1. DoLS v Guardianship: Applications, criteria and powers

Having written my thank you card to blog readers, I thought I’d kick of 2012 with a long-promised post on the deprivation of liberty safeguards (DoLS) and Guardianship. In true Small Places style, this will be not be a short one... in fact, it’s such a monster that I’ve decided to break things up into three separate posts. The first (this one) is on the criteria and powers of DoLS and guardianship; the second is on the appeal mechanism.  These are very long, and if you're pressed for time I recommend you skip to the third and final post which provides a summary and a general discussion of the issues.  Even so, I have a horrible feeling this might be one of those posts I spend hours on, and about 3 geeks like me actually read... Still, all fodder for the thesis, and this way I can kid myself I’m working.  I'm very aware that I'm much more familiar with DoLS than I am guardianship, so please alert me to any glaring errors, and I'd also welcome any guest posts or comments on people's views on these issues.

The question: Would guardianship have been better than the DoLS? 

The DoLS, as readers will know, were introduced to close the ‘Bournewood gap’ – to provide safeguards against arbitrary detention for incapacitated adults in hospitals and community settings. The DoLS are only three years old in April, but they have had a troubled early life. From their inception, observers like Roger Hargreaves wondered early on if they were a ‘bureaucratic monster’, and recently declared them ‘not fit for purpose’ in a draft report for the Mental Health Alliance. Barrister Paul Bowen (who acted in the original Bournewood case) has called them: 

‘decidedly inelegant... the provisions so labyrinthine and bureaucratic that those responsible for administering them are likely to take every opportunity to avoid using them.’ (Paul Bowen, Barrister and author of Blackstone’s Guide to the Mental Health Act 2007)

And Richard Jones – author of ‘the bible’ on the Mental Capacity Act, has described them as: 

‘a procedure has been created that is hugely complex, voluminous, overly bureaucratic and difficult to understand, and yet provides mentally incapacitated people with minimum safeguards’ (Richard Jones, Solicitor and author of the Mental Capacity Act Manual)

It has been suggested – by Richard Jones, by the Mental Health Lawyers Association (MHLA) – that an amended form of guardianship under the Mental Health Act 1983 would have provided a better alternative. In their response to the legal aid consultation, the MHLA wrote: 

‘The DOL Safeguards are cumbersome and unpopular with nearly all those who have to deal with them. The only appeal avenue is to the Court of Protection at considerable expense. Often, the case has to be heard in London, a long way from where the person subject to the DOLS (and their family and the professionals concerned) is based. DOLS cases are mainly still heard before the senior judiciary, with the effect that counsel is often relied upon. The costs are often many times in excess of the costs of even the most complex tribunal. The Safeguards could be replaced by amendments to the Guardianship scheme under the MHA 1983, with the Tribunal service dealing with appeals rather than the Court.'

In a 2007 article for the Journal of Mental Health Law (no link sorry, paywalled), Richard Jones suggested that a person subject to guardianship powers was very likely to be deprived of their liberty: 

'Put bluntly, a person under guardianship can be forced to leave his or her home to go to a place where he or she does not want to go to, can be required to stay in that place and can be returned to that place if he or she leaves without being given permission to do so. Given the interpretation that the European Court of Human Rights and the High Court have given to the meaning of a deprivation of liberty, how can it possibly be argued that a person who is subject to the operation of such powers is not being deprived of his or her liberty? Such a person is clearly subject to the continuous control of the guardian and is not free to leave the specified place of residence. The fact that the provisions of the MHA that relate to guardianship do not specifically state that guardianship can be used to authorise the deprivation of a patient's liberty is not legally relevant to the question of whether the provisions can have that effect."

I’m not sure whether, following the ruling in Cheshire, he would still hold this to be the case; Jones reasons on the basis of JE v DE, a case where there was an alternative residence – but a person could object to their confinement and attempt to escape even if there is no existing alternative. This might be particularly the case for older people whose homes have been sold to pay for care home fees.  I think that’s a bit of a blind spot in the Cheshire ruling, but that’s another issue for another day. 

