England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
ADMINISTRATIVE COURT
OWEN J
[2011] EWHC 2986 (Admin)

		Royal Courts of Justice Strand, London, WC2A 2LL
		19/04/2012

Lady Justice Arden:

1. This is the judgment of the court to which all members of the court have contributed.
2. This appeal is brought by the Joint Committee of Primary Care Trusts ("JCPCT") from the order of Owen J dated 7 November 2011 quashing the major consultation which it had conducted into the reconfiguration of national paediatric cardiac surgical services by means of a public consultation document entitled Safe and Sustainable – A new vision for children's congenital heart services in England ("the consultation document"), issued in March 2011.
3. The consultation document set out a number of options for the whole of England. We can leave aside the options for areas other than London, with which we are not principally concerned. The main point was that in the consultation document the JCPCT expressed the preferred option that there should be two centres for London and that those two centres should be Evelina Children's Hospital (at Guys and St Thomas' Hospital) and the Great Ormond Street Hospital ("GOSH").
4. This preferred option did not include the respondent ("Royal Brompton"), notwithstanding that its excellence and its place as a world-leading research institution have never been in doubt. Royal Brompton is the largest specialist heart and lung centre in the UK and among the largest centres in Europe. It has, for many decades, been at the forefront of specialised treatment for complex heart and lung disease. It provides a specialist service for children's heart and lung disease and comprehensive paediatric critical care services. It has the second largest paediatric intensive care unit ("PICU") in England. It is one of the largest centres for clinical research into cardiological disease in the country. Its work is highly regarded. A member of the public might well find it difficult to understand why a centre of the standard of Royal Brompton should cease to be a centre for paediatric cardiac surgical services under the configuration exercise.
5. Royal Brompton was the applicant for the quashing order. It succeeded before the judge only on the last of five grounds on which it sought to have the consultation process set aside. In particular, Royal Brompton failed to establish that the preferred option precluded consultees from responding to the consultation that it ought to be included. Royal Brompton seeks to uphold the judge's order on a number of additional grounds.

Legal framework for the consultation exercise

6. PCTs are statutorily obliged to consult with users of their services on any change to the way those services are provided. As the judge explained

"8. …. Sections 1 and 3 of the National Health Service Act 2006 (the "Act"), oblige the Secretary of State for Health to provide or secure certain medical services. By regulation 3 of the National Health Service (Functions of Strategic Health Authorities and Primary Care Trusts and Administration Arrangements) (England) Regulations 2002 (SI 2002/2375) (the "2002 Regulations"), as amended, that function has for the most part been delegated to Primary Care Trusts ("PCTs"), of which there are 152 in England.

9. PCTs commission services from "providers", including NHS Foundation Trusts to meet the needs of the populations for which they are responsible.

10. Section 242 (2) (b) of the Act imposes a duty on each body to which it applies, which includes PCTs, to consult persons to whom services are being or may be provided on "the development and consideration of proposals for changes in the way those services are provided".

7. Users of services in this case include the parents of the children for whom paediatric cardiac surgical services are provided. They must be involved in the changes through the process of consultation.
8. Apart from the statutory framework, the general law must be considered. We shall deal later in this judgment with the correct approach to an application to prevent a consultation process from taking place. At this stage, it is sufficient to describe the obligation of fairness which the law imposes on any public consultation exercise. The leading authority on this is the judgment of this court in R v North and East Devon Health Authority ex parte Coughlan (Lord Woolf MR, Mummery and Sedley LJJ) [2001] QB 213:

"108. It is common ground that, whether or not consultation of interested parties and the public is a legal requirement, if it is embarked upon it must be carried out properly. To be proper, consultation must be undertaken at a time when proposals are still at a formative stage; it must include sufficient reasons for particular proposals to allow those consulted to give intelligent consideration and an intelligent response; adequate time must be given for this purpose; and the product of consultation must be conscientiously taken into account when the ultimate decision is taken (R v Brent London BC, ex p Gunning (1985) 84 LGR 168)."

9. The Coughlan formula is a prescription for fairness. It is an aspect of fairness that a consultation document presents the issues in a way that facilitates an effective response: see, for example, R (Capenhurst) v Leicester City Council [2004] EWHC 2124 (Admin), [2004] A.C.D. 93. No doubt for that reason, as will appear below, the consultation document in this case explains at length the successive criteria for change that the JCPCT applied in this case. The consultation document must be clear to the general body of applicants: see R v Secretary of State for Transport ex parte Richmond upon Thames LBC (No.2) [1995] Env L R 390.
10. Another aspect of fairness is that it must present the available information fairly. In this case, because the JCPCT had to collect information from the centres to present the available information it would have to make clear to the centres what information it needed. A further aspect of fairness lies in the presentation of the information on which the views of consultees should be sought. The options for change must be fairly presented. Nonetheless, a decision-maker may properly decide to present his preferred options in the consultation document, provided it is clear what the other options are: Nichol v Gateshead Metropolitan Borough Council (1988) 87 LGR 435.
11. The object of requiring fairness is to ensure high standards in decision-making by public bodies, and to enable responses to be made which will best facilitate a sound decision as a result. In addition, it must achieve the statutory objective of section 242(2)(b) of the National Health Service Act 2006 of engaging users.
12. If the presentation of information inaccurately would have no material adverse effect on the process of consultation, perhaps because the error is patent, the error is unlikely to amount to unfairness when taken on its own (see generally R v Secretary of State for Transport ex parte Richmond-upon-Thames LBC (No.3) [1995] Env L R 409). However, aspects of alleged unfairness should be reviewed both individually and in aggregate. An individual aspect of unfairness may seem trivial on its own but when seen with other aspects of unfairness it may acquire greater significance.
13. If it is alleged that a consultation process is unfair, clear unfairness must be shown. As Sullivan J pointed out in R(o/a Greenpeace Ltd) v Secretary of State for Industry [2007] EWHC 311(Admin), it must be shown that the error is such that there can be no proper consultation and that "something [has] gone clearly and radically wrong".
14. On the other hand, it is sufficient to show that the unfairness affects only a group of the persons affected by the consultation: see R(Medway Council and ors) v Secretary of State for the Environment [2002] EWCA 2516 (Admin). Unfairness to the general body of consultees is not required.
15. In this case, the judge found that Royal Brompton had a "legitimate expectation" that its research information would be used in a certain way. A legitimate expectation arises where a public body such as the JCPCT makes a promise which has conferred on a person an expectation that it will act in a particular manner. The public body may be precluded from acting inconsistently with that expectation if it would be unfair for it to do so. Since legitimate expectation is relied on here to support a claim of unfairness, Mr Maclean agrees that it can be subsumed in the broader issue of unfairness.

Provisional decision-making and preparation for consultation

(a) History: the Bristol Royal Infirmary Report and the Monro Report

16. There have been a number of reports on paediatric cardiac surgery in recent years. These include the Report of the Inquiry into deaths at the Bristol Royal Infirmary chaired by Professor Sir Ian Kennedy (2001), the Report of the Paediatric and Congenital Cardiac Services Review Group chaired by James Monro (Department of Health 2003) and Surgery for Children – delivering a first class service (Royal College of Surgeons, 2007). More details about these reports can be found in §§ 26 to 29 of the judge's judgment.

(b) The Safe and Sustainable Review 2008 and continuing

17. Following the concerns expressed in these reports, on 29 May 2008 the National Health Service Medical Director, Sir Bruce Keogh, acting on behalf of the National Health Service Management Board, requested the NHS National Specialised Commissioning Group ("NSCG") to review the provision of paediatric congenital cardiac services with a view to their reconfiguration. The review became known as the "Safe and Sustainable Review" ("the Review"). In 2010 the JCPCT was established as the formal consulting body with responsibility for the conduct of the consultation on the Review and for taking decisions on issues which are the subject of the present consultation exercise. One of the aims of the Review was to develop a national service that achieved better results and which resulted in improved communications between parents and all the services in the network that saw their child. The judge set out the aims of the Review and its guiding principles in §§ 32 and 33 of his judgment.

(c) The role of the JCPCT and SCGs

18. Paragraph 10.3.2 of the Department of Health's Overview and Scrutiny of Health Guidance provides that PCTs must set up a joint committee where changes are proposed to the provision of services which span more than one PCT to make the decisions on their behalf. As the judge explained, specialised paediatric cardiology and cardiac surgical services are "specialised services", as defined in the National Specialised Services Definition Set. Specialised services are commissioned regionally by Specialised Commissioning Groups ("SCGs"), which are constituted as joint committees of the PCTs in their catchment area. There are ten SCGs in England corresponding to the ten Strategic Health Authorities. The NSCG coordinates the work of the ten SCGs and oversees pan-regional commissioning where a specialised service has a catchment area or population greater than that of a single SCG.

