England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE ADMINISTRATIVE COURT
Mr Justice Mostyn
CO/1587/2017

		Royal Courts of Justice Strand, London, WC2A 2LL
		11/04/2019

Lord Justice McCombe:

(A) Introduction

1. The appellant, "Keep the Horton General" (a campaign group which acts by one of its members, Mr Keith Strangwood) appeals, with permission granted by Newey LJ by his order of 28 October 2018, from the order of Mostyn J of 21 December 2017 dismissing the judicial review claim of the Interested Parties brought against the respondent Oxfordshire Clinical Commissioning Group ("the CCG"). The claim was brought by four local government authorities as claimants, challenging the lawfulness of a public consultation launched by the CCG in January 2017 about proposals for changes in the provision of hospital and other health care services in the Oxfordshire area. The appellant group was joined as an interested party to the claim. The claim form was issued on behalf of the four claimant authorities on 30 March 2017.

(B) The Consultation: overview

2. The main public consultation document, issued on 16 January 2017, was entitled "The Big Consultation: Best Care, Best Outcomes and Best Value for Everyone in Oxfordshire". It was planned by the CCG as the first part of a two-phase consultation exercise and was conducted pursuant to the CCG's statutory duty set out in section 14Z2 of the National Health Act 2006. In its material parts, that section provides as follows:

"14Z2 Public involvement and consultation by clinical commissioning groups

(1) This section applies in relation to any health services which are, or are to be, provided pursuant to arrangements made by a clinical commissioning group in the exercise of its functions ("commissioning arrangements").

(2) The clinical commissioning group must make arrangements to secure that individuals to whom the services are being or may be provided are involved (whether by being consulted or provided with information or in other ways)—

(a) in the planning of the commissioning arrangements by the group,

(b) in the development and consideration of proposals by the group for changes in the commissioning arrangements where the implementation of the proposals would have an impact on the manner in which the services are delivered to the individuals or the range of health services available to them, and

(c) in decisions of the group affecting the operation of the commissioning arrangements where the implementation of the decisions would (if made) have such an impact.

(3) The clinical commissioning group must include in its constitution—

(a) a description of the arrangements made by it under subsection (2), and

(b) a statement of the principles which it will follow in implementing those arrangements.

(4) The Board may publish guidance for clinical commissioning groups on the discharge of their functions under this section.

(5) A clinical commissioning group must have regard to any guidance published by the Board under subsection (4)."

3. The five principal proposals upon which the consultation was carried out were these: (1) Following the temporary closure of 146 acute hospital beds, the CCG perceived a reduced need for such beds and proposed their permanent closure; (2) More planned diagnostic, outpatient and elective surgery services were to be provided at the Horton General Hospital ("the HGH") in Banbury; (3) Patients diagnosed with acute stroke conditions would be taken immediately to the hyper acute stroke unit at the John Radcliffe Hospital ("the JRH") in Oxford; (4) The HGH would continue to have a critical care unit, but the sickest critical care patients would be treated at the JRH; (5) The temporary closure (implemented in October 2016) of the obstetric unit at the HGH would be made permanent. Obstetric services and emergency gynaecological services would be provided in the future at the JRH and a "Midwife Led Unit" (or "MLU") would be established and maintained at the HGH. This first part of the consultation sought responses by 9 April 2017.
4. As for the second phase of the consultation, the document said this:

"Phase 2 consultation

During the next phase of consultation, we are expecting to invite your views on proposed changes to the following services in Oxfordshire:

Acute hospital services:

A&Es in Oxfordshire

Children's services

Community hospitals including MLUs

During this second phase we will also be looking in more detail at plans to develop primary care, which will underpin all our other changes (primary care services include GPs, nurses, healthcare assistants, community nurses and other clinicians).

…

These proposals set out in phase 1 would involve investment in some areas and would not be at the cost of other proposals we will be discussing in the consultation for phase 2."

It was not said when the second phase would be undertaken, although I understand that, by the time of the hearing before Mostyn J in December 2017, it was envisaged that Phase 2 would take place during 2018.

5. The consultation document stated expressly that the aim was to keep the HGH open and to "develop [it] to become a hospital fit for the 21^st century". It was said there were plans "to invest significantly in the hospital so that it can continue to develop and change as healthcare evolves and meet the needs of local people". In contrast, in the appellant's skeleton argument for this appeal it was said that:

"The Appellant was (and remains) concerned that the service changes, incrementally proposed, will in due course spell the end of the HGH as a general hospital because the elimination of some services will in due course make other services unviable."

