Watts, R (on the application of) v Bedford Primary Care Trust & Ors

This case is about National Health Service waiting lists. It represents an attempt by the claimant to escape from what she saw as the almost unbearable consequences of what she says was unacceptable delay by the National Health Service (“the NHS”) in providing her with hip replacement surgery. She seeks to compel the authorities to reimburse her for the cost of appropriate treatment that in the event she obtained in France. If the claimant is correct the implications for the NHS and its patients may be profound.

Essentially the claimant’s case is founded on the assertion that she has enforceable rights under European Community law to be treated in another Member State if the NHS is unable to provide appropriate treatment within the timescale indicated by the patient’s medical condition and medical needs. Specifically she relies upon Article 49 (previously Article 59) of the European Community Treaty and, separately, upon Article 22 of Council Regulation No 1408/71.

I shall have in due course to consider both provisions in more detail. For the moment it suffices to indicate that the effect of Article 22 is to confer a right to be treated in another Member State at public expense:

“where the treatment in question is among the benefits provided for by the legislation of the Member State on whose territory the person concerned resides and where he cannot be given such treatment within the time normally necessary for obtaining the treatment in question in the Member State of residence, taking account of his current state of health and the probable course of the disease.”

The Secretary of State for Health has always accepted that, in certain circumstances, Article 22 is engaged. The mechanism by which a NHS patient invokes her rights under Article 22 is by means of an application for a Certificate in Form E112. However the Secretary of State has always consistently maintained that Article 49 has no application in such a case.

The Department of Health’s understanding of the position under Community Law (particularly in the light of the European Court of Justice’s decision in Geraets-Smits v Stichtung Ziekenfonds VGZ, Peerbooms v Stichtung CZ Groep Zorgverzekeringen (Case C-157/99) [2002] QB 409) is set out in an internal ‘Background Note – “Undue Delay”’ which I should set out in full:

“The Decision of the European Court of Justice (ECJ) in the case of Geraets-Smits/Peerbooms did in fact endorse the need for prior approval under the E112 arrangements in terms of control and management of healthcare and budgets across the EU.

In considering the national legislation of the Netherlands which was at issue in this case, the ECJ said that Netherlands healthcare system, which required prior authorisation by the patient’s sickness insurance fund for treatment in other EEA countries, was contrary to the freedom to provide services provisions of the Treaty Establishing the European Community. However, the ECJ said that, provided certain conditions were met, such a rule was justified in the interests of maintaining the financial stability of the Dutch national health system.

One of the rules of the Netherlands system was that it had to be proved that the patient’s medical condition required the service in question. The Court said that was justifiable provided that it was construed to mean that authorisation to receive treatment in another member state may be refused on grounds of necessity only if the same or equally effective treatment can be obtained without undue delay in the home state.

The ECJ did not define what it meant by “undue delay”, by implication leaving that as a matter for national law and practice. The Department of Health takes the view that where an application for treatment is made abroad under the E112 arrangements on the grounds of undue delay, account should be taken of the national targets for waiting times as published in the NHS Plan, and of the individual patient’s clinical need. Ministers stated this publicly last year in answer to a Parliamentary Question.

Patients do not have an automatic right to be referred abroad for treatment on the grounds that treatment could be made available more quickly elsewhere. The E112 procedures published in the leaflet “Health Advice for Travellers” do provide that the recommendation of a UK NHS consultant, rather than a GP, should be obtained. If the wait to see a consultant is in excess of the NHS target maximum outpatient waiting time, or the waiting time for an operation is in excess of the maximum inpatient waiting time for surgery (currently 15 months), this would be prima facie evidence of undue delay.”

Publicly the Department’s position is explained in two documents. The first is ‘Health advice for Travellers’ last updated on 25 March 2002. The relevant part, in a passage headed ‘Non-emergency treatment in an EEA country: Form E112’, reads as follows:

“Form E111 will not cover you for free or reduced-cost treatment if you are going to another EEA country specifically for medical care, or if you require ongoing treatment for a pre-existing condition. For this, you will need Form E112 which is not issued automatically but requires authorisation from the Department of Health. Unless you have an E112, you will have to pay for the treatment you receive and you will not be able to obtain a refund of the costs.

For pre-existing conditions that will require treatment during your time abroad – eg, blood tests, medication or injections, etc. – apply by letter enclosing details, including dates of treatment, supporting medical evidence from your GP and a copy of your E111 to the following address(es): …

Please note, however, that form E112 is only available where a clear need for on-going treatment is established. The form is not available on a ‘just in case’ basis.

If you need maternity care in another EEA country, write to the above address(es) explaining why you want care outside the UK and enclose a copy of your E111 and evidence from your GP or midwife of your expected date of confinement.