(d) The organisational structure established to carry out the Review

19. A number of committees were needed to complete the Review. The judge provided a helpful description of these. We will adopt the term "the Kennedy Panel", which was used in argument, for the term "the Independent Assessment Panel" used by the judge:

"35. Day to day management of the Review has been led by a project team of the NSCG (the "NSCG Team"), assisted by a number of specialist working groups, in particular:

1. a Steering Group;

2. a Standards Working Group (a sub-group of the Steering Group) and

3. an Independent Assessment Panel ("the [Kennedy] Panel")

The Steering Group

36. The Steering Group was chaired by Dr Patricia Hamilton, past President of the Royal College of Paediatrics and Child Health and Director of Medical Education in England. It comprised about 25 – 30 members drawn from professional and lay associations and commissioners representing a broad geographical spread. The original membership included Dr (now Professor) Shakeel Qureshi, a consultant paediatric cardiologist at the Evelina and then President elect of the British Congenital Cardiac Association (BCCA). It was subsequently expanded to include Professor Martin Elliott, a consultant paediatric cardiothoracic surgeon at GOSH, and a senior member of the BCCA.

37. The role of the Steering Group was originally to steer the development of proposals, reporting to the NSCG on, inter alia, the appropriate model of care, standards, and criteria for the designation of services.

38. Proposals for reconfiguration were initially to be developed by the SCGs organised into four regional zonal teams ("The SCG Collaboratives") reporting to the Steering Group. London was included within the South Eastern Zone which also comprised the East of England and SE Coast SCGs. The SCGs Collaboratives were charged with identifying reconfiguration options within their zones.

The Standards Working Group

39. The Standards Working Group was a multi-disciplinary panel of experts, set up as a sub-group of the Steering Group, to research and develop a framework of clinical and service standards. Draft Standards were to be presented to the Steering Group, then to the NSCG for endorsement. Once agreed, they were to be used to assess the existing 11 centres and their ability to provide a high quality service in the future.

The [Kennedy] Panel

40. The Kennedy Panel, chaired by Professor Sir Ian Kennedy, was tasked with reviewing each of the existing 11 providers of PCCS services and evaluating their compliance with the proposed service standards. Panel membership comprised experts in paediatric cardiac surgery, paediatric cardiology, paediatric anaesthesia/paediatric intensive care, paediatric nursing, paediatrics and child health, together with lay representatives and NHS commissioners. It was a requirement that members should have no existing or direct relationship with any of the 11 current providers.

41. In the Spring of 2009 concerns emerged as to how the arrangements for the Review would work in practice. It was considered that the process by which SCG Collaboratives would recommend centres within their zones might not result in an appropriate distribution of services. Secondly there was a question as to whether there was a body with authority to take decisions as to implementation of the Review.

42. At the end of 2009, and in the light of such concerns, the governance structure of the Review was revised. The SCG Collaboratives were disbanded. Secondly the NSCG recommended the establishment of a Joint Committee of Primary Care Trusts to act as a single body with delegated powers of consultation and decision making. In April 2010 the NHS Operations Board endorsed the proposed JCPCT subject to ministerial approval which was obtained in July 2010. Although the JCPCT was not formally constituted until it received ministerial approval, I shall refer to it throughout as the JCPCT.

43. With the creation of the JCPCT, the Steering Group's mandate was no longer to "steer" the Review, but to advise the JCPCT, the sole decision maker acting on behalf of all English PCTs, on clinical matters, including the design of the proposed congenital heart networks. The change was reflected in the Steering Group's revised Terms of Reference published in June 2010."

(e) Establishing the stages for the Review

20. The methodology of the Review was in outline as follows:

i) In January 2010, the centres which provided paediatric cardiac surgical services were asked to provide "baseline" information. This sought information as to each centre's role in research, including a list of the areas in which its research interests lay and the number of papers in peer reviewed journals published by members of the centre's paediatric cardiac surgical centre in 2008/9. According to the second witness statement of Mr Jeremy Glyde of the NSCG, the responses, together with corresponding SCG commentaries and the earlier baseline information, submitted by the centres were provided to the Kennedy Panel for information only and not for the purpose of scoring.

ii) According to the same witness statement, if there were two centres in London, these being GOSH and Evelina Children's Hospital, and Royal Brompton's existing caseload were distributed on the same lines as current distribution patterns, of the 1,482 projected procedures per year, GOSH would do 910 procedures and Evelina Children's Hospital 572.

iii) In March to April 2010, the Standards Working Group published their proposed national quality standards. They endorsed the concentration of specialist expertise into larger teams at Specialist Surgical Centres, recommending that each such centre should meet certain threshold criteria:

"C4 . . . must be staffed by a minimum of four full time consultant congenital cardiac surgeons;

C6 . . . must undertake a minimum of 400 paediatric surgical procedures per year to avoid 'occasional practice';

C7 . . . should perform a minimum of 500 paediatric surgical procedures each year."

21. The Review then proceeded through three further stages before the consultation document was issued:

1. self-assessment;

2. an assessment by the Kennedy Panel;

3. a 'configuration options assessment' to establish a shortlist of options.

22. Each of those stages needs to be considered.

(f) Stage 1: Self-assessment

23. This stage took place between March and May 2010.
24. In March 2010, each centre which wished to be designated as a Specialist Surgical Centre was required to complete a self-assessment template. This was intended as a means of obtaining evidence that the centre met specified core criteria, such as "deliverability and achievability", that would be used to inform the final recommendation. The covering letter invited the centres to ask for any clarification which they required.
25. The self-assessment template contained the following passages relevant to Royal Brompton's submissions on this appeal:

"Safe and Sustainable Self Assessment Template

Overview

1. Introduction and process

At the request of the NHS Management Board, the NHS Medical Director has asked the National Specialised Commissioning Group (NSCG) to undertake a review of the provisions of paediatric cardiac surgical services in England with a view to reconfiguration.

Safe and Sustainable has been set up to take this forward.

The objectives of the programme are:

To ensure a safe paediatric cardiac service now and in the future.

To ensure equality of service provision across England, where patient access to services is reasonable and appropriate.

This will be achieved by:

…

Developing criteria for the designation of specialist paediatric congenital cardiac services.

…

2. Evaluating process and scoring

Evaluation process

The evidence you supply in this exercise will be assessed as part of the evaluation process we will undertake, and therefore will ultimately inform the final recommendation.

The entire evaluation process has 2 discrete stages- Assessment Evaluation and Configuration Evaluation. This process will fulfil the first stage of the assessment evaluation.

The second stage of the Assessment Evaluation will be visits by the Assessment Panel to each centre. …

…

It should be noted that the criteria and scoring process for the Configuration Evaluation have not yet been determined. This will be communicated to all stakeholders in due course. However, the criteria and scoring for the Configuration Evaluation is separate from the Assessment Evaluation. The information supplied in the assessment stage of the process will not have any direct bearing on the scoring of the configuration evaluation process." (emboldening added)

26. There then followed a flowchart which demonstrated the way in which individual scores would be determined and the part those individual scores would play thereafter in the reconfiguration of paediatric cardiac surgical services. The flowchart showed the responses to the self-assessment template being scored and the scores being adjusted if appropriate as a result of the visit of the Kennedy Panel. This would result in the individual scores of the centres being fixed.
27. The flowchart showed that the next stage would be to determine the criteria for the configuration of paediatric cardiac surgical services. The individual scores would then be used to identify the reconfiguration criteria. These criteria would be tested against ease of access, effect on other services and other criteria.
28. The narrative gave more detail about the range of scores used to achieve individual scores:

"For the Self Assessment Evaluation Stage, each question within the 9 self assessment criteria will be scored individually, as indicated below:

1	Inadequate (no evidence to assure panel members)
2	Poor (limited evidence supplied)
3	Acceptable (evidence supplied is adequate, but some questions remain unanswered or incomplete)
4	Good (evidence supplied is good and the panel are assured that the centre has a good grasp of the issue)
5	Excellent (evidence is exemplary)




Each question within that criterion will then be weighted according to the stated multiplier, in order to reach a final score for each question. The sum of these final scores will be the total score for that criteria.

The total scores for each criterion will come together as a final score for each centre.

It should be noted that a score of below 3 for any question may raise concerns about the centre's ability to be successful during the Configuration Exercise."

29. The self-assessment template explained what criteria were being used to ascertain individual scores and what evidence was being sought in respect of them. There were nine criteria. The first was "Leadership and Vision". The evidence here included information as to the way the paediatric care team "works to learn, develop and grow", taking into account learning from practice, national and international research evidence, best practice and multidisciplinary working. Centres were asked to include an example of innovative working that the centre had undertaken and how this had had benefits in clinical care.
30. The eighth criterion for individual scores was "Ensuring Excellent Care". This was subtitled "core requirement 7" and was described as follows:

"Each Tertiary Centre must have a dedicated management group for the internal management and coordination of service delivery. The group must comprise the different departments and disciplines delivering the service.

