, particularly if P has capacity. That said, as noted above, P may well be in a situation that creates vulnerability through a power imbalance. There might well be distinct advantages to P to having a formal RPR in place, even when he or she has capacity. Further, there may be an issue of reduction of stigma if everyone gets an RPR, rather than only those whose disability is perceived to warrant one. Finally, the House of Lords Committee noted a potential conflict of interest in the current DOLS system, where the local authority appoints the RPR, essentially the authority appointing the person who is to keep an eye on the local authority. [59] This problem would be avoided if P were still competent and able to appoint his or her own RPR at the time of admission.

The second stage of the inquiry is more difficult to envisage, in part because as noted above the criteria for the provision of safeguards need to be determined through consultation. Presumably, some gateways to more stringent safeguards will be based on objective factors, such as restrictions to contact with family or friends, particularly intrusive medical treatments (eg., those with sedative effects), ongoing restraint, where a change of care home is proposed, or restrictions on coming and going from the care home. If we understand the post-DOLS system in the context of safeguarding (in the 'No Secrets' and Care Act sense, as discussed above), it may make sense to look to that literature for guidance on objective criteria. Others will be personal: does P object to the care, or (particularly if the MCA model is followed) would he or she so object if capable? As noted above, P may be in a position to object, and the mechanism to access a second stage safeguard to be triggered by family members, the RPR, an advocate, or perhaps anyone to raise concerns.

Quite what the safeguard would entail is, again, a matter for consultation. One possibility would be to rely on advocacy at least for initial stages, since it can provide a relatively inexpensive and flexible response by someone who should have the skills both to identify abuse and to identify appropriate procedures to address the problem. That might be to take no action: even if one of the objective factors that triggers a safeguard is identified and an investigation is warranted, it does not follow that something has gone wrong or that P's rights have been violated. Similarly, if a complaint is made (particularly by someone other than the RPR or an immediate family member of P), it does not follow that a court application is the appropriate response; but it may well be that someone removed from the situation should be called in to have a look at what is occurring. The advantage of using an IMCA or similar advocate in this way is that he or she should be well-versed in the law and relevant policy, and understand the context of the care. In the event that the advocate considered there to be a problem, the next steps could vary according to circumstance, from the provision of rights advice or the outlining of different options for care and reaching an informal resolution, to a request for a Court of Protection Visitor, to more formal legal measures if required.

Using advocacy as the entry level to the provision of safeguards thus has the advantage that it can be tailored to the needs of individual service users. It further engages with the existing provisions of both the Care Act 2014 and the MCA, and can be made to dovetail with the court system more generally. It would however require some re-thinking of how advocacy is structured. Advocates would need access to additional support - for example from independent medical experts, in the event that medical matters were in dispute. It would further require the assurance of certain standards in the provision of advocacy. This is consistent with the House of Lords Committee report, that proposed more systematic training and control of standards for advocates. [60] Some re-thinking might also be necessary as to how advocates are appointed. Currently, advocacy services are commissioned by the local authorities, and if they are in turn likely to be parties to much of the advocacy, this creates a potential conflict of interest. In the current MCA system, where in many cases provision of an IMCA to the individual may be discretionary on the part of the local authority, the conflict is even more stark.

Increased reliance on advocacy also requires clarification of the role of the advocate. Is the advocate to represent P's current wishes and feelings, or is the role of the advocate to serve as an impartial arbitrator in dealings with the care home management and the local authority? If P lacks capacity, is the role of the advocate to press for P's stated wishes and feelings, or to act according to the statutory best interests test in the MCA (and therefore, perhaps, in express contradiction of P's stated wishes and feelings)? There are clearly different roles, and any increased focus on advocacy within post-DOLS safeguards would need to work through these conflicts with care.

One approach to take account of this would be to divide the roles: the advocate acts on behalf of P, but the actual decision is taken by someone else. This is how the current DOLS system is intended to work: P often has an advocate, but the actual decision is made following the investigations of assessors appointed for the purpose. That model has potential advantages, but correspondingly increases the number of people involved in the process, and thus the cost.

The House of Lords Committee proposed that mediation be introduced in MCA matters, [61] and it might well warrant consideration for the post-DOLS system as well. Court processes are expensive and in the end inflexible; it is no doubt desirable to avoid them if good results can be reached through other means.

While it is at least arguable that advocates should have an expanded role in the system, it does not follow that they should be the only gateway to the Court or perhaps some other safeguards. Near relatives and the RPR, for example, should presumably also have some rights to ensure that appropriate standards are being met, or some additional options if they are unsatisfied with the direction taken by the advocate.

The relevant legal mechanisms also need to be able to deal with all the issues that may arise in one set of proceedings, and it is not yet obvious that this is the case in the current system. Certainly, the Court of Protection has become more adept at judges 'wearing two hats' so that proceedings relating to the MCA and other law can be heard together, but it remains somewhat awkward. For cases where the issue is between the MCA and the Mental Health Act 1983, it is likely to be impossible, since cases under the latter are by statute considered in a tribunal and cannot therefore be heard jointly with MCA matters. If, as may well be the case, many of the issues in a post-DOLS universe concern what a local authority will fund, it does seem essential for the relevant forum to be able to consider that matter for any new system to succeed.

4. Conclusion

If a post-DOLS system of safeguards is to be effective, we must be clearer as to what it is intended to achieve. For the reasons indicated herein, the problem should not be approached as a technical fix to the technical legal problem created by HL. Rather, the question to be asked what problems warrant provision of safeguards, and what processes will actually engage with those problems to make things better for P in his or her day-to-day life. That is the only way in which the required engagement with care providers will be obtained, and the only way in which the actual benefits will be obtained for P. Developing such a system will take time and considerable consultation. We are at the beginning of that conversation, not the end.

Certainly, at some point, it will be necessary to re-engage with the Strasbourg jurisprudence to ensure that a reasonable case can be made for compliance. This should not necessarily strike fear into our hearts. While the Strasbourg court clearly has an active and developing interest in matters related to mental disability, it is hard to imagine that they are unaware of the practical problems involved, and they too are likely to understand that human rights is at its most important when it is effective in real situations to improve peoples' lives. Certainly they will want standards and processes to be in place, and certainly court processes cannot ultimately be excluded; but there is nothing obvious in the Strasbourg jurisprudence that precludes other approaches to the relevant problems.

That suggests that the questions for discussion should at this time be broad: to make a real difference for P, when does care become sufficiently intrusive that safeguards are required, and what should the safeguards look like?