Costing the deprivation of liberty safeguards

Even though the number of applications are far lower than anticipated, even though the proportion of those appealing is far lower than anticipated, the cost of the DoLS is ballooning far beyond the predictions of the impact assessment.

One of the criticisms that are frequently levied against the deprivation of liberty safeguards (DoLS) is that they are enormously expensive.  However, there appears to be no official data on how much they are actually costing in contrast with predictions, and so it is hard to know whether this criticism is fair.  Certainly a variety of significant rulings on the scope of Article 5 (and hence the DoLS) have mentioned the wider resource implications of finding that a particular situation amounts to a deprivation of liberty (e.g. Re RK [2011] [8], [44], [45]; Cheshire West and Chester Council v P [2011] (on costs) [8]; P & Q v Surrey County Council [2011] [4] –[5]).  And many of these hearings on deprivation of liberty are currently being heard in the High Court; a function which they have been allocated no additional resources for.  It seems likely that this additional workload must have knock-on consequences for the ordinary family law work of the court, and I expect the judiciary must be feeling the strain.  Because of the dearth of official data on the cost of the DoLS, I have tried to put together some costings on the legal process.  And what I found is not pretty at all.  The headline is this: even though the number of applications are far lower than anticipated, even though the proportion of those appealing is far lower than anticipated, the cost of the DoLS is ballooning far beyond the predictions of the impact assessment.  Furthermore, whereas the impact assessment anticipated that the number of applications, authorisations and appeals would decrease after the first year, all the evidence is that applications and appeals - and their associated costs - will continue to rise.  The overall cost to the public purse is not being monitored centrally, but if these estimates are correct then it could be very significant indeed.

The cost of the assessment process at local level

Shah and colleagues have shown that the cost of the assessment process under the deprivation of liberty safeguards far exceeds the predictions of the impact assessment. They found that whereas the impact assessment catered for 20,000 assessments for DoLS (in England, another 1000 in Wales) at a cost of £600 per assessment, it was in fact costing around £1277 per assessment. However, the authors concluded that the budget allocated for assessments is likely to be adequate because the number of applications has been so low.  But since they wrote, the number of applications - and hence assessments has been rising.  If Shah’s estimates are applied to the more recent DoLS annual statistics, and extrapolated out from the  data for 2011-12 so far, we can see that the cost of assessments now far exceeds predictions.  Contrary to a projected year-on-year decrease in the use of the DoLS, the number of applications – and associated costs - may continue to rise. 

If this is how much more applications and assessments are costing - what about the cost of the legal process?

The impact assessment predicted that 25% of all DoLS applications would result in authorisation (5250 for the first year); that 25% of authorised cases (approx 1313) would result in a request for legal advice funded by the Legal Services Commission (LSC), and that 2.5% of all DoLS authorisation would result in an appeal (approx 131 per annum).  The number of applications, authorisations and appeals was predicted to fall year on year, and reach a steady-state in 2014/15.  They estimated the typical cost of a legal aid certificate for advice to be £141, and the typical legal costs of a DoLS appeal – including legal aid and court costs – to be £9k. It is unclear from these estimates whether they also factored in the legal costs to supervisory bodies of appeals.  The costs of legal advice and appeals were consequently predicted to be:

• 2009/10 - £1.37m
• 2010/11 - £1.18m
• 2011/12 - £1m
• 2012/13 - £0.82m
• 2014/15 - £0.64m
• 2015/16 - £0.46m

What are appeals actually costing under the DoLS? 

I asked the LSC for the costs of closed legal aid certificates for advice on s21A MCA; the average cost was £388.54 (2.75x greater than predicted). The average cost of a legal aid certificate for legal representation was £10,993.30. The Ministry of Justice says it does not collect data on court costs for DOLS, but already the cost of representation for a single litigant is greater than the total £9k predicted by the impact assessment for the total legal costs of an appeal. It would normally be anticipated that at least two legal aid certificates would be issued for a s21A appeal: one for P and one for P’s representative (bringing the total legal aid cost to £21,987). I have also obtained estimates of cost breakdowns from solicitors; however the cost estimates of the two solicitors I asked were much higher than the legal aid certificates, ranging between around £12-23k per client. I assume this may refer to privately funded litigants, where solicitors’ fees may be higher – so I have not based the following calculations on data from solicitors, only from local authorities. 