All clinical teams will operate within a robust and documented clinical governance framework that includes clinical audit, including in outreach centres.

Each Tertiary Centre must have, and regularly update, a research strategy and programme that documents current and planned research activity, the resource needs to support the activity and objectives for development. The research strategy must include a commitment to working in partnership with other centres in research activity which aims to address research issues that are important for the further development and improvement of clinical practice, for the benefit of children and their families.

31. Centres were then asked to provide evidence of their research programmes:

"Please attach the following documents in support of this core requirement:

- Clinical governance framework and process

- Research strategy and programme"


32. Royal Brompton submitted its self-assessment in May 2010. It included over 100 supporting documents grouped into 20 appendices. In response to Core Requirement 7 it said, inter alia:

"[Royal Brompton] has a clear and accountable research strategy and infrastructure (appendix 20e). Our willingness to work with other centres is evidenced by several of our recent studies including several national epi-immunological studies in congenital heart disease and the National multi-centred NIHR-funded "Chip" trial which ran at Royal Brompton …. The Trust has recently restructured its research and development arrangements including the recruitment of a new Associate Director of Research. A key aim of these changes is to improve the alignment of the Trust research activity with the objectives of the NHS at large."

33. Appendix 20e to the response contained "The Trust Research Strategy." This was a document extending to forty-five pages. It was expressed to contain an evaluation of the research activity of the Trust.
34. The centres were later sent a second template concerning the specialised nationally commissioned services that rely on cardiac surgery: paediatric heart and lung transplantation, complex tracheal surgery and respiratory extracorporeal membrane oxygenation ("ECMO"). Centres which did not currently provide such services were asked whether they wished to seek designation to do so following reconfiguration. Royal Brompton does not provide such services and did not seek designation to do so in future.
35. Under a section of the second template headed "Other implications for reconfiguration", information was requested about "the likely impact on PICU if your centre was not designated".
36. The judge refers to this template at §§ 49 and 50 of his judgment, but not to any response to it. Our attention was not drawn to any such response.

(g) Stage 2: Assessment by the Kennedy Panel

37. On receipt of the self-assessments, the Kennedy Panel agreed initial scores for each centre. The Kennedy Panel undertook a round of visits to the centres in May/June 2010, visiting Royal Brompton on 9 June 2010. Following the visits, the self-assessments and scoring were reviewed, and each centre was given a score measuring its current and future compliance against the criteria.
38. In its report, the Kennedy Panel noted that Royal Brompton had a good track record with clinical research but "the panel felt that this had recently slipped and the research undertaken by the two [Biomedical Research Units ("BRUs")] at the Trust [was] not relevant to paediatric cardiac surgery."
39. However the Kennedy Panel was unable to assess deliverability and achievability. This matter was left to the JCPCT.
40. In the report of the Kennedy Panel, Evelina Children's Hospital received the highest score of 535 of all the eleven centres in England. Royal Brompton and GOSH were equal fourth with a score of 464 after Southampton General Hospital and Birmingham Children's Hospital.

(h) Configuration options assessment from short listing to finalisation of the options

(i) Short listing the options

41. There were over 2,000 possibilities for reconfiguration in theory. The JCPCT eventually produced a short list of some 12 options, which was increased to 14 before the consultation document was issued. It is not necessary to go through the short listing process which the JCPCT adopted for the purposes of this appeal save to note that in July 2010 it agreed inter alia that there should be at least two centres in London and also that the number of options was reduced to eight so as to exclude Royal Brompton as it had no advantage (say) of access. However, the JCPCT determined that Royal Brompton should be included for the purpose of evaluating the London centres against the evaluation criteria.
42. The twelve options on the JCPCT's original short list comprised:

i) Four seven site options with two centres in London (as at the previous meeting).

ii) Four six site options with two centres in London (one of which had been presented at the last meeting).

iii) Four three London centre options.

(ii) Setting the weighted evaluation criteria

43. In June/July 2010 the NSC team, acting on the advice of the Steering Group, consulted stakeholder groups both as to the proposed criteria and as to the weightings to apply to such criteria for the purpose of the Configuration Evaluation. The stakeholders included SCG directors, parents who had registered for one of the 2010 engagement events and five clinicians nominated by the current surgical centres. They were notified that it would be for the JCPCT to agree the evaluation criteria and the weightings to be applied to them.
44. The evaluation criteria agreed by the JCPCT for assessing the options were as follows:

"(1) Quality: (a) centres will deliver a high quality service; (b) innovation and research are present; (c) clinical networks are manageable;

(2) Deliverability: (a) high quality NCSs will be provided; (b) the negative impact on other interdependent services will be kept to a minimum, as will negative impacts on the workforce;

(3) Sustainability: centres are likely to perform at least 400-500 procedures; will not be overburdened and will be able to recruit and retain newly qualified staff.

(4) Access and travel times: negative impact of travel times for elective admissions are kept to a minimum; retrieval standards are complied with."

45. Innovation and research had originally been factors taken into account by the Kennedy Panel when assessing "Leadership and Strategic Vision" and "Ensuring Excellent Care", but had not been given a discrete score. The Kennedy Panel was therefore asked to reconvene, and separately to assess the capacity for research and innovation of each of the centres.
46. The panel met for this purpose on 14 December 2010. The Kennedy Panel received advice from KPMG. The Kennedy Panel arrived at the following scores for research and innovation for the three London centres:

Evelina 5

GOSH 5

Royal Brompton 2

47. In its assessment as of December 2010, the Kennedy Panel stated that it felt that "not all the research undertaken and referred to by [Royal Brompton] during the assessment visit applied to paediatric cardiac surgery" and that its "research strategy had insufficient reference to paediatric cardiac surgery." On the other hand the research at Royal Brompton into foetal medicine was regarded as strong.
48. The JCPCT then approved the weightings to be given to the evaluation criteria. The weighting for access and travel times was 14, for quality 39, for deliverability 22 and for sustainability 25. The high weighting given to quality shows the importance attributed to that criterion.

(iii) Identifying the best options: determining the preferred two-centre option for London

49. At the meeting of the JCPCT on 11 January 2011, fourteen options and supporting analysis were presented to the JCPCT and were examined in detail, two further options to the twelve before the committee on 28 September having been added by the NSC team at the request of the JCPCT. (At a later stage, the JCPCT narrowed the options to eight in number, and then to six and then to four).
50. The JCPCT determined that the consultation should proceed on the basis that proposals incorporating two sites in London were preferred. The next issue was which two London centres.

(iii) Determining which two of the London centres was to be included in the preferred two-centre option

51. For this purpose the London centres were scored against each other without reference back to them. Evelina Children's Hospital and GOSH received the highest scores. The JCPCT went on to determine that the consultation should proceed on the basis of an expressed preference for GOSH and Evelina Children's Hospital over Royal Brompton as the London centres.
52. That decision was arrived at by applying the scoring of the London centres by the Kennedy Panel against the four weighted evaluation criteria: Quality, Deliverability, Sustainability and Access and Travel times, the weighted criteria and the scoring having received the approval of the JCPCT.
53. The centres received different scores only in "Quality" and "Deliverability". The difference in "Quality" was attributable to Evelina Children's Hospital's higher overall score by the Kennedy Panel (ranked first amongst the 11 centres). In "research and innovation" both Evelina Children's Hospital and GOSH had been scored the maximum of 5 by the Kennedy Panel, whereas Royal Brompton had scored 2. However, research and innovation was not the only component taken into account in establishing Quality because the excellence of the service provided by each centre was also taken into account.
54. Under Deliverability, the difference in scores was attributable to two elements; first the benefit to the country of maintaining the provision of three nationally commissioned services at GOSH, GOSH being one of three centres providing ECMO, one of two providing transplantation services and the sole provider of complex tracheal surgery. The second element was the assessment that the loss of Royal Brompton's PICU, supporting predominantly cardiac patients, would present a limited risk to local and national PICU provision.
55. The overall result of the scoring against the weighted criteria for the London Centres was:

Evelina 364

GOSH 347

Royal Brompton 264

(iv) Finalising the decision to put the options out to consultation

56. On 16 February 2011, the JCPCT met in public to discuss and finally to agree the preferred options to be put out to consultation, the consultation document, and the form the consultation was to take. Before inviting questions, Sir Neil McKay concluded the formal session by saying:

"Let me say categorically, the consultation exercise is what it says on the tin. We are open minded about the outcome, we are prepared to listen to alternative views, as we said on three occasions during the course of the afternoon, and we will move forward with further discussions in the autumn ..."