Jones regards the guardianship regime as preferable to the MCA process for detention (ie. what became the DoLS). But it is clear from a more recent ruling of the Court of Appeal that if the effect of Guardianship is to detain a person, then as things stand it cannot be regarded as satisfying the requirements of Article 5(4). In The Secretary of State for Justice v RB & Anor [2011] Arden LJ found that a tribunal cannot order a conditional discharge for a restricted patient (s73 MHA) on terms that would create a deprivation of liberty: the review mechanism was inadequate, the criteria not set out, parliament had not intended s73 to be used for this purpose, and it forsaw detention for the purpose of ‘containment’ rather than treatment, which was contrary to the policy of the MHA [44], [54], [57]. I think similar difficulties would hold for Guardianship, since it was never intended by parliament to be a mechanism for detention, and the government explicitly rejected such a use in the Bournewood Consultation report. But, it does seem possible that in the light of widespread criticisms and the spiralling costs of the DoLS (albeit costs that the Ministry of Justice seems reluctant to keep track of), the government may return to this issue and re-consider the use of guardianship. And so, in this post, I want to explore how DoLS and guardianship compare, and in what ways guardianship (and indeed the DoLS) might need to be amended to comply with Article 5(4) and offer robust safeguards against detention. I will look to at: criteria for detention; coercive powers; the appeal and review mechanisms; the role of family, friends and advocates; terminology and stigma; and monitoring and enforcement. 

Criteria for detention 

In their work on mental incapacity in the 1990’s, the Law Commission notes that historically guardianship was rarely used, even though it was the only framework providing real authority for personal welfare decisions for people who lacked capacity. The already minimal use of guardianship dwindled further after it was reformed under the Mental Health Act 1983 (as amended by the Mental Health Act 2007, henceforth MHA), which reduced the eligibility criteria, and in particular excluded from scope adults with learning disabilities who are not ‘abnormally aggressive’ or ‘seriously irresponsible’. As Bartlett and Sandland put it:

‘...it means that the majority of persons with learning disabilities, for whom guardianship was designed, who are not normally aggressive nor irresponsible, are not eligible for guardianship’ (p489-90).

This problem was highlighted by several respondents to the government’s Bournewood Consultation, which canvassed views on whether guardianship could provide an appropriate solution to the ‘Bournewood gap’: The Law Society, the Healthcare Commission, Bevan Brittan (incidentally, I can’t link to these as I only have paper copies of these responses – requested from the Department of Health under the FOIA). 

By contrast, the DoLS are more flexible with respect to population: it does require that a person is suffering from mental disorder in the meaning of the Mental Health Act, but takes out the restrictions on learning disabilities (last year 14% of people detained under the DoLS had learning disabilities). And furthermore, whereas the criteria for guardianship are based on a person’s ‘status’ as mentally disordered, the DoLS have the additional ‘functional’ requirement that a person must lack capacity with respect to whether he should be accommodated in a particular place for the purposes of care or treatment. A person considered to have the capacity to decide about these matters could – in theory – still be coerced under guardianship. And so, if guardianship was to replace DoLS, I think it would have to be amended not only to remove the learning disabilities exclusions, but also to include a capacity criterion so it does not become an outmoded ‘status’ based tool of coercion. 

The key difference, in my view, comes down to the ‘welfare’ requirement under guardianship and the ‘best interests’ requirement under the DoLS. Aside from the mental health requirement, reception into guardianship requires that: 

...it is necessary, in the interests of the welfare of the patient or for the protection of other persons, that the patient should be so received.

Meanwhile, detention under the DoLS requires that: 

• it is in the best interests of the relevant person to be deprived of liberty 
• it is necessary for them to be deprived of liberty in order to prevent harm to themselves, and 
• deprivation of liberty is a proportionate response to the likelihood of the relevant person suffering harm and the seriousness of that harm. 