6. As part of the processes of decision making and implementation of changes to healthcare provision CCGs are required to follow guidance issued by NHS England in implementing the government's "mandate" to that body: see section 14Z2(4) and (5) above. The relevant guidance informs CCGs (among with other matters) that they must provide to NHS England assurance that certain tests of "service reconfiguration" are satisfied before they proceed to the statutory public consultation. As the guidance provided, in the form it was in prior to this consultation, the four tests were: 1. "Strong public and patient engagement"; 2. "Appropriate availability of choice"; 3. "Clear clinical evidence base"; and 4. "Clinical support". Initial approval of the proposals upon which consultation was to take place was given by NHS England on 5 December 2016 in approving the CCG's "pre-consultation business case" ("PCBC").
7. The consultation document informed readers that the CCG had to satisfy the four tests identified by NHS England and that it had stated in the PCBC how the tests had been met. There was a cross-reference to the web link for the PCBC where these matters were addressed. In the course of the consultation, however, on 3 March 2017, NHS England announced a further assurance test in addition to the original four, which would have to be satisfied in a case of a proposal to close hospital beds. This additional test was to apply from 1 April 2017 and provided for three new conditions as follows:

"Demonstrate that sufficient alternative provision, such as increased GP or community services, is being put in place alongside or ahead of bed closures, and that the new workforce will be there to deliver it; and/or

Show that specific new treatments or therapies, such as new anti-coagulation drugs used to treat strokes, will reduce specific categories of admissions; or

Where a hospital has been using beds less efficiently than the national average, that it has a credible plan to improve performance without affecting patient care (for example in line with the Getting it Right First Time programme)."

8. Following the announcement of the prospective new test, there was no public communication to consultees in this case on the subject of the new conditions and no invitation to comment upon them was made. This is the subject of a specific ground of appeal to which I will return.
9. Following the consultation and while the judicial review claim was pending, on 10 August 2017, the CCG resolved to implement the proposals. On 21 December 2017, Mostyn J dismissed the claim. The appellant (but not the claimant councils) sought permission to appeal from the judge's order. While that permission application was pending, on 9 February 2018, the Independent Review Panel, advising the Secretary of State upon changes such as these, advised that further action was required before the proposals relating to obstetric services at the HGH could be implemented. On 29 March 2018, the CCG resolved to cancel Phase 2 of the consultation. The appellant sought permission to adduce in evidence on the appeal the documentation relating to these new events. Permission to do so was granted by Newey LJ in his permission to appeal order of 24 October 2018.
10. The appeal is narrower in scope than the initial challenge brought by the councils and relates solely to the "bed closure" aspect of the CCG's proposals.

(C) The Appeal

11. There are four grounds of appeal. First, it is said that the judge erred in failing to consider properly the fairness of the consultation in view of the bed closure proposals being the subject of phase 1 while the subject of community provision (including such provision as alternative to hospital care) was not to be consulted upon until phase 2. Secondly, it is argued that, while the judge found a flaw in the consultation in its failure to put NHS England's new "bed closure" test to the consultees, he did not find that the flaw was "sufficiently serious" to render the consultation unlawful. Thirdly, it is submitted that the CCG failed properly to set out in the consultation materials the "pros" and "cons" of the bed closure proposals. While stating the "pros", it did not identify the "cons". Fourthly, it is said that the judge improperly admitted into evidence a new witness statement from the CCG on the afternoon of the second day of the two day hearing and after the parties had concluded their arguments.
12. For the purposes of examining these grounds of appeal, it is necessary to refer to some further details of the consultation document.
13. In the Introduction, in dealing with the Phase 1 proposals, the Clinical Chair of the CCG referred to an earlier "Big Health and Care Conversation" involving the public undertaken in 2016, he wrote this about the first phase and the later second phase:

"During the Big Health and Care Conversation, the listening exercise we carried out in 2016, many people took the time to tell us what they thought and we have used your feedback while we were developing the proposals set out in this document. It is clear that the NHS is greatly valued and that people also understand the pressures we are facing. We had many examples of people's own experiences and many ideas and suggestions for improving care. Thank you to everyone who took the time to share their views, attend events and respond to the survey.

We have now reached a point where we want to ask the public and our partners questions and seek feedback on some more specific proposals for change. In this document you will find proposals for changes to the following services.

Changing the way we use our hospital beds and increasing care closer to home

Planned care services at the Horton General Hospital

Acute stroke services

Critical care

Maternity

These changes are being considered now because the quality of care for patients will be affected if we delay making decisions. Furthermore, some of these services do not meet national clinical best practice recommendations.

A further set of proposed changes will be presented in a Phase 2 consultation but more work is needed to develop these options before a second consultation can be launched."

In the body of the document, there appeared the following about the phasing of the consultation:

"During the second phase we will also be looking in more detail at plans to develop primary care, which will underpin all our other changes (primary care services include GPs, nurses, healthcare assistants, community nurses and other clinicians).

This document focuses on Phase 1 only. It includes proposals for formal public consultation on:

changes to acute hospital bed numbers in Oxfordshire as part of a plan to provide more care out of hospital

more planned care at the Horton General Hospital in Banbury (planned care is a term for Healthcare which has been planned in advance and which is not urgent or an emergency, such as diagnostic tests, outpatient appointments and surgery)

stroke services in Oxfordshire

critical care at Horton General Hospital (critical care helps people with life-threatening or very serious injuries and illnesses)

maternity and obstetric care including obstetrics, the Special Care Baby Unit (SCBU) and emergency gynaecology inpatient services at the Horton General Hospital

These proposals set out in Phase 1 would involve investment in some areas and would not be at the cost of other proposals we will be discussing in the consultation for Phase 2."