The consultation document

57. The consultation document was issued on 1 March 2011. Consultees were asked to consider four out of the fourteen options that had been considered by the JCPCT. However there was no option for more than 2 London centres and the London centres in the options put forward were Evelina Children's Hospital and GOSH.
58. The consultation document explained in some depth the process whereby the four options had been arrived at. It set out how the centres received their individual scores, and how a short list of options had been drawn up. It noted that one of the principles adopted was that London required at least two centres due to the size of the population. It stated that the population served by London included the East of London and South East England.
59. The consultation document also explained how of the fourteen options considered only six were viable. Two of these involved three sites for London. Next, the consultation document explained how the weighted evaluation criteria had been arrived at and showed the result of applying those criteria to the six viable options:

	OPTION 2 [7 sites – 2 for London]	OPTION 6 [6 sites- 2 for London]	OPTION 8 [6 sites- 2 for London]	OPTION 10 [7 sites- 3 for London]	OPTION 12 [7 sites- 3 for London]	OPTION 14 [Top 7 sites scoring 2 for London]
Access and travel	56	14	42	14	42	14
Quality	117	117	117	117	117	156
Deliverability	66	44	22	44	22	66
Sustainability	75	75	50	75	50	50
TOTAL	314	250	213 [error: should have been 231]	250	213 [error: should have been 231]	286

60. The two lowest scoring options were options 8 and 12. The options which provided for three London centres, i.e. options 10 and 12, were then eliminated so that there were then four recommended options for consultation, which were described as follows:

"The final recommended options for consultation are:

Option 2 is viable as it is consistently the highest scoring potential option

Option 14 is retained…

Option 6 is viable

Option 8 is viable"

61. The consultation document then set out in a box ("the London box") on the same page the reasons for preferring the two-centre option for London:

"LONDON

It was recommended to the Joint Committee of Primary Care Trusts that Options 10 and 12 (which included 3 centres in London) should not form part of the public consultation for the following reasons:

The Joint Committee of Primary Care Trusts recommends that two designated centres is the ideal configuration for the population of London, East of England and South East England. The question of whether two centres in London is the right number will be asked during consultation.

The forecast activity levels for London and its catchment area (currently around 1,250 paediatric procedures per year) mean that two centres would be well placed to meet the proposed ideal number of 500 procedures a year. This could only happen with three London centres if patients were diverted from neighbouring catchment areas into London. Our analysis shows this would significantly, and unjustifiably, increase travel times and impact on access for patients outside of London, South East and East of England.

The advice of the Safe and Sustainable Steering Group is that two centres, rather than three, are better placed to develop and lead a congenital heart network for London, South East and East of England according to the Safe and Sustainable model of care."

62. The consultation document then turned to the question "Which 2 centres in London?". It set out (on pages 95-96) a table stretching across two pages, giving the scores of the three London centres, weighted as explained above, which had been the basis of its decision as to which two London centres to recommend. That table is set out below but the columns containing (1) an analysis of the evaluation criteria, (2) the individual scores given to the centres by the Kennedy Panel and (3) the weightings, have been omitted because that information has been set out above.

	GOSH weighted score	Evelina Children's Hospital weighted score	Royal Brompton weighted score
Access and travel times	42	42	42
Quality	117	156	78
Deliverability	88	66	44
Sustainability	100	100	100
Total	347	364	264

63. Evelina Children's Hospital was the clear leader. GOSH's score exceeded that of Royal Brompton's only for Quality and Deliverability. The difference was explained thus in the consultation document:

"QUALITY

The proposed score for the Evelina Children's Hospital reflects the results of Sir Ian Kennedy's panel assessment of its capacity for 'research and innovation' (refer to map on page [102]).

Similarly Great Ormond Street Hospital and the Royal Brompton Hospital were ranked equally by the panel, but the higher score for Great Ormond Street is due to its capacity for 'research and innovation'. Because they are already close together, there is unlikely to be an impact on the sub-criterion of 'manageable networks'.

DELIVERABILITY

As Great Ormond Street Hospital would retain three nationally commissioned services in their current location (cardiothoracic transplantation, ECMO and complex tracheal surgery) we recommend it scores higher in potential configuration options. Because the PICU at the Royal Brompton Hospital exists predominantly to support cardiac surgery, we propose it is scored lower than the Evelina Children's Hospital on the sub-criterion involving 'the negative impact for the provision of paediatric intensive care and other interdependent services is kept to a minimum'."

64. There was more detailed analysis of the evaluation criteria as they applied to all the centres on the pages which followed. Those included on page 102 a map of England and Wales showing scores for research and innovation. This showed that Royal Brompton had achieved a score of 2 and there was a key to the map which showed this meant "Poor - limited evidence received".
65. The consultation document explained that the effect of the preferred option for two London centres, not including Royal Brompton, was that the PICU would have to be closed. However, as this predominantly supported cardiac surgery and because there was existing PICU provision in London, it considered that this involved little risk to local and national paediatric intensive care provision.

The response form

66. The form stated:

"Q7 Before answering this question, please read pages 93-96 in the Safe and Sustainable Consultation Document. Do you support the proposal for two Specialist Surgical Centres in London?

PLEASE TICK ... ONE BOX ONLY

... Yes – support the proposal for two Specialist Surgical Centres in London

... No – do NOT support the proposal for two Specialist Surgical Centres in London

... Don't know

ALL TO ANSWER

Q8 What, if any, comments do you have on the number of Specialist Surgical Centres in London?

PLEASE SUMMARISE YOUR KEY COMMENTS IN THE BOX BELOW

ALL TO ANSWER

Q9 Before answering this question, please read pages 93-96 in the Safe and Sustainable Consultation Document

It is proposed that the two Specialist Surgical Centres in London will be Great Ormond Street Hospital for Children NHS Trust (GOSH) and Evelina Children's Hospital – Guy's and St Thomas' NHS Foundation Trust.

If there were to be only two Specialist Surgical Centres in London, please indicate whether you support this choice (i.e. GOSH and Evelina Children's Hospital), or whether you think that the Royal Brompton & Harefield NHS Foundation Trust should replace one of these other two London hospitals?"

Further consultation document and processes during the consultation period

67. In May 2011, the JCPCT published a further paper "Safe and Sustainable – Improving children's congenital heart services in London". The introduction contained the following paragraph:

"At this half way stage in the public consultation on the future of children's congential cardiac services, now is an appropriate time to look at the issues that have been raised so far and focus on the unique situation in London. Every other surgical centre is the sole centre in its city or region; London has three centres close together."

68. The paper went on to set out the case for two children's heart surgical centres in London.
69. The consultation formally concluded on 1 July 2011. During the four month consultation period, about 50 public events were held and approximately 55,000 written responses were received.

Responses to consultation

70. Some 77,000 responses were received in response to the consultation. According to an analysis dated 24 August 2011 prepared by Ipsos MORI, about three quarters of respondents supported the proposal for two specialist surgical centres in London. However this dropped to just under half of the individuals in London, with many of these suggesting that all three hospitals in London should retain heart surgical services for children. They noted that all three hospitals provided high quality care, and stated that they would like to see them work together to provide services. Some had concerns that two centres in London would not be able to cope with the demands of its population. If there were to be two centres in London, the majority of consultees supported GOSH and Evelina Children's Hospital. Consultees were positive about Royal Brompton and the quality of its research. On the question which two centres should be chosen for London, the JCPCT's preferred option was the most popular both when all responses were considered and when those of individuals were considered.