For people going to another EEA country specifically for medical treatment or an operation, the issue of Form E112 will usually only be considered if:

•

your UK NHS Consultant recommends treatment in the other country;

•

your Health Authority in England and Wales … agrees to meet the cost of the treatment;

•

the treatment is available under the other country’s health insurance scheme;

•

you are entitled to Form E111.

To apply for an E112 in these circumstances, write a letter explaining your reasons for seeking treatment outside the UK. Send this letter to your health authority … together with a letter from your NHS consultant explaining why he or she is recommending treatment in another EEA country. The health authority … should then send your application and their agreement to fund the treatment to the appropriate address.”

The other public explanation is in a ‘Draft for consultation’ issued in July 2002 of a document entitled ‘Treating more patients and extending choice: Overseas treatment for NHS patients – Guidance for Primary Care and Acute Trusts’. The relevant part is to be found in paragraphs 1.14-1.18:

“1.14

Direct commissioning does not affect the long-established arrangements for referral of patients under European Community legislation – the so-called E112 system contained in Regulation (EEC) 1408/71. This Regulation contains provisions which permit the referral of patients specifically for treatment for a pre-existing medical condition. Benefits are provided on the same terms as are available to the insured people in the host state.

1.15

Prior authorisation is required where E112 referral is sought. Patients should first seek the opinion of their NHS Consultant and home PCT. Final approval of the Department of Health then needs to be obtained. It should be noted that the Regulations say that authorisation for E112 referrals cannot be refused if the treatment in question is a benefit provided under the state health care scheme of the home country and cannot be provided “within the time normally necessary for obtaining the treatment in question taking account of the patient’s current state of health and the probable course of the disease”.

1.16

However, even if the conditions leading to mandatory authorisation are not met there is still discretion to refer if a PCT and the Department believe this is warranted.

1.17

Further information is given in Department of Health leaflet “Health Advice for Travellers” available from Post Offices, and on the internet at www.doh.gov.uk/traveladvice.

1.18

The direct contracting and E112 schemes operate in tandem and PCTs may use whichever is most convenient. The E112 arrangements may be particularly useful where treatment has to be sought from a facility where the lead commissioners have no contract.”

So much for the general background. I turn now to consider the facts of the present case.

The claimant, Mrs Yvonne Watts, is 72 years old. In September 2002 she was diagnosed by her GP as having osteoarthritis in both hips. On 16 September 2002 the GP wrote to a consultant orthopaedic surgeon, Mr Philip Edge, saying that he felt she required urgent hip replacements and requesting a private appointment. The private appointment was suggested by the GP because the waiting time to see Mr Edge as a NHS consultant was at that time between 19 and 21 weeks. Mr Edge saw Mrs Watts at his private clinic on 1 October 2002.

In the meantime, on 13 September 2002 Mrs Watts’s daughter, Ms Julie Harding, telephoned Bedford Primary Care Trust (“the PCT”) to speak to the PCT’s Public Liaison Manger. She was put in touch with the PCT’s Assistant Director of Commissioning and Performance Management, Ms Hilary Laughton, who telephoned her back the same day. Ms Harding told Ms Laughton that she would like the PCT to support an application by her mother to have bilateral hip surgery overseas using Form E112. Ms Laughton said she would find out what the scheme entailed and what were the relevant criteria.

On 24 October 2002 Ms Harding sent Ms Laughton an e-mail “applying for an E112 certificate on behalf of my mother”. She said that the operation could be carried out in Belgium within two weeks at a cost of £4,845. She referred Ms Laughton to the decisions of the European Court of Justice in Vanbraekel v Alliance nationale des mutualities chretiennes (Case C-368/98) [2001] ECR I-5363 and Geraets-Smits v Stichtung Ziekenfonds VGZ, Peerbooms v Stichtung CZ Groep Zorgverzekeringen (Case C-157/99) [2002] QB 409. She asked to be told as soon as possible “if [the PCT] will write to the Department of Health supporting my mother’s application or if not, the reasons for refusal.”

On 28 October 2002 Mr Edge wrote to Ms Laughton. I think I should set out the letter in full:

“This lady was seen by me on 1.10.02 with severe bilateral hip pain. She has experienced severe deterioration in her function over the course of the last three months. She now has to use two walking sticks to mobilise.

Clinical examination and X-ray examination show her to have severe arthritis of both hips.

This lady requires bilateral total hip replacements. She is battling tremendously with her mobility and is in constant pain.

This lady is certainly deserving of bilateral hip replacements. She is as deserving as any of my other patients who are waiting such surgery.

Unfortunately my waiting list is approximately one year. As this lady’s function is severely hampered by her arthritis she wishes to enquire whether this surgery can be performed abroad at the cost of the NHS. She is certainly as deserving as any of the other patients of mine with severe arthritis who are on my waiting list.