Although these criteria have broad similarities, it strikes me that the best interests requirement is potentially much more stringent than the ‘welfare’ requirement for guardianship. For a start, it explicitly requires the harm that would arise if a person wasn’t detained to be spelled out, whereas harm is only implicitly referred to under the welfare requirement. It carries an explicit requirement for ‘proportionality’, perhaps as a nod towards European Convention on Human Rights case law on Article 8. And most importantly, it engages the ‘best interests’ checklist procedure of the MCA. This requires that a range of issues - including the person's own past and present views are considered.  And also that, unless there are good reasons to the contrary, a wide range of people with an interest in P’s care and wellbeing must be consulted. It is entirely possible that similar reasoning might be applied under guardianship, or that those same people would in practice be consulted during an application for guardianship (I will come on to discuss the role of the Nearest Relative, below), but it doesn’t seem to me to be a clear legal requirement. DoLS also engages emerging ‘best interests’ case law which places a high value on ‘social’ as well as ‘medical’ and safety considerations (e.g. Munby LJ ‘What good is it making someone safer if it merely makes them miserable?’ in Re MM (2007) [120]). I’m not well placed to comment as I’m not so well read in guardianship case law, but it strikes me that in the context of the MHA ‘welfare’ may place greater emphasis on medical matters – I’d be interested in your views on this. Having said that, the tribunal panel is also composed of a lay member, and so perhaps they would inject greater ‘social’ reasoning to counterbalance any tendencies to medicalisation of criteria. 

The ‘no refusals’ requirement of the DoLS says that a person cannot be detained to a hospital or care home for the purpose of treatment prohibited by an advance directive or a decision of a deputy acting within their authority. This would presumably also apply to guardianship, since guardianship does not create powers to consent to medical treatment (on which more below). Under the DoLS, ‘If the donee of an LPA or deputy who has authority to determine where P will live objects to the proposed accommodation, the no refusals requirement is not met...’ (Bartlett, [4.77]; see Schedule A1 s20(1) MCA). It’s worth bearing in mind, though, that applications for deputyship can be costly, and are not very often granted, so this may not be as empowering to family members as at first appears. By contrast, guardianship ‘trumps’ the decision of a deputy: 

While the reception of a patient into guardianship does not affect the continued authority of an attorney or deputy appointed under the MCA, such attorneys and deputies will not be able to take decisions about where a guardianship patient is to reside, or take any other decisions which conflict with those of the guardian. [MHA 1983 code of practice, [2.26]]

Whether or not one feels this is a good thing is probably a matter of perspective. I’ll come onto discuss more issues around potential conflicts between families and professionals under the different regimes below. 

It is also worth noting in passing that guardianship can be applied to minors over the age of 16 (see s7(1) MHA 1983), whereas the DoLS have a minimum age requirement of 18. Greater flexibility over age could help remedy some of the issues that have started to emerge about older children who may be detained unlawfully in children’s homes (e.g. Re RK), residential schools (e.g. A Local Authority v C) and even care homes (e.g. MIG & MEG). 

Powers

The effect of an accepted guardianship application is to: 

...confer on the authority or person named in the application as guardian, to the exclusion of any other person—

(a) the power to require the patient to reside at a place specified by the authority or person named as guardian;

(b) the power to require the patient to attend at places and times so specified for the purpose of medical treatment, occupation, education or training;

(c) the power to require access to the patient to be given, at any place where the patient is residing, to any registered medical practitioner, approved mental health professional or other person so specified. (s8(1) MHA 1983)

I have to confess, I’ve never really got the point of these powers, given that the MCA provides a framework for making substitute decisions over residence (and indeed, further authority that guardianship lacks regarding payment for residential services or entry into a tenancy if need be), and attendance for various activities. Furthermore, guardianship can require attendance for medical treatment, but (unlike under the MCA) cannot offer consent to it on a person’s behalf. Bartlett and Sandland write: 

‘...in practice, the paradoxical situation is that, to be effective, these ‘compulsory’ powers rely on the cooperation, or at least absence of positive resistance, of patients.’ (p489)

The language of guardianship is notably more ‘bossy’ sounding than the MCA, the patient is “required” to reside at a place, attend various activities, allow access to professionals. But the reality is that however ‘fluffy’ the language of the MCA in comparison, it still permits coercive acts.  However, it does so in a much more nuanced fashion, allowing for capacity and best interests to be determined separately for each specific issue.  The key difference, in my view, is that guardianship has greater safeguards than the main provisions of the MCA, which may be why it is still used by some local authorities who want to offer people who are not detained greater legal protection.  Following Cheshire, that might be a very wise move.