The judge's judgment

71. As already explained, Royal Brompton only succeeded on the last of five grounds.
72. Pre-determination: The first point was whether the two-centre option comprising GOSH and Evelina Children's Hospital had been pre-determined so that there was no proper consultation on this issue. The judge held that it was clear from Q7 and Q8 of the response form that it was open to a consultee to take issue with the proposal for two London centres and from Q9 that a consultee could take issue with the exclusion of Royal Brompton. So, on a fair reading of the consultation document, the issue had not been pre-determined against Royal Brompton. Other documentary evidence and witness statements relied on by Royal Brompton were to the same effect.
73. Irrationality: Royal Brompton's second ground for challenging the consultation was irrationality in the rejection of the three-centre option for London. The judge held that this also failed for two reasons. First the decision to make the two-centre option the preferred option was not justiciable since the proposal was still at a formative stage. Secondly, there was no obligation to consult on all the viable options. The JCPCT could properly have a preferred option.
74. Misinformation: The third ground of challenge was that passages in the consultation document so distorted the position as a result of inaccuracies as to preclude a properly informed response and thus to make the consultation process unfair. In particular reliance was placed on the London box and the statement that with two centres London would be well placed to meet the proposed ideal number of 500 procedures per year. In fact the number of procedures per year in London was understated in the London box. Elsewhere in the consultation document there were stated to be a projected 1,485 procedures per year in London. That meant that if there were three centres in London the criterion that centres should ideally perform 400 to 500 procedures per year would still be met. The judge did not consider that this was misleading. The intention was to state the likely caseload from the resident population. Likewise the judge did not accept that a statement in the consultation document that two centres was the ideal configuration for London suggested that the three-centre option was not viable.
75. The judge also rejected the argument that Royal Brompton should have been scored more highly on Deliverability because its PICU would suffer as a result of the closure of its centre. The JCPCT had commissioned a report on this from Mr Adrian Pollitt. This report confirmed that the closure of Royal Brompton's PICU would not cause the closure of Royal Brompton's respiratory services.
76. Bias: The fourth ground of challenge was that the Steering Group, which had recommended the two-centre option, had included Professor Qureshi, a consultant paediatric cardiologist at Evelina Children's Hospital, and Professor Elliott, a consultant paediatric cardio-thoracic surgeon at GOSH (see §36 of the judge's judgment quoted in paragraph 19 above). The judge rejected this challenge on the grounds that the role of the Steering Group was to make recommendations to the JCPCT. The judge found that there was no doubt that the Steering Group limited its role to giving advice from the clinical perspective. He distinguished the decision of this court in R (Goldsmith) v Wandsworth LBC [2004] EWCA Civ 1170.
77. Legitimate expectation: Royal Brompton complained of a failure of the JCPCT to meet a legitimate expectation derived from the self-assessment template that the criteria and scoring in the evaluation undertaken by the Kennedy Panel would be "separate" from the configuration evaluation and that it would have "no direct bearing" on it (see the second passage emboldened in paragraph 25 above). The judge held that there was a clear and unequivocal representation in the self-assessment template. In the judge's judgment, the scores reached by the Kennedy Panel on innovation and research had clearly been used in, and had therefore a direct bearing on, the scoring for the purposes of the configuration evaluation.
78. The judge rejected an argument (which the Appellants say was not advanced) that, if Royal Brompton had been given a further opportunity to respond, it would not have made any difference. He set out a long passage from the witness statement of Dr Duncan Macrae, paediatric intensivist and Director of Paediatric Intensive Care at Royal Brompton and the President of the International Paediatric Cardiac Intensive Care Society. In it, Dr Macrae explains that Royal Brompton took the view that the information sought in the self-assessment template was not about the content of Royal Brompton's research but about its governance. The judge held that Dr Macrae's understanding of the issues at which the self-assessment template was directed was fully justified, and that the assessment of research and innovation did not fully reflect the response of Royal Brompton to the self-assessment template.
79. The judge also referred to the evidence of Professor Timothy Evans, a consultant of Intensive Care and Thoracic Medicine at Royal Brompton and professor of Intensive Care Medicine at Imperial College, which stated that there could be a benefit to paediatric cardiac surgical services from wider research. The judge held that it was not for him to assess the research done by Royal Brompton. However, on the basis of the evidence before him he rejected the argument (which the Appellants say was not advanced) that it would not have made any difference to the assessment of Royal Brompton by the Kennedy Panel to have asked for further information and explanation about Royal Brompton's research. Royal Brompton might, if full information about its research had been available to the Kennedy Panel, have been rated equally with GOSH.
80. The judge held that it was no answer that Royal Brompton would have been placed third in the individual scoring of the London centres against the configuration criteria. Consultees would have seen, for instance, that Royal Brompton's research had been described as "poor" in the consultation document. On that basis the failure to meet Royal Brompton's legitimate expectation had led to the consultation process being seriously distorted. The judge went on to hold that, if consultees had been given the information that Royal Brompton's research was rated equally with that of GOSH, they might have favoured a three-centre option. This might have happened despite the fact, even taking 1,482 as the aggregate annual caseload, two out of the three centres would have had a caseload of between 400 and 500 each year.
81. In short, the way in which the scoring of "Quality" had been carried out rendered the consultation exercise unfair to Royal Brompton. Accordingly, the consultation exercise was unlawful and had to be quashed.

Rescoring of research by the Kennedy Panel – February 2012

82. Following the judge's judgment, the JCPCT invited the centres to submit additional evidence on research and innovation. Royal Brompton was among the centres that submitted further research material. The Kennedy Panel increased its score to 3. In the course of its assessment it noted that:

"While recognising [Royal Brompton's] reputation in the field of clinical research, in the panel's opinion the evidence submitted by [Royal Brompton] is limited in its references to paediatric cardiac surgical services and paediatric interventional cardiology services."

83. The new score would have increased its score in the "which centre?" table to 299, but still left the weighted score in that table for its paediatric cardiac surgical services behind that of Evelina Children's Hospital and GOSH.

The issues on this appeal and the respondent's notice

84. The JCPCT contends that the self-assessment template did not give rise to a legitimate expectation that the responses would not be used in the configuration criteria. The self-assessment template sought information about research and made it clear that the configuration criteria had not then been established. The judge was wrong to say that it would have made a difference if Royal Brompton had produced more information about its research. The judge went into the question whether the research by Royal Brompton more generally into cardiac surgery was relevant to paediatric cardiac surgery and should not have done so.
85. The JCPCT also argues that, even if there was a such legitimate expectation, there was no unfairness to Royal Brompton since it had supplied information in its response to the first self-assessment template, all centres were treated alike and the information was used only to inform an assessment relied on in the consultation document. In other words, Royal Brompton had the opportunity to challenge the assessment before any decision was made.
86. As explained below, Royal Brompton has served a respondent's notice and it seeks to uphold the judge's judgment on five additional grounds, as summarised below.

Discussions and conclusions

87. The arguments put to us on the appeal and the respondent's notice can now be succinctly stated in the light of the very detailed exposition of the facts and other matters given above. They are all subject to the same threshold objection: the act challenged was a consultation process, not the final decision of a public body. One of the functions of a consultation process is to winnow out errors in the decision-maker's provisional thinking. The JCPCT owes a public law duty to reconsider matters in the light of responses. True consultation is not a matter of simply "counting heads": it is not a matter of how many people object to proposals but how soundly based their objections are.
88. Moreover, the process of reconsideration is a public and transparent one. If a public body fails to consider a significant matter or to reach a reasonable result by doing so, its further decision is liable to be the subject of challenge. A well-governed public body, with the public interest at heart, will be one that is able to concede in appropriate circumstances that its provisional thinking has been proved to be wrong. Nor should the court overlook the possibility that, following receipt of consultation responses, the decision-maker may conclude that no decision is yet possible. That would mean that nothing may be gained by complex litigation at the consultation stage.
89. It is of course difficult to know at the earlier stage whether the decision will be persisted in after consultation. Intervention at the earlier stage may also cause wasteful, harmful or avoidable delay, particularly where consultation is conducted on the scale on which it was conducted in this case. On the other hand, there will be cases where it is appropriate to grant some form of relief in relation to a consultation process, not least because applications for judicial review must be made promptly. Nonetheless, the judge may properly conclude that, even though there has been a public law wrong, the matter is best dealt with by refusing relief and allowing the decision-maker to consider the matter following completion of the consultation and an opportunity to take the appropriate action at that stage.
90. A further reason for caution was suggested by Miss Dinah Rose QC, for the JCPCT. The decision-maker has to balance the interests of several different groups, not simply those represented before the court. The decision-maker may be in a better position to do this effectively and in such as way as to prevent the interests of one particular group receiving inappropriate precedence over the interests of other groups.
91. Furthermore, the JCPCT, like any other public body, has a toolbox full of tools at its disposal to deal with objections that need further consideration. It can engage in further consultation, as it did in May 2011. It can commission a further expert report as it did in the case of the Pollitt report. It can return to the centres and ask them for more information, as it did in the rescoring exercise. It can ask the Kennedy Panel to undertake further work, as it did prior to completing the configuration scoring exercise, and so on. None of these tools in the JCPCT's toolbox needs or ought to need an order of the court.
92. Not all objections to the accuracy of a consultation process will lead to a full reconsideration of provisional decisions. It is not enough, therefore, for a party seeking to quash a consultation exercise to point to some facts that are inaccurately presented. Their inaccuracy may on reflection lead only to a minor and immaterial scaling down of the case supporting the provisional proposal for change. The arguments for change will not then be as black and white as they appeared in the consultation document, but different shades of grey. Determining the strength of those shades of grey is generally not a matter for the court but the decision-maker.
93. In short, it is inherent in the consultation process that it is capable of being self-correcting. This has to be borne clearly in mind. For the various reasons already indicated, the courts should therefore avoid the danger of stepping in too quickly and impeding the natural evolution of the consultation process through the grant of public law remedies and perhaps being led into areas for the professional judgment of the decision-maker. It should, in general, do so only if there is some irretrievable flaw in the consultation process.
94. In this case, the JCPCT took action following the judgment of the judge. The JCPCT very properly gave Royal Brompton the opportunity it had sought of providing further information about its research. It did so. Its scoring was increased. While the rescoring was not sufficient to alter the comparative scoring with the other London centres, it demonstrated that the research and innovation of Royal Brompton was not as stated in the consultation document "poor". Its research and innovation proved to be acceptable, despite being, in the assessment of the Kennedy Panel, peripheral in its subject matter to paediatric cardiac surgical services. The reaction of the JCPCT shows that it is, as its public responsibility demands, prepared to be flexible in implementing the Review. If the JCPCT had taken that position before the judge, it might well have been that the appropriate course was for the case to be adjourned to allow the rescoring to be carried out. As it is, the rescoring has been done in the period pending appeal.
95. Turning to the arguments addressed on the appeal, it is clear that they must be addressed on the basis that it is no longer asserted that the consultation document pre-empted any consideration of the three-centre option for London or that there should be any combination of two which included Royal Brompton. It is clear from the passages in the consultation document and from the response form that consultees could put forward the three-centre solution if they wished to do so.