14. Specifically dealing with bed closures, the consultation document stated this:

"One of our key aims over the next few years is to reduce the time patients spend in hospital for care in an emergency and increase care for people in the community or at home."

It was also explained that there were special considerations that made it desirable for elderly patients to be cared for at home and away from hospitals if possible. Ms Broadfoot QC, at the outset of her helpful argument for the appellant, indicated that these aims were entirely uncontroversial so far as her clients were concerned.

15. The document then stated that in the summer of 2015 there were 150 people in hospital in Oxfordshire who could have been better cared for elsewhere. It then outlined the reasons for this, together with "some new approaches" which had been adopted as follows:

"We have piloted some new approaches that have resulted in fewer hospital beds being needed. Staff came up with innovative ideas to tackle the problem in the short and long term. Not all of these changes happened at once and some were put in place as we learned what worked best for patients.

A 'liaison hub' was set up which brought together experienced nurses and other staff from care organisations. Its role is to make sure that when patients are ready to leave hospital, the right care is available for them at the right time.

Patients were moved from hospital to nursing home beds with additional therapy support and cared for by teams which included GPs, doctors and nurses and therapy and social care staff. This continued until patients were ready to either remain in a nursing home or return to their home with or without care.

A recruitment drive was launched for care workers to support people in their own homes.

These changes mean that patients can be cared for in a range of places which are better for them than being in a busy acute hospital ward."

16. The document went on to explain further proposals to reduce the need for in-patient hospital care and stated that the need for hospital beds had been reduced. It said this:

"As a result, the number of hospital beds we need reduced and we closed 146 acute hospital beds on a temporary basis. Initially 76 beds were temporarily closed in the winter of 2015/16, then in September 2016 a further 70 beds were temporarily closed. These beds were in Oxford (101 beds) and Banbury (45 beds) from areas including post-acute and surgical emergency units, general medicine, elective surgery, orthopaedics, and other wards at the John Radcliffe Hospital."

Ms Broadfoot took issue with the Banbury figure which she said involved 23 closures during the winter of 2015/6 and a further 28 on 1 October 2016 at the HGH, the latter being only just before the consultation began in January. In my view, however, this small discrepancy in numbers does not affect any of the issues on the appeal.

17. The document went on to give some details about an evaluation of 483 patients discharged from hospital into nursing home care between December 2015 and August 2016. A survey of such patients was described which had yielded a high level of satisfaction with the process. It was said that changes to the acute beds provision was expected to result in savings of £4.9 million the bulk of which would be invested in the new services described. Again, cross-reference was made to the web link to the PCBC.

(D) The Law

18. The law was not in dispute between counsel and no reported cases upon the main principles relating to the law's requirements for a public consultation had to be cited to us, either in written or in oral argument. It appeared to be common ground that "fairness" underpins all; to be lawful a consultation must be fair, but fairness does not require perfection. A challenge will not necessarily succeed simply by pointing out a way in which the consultation could have been better, unless the failure to proceed in that way has led to real unfairness.
19. It is sufficient for our present purposes to recall certain passages from the judgments of the Supreme Court in R (Moseley) v Haringey LBC [2014] 1 WLR 3947. The well-known principles underlying the common law on the subject of lawful consultation appear in the judgment of Lord Wilson of Culworth (with whom Lord Kerr of Tonaghmore agreed) at paragraph 25 as follows:

"25.  In R v Brent London Borough Council, Ex p Gunning (1985) 84 LGR 168 Hodgson J quashed Brent's decision to close two schools on the ground that the manner of its prior consultation, particularly with the parents, had been unlawful. He said, at p 189:

"Mr Sedley submits that these basic requirements are essential if the consultation process is to have a sensible content. First, that consultation must be at a time when proposals are still at a formative stage. Second, that the proposer must give sufficient reasons for any proposal to permit of intelligent consideration and response. Third … that adequate time must be given for consideration and response and, finally, fourth, that the product of consultation must be conscientiously taken into account in finalising any statutory proposals."

Clearly Hodgson J accepted Mr Stephen Sedley QC's submission. It is hard to see how any of his four suggested requirements could be rejected or indeed improved. The Court of Appeal expressly endorsed them, first in Ex p Baker [1995] 1 All ER 73, cited above (see pp 91 and 87), and then in R v North and East Devon Health Authority, Ex p Coughlan [2001] QB 213, para 108. In Ex p Coughlan, which concerned the closure of a home for the disabled, the Court of Appeal, in a judgment delivered by Lord Woolf MR, elaborated, at para 112:

"It has to be remembered that consultation is not litigation: the consulting authority is not required to publicise every submission it receives or (absent some statutory obligation) to disclose all its advice. Its obligation is to let those who have a potential interest in the subject matter know in clear terms what the proposal is and exactly why it is under positive consideration, telling them enough (which may be a good deal) to enable them to make an intelligent response. The obligation, although it may be quite onerous, goes no further than this."