The MHA code of practice says that in general a move to residential care can be effected using s5MCA or the DoLS, but guardianship ‘may still be appropriate’ if: 

• ‘there are other reasons – unconnected to the move to residential care – to think that the patient might benefit from the attention and authority of a guardian; 
• there is a particular need to have explicit statutory authority for the patient to be returned to the place where the patient is to live should they go absent; 
• or it is thought to be important that decisions about where the patient is to live are placed in the hands of a single person or authority – for example, where there have been long-running or particularly difficult disputes about where the person should live’ [MHA code of practice [26.10-12]]. 

If there are long running disputes about where an incapacitated person should live, I would have thought the appropriate channel was an application to the Court of Protection, not the invocation of guardianship.  In fact, the invocation of guardianship is a bit sneaky, as the MHA trumps the Court of Protection's powers.  However, guardianship does invest the nearest relative with considerable powers, which I'll come on to.

As Jones points out in his 2007 article, there are no explicit powers for conveying under the DoLS. But the Court of Protection has now twice confirmed that a standard authorisation implicitly creates a power to coercively return a person to a care home or hospital they are detained to (Re P (Scope of Schedule A1) (2010), DCC v KH (2009)). A person under guardianship, who is absent without leave from the place they are required to live at, may be forcibly returned by a social worker (or anyone authorised in writing by them) or police officer under s18(3) MHA. This power looks a bit odd in the light of DoLS, and I think consideration would need to be given to whether coercively returning a person to a place against their objections engaged Article 5; the present state of the case law makes this rather tricky. Cases like Storck v Germany strongly suggest it would engage Article 5, as do the ‘objections’ issues discussed in P & Q (2011), but following Cheshire I am unclear whether a person would be considered detained by this if they were equally likely to be coercively returned to any place they resided at.

It’s also worth noting that under the DoLS, authority to detain can only be given to care homes or hospitals, whereas under guardianship there is no reason a person could not be required to live in a different setting – for example supported living, or even the family home. This locational flexibility places guardianship at a distinct advantage over DoLS, which have run into serious difficulties around supported living, and potentially other settings like residential schools. Nothing under guardianship prevents a person being admitted to hospital in the normal way, and it is compatible with being detained to hospital under the DoLS or s2 MHA (but not s3 MHA).  If guardianship were amended to permit detention, it could presumably be used in the same way as DoLS to permit detention in hospital for treatments ineligible for the MHA.

The role of family under the different regimes 

Depending on one’s perspective, one of the strengths of guardianship in relation to DoLS may be that it offers far greater power to family members. Under the DoLS the detainee is appointed a representative, and according to regulations this should be chosen by the person themselves if they have capacity, then if not by their deputy or LPA, and then – in the last resort – by the best interests assessor. As the Mental Health Alliance have noted, the possibility for the supervisory body to select a representative who does not oppose the detention is rife with potential for conflicts of interest. There is very little monitoring or accessible means of challenge to ensure that P is allowed to choose their representative when capable, and there is very little to stop the best interests assessor choosing a more compliant representative than a more appropriate or obvious candidate. This compares poorly with the selection of the comparable person – the ‘nearest relative’ – under the MHA. The nearest relative is automatically selected from a list (s26 MHA), and the approved mental health professional (AMHP, roughly analogous with the ‘best interests assessor’ under the DoLS) can displace them by application to court. Presumably this is a guard against conflicts of interest in their appointment, and I am intrigued that the drafters of the DoLS did not consider this issue. However, one shortcoming of the MHA regime, of course, is that if the person themselves has capacity, they might end up in a situation which, to quote David Hewitt, their nearest isn’t their dearest. In such circumstances, it would be incumbent upon them (or the AMHP) to apply to the court to have them displaced, which rather raises the inverse problem that if their nearest relative supports the detention then they may find it hard to recruit support to exercise this right. One option might be to allow a person with capacity to select their nearest relative/representative, and then if they lack capacity to refer to a decision by a deputy, or finally list system if they have none.