Legitimate expectation

96. On the judge's holding on legitimate expectation, Miss Rose submits that the self-assessment template did not contain a clear representation that the response would not be used as part of the scoring for the purposes of the configuration evaluation. It was clear that the centre had to send in details of its research as it was asked to attach its research strategy and programme. If Royal Brompton was in any doubt as to what was required it could have asked for clarification as stated in the covering letter. The statement about the configuration evaluation being a separate exercise and the response not having "any direct bearing" on the scoring of the configuration exercise were ambiguous and equivocal statements.
97. This statement in the self-assessment template had to be read with the first paragraph under the cross-heading "Evaluation process" which made it clear that the response would inform the final recommendation. The two processes of scoring individual centres and scoring options for the purpose of the reconfiguration were in any event separate exercises. The position surrounding the use of the data provided by the centres was in any event complex. The configuration criteria were derived from the data produced by the centres as a whole. Individual scores were only used at a second stage after it was determined whether there should be two or three centres for London. There could be no promise at the time of the self-assessment template that the responses would not be used for the purposes of reconfiguring the services because the configuration evaluation criteria had not then been established.
98. In any event, Miss Rose accepted that legitimate expectation in the present context was simply a subset of the wider concept of fairness. So Royal Brompton had to show that some unfairness was caused. In the present case, any breach of a legitimate expectation did not lead to any unfairness. The initial analysis of responses to the consultation prepared by Ipsos MORI showed the contrary. Consultees responding to the consultation document were prepared to express a preference for a solution that included Royal Brompton notwithstanding the way in which its research and innovation had been described in the consultation document. So the judge was wrong to suppose that it was unrealistic to expect consultees to support Royal Brompton, given the way in which its score for research and innovation was described in the consultation document. The Ipsos MORI analysis showed that the three-centre London option was not regarded as precluded by the poor scoring attributed to Royal Brompton's research and innovation score.
99. In any event, all consultees had been treated alike. Furthermore, Royal Brompton had now had the opportunity to put in further research and have that research rescored. However, the result of the rescoring had not been such as to alter its position vis-à-vis Evelina Children's Hospital and GOSH. Their research was considered to have greater relevance to paediatric cardiac surgical services. Royal Brompton had also had the opportunity of making all its points in the consultation process. This was, on Miss Rose's submission, the route Royal Brompton should have chosen for pursuing its objections to the preferred options of the JCPCT. Moreover on her submission, the question how Royal Brompton's research should be scored was a matter for the Kennedy Panel and the court should not enter into the debate whether or not it was peripheral to paediatric cardiac surgical services.
100. Mr Alan Maclean QC, for Royal Brompton, was quick to point out that GOSH and Royal Brompton both scored 464 on the original assessment by the Kennedy Panel. He submits that research and innovation had been used so as to separate the two centres and that the responses received from centres had clearly been used in the evaluation of the configuration options because there was no other source for the research and innovation criterion used to decide which combination of two London centres was to be preferred. Therefore the responses to the self-assessment template had a direct bearing on the formation and application of the configuration evaluation criteria. Moreover, he submits that research and innovation were the sole component of the criterion for quality used for deciding which two London centres were preferred. In fact, the evidence shows that that is not the case.
101. Mr Maclean submits that it is no answer that all the centres were treated in the same way. That factor would not prevent there being unfairness.
102. In the light of the express request for documents made by the self-assessment template, we accept the argument that it was reasonably clear from the self-assessment template that centres were required to provide details of the content of its research programme with their response to the self-assessment template. They were indeed expressly asked to send in their research programmes. The criteria for such research programmes were described in the passage set out in paragraph 30 above. The fact that one centre, Royal Brompton, interpreted the requirements in a different way was not the test. In fact, Royal Brompton did send in a substantial response although it is clear from its later submission for the purposes of the rescoring that there was further information that it could have sent in response to the self-assessment template but did not submit originally. Moreover, the details of Royal Brompton's research sent in response to the baseline self-assessment template were also provided to the Kennedy Panel for information. In the context of a reasonably clear request to send in relevant research information followed by the submission of a substantial document on research, the consultation process cannot be said to be unfair. In those circumstances, the JCPCT was entitled to proceed on the basis that Royal Brompton had said what it wished to say on the subject. There was nothing to suggest that it had not done so. Contrary to Mr Maclean's submission, it makes no difference that the Kennedy Panel did not ask about research on their visit to Royal Brompton.
103. Miss Rose suggests that the judge looked at the matter subjectively rather than objectively in § 167 of his judgment. We do not agree. The judge came to the same conclusion as Dr Macrae, which we respectfully consider to be erroneous about what the self-assessment template requested. The meaning of the self-assessment template is a mixed question of law and fact but, in so far as it is a question of fact, this is not one of those situations in which an appellate court defers to the finding of the trial judge. That occurs only if the judge has had the benefit of hearing witnesses give oral evidence.
104. For a legitimate expectation to arise in public law, a public body must make a clear and unequivocal representation. There was not, as we see it, a representation of the requisite clarity made by the self-assessment template about non-use of the information provided for the purposes of evaluating the configuration options.
105. The exercises of individual scoring and configuration evaluation were certainly "separate" exercises. The only representation that could have given rise to a legitimate expectation was the statement that the responses would have no "direct" bearing on the configuration evaluation criteria or scoring. However, it is obscure as to what constitutes a "direct" bearing. That expression had to be read in the context of the statement made in the same part of the self-assessment template that the evidence supplied by the centre would "ultimately inform the final recommendation."
106. The mere fact that the assessment of the Kennedy Panel intervened did not, as suggested, mean that the evidence had no direct bearing: no reasonable reader would read it in that way. What is, however, clear is that the research and innovation response to the self-assessment template, as moderated following the visit to the centre by the Kennedy Panel, was a component of one of the criteria for the configuration exercise, namely the criterion of Quality. Therefore the response to the self-assessment template did have a substantial influence on the ultimate result.
107. We would therefore agree that, had the statement stood alone, there was a representation that the information on research supplied pursuant to the self-assessment template would not be used in the significant way that it was used for the configuration evaluation exercise. But the statement does not stand alone. It has to be read in the context of the further statement that the information provided would inform the final recommendation. The flowchart described above was to the same effect. The statement also has to be read against the background that research would obviously be a factor in the configuration exercise for paediatric cardiac surgery services, and that the JCPCT would be unlikely to ask for the same information twice.
108. However, if there was a representation sufficient to give rise to a legitimate expectation (and assuming at this stage of our assessment of the issues on this appeal that there were no materially misleading statements in the consultation document), the position following the issue of the consultation document was not irretrievable. Royal Brompton could have submitted the further information that it had about research and innovation to the JCPCT as part of its response on consultation. Certainly no relief could be granted at this stage as the JCPCT has very sensibly rescored the research.
109. Accordingly, in our judgment, the judge was wrong on legitimate expectation, which was the ground on which Royal Brompton won below.
110. It must be emphasised that this court is concerned only with the process adopted by the JCPCT. It is not concerned with any substantive objection which Royal Brompton may residually have to the assessment of its research programme: that assessment was a matter for the Kennedy Panel.

Bias

(i) Professor Qureshi and Professor Elliott

111. Mr Maclean's skeleton argument introduces his case on bias in this way:

"The consultation was unfair in that it was tainted by apparent bias arising from the involvement in the Steering Group (the recommendations of which were accepted by the JCPCT) of senior consultants from the two London hospitals which were ultimately favoured by the JCPCT, i.e. Evelina and GOSH. The Royal Brompton tried its best to secure a place on the Steering Group for one of its clinicians, but these efforts were rebuffed. There is evidence that some members of the Steering Group were concerned about the bias aspect, but these concerns were swept aside."