The time has come for this court also to endorse the Sedley criteria. They are, as the Court of Appeal said in R (Royal Brompton and Harefield NHS Foundation Trust) v Joint Committee of Primary Care Trusts (2012) 126 BMLR 134, para 9, "a prescription for fairness"."

20. The Moseley case, like the present, involved a statutory consultation, but it was a case involving statutory consultation about a scheme for council tax reduction, replacing council tax benefit. Lord Reed was at pains to pin his judgment upon "the statutory context and purpose of the particular duty of consultation…" in question, rather than upon the common law: paragraph 34. At paragraphs 38 and 39, Lord Reed said this:

"38.  Such wide-ranging consultation, in respect of the exercise of a local authority's exercise of a general power in relation to finance, is far removed in context and scope from the situations in which the common law has recognised a duty of procedural fairness. The purpose of public consultation in that context is in my opinion not to ensure procedural fairness in the treatment of persons whose legally protected interests may be adversely affected, as the common law seeks to do. The purpose of this particular statutory duty to consult must, in my opinion, be to ensure public participation in the local authority's decision-making process.

39.  In order for the consultation to achieve that objective, it must fulfil certain minimum requirements. Meaningful public participation in this particular decision-making process, in a context with which the general public cannot be expected to be familiar, requires that the consultees should be provided not only with information about the draft scheme, but also with an outline of the realistic alternatives, and an indication of the main reasons for the authority's adoption of the draft scheme. That follows, in this context, from the general obligation to let consultees know "what the proposal is and exactly why it is under positive consideration, telling them enough (which may be a good deal) to enable them to make an intelligent response": R v North and East Devon Health Authority, Ex p Coughlan [2001] QB 213, para 112, per Lord Woolf MR."

21. In the light of this nuance in approach between Lord Wilson and Lord Reed, Baroness Hale of Richmond and Lord Clarke of Stone-cum-Ebony added this at paragraph 44:

"44.  We agree that the appeal should be disposed of as indicated by Lord Wilson and Lord Reed JJSC. There appears to us to be very little between them as to the correct approach. We agree with Lord Reed JSC that the court must have regard to the statutory context and that, as he puts it, in the particular statutory context, the duty of the local authority was to ensure public participation in the decision-making process. It seems to us that in order to do so it must act fairly by taking the specific steps set out by Lord Reed JSC, in para 39. In these circumstances we can we think safely agree with both judgments."

22. In the present case, it was not suggested that any such nuance of approach was required. Both Ms Broadfoot and Ms Morris adopted the yardstick of "fairness" in their submissions and I am content to adopt their common approach, while noting the slight differences appearing in the judgments of Lords Wilson and Reed in the Moseley case, in a rather different statutory context from the present.

(E) The Grounds of Appeal and my Conclusions

23. The ground of appeal upon which Ms Broadfoot concentrated was ground 1 (inter-dependency of the two phases of consultation) although, to my mind, all four grounds were to a degree linked. I will take grounds 1 and 3 together.

Grounds 1 and 3

24. Ground 1 contends that the judge was wrong in failing to find the consultation unfair in failing to bring into the "bed closure" issue the question of alternative services in the community which was intended to be part of phase 2. It was submitted that consultees could not sensibly consider how to respond to the bed closure points without knowing what was to be proposed about future community care overall. Linked into that issue is ground 3, in which the appellant complains that the consultation materials failed to provide a sufficient statement of the arguments against the bed closure proposals (the "cons") as distinct from the "pros", namely that the CCG had reached its own conclusion that "the number of hospital beds we need has reduced".
25. Ms Morris QC for the CCG argued that the "cons" were entirely obvious and those consulted clearly appreciated that. She argued that there was a wealth of material backing up the main document on the CCG website, to which the public's attention was drawn in many places in the consultation paper. This included material from the public feedback during the earlier public engagement in 2015 to 2016.
26. We were also referred by Ms Morris to various extracts from a report to the CCG from independent consultants, called Qa Research, analysing the feedback from the consultation itself.
27. The first passage most directly relevant to the bed closure issue was this:

"Changing use of acute hospital beds across Oxfordshire; increasing care closer to home

At least three-quarters of survey respondents (more in some cases) agreed with five out of each of the six statements relating to the way hospital beds are used and providing care closer to home.

A majority agreed that care closer to home is best; that a hospital bed is not necessarily the best place for an elderly person to be cared for and that some hospital stays can be unnecessarily long due to care at home or in the community not being available. A majority also agreed that organisations don't always work together to find the right support for patients outside of hospital.

The one statement which less than three-quarters of respondents agreed with related to whether too many people are admitted to hospital when assessment, treatment and support could have potentially have been provided elsewhere, including at home (67% agreement). This statement resulted in a greater neutral response than others (17%).