I also asked local authorities to give me estimates of how much appeals were costing them. I have heard back from only a few, but here is a table of the costs they cited:

Some local authorities estimated a range for individual costs, so I put an entry into this table for a case at the upper and lower range for that local authority (e.g. 3a, and 3b are the upper and lower range estimated by local authority 3).  Others cited the costs of particular cases, some still ongoing, so those estimates have a line each.  Some costs were only provided by some local authorities - very few quantified the costs to their in-house legal staff and DoLS teams.  I took an average for the costs of experts, counsel, legal staff, DoLS staff, and court fees, and summed them to produce an average estimated cost of a DoLS hearing for a local authority: £23.6k (range: £4.8k-£59k).

To produce an estimate of the overall cost to the public purse of a single hearing in the Court of Protection about a deprivation of liberty, I assumed that it would involve two legally aided litigants (P, and P's representative) and a supervisory body.  Often a managing authority would be listed as a defendant as well, but I have not considered these costs as they will not usually be met from the public purse (not directly, at least).  One local authority commented that there were also cases where a health authority might be listed as well, if they were funding the care home fees - I have not included such eventualities in my estimates.    Neither have I been able to consider the costs to the court services of a hearing, nor the costs to the Office of the Official Solicitor for their time, as no such data is collected.  In short, the estimates I am basing the following projections on are likely to be very conservative.  Based on these assumptions, a conservative estimate of the typical cost of a DoLS hearing to the public purse would be £45.6k, not £9k, as anticipated by the impact assessment.

Factoring in the actual number of appeals

Fortunately for the public purse, although not for detainees, the DoLS ‘have barely begun to function’.  The impact assessment predicted 2.5% of all authorisation would result in an appeal under s21A. In the first year, there were only 19 appeals under s21A (0.56% of all authorisations); in the second year there were 40 (1.52% of all authorisations). However, following the case London Borough of Hillingdon v Neary [2011] there is a visible ‘spike’ in appeals under s21A as well as court hearings relating to deprivation of liberty under other parts of the MCA: 

If we factor in the estimated average cost of a s21A MCA appeal, with the number of appeals, and compare it with the impact assessment predictions, it may look something like this: 

[* Estimates for this year extrapolated out from data on appeals to October 2011]

But of course, what the impact assessment did not take into account was that a large number of cases in the Court of Protection relating to deprivation of liberty (in fact, the majority) are not brought under s21A MCA.  This is because they are either cases where a public authority has had to seek authorisation directly from the court, and annual reviews, because the DoLS did not take into account the full range of contexts where a person might be deprived of liberty (a rather costly oversight, it would now seem).  Or they are cases where, for one reason or another either P or P's representative has been unable to trigger a s21A MCA appeal, but there is an ongoing dispute. And so, following Neary, supervisory bodies are acting on advice to refer disputes involving deprivation of liberty to the Court of Protection themselves where the appeal process has failed.  They may well do so under s15 and/or s16 MCA.  So, taking into account these cases as well, the costs of all cases in the Court of Protection concerning deprivation of liberty may look more like this:

I reiterate: This is not taking into account the costs of managing authorities, the courts, the Official Solicitor; nor cases where a second public authority is a defendant, nor any damages payable for unlawful detention. And, as the number of appeals looks as if it is growing, the costs will do so also.

The rising costs of DoLS in the future?

If appeals continue to rise, and the costs of a Court of Protection case relating to deprivation of liberty do not decrease, there are several possible outcomes: 

• The cost to the public purse – to local authorities (and PCTs, until they are abolished), to the courts, to the Official Solicitor, the Ministry of Justice -may become intolerable. 
• In response, supervisory bodies may seek ways to avoid authorisations and disputes coming to court; which may be incompatible with the Article 5 rights of detainees, and the Article 8 rights of detainees and their families. 
• Likewise, the pressure on the Official Solicitor may require other litigation friends to become available; one option is to use paid representatives.  Their pay will also have to reflect the true amount of time of case management for a litigation friend; however, their time is likely to be less expensive than the legally qualified case managers that work for the Official Solicitor. 
• Although the interpretation of deprivation of liberty should not reflect the burden these cases place on the legal system, we must consider that there is a danger that this may still influence judicial reasoning. Certainly, many cases make reference to this issue, indicating at least that public authorities seem to view it as pertinent to the scope of Article 5 when presenting their case to the court.