112. Before going any further, it is necessary to unpack this allegation. If a deliberative body's advice is to be vitiated by the apparent bias of one or more of its members, the bias cannot depend on the outcome. Outcome goes only to entitlement to relief. In principle, any complaint of apparent bias which can be made in retrospect is one which could equally have been advanced prospectively, always provided the facts were known.
113. Thus, for example, if Evelina Children's Hospital or GOSH had claimed to have objective reason to fear that a panel containing one of their own consultants might bend over backwards, or might cause the JCPCT to bend over backwards, in order to show them no favour whatever, they too would have been able to challenge the composition of the panel for apparent bias. But to have placed a Royal Brompton consultant on the Steering Group, as if it were a negotiating body representing antagonistic interests, would have compounded, not resolved, the difficulty. The claimant cannot complain simultaneously that the presence on the Steering Group of consultants from the other two London hospitals gave rise to apparent bias and seek a place on that body for a clinician of its own.
114. The question has accordingly to be whether, given what is now known, the presence on the Steering Group of Professors Qureshi and Elliott gave rise to a legitimate perception of potential bias; not whether the exclusion of a Royal Brompton clinician did so.
115. The Steering Group's initial composition and role were described by the judge in §§ 36 and 37 of his judgment. These paragraphs are set out in paragraph 19 above.
116. In his written evidence, Jeremy Glyde, the programme manager of the NSCG, describes in paragraphs 71-77 of his second witness statement the basis and mode of appointment of Professors Elliott and Qureshi. For reasons to which we now turn, however, their impeccable qualifications and high professional standing will not assist the JCPCT if their participation brought with it the appearance of bias in the JCPCT's decision-making.
117. One of the Steering Group's roles was initially to consider and advise the NSCG on the clinical and service standards to be drawn up by the Standards Working Group and to form the criteria by which the Kennedy Panel would assess the existing 11 centres. In mid-2010 the primary care trusts pooled their functions by setting up (albeit without separate legal status) the JCPCT. The Steering Group's function switched accordingly to advising that body on clinical matters, including the design of the proposed congenital heart surgery networks. It also advised the NSCG Team to consult stakeholder groups about the proposed criteria and weightings which were to be used for the configuration evaluation.
118. Following each hospital's self-assessment, the Kennedy Panel made its own comparative assessments. While that process is subject to other challenges in these proceedings, it was here that the Steering Group's involvement ended. It had advised on what became the Kennedy Panel's assessment criteria – the criteria which in turn led to the two-centre option for London and the preference of Evelina Children's Hospital and GOSH as the two centres.
119. Owen J took the view that the bias allegation fell at the first fence: the Steering Group's role was only advisory and so not open to challenge for bias. He distinguished the decision of this court in R (Goldsmith) v Wandsworth LBC on the ground that what had vitiated the council's decision in that case was its uncritical acceptance of legally deficient advice. Here, the judge held, it was not and could not be contended that the advice given by the Steering Group was legally defective: it was based, he pointed out, on an analysis carried out by the NSC team with the advice of management consultants.
120. Pausing here, it seems to us, with respect, that this finding leaves open the question whether the advice was, as Mr Maclean alleges, legally deficient because of apparent bias in its authorship. If it was, the question arises whether it may have tainted the decision to which it contributed.
121. The judge's conclusion on this issue, however, had a second ground:

"it is clear from the minutes of the meetings of the JCPCT and from the witness statement of Sir Neil McKay that it arrived at its decision as to its preferred options after a full and proper consideration of the material before it, and was not simply rubber-stamping the recommendations of the Steering Group."

122. This is a material and weighty finding of fact; but it does not by itself foreclose the possibility of a rational perception that bias, if there was any, had travelled from the advisory into the decision-making process.
123. Where, as in Goldsmith (above), the decision-making body has surrendered its judgment to the advisory body, the former's decision is in law no decision at all. On no view is this the present case. But the problem of possibly tainted advice given to an independent decision-maker cannot be dismissed as readily as Miss Rose invites us to dismiss it, on the bare grounds that it was only advice and that the JCPCT knew perfectly well the membership of the group which had tendered it.
124. At the root of the issue is the presence on the Steering Group of consultants from two of the three London hospitals under scrutiny. Legally it cannot matter which two. The vice, if there was one, was giving two of the three London institutions an apparent inside track in the evaluation exercise.
125. The question which the hypothetical observer will be thinking about in such a situation has, in our judgment, to be adapted (rather than simply adopted) from the now conventional question relating to perceived bias in a decision-maker. If, as was said by this court in In re Medicaments [2001] 1 WLR 700, bias in a judicial decision-maker is "an attitude of mind which prevents the judge from making an objective determination of the issues that he has to resolve"; and if, as Lord Hope put it in Porter v McGill [2001] UKHL 67, the issue for the court is "whether the fair-minded and informed observer, having considered the facts, would conclude that there was a real possibility that the tribunal was biased"; it seems to us that, where the issue relates to the effect on a decision of potentially biased advice, the question has to be whether the observer, knowing the composition and remit of both the advisory body and the deciding body, would perceive a real possibility both of bias in the advice and of its infecting the decision.
126. We think it necessary to pose the question in this way, first because it is not possible to exclude a priori the possibility that legally bad advice will affect a decision based in part on it, and secondly because it is not possible to draw a principled line between advice which is bad because of misinformation and advice which is bad because of bias. What matters here, as elsewhere, is the matrix of fact in which the risk is said to have taken shape. That includes, but is not confined to, the composition and method of working of each of the bodies concerned. But the answer cannot depend on what the decision was: as we have said, that goes only to the entitlement to relief. It depends on the evidence; and, as with all evidence, it is for the party which asserts a fact to prove it.
127. In the present case the material decision was not how the future provision of paediatric cardiac care in and outside London should be configured. It was what questions and information to put out to public consultation. Accepting Mr Maclean's point that what answers you get will depend on what information you give and what questions you ask, it remains the case that the decision itself was not going to be a final one. That is the first point. The second point is that the body which decided on the form and content of the consultation, the JCPCT, was itself composed of representatives of all but one of the 152 English primary care trusts. The third point is that the Steering Group which advised it was a clinically expert body whose remit was among other things to advise the JCPCT on the appropriate model of care, on the workforce and patient volume required for a safe and sustainable service, on the relation of these to other services, on the designation criteria for paediatric cardiac surgery and on the assessment process. It had no decision-making role.
128. In our judgment the objective observer, knowing all this, would not have been disturbed at the presence of Professor Elliott and Professor Qureshi on the Steering Group. She would have known and borne in mind that the group was chaired by Dr Patricia Hamilton CBE, a distinguished neonatal paediatrician and immediate past president of the Royal College of Paediatrics and Child Health; that it consisted of 29 experts, more than a third of them practising clinicians; and (though perhaps less influentially) that all of them knew that they had been appointed to represent their disciplines and professional bodies, not their hospitals. The observer would recall next that the Steering Group's advice was to be tendered to a nationally representative body composed of people with specialised knowledge and experience of the health service and fully capable of noting and allowing for the fact that the two preferred London centres on which it was proposed that they should consult had had consultants on the Steering Group. She would recall, finally, that all the substantive issues on which the Royal Brompton would wish to make its case, including if need be the composition of the body which had advised the contentious London configuration, would remain at large in the consultation process.
129. The bare fact the JCPCT in the event adopted the Steering Group's recommendation is neither here nor there in determining the presence of apparent bias in the Steering Group. It is also of limited relevance in determining whether any perceived bias had ostensibly travelled from advice to decision: what matters at least as much is the judge's finding about the quality and method of the JCPCT's own deliberations.
130. There was little or nothing to support Mr Maclean's charge that the JCPCT simply rubber-stamped the Steering Group's advice. The statement in the consultation document itself that the Group had "actively steered the review" goes no distance in this regard. That was its job. The verbatim record of the JCPCT's meeting in February 2011 does not reveal any discussion of the possibility of bias in the advice which the Steering Group had given, but one would not expect this to be discussed ad hominem. It does not, however, answer the issue now before the court.
131. We are inclined to agree with Mr Maclean that the objective observer would not think it entirely realistic to try to limit the capacity of the two consultants to that of representatives of their professional associations. She might well think this less relevant than the fact that they held senior posts in two of the three London hospitals. But she would not, in our judgment, regard it as a real risk that they would be both inclined and able to use their position on the Steering Group to promote the interests of their own hospitals. She would no doubt keep in mind Professor Elliott's letter of 9 March 2011 to a colleague at GOSH about the need for the two hospitals to be ready to take over the Royal Brompton's cystic fibrosis service. But by that date the decisions, so far as relevant to this case, had been taken; the public consultation had begun, and the transfer of paediatric interventions from the Royal Brompton was now in prospect.
132. The objective observer would conclude, in our opinion, that there was no real risk that the judgment of the JCPCT as to the preferred options on which it was to consult the public would be distorted by the advice of a Steering Group which included consultants from two of the three hospitals potentially at risk in the eventual reorganisation.