Other public and stakeholder consultation responses show that although there is support for the principles behind these proposals there was significant concern that the impact on adult social care resourcing had not been fully explored within the proposals particularly in the context of the existing pressures on the social care workforce and the likely impact on carers."

The document also referred to the activities of the appellant group in these terms:

"The local campaign group 'keep the Horton General' participated in the consultation in a number of ways. Members of the group attended every public consultation meeting and distributed material about the consultation and their campaign to those attending and to households across the area. They provided information via their website and Facebook page. They expressed their concern about the consultation survey and encouraged people to seek advice from them before completing it. They also provided a template letter that people concerned about the proposals could use to respond to the consultation."

28. A later passage in the paper referred to answers from respondents to the specific question whether they agreed with the bed closure proposal. It was reported that 50% did not agree with it. The reasons for disagreement were summarised as follows:

"Reasons for disagreeing with the proposal focused on a belief that the hospital was already stretched, with more beds needed not less. These was also a concern about the knock on effect of closing beds; which other services would it have an impact on? Some felt closing the beds simply wouldn't solve 'the problem' and that the 'alternative' model of care needs to be in place and fine-tuned before any beds are closed. Others commented that it was too difficult or took too long to get to Oxford."

Similar comments were reported from public consultation events. The complaint as to the splitting of the consultation into two phases was also reported.

29. Ms Morris submitted that this material shows that the public, including the appellant, were well aware of the arguments against the bed closures and stated them; there was, she argued, no more that could reasonably have been said to the public to enable them to have a better ability to respond to the phase 1 consultation.
30. Ms Morris went on to take us to passages in the "Decision Making Business Case" ("DMBC") which was placed before the CCG when it took its decision on the proposals on 10 August 2017, in the light of all the materials including the responses to the consultation. She argued that these extracts showed that the public had responded fully to the consultation and had not been hampered in doing so by a lack of information.
31. The DMBC made a modified recommendation for bed closures, proposing a "staggered" implementation: 110 beds were to remain closed; the remaining 36 would only be permanently closed "…when the system has made significant progress in reducing the numbers of delayed transfers care". Permanent closure of these further beds would be "subject to further Thames Valley Clinical Senate review and NHS England assurance".
32. The DMBC outlined the alternative provision already being made to meet the bed closures proposed (quite apart from anything to be proposed in Phase 2) and said, with reference to a diagram, that:

"Overall, the number of beds in the system has not reduced markedly, but these beds are used in different ways to ensure that when patients are medically fit for discharge (but are still awaiting further care) they are in a more appropriate environment. As can be seen from the diagram below, the bed changes have been accompanied by a significant increase in the capacity and activity levels in ambulatory assessment. Other non-bed-based services have also been expanded."

The document also recorded the fact that 50% of the respondents to the consultation did not agree with the closure proposals.

33. Importantly, the DMBC also addressed the question of the new NHS England bed closure test and recited that body's response to the proposals in the light of the additional test. It said this:

"NHS England received the report from the Thames Valley Clinical Senate setting out their review of Phase One proposals for bed closures against the 5^th test.

The Senate recommended that the conditions for the NHS Bed Test had been met subject to the following:

1) The delays associated with patients being referred to HART need to be resolved and there needs to be sufficient capacity for HART to discharge once their element of service provision is completed. The Senate was advised that this is currently a problem for HART.

2) OCCG should monitor the system and take action to ensure that delays do not build with regard to the discharge to domiciliary care.

3) The Senate retrospective review was based on the current closure of 110 beds. It did not consider any future closures.

NHS England, 31 July 2017, confirmed that it is content to accept the recommendations of the Senate as set out above regarding the review and compliance against the 5^th test based on the closure of 110 beds. Any proposal to further reduce beds would need to be reviewed by the Senate."

34. As already mentioned, the decision of August 2017 adopted the recommendation of a "staggered" closure of the 146 beds.
35. In the light of these materials, which I have sought to outline, both grounds 1 and 3 raise quite short points.
36. Ms Broadfoot set out six propositions in which she noted the statutory duty to "involve" individuals to whom services were being or might be provided (whether by consultation or being provided with information in other ways). Formal consultation was chosen and therefore the process had to be effective and give the consultees an opportunity to influence the outcome.
37. In my judgment, looking at ground 3 first, and assessing whether the information provided enabled the public to respond effectively to the bed closure issue, I believe that the material provided in the consultation (which I have outlined above) was adequate. It is clear that there was significant response to the bed closure proposal and it largely questioned the ability of alternative care proposals outlined in the consultation to meet the need. I believe that Ms Morris was correct in her argument that the potential disadvantages of the proposal (the "cons") were obvious and that is what the respondents pointed out. The response clearly influenced the decision made.
38. Turning to ground 1, Mostyn J questioned the desirability of splitting of the consultation but found that the data provided in the late evidence showed that the decisions in Phase 1 would have no material effect on Phase 2. He said this at paragraphs 25 and 26 of his judgment:

"25. The conclusions I have reached thus far should not be taken to signify that I personally approve of the decision to split this consultation. It was said that the reason it was done in this way was because of the urgency of the matters covered by phase 1. But they were not urgent. The obstetric unit had already been closed, albeit temporarily. The number of Level 3 critical care and stroke victims was tiny compared to overall activity. And in any event, it proposed that phase 2 should follow very shortly after phase 1 – the papers mention the consultation for phase 2 beginning in April 2017. Miss Morris QC argued that to leave the obstetric unit temporarily closed without a definitive decision was bad for morale, but that was mere assertion and did not, in my opinion, justify taking the risks in splitting which I have mentioned above.