It is of course possible that these costs are an overestimate; although I feel they are more likely to be overly conservative.  It is also possible that costs will decrease in time, as case law becomes clearer and cases are dealt with in lower courts.  However, my feeling is that the technical issues in the DoLS have not yet all been resolved.  And I think, talking to DoLS teams, that cases like Neary may only be the tip of the iceberg in terms of disputes; we should expect more.  Futhermore, although the ruling in Cheshire in no small part attempted to simplify decisions around whether a deprivation of liberty is actually occurring, I think it may simply have shifted the debates from the 'Sorites paradox' (Clements, 2011) of where restrictions on liberty shade into deprivation of liberty, to the question of what is 'the appropriate comparator' for any given individual, and just how real does an alternative residence need to be to engage Article 5?  Because so many safeguards and benefits attach to the scope of Article 5, I don't think we'll see litigants letting this issue go so easily.  Such is the complexity of social care and people's lives, that determining issues on the basis of Cheshire is unlikely to be as simple as many lawyers seem to expect.  In short, whilst I think it is possible that Cheshire will offer a means for supervisory bodies to avoid authorising DoLS applications, hence diminishing protections in the community, I don't think it will resolve the conundrum facing the legal system of how to deal with the ballooning costs of appeals.

This cannot be sustainable in the long term. In part the costs derive from having to seek authorisation and review for cases outside the scope of the DoLS; amending the framework to allow for authorisations in settings other than care homes and hospitals would certainly reduce the number of review cases.  In part costs may be so high because the DoLS themselves are very complex (‘labyrinthine’, ‘byzantine’, ‘bureaucratic’), necessitating more complex legal arguments.  To remedy this issue would require a total revision of the framework; and of course this would have significant attendant costs.  Some disputes that could have been resolved earlier may be reaching court because of poor understanding and application of the DoLS (certainly, one hears anecdotally of many 'interesting' practices by DoLS teams); a revision of the code of practice to include important case law could perhaps stem some of these issues.  Another idea that I have been toying with would be to switch to a statutory definition of 'deprivation of liberty', which could be achieved by means of an amendment to the DoLS.  Ideally this definition would be subject to public consultation so we could hear reasoned and evidenced arguments about who needs the protection of the DoLS, rather than a mish-mash approach based on case law that is highly fact specific and problematic when applied to cases outside those facts.  Such a statutory definition would reduce the amount of time taken by public authorities and courts deliberating and debating the scope of Article 5, and potentially save on legal argument (and hence, costs).  However it is possible that greater certainly might increase the use of DoLS if currently managing authorities and supervisory bodies are exploiting this uncertainty and under-applying the framework.  Clearly greater legal certainty and consistency of approach would be a good thing, although if it ends up costing more it might be hard to persuade governments to consider this.  And then, of course, are the perennial debates about whether DoLS appeals should be more like tribunals, or even held by tribunals, although my view is that nobody has yet conducted a proper analysis of why the cases are taking so much time, and what 'efficiencies' of approach could be found whilst still ensuring people's rights are upheld.

Overall, I think if the argument that DoLS aren't 'fit for purpose' doesn't persuade the authorities that it's time to go back to the drawing board, then maybe the growing costs should persuade parliament that it would be prudent to engage in thorough post-legislative scrutiny. However, my concern is that high though these costs are, significant reforms would almost certainly cost more.  Instead what I suspect will happen is that the DoLS will continue to put a significant strain on the court system and public authorities, who will come under pressure to avoid using them wherever possible.  I think it is possible that only a repeat of Bournewood, embarrassing litigation in Europe where the DoLS have failed to protect people's rights, will prompt serious revisions to the framework.  With rulings like C v Blackburn with Darwen, we creep closer every day.