(ii) Mr James Monro

133. A separate but related issue has arisen since judgment was given by Owen J. It concerns a member of the Kennedy Panel, James Monro, an emeritus consultant cardiac surgeon of great distinction who in 2003 had chaired the departmental review which reported on the reorganisation of paediatric cardiac surgery (see paragraph 16 above). He had been nominated for membership of the panel by the Society for Cardiothoracic Surgery. The panel's task was to evaluate the self-assessments of the 11 hospitals, to visit each centre and then to submit the JCPCT its scores against the evaluation criteria.
134. Of the four options eventually canvassed in the consultation document, only one (Option B) proposed retaining Southampton as a paediatric surgical centre. In the course of June 2011 Mr Monro lent his name to a public campaign to save the Southampton unit, to which he had devoted most of his career, from the risk of closure in the wake of the ongoing consultation. His participation left no doubt about his view: "It would be crazy to close a very successful unit with a dedicated team which has taken years to build up." Sir Ian Kennedy's reaction, while expressing confidence that Mr Monro had thus far discharged his duties without bias, was to ask Mr Monro not to attend the September and October 2011 meetings of the panel.
135. Mr Maclean on behalf of Royal Brompton wishes to introduce this evidence, which he submits meets the test in Ladd v Marshall [1954] 1 WLR 1489, in order to submit that Mr Monro's self-evident partiality must relate back to the formative stage of the consultation and must vitiate it.
136. It is not desirable to shut out this evidence even if it might arguably have been obtained sooner. But it is necessary to look at it in context. The context is an assault on the JCPCT's decision to go out to consultation on a preferred option of two London centres, neither of them Royal Brompton. The submission in relation to Mr Monro has to be that an objective observer would deduce from his subsequent partisanship on behalf of the Southampton unit a real risk that, in helping to evaluate the 11 national centres, he will have undervalued the other institutions and overvalued his own. If that is right, and if it may have infected the consultation, it arguably does not matter that its impact on the prospects of Royal Brompton is at best oblique. The claimant will have a sound case for stopping the consultation and rewinding the film.
137. The panel, in addition to Professor Sir Ian Kennedy, included a lay member, a paediatric nursing adviser, a health service administrator and three senior paediatricians. Of these, only Mr Monro was associated with one of the 11 hospitals under scrutiny. It is unsurprising in these circumstances that Sir Ian took the precautionary step he did. But this does not answer the question facing the court – the same question, mutatis mutandis, as in relation to the Steering Group.
138. We do not consider that Mr Monro's support for his unit at Southampton, in the context of the consultation, can tenably be related back to the pre-consultation period when he was participating in the evaluation of the 11 English centres. It is one thing, and a perfectly intelligible thing, to object to a proposed outcome to which you have yourself contributed. It is another thing, and a very serious one, to have attempted by stealth to obviate that outcome. To deduce the second from the first in the absence of firm evidence requires an assumption that the individual concerned was prepared to forfeit professional objectivity in favour of partisanship. That is not an assumption which in our judgment the fair minded and objective observer would be prepared to make. She would look for evidence of the assumed link, and in the present case she would find none.

Royal Brompton's additional grounds for upholding the judge's order

139. First additional ground: Mr Maclean submits that the judge was wrong to hold that the decision by the JCPCT to adopt a two-centre preferred option was non-justiciable. He relies on the decision of Sedley J as he then was in R v Secretary of State for Transport, above, where, on the basis of R v Secretary of State for Employment ex parte the Equal Opportunities Commission [1995] AC 1, he accepted the submission of Mr Maclean that "the want of an identifiable decision is not fatal to an application for judicial review".
140. Our answer to this ground is contained in our observations about whether it is appropriate for the court to intervene in a consultation exercise. We consider that there could have been no objection to the judge's conclusion on this point if he had expressed the view that, in the particular circumstances of this case, that particular challenge was premature. It was the type of challenge which could, so far as this point is concerned, have been made in the response to consultation, and in that sense the challenge was premature. There will, however, will be other cases where the court has to grant relief in relation to a consultation process.
141. Second additional ground: Mr Maclean submits that the exclusion of the three-centre option was irrational. The JCPCT was satisfied that there were two three-centre options which were viable. These should have been the subject of express options in the consultation document. Our answer to that point is that it was open to the JCPCT to have preferred options provided it was clear to consultees, as it was, what the other options were. The response form gave consultees the opportunity to support an option for three London centres. The rescoring of Royal Brompton's research and innovation means that the position as regards those matters in the field of paediatric cardiac surgery is now known. It is clear that it was not presented in such a way as to distort the picture which consultees were given in the consultation document. This point is reinforced by the analysis of responses by Ipsos Mori.
142. Miss Rose submits that the judge was right to reject Mr Maclean's arguments. It is clearly open to a decision-maker conducting a consultation to identify his preferred options: the decision-maker is not obliged to consult on all possible options.
143. In any event, as Miss Rose points out, Royal Brompton accepted below that if there were only two London centres those centres would be Evelina Children's Hospital on the grounds that it scored highest, and GOSH on the grounds that it was the only centre in the country to provide complex tracheal surgery. It is inappropriate for it to argue for a different combination on this appeal.
144. Mr Maclean also relies on the underestimate of projected procedures per year in London (see paragraph 74 above). The lower figure given in the London box does not in any event include overseas (private) patients. The position was seriously misstated.
145. Miss Rose submits that essentially the grounds put forward by Royal Brompton amount to challenging the reasons for preferring a two-centre option. The Steering Group had advised that two centres rather than three were better placed to develop the network of services to support paediatric cardiac surgical services. It also considered that ideally each centre should perform 500 procedures per year. The judge was also correct to hold that the two-centre recommendation was as valid with 1,500 procedures per year as it was with 1,250, the number given in the London box. The volume of procedures could only be raised to the ideal by diverting work from neighbouring catchment areas, which might have an adverse effect on access and travel and retrieval times. It was for the JCPCT to determine whether it was better to have larger numbers of procedures being dealt with by two London centres than to have three centres. The same point applied to criticisms which Mr Maclean made of the JCPCT's assessment of travel and retrieval times.
146. The current annual figure for procedures in London given in the London box was 1,250. It is clear from the table setting out all the options that this was below the number of projected procedures. The information given in that table about the viable options was as follows:

OPTION	2	6	8	10	12	14
	7 sites 2 London	6 sites	6 sites	7 sites 3 London	7 sites 3 London	Top 7 scoring
London (per centre)	721	741	741	494	494	580

147. It would thus have been clear to consultees that the two-centre option had been preferred notwithstanding this projected increase in caseload. It remained the case that it was open to the decision-maker to decide whether to have a preferred option. It did so. It gave an opportunity for consultees to state whether they agreed with its preferred option. It was not obliged to include options 10 and 12 as options for the purposes of consultation. If a consultee considered that the figures for projected procedures in the future undermine the provisional decision made by the decision-maker, this was a point to raise in response to the consultation. It is not one where it is necessary or appropriate for the court to intervene in the course of the consultation.
148. The same applies to any criticisms which Royal Brompton wishes to make of travel and retrieval times. There were other criticisms made by Mr Maclean. He criticised the JCPCT's failure to allow the Kennedy Panel to use the information supplied in response to the baseline template, but we cannot take that point further as Royal Brompton's response to the baseline template is not in evidence in these proceedings. He criticised the way in which the research of different centres had been scored. These are points of importance to Royal Brompton but they do not, even when they are all added together, tip the consultation exercise in this case from one which is fair and lawful into one which is unfair and unlawful.
149. Third additional ground: This ground relates to Deliverability. Mr Maclean submits that the scoring for this took insufficient account of the fact that, if the two-centre option were implemented, Royal Brompton's PICU would have to close. Royal Brompton's score on Deliverability should have been increased to compensate for the loss of this. Consultees were misled into thinking that the adverse effects of the closure of its PICU would be less than if that of GOSH or Evelina Children's Hospital was closed when this was not the case.
150. Miss Rose disagrees. The JCPCT conducted its own investigation and came to the conclusion on the lack of risk to local and national provision due to the loss of Royal Brompton's PICU (see paragraph 63 above). Miss Rose submits that in any event, Royal Brompton could challenge this in its response to the consultation document. The Pollitt report had in effect discounted any suggestion that the closure of the PICU would render its specialised respiratory and cystic fibrosis services unviable.
151. We agree with Miss Rose's submission here. Royal Brompton's point may be accepted to be correct but the proper place to make it is to the JCPCT as part of its response to the consultation document. It is not a ground which would, having regard to the principles outlined above, justify the court interfering with the consultation process. In addition, the effect of closure was a matter for the expert opinion of the decision-maker and the court should not pre-empt his decision.
152. Fourth additional ground: this relates to bias and has been dealt with above.
153. Fifth additional ground: Mr Maclean argued in his skeleton argument that the judge gave excessive weight to the opinion of Doctor Pinto-Dutchinsky. This matter was not addressed by either counsel in oral submissions. As this judgment does not rely on that evidence it is unnecessary to say more about that ground.

Relief granted by the judge

154. Miss Rose reserved her position on relief should this appeal fail. In essence she submits that the judge should simply have directed the re-scoring of research and innovation. We wish to leave open the question whether, even on the judge's conclusions, it was correct for the judge to quash the whole of the consultation.

Order

155. For the reasons given above we would allow this appeal, and dismiss the respondent's notice.