26. I can well see why in the absence of hard data the claimants and the interested party would assert that as a matter of principle decisions made following phase 1 would queer the pitch when the phase 2 consultation came around. However, as I have demonstrated, the hard data shows quite clearly that the decisions on the very small number of cases involved will have no material effect on the scope of the phase 2 consultation. It is a mystery to me why that data was not supplied sooner."

39. Before the judge, however, the concern as to the splitting of the consultation arose principally in relation to issues other than bed closure, i.e. obstetrics and gynaecology, anaesthetics and accident and emergency. The preceding passage of the judge's judgment (paragraphs 17-24) shows this and indicates that in the judge's mind the new evidence related largely (if not exclusively) to these other issues. It dispelled his initial anxieties as to the split in the consultation phases.
40. Ground 1 attacks the splitting of the consultation on the basis that the judge failed to consider it at all in relation to the specific issue of bed closures. It might have been better if the judge had specifically addressed that question in his judgment. However, as I see it, he saw other issues as far more material to the wider challenge that was before him on this point. Looking at the issue in the light of all the material to which our attention was drawn, I think he was right to do so.
41. The issue of bed closure was largely a question of how the measures already piloted and in place were coping with the change caused by the temporary closures and whether it was acceptable to proceed to permanent closure. A significant number of respondents understandably questioned this. However, these existing coping strategies were there and were not phase 2 matters. The evidence showed that the concerns were considered (in the light of the existing steps already taken) by the Clinical Senate, NHS England (in the light of the new fifth test), in the DMBC and by the CCG. The result was the staggered closure recommendation which the CCG adopted. Given the limited ambit of the bed closure proposal and the existence of remedial steps already in place, I do not see that the consultation was unfair in addressing this issue in phase 1 and before the phase 2 proposals had been formulated.

Ground 2

42. Mostyn J found that the consultation was formally flawed by the failure of the CCG to seek the views of the public in the light of the new test introduced by NHS England in April 2017, following the announcement in March 2017.
43. It is right that the main consultation paper identified, as a requirement on the CCG, that it should "pass" the original four tests: see the passage quoted above. It would no doubt have been natural (if the new fifth test had been promulgated earlier) to include that test in the paper also. However, in my judgment, Ms Morris was correct in submitting that the statutory guidance makes it clear that satisfaction of these tests is a matter to be addressed as part of the PCBC and as part of the process of getting approval of the proposals from NHS England prior to public consultation. This appears in a number of places in the guidance document, particularly at page 21. Further, it is quite clear that that is what in fact occurred in this case: see the DMBC p. 3 (Appeal Bundle, Tab 30, page 324).
44. For my part, I also agree with the view of Sir Stephen Silber (sitting as a judge of the High Court) in R (Hinsull) v NHS Dorset CCG [2018] EWHC 2331 (Admin), paragraph 119, that there is no obligation upon a CCG to consult as to whether the four (now five) tests have been satisfied. It is NHS England that has to be satisfied about them before they will allow proposals to go forward. It is not for CCGs thereafter to invite the public to say whether they have been satisfied or not. In so far as the judge in the present case decided otherwise, I respectfully consider that he was wrong. However, that is not to say that any member of the public would not be quite entitled, in objecting to a proposal, to make the argument that NHS England should not have been satisfied as to any one or more of the five tests. Such argument might carry little weight when the "expert" body (as Sir Stephen called it in Hinsull's case), which set the tests in the first place, has expressed itself as content.
45. In any event, it is hard to see how this additional material could sensibly have been inserted into the consultation process, given the late stage in that process at which it emerged. In addition, the CCG did refer its proposals to NHS England, before it made its decision, as already mentioned and approval was given on 31 July 2017. In the circumstances, I would reject ground 2 of the grounds of appeal.