In comparison with the representative under the DoLS, the nearest relative has considerable powers. The nearest relative can oppose an application for reception into guardianship, and they also have powers to discharge a person from guardianship simply by writing to the local authority. However, the AMHP can – as stated – apply to have the nearest relative removed from that position by the court, if they feel they are likely to inappropriately exercise these rights. In their response to the Bournewood Consultation, the Law Society noted that sometimes the threat of displacement may discourage a nearest relative from ‘acting vigorously on behalf of their family member’. One solution to this issue may be to support the nearest relative with independent advocacy, and ensure they have access to good quality information about their rights.  I had difficulty finding equivalent guidance for nearest relatives as the Department of Health supplies for representatives for the DoLS, but perhaps there is some out there (certainly I don't think they should have to cough up for a handbook; perhaps the detaining authorities could buy copies for them?!). 

In contrast, the representative under the DoLS has no right to oppose the detention (unless they are also the donee of an LPA, or their deputy), nor of discharge; their only right is to request review (which can be refused) or to appeal to the Court of Protection under s21A MCA. The weakness of this position will be discussed more in Part 2, but for now it’s worth noting that the balance of power between family representative and professionals is entirely shifted with a switch from DoLS to guardianship. Whereas in the DoLS the representative has to make the case that a person should not be detained, and in any case it is relatively easy to avoid giving significant powers to any oppositional family member, under guardianship, where the nearest relative opposes the detention, the shoe is on the other foot. To put this in perspective, had Steven Neary been detained under a guardianship-like regime, his father could have opposed his placement in the care home from the start, or discharged him by merely writing a letter to Hillingdon Council. Had the council wanted to oppose this, it would require them to refer the case to court to displace his father as the nearest relative, and as history tells us, it is unlikely they would have succeeded. 

The roles of different professionals under the different regimes 

Under the MHA, an application for reception into guardianship must be approved by two medical professionals. Usually the application is made by an AMHP working for a local authority social services team, which means that a professional who is likely to have a social care background is involved in the process. Although AMHP’s are making the application on behalf of the local authority, the code of practice is careful to stress that ‘the decision to make an application lies with the AMHP personally’ [8.14]. They are expected to be independent, and as the instigator of the application arguably have more power than the medical professionals – who merely exercise a right of veto. However, in the responses to the Bournewood Consultation, law firm Bevan Brittan argued that it might be better if one of the professionals who must approve applications for reception into guardianship had a social care background. I would probably support this view, as I think that the expertise of social care professionals is very relevant to community based forms of detention. 

Under the DoLS, an assessment must be made by both a doctor (the mental health assessor) and a best interests assessor. The requirements for the best interests assessor are somewhat vague; they must not be related to P, must be independent from his care planning, and must have no conflict of interest. The code of practice says the supervisory body ‘must be satisfied in each case that the assessors have the skills, experience, qualifications and training required by regulations to perform the function effectively’ [4.15]. Although best interests assessors often are social workers and AMHP’s, there is no legal requirement for them to be. This means that the degree of experience and training in legal frameworks for detention is potentially much lower among best interests assessors than AMHPs involved in guardianship. Given the requirement that the best interests assessor also explores the complex issue of the MHA/MCA interface in the eligibility assessment, a background as an AMHP could be very useful. If I were reforming DoLS, I would almost certainly make it a requirement that best interests assessors were AMHP’s, and if I were in charge of AMHP’s training I would probably make it a requirement that they learn about DoLS so there is a more standardised level of expertise than ‘appropriate training’ among assessors.

I now want to go on and discuss appeal and review mechanisms under each regime in Part 2.