Ground 4

46. As already mentioned, Mostyn J admitted into evidence, at a very late stage in his hearing, the additional witness statement of Mr David Smith of the CCG, supplying the more detailed data which opponents to its proposals had been seeking for a considerable time. As the judge noted, it was a mystery to him why it had not been supplied earlier.
47. We were told that the statement (dated 6 December 2017) was submitted, without prior warning to other parties, at about 3 p.m. on the second day of the hearing, Thursday, 7 December 2017. In the face of objection from the claimants and the present appellant, the judge made his decision to admit the statement, but he allowed those parties to file evidence and argument in response (which they did on Monday 11 December). There was no application for an adjournment.
48. Ms Broadfoot informed us, however, that the judge told the parties, perhaps unwisely, that he planned to begin writing his judgment on the Friday (8 December) and, we were told, a draft of the judgment was supplied to the parties on Tuesday, 12 December. The perception of hasty acceptance of the new material created in this way was perhaps unfortunate.
49. The manner in which this late material was produced was clearly highly unsatisfactory and, if it was to be admitted, the other parties had to be given the opportunity to make an adequate response. If an adjournment for that purpose had been sought, it seems to me that it would have been irresistible. However, no such application was made.
50. Ms Broadfoot submitted that the new evidence was an important feature in the judge reaching the decision that he did, as was shown by his statement in paragraph 25 of the judgment that he was not to be taken as personally approving the decision to split the consultation. She argued that it was what judge saw to be the "hard data" in the new statement that prevented the opponents of the proposals succeeding in their challenge based upon the phasing arrangement.
51. It seems to me, however, that the failure to seek an adjournment when this new material was presented is fatal to this ground of appeal. A focussed request for more time to respond and, perhaps, for a re-convening of the hearing when the response had been prepared would have carried weight. As it is, the judge gave time for a response and there is no indication in the responses that more might have been forthcoming if more time had been allowed. Further, the material advanced on each side was not bulky and the judge would have had time to digest what had been submitted, at a time when he was immersed in the detail of the case and before circulating his draft judgment on Tuesday, 12 December.
52. Moreover, as already mentioned, the new evidence did not address further the bed closure proposals to any great extent: for example, there was one paragraph in Dr Fisher's response statement, on behalf of the appellant, touching directly on the question to which Ms Morris drew our attention.
53. As I have said, I would reject ground 4.
54. I would add that on the morning of the hearing of the appeal, the CCG again presented last minute evidence, in a witness statement and exhibit dated 13 March 2019, which was said to be directed to "bringing matters up to date" and might be relevant to any relief to be granted if the appeal were successful. This new statement too seems to have been an unheralded development and was not accompanied by any formal application notice applying for permission to adduce fresh evidence.
55. Ms Broadfoot naturally objected to our receiving this evidence which she said she had had the chance only to "skim-read".
56. After briefly considering the matter, we decided to proceed with the hearing of the appeal, without making a formal ruling on this unorthodox application to adduce fresh evidence, preferring to see whether any real issue arose to which the new material could be said to be directly relevant. In the end, we were able to consider the arguments without reference to the new statement.
57. I mention the matter only because, in my judgment, the CCG and its advisers, and other litigants, should be reminded that applications to adduce fresh evidence on appeal should proceed in the ordinary way by formal application and with proper notice to the opposing parties, addressing the customary criteria relevant to that question. It is only fair that that procedure is used.
58. What happened shortly before our hearing seemed to me to be a repeat of the unsatisfactory way in which new evidence, that had been available to it for some time, was produced by this CCG before Mostyn J.

(F) Overall Conclusion

59. For the reasons given above, however, I would dismiss the appeal.

Lord Justice Lindblom:

60. I agree with both judgments, which I have seen in draft.

Sir Terence Etherton MR:

61. I agree with the judgment of Lord Justice McCombe. I am adding a short concurring judgment of my own in recognition of the importance of this issue to the campaign group, acting by the appellant, and in deference to the high quality submissions made to us by counsel. I have used the same abbreviations as Lord Justice McCombe.

Appeal Grounds 1 and 3

62. As Lord Justice McCombe observed, appeal grounds 1 and 3 stand or fall together.
63. There is obvious force in the appellant's contention that there was an important interdependency between the proposed bed closures and the provision of community facilities. Indeed, NHS England has itself made that interdependency explicit in its patient care test announced in March 2017 for hospital bed closures after 1 April 2017, one of the new conditions in the test being to: "Demonstrate that sufficient alternative provision, such as increased GP or community services, is being put in place alongside or ahead of bed closures, and that the new workforce will be there to deliver it".
64. That interdependency is also implicit in the principal consultation document published on 16 January 2017 ("The Big Consultation: Best Care, Best Outcomes and Best Value for Everyone in Oxfordshire"), which makes clear that the issue of the range and nature of appropriate community facilities and care would be dealt with in Phase 2, identifying in that context community hospitals, including MLUs, and the development of primary care, including GPs, nurses, healthcare assistants, community nurses and other clinicians.
65. The DMBC stated that: "The decision to split the Oxfordshire Transformation Programme into … two phases was taken based on advice from the Joint Health Overview and Scrutiny Committee ("JHOSC")". Nevertheless, in view of the interdependency between the merit or otherwise of bed closures and the provision of community facilities, it was entirely logical for Ms Samantha Broadfoot QC, for the appellant, to submit that either the consultation should have dealt with everything at one and the same time or, alternatively, the decision as to what to do about the closure of hospital beds should await the outcome of the consultation on Phase 2. She was also correct to highlight that the Judge did not expressly address in his judgment the issue of the absence of that interdependency in the consultation.
66. As Lord Justice McCombe has emphasised, however, and indeed was common ground before us, the consultation will only have been unlawful if, in the actual circumstances of the case, including the statutory context, it was unfair: R (Moseley) v Haringey LBC [2014] UKSC 56, [2014] 1 WLR 3947. The mere fact that it was not perfect or could have been improved is not enough to make the consultation unlawful if, in all the circumstances, it provided a fair opportunity for those to whom the consultation was directed adequately to address the issue in question.
67. In the circumstances of the present case, there are several reasons why the consultation was sufficiently fair to have been lawful.
68. Firstly, it was obvious and inevitable that the issue of whether there was appropriate and sufficient community care to counter the consequences of the bed closures was bound to be a concern of consultees and would be addressed by them. That was reflected in Q4 of the consultation survey, mentioned below.
69. Secondly, the consultation document said that consultees could read more about the vision for healthcare services in Oxfordshire in the Transformation Programme Pre-Consultation Business case on the website, the address of which was given.
70. Thirdly, Q4 of the consultation survey asked the specific question (at (e)) whether: "Too many people are admitted to hospital in the first place when they could have been assessed, treated and supported at home or in community settings such as a community hospital, care home or at home".
71. Fourthly, it is clear that consultees did in fact express their concerns about the adequacy of community services. There were various different ways in which members of the public, including patients, were engaged in the consultation, including events and meetings of various kinds, Qa Research Consultancy ("Qa Research") prepared a report analysing for the CCG the responses that were made, totalling some 9248 letters and emails from individual members of the public. The Qa Research report said that: "Uncertainty and a lack of confidence was expressed as to whether the new model of provision of out-of-hospital support would actually work and some people suggested it was high-risk to close hospital beds until it has been further proven" and "There was significant interest in ensuring the adequate provision of intermediate care beds. People were concerned about the numbers reducing and having to travel further to receive this type of care".
72. Fifthly, this aspect of the public response was addressed in the CCG's DMBC. In relation to the consultation, it was recorded in 4.1 of the DMBC that more than 10,000 individual responses were received by the CCG and more than 1,400 people attended the public meetings. It said that the CCG accepted comments in any form people wished to use and all feedback was passed to Qa Research, which analysed the responses and produced its report. At 9.4 of the DMBC the issues raised in the consultation and the impact assessment from the Board of the CCG were set out and a response given on behalf of the Transformation Programme. The first identified issue was: the "Capacity of community care (care homes, care at home, carers) to cope with existing and additional demand". The second identified issue was: "Wider implications of proposals on the 'whole system'". The response to each of those issues referred to Phase 2 but, for the purposes of this appeal, the important point is that the issue was raised in response to the consultation, as was inevitable.
73. Sixthly, by the time the consultation took place, it would have been apparent to those interested what the consequences of the bed closures would be. Twenty-three beds had been temporarily closed in December 2015, over a year before the consultation began in early 2017 and even longer before the consultation closed in April 2017. A further 28 beds were temporarily closed in October 2016, just before the winter months when demand would have been high for the elderly and vulnerable. The proposal was not to close more beds but to make permanent those temporary closures. Accordingly, any insufficiency of community care as a result of those closures would have been uppermost in the mind of the consultees. I, therefore, do not accept Ms Broadfoot's submission that people could not give an intelligible and meaningful response on the required nature and extent of the care to be provided in the community before the consultation on Phase 2.

Appeal ground 3

74. I am inclined to agree with Ms Fenella Morris QC, for the CCG, that NHS England's new bed test probably came too late to be required to be the subject of consultation, either by extending the then consultation period or by a further consultation.
75. It is not necessary to reach a conclusion on that point, however, for the following reasons. The four original tests were: strong public and patient engagement, appropriate availability of choice, clear, clinical evidence base, and clinical support. NHS England itself had to be satisfied that those tests were met before any changes could be made. Furthermore, satisfaction of those tests had been demonstrated in the PCBC before the consultation had begun. The purpose of that document was to inform assessment of the proposals against the four tests and NHS England's best practice checks.
76. Satisfaction of NHS England's new bed closure test of 3 March 2017 was equally something of which NHS England needed to be assured before changes could go ahead. The responsibility for judging whether the test was satisfied fell on NHS England and not the CCG. There was, therefore, no requirement for the CCG to consult on it. I agree with Lord Justice McCombe that Sir Stephen Silber in R (Hinsull) v NHS Dorset CCG [2018] 2331 (Admin) at [119] was correct on this point. On 31 July 2017 NHS England indicated that it was satisfied.
77. It is to be noted that, when the JHOSC wrote to the Secretary of State on 30 August 2017 seeking a review, it limited its request to a review of the decision of the CCG permanently to close consultant-led maternity services at HGH. It made no complaint about the failure to consult on satisfaction of NHS England's new bed test.

Appeal ground 4

78. I have nothing to add to what Lord Justice McCombe has said on this ground of appeal.