I note also that she has a valve replacement in her heart and is currently on Warfarin. She will therefore need to be admitted to hospital several days prior to surgery and her anti-coagulation levels monitored peri-operatively.

I am able to confirm that the surgery is technically possible at her local hospital but the waiting list is approximately one year.

Thank you for your assessment of this case.”

On 6 November 2002 Ms Laughton telephoned Mr Peter McConn, section head of the International Branch of the Department of Health and the official responsible for policy on the United Kingdom’s system of overseas reciprocal health agreements, including oversight of the arrangements under Article 22. Mr McConn explained to Ms Laughton the Department’s understanding of the operation of the Form E112 scheme in the light of the European case-law on Article 22.

On 11 November 2002 Ms Laughton wrote to Ms Harding in response to her e-mail of 24 October 2002. She said that she was following up Ms Harding’s request for the procedure to be undertaken through the E112 certificate process with the Department of Health and would write to her again “when I receive confirmation whether or not this procedure would qualify under the E112 certificate system.” Ms Harding responded on 15 November 2002 with a letter in which she enclosed the Department of Health’s draft ‘Guidance’ document and drew attention to paragraphs 1.14-1.18.

On 21 November 2002 Ms Laughton wrote to Ms Harding explaining why the PCT was unable to support the claimant’s request for treatment under Form E112. Ms Laughton dealt at some length with the position under Community Law as she understood it, referring to the Department of Health’s published guidance. Summarising what Mr Edge had said in his letter of 28 October 2002, she said that Mr Edge saw the claimant “as a routine case”. The PCT’s decision was set out in the following passage:

“Although Bedford PCT is sympathetic to your request on behalf of your mother, that Bedford PCT should agree to support your proposition that your mother should be able to have a bilateral total hip replacement performed overseas on the E112 scheme, Bedford PCT is unable to support this request for the following reasons:

1.

Mr Edge, your mother’s NHS Consultant does not specifically support treatment overseas. He states that your mother’s clinical condition is “certainly as deserving as any of the patients of mine with severe arthritis who are on my waiting list”.

Mr Edge’s letter dated 1st October 2002 does not indicate that your mother requires treatment outside of the draft NHS guidelines for overseas treatment which states that overseas treatment can only be supported “if treatment cannot be provided within the time normally necessary for obtaining the treatment in question taking account of the patient’s current state of health and the probable course of disease”.

The treatment your mother requires “can be provided locally within the time normally necessary for obtaining the treatment in question taking into account the patient’s current state of health”.

2.

The ruling above states that “authorisation can be refused on the ground of lack of medical necessity only if the same or equally effective treatment can be obtained without undue delay at an establishment having a contractual arrangement with the insured person’s sickness insurance fund”.

Bedford PCT interpret ‘undue delay’ as meaning, within the Governments NHS Plan targets, and will endeavour to ensure (to the best of their ability) that your mother will be treated within the NHS Plan target time for access to inpatient treatment of 12 months, either at Bedford Hospital, or by another provider if Bedford hospital are not able to offer your mother an appointment for surgery within the 12 month inpatient waiting time target.

Bedford PCT is committed to providing the best possible care for all its resident population, within the resources available to provide an equitable range of healthcare services and treatments.”

In the meantime Ms Harding had been in communication with the Department of Health herself. I need not go into the detail of these interchanges, which began with an e-mail from Ms Harding on 7 November 2002 and concluded with an e-mail from the Department on 3 December 2002 acknowledged by Ms Harding the same day. I should however make two points. The first is that Ms Harding was merely seeking information or expressing her views: at no stage did she ever make any application to the Department on behalf of her mother for authorisation whether under Form E112 or otherwise. The other is that, as Mr McConn frankly admitted in a subsequent witness statement, an official in the Department gave Ms Harding inaccurate advice. In her e-mail of 7 November 2002 Ms Harding, having mentioned the fact that the PCT had turned down her mother’s application, asked “Is there any mechanism for appealing directly to the Department of Health over this issue?” The response to this question, in an e-mail from a different section of the Department dated 3 December 2002 was:

“I am afraid that it is up to the discretion of the primary care trust to decide how it commissions care, and therefore which patients are eligible for treatment abroad. There is no system of appeal against any decision that the PCT may take.”

During the week commencing 9 December 2002 Ms Laughton spoke to Mr McConn about the letter she had sent to Ms Harding on 22 November 2002. Their recollections of what was said are not entirely clear, but since in my judgment nothing turns on the point I need say no more about it. More importantly, however, at some time during the same week Ms Laughton spoke to Mr Edge. “He told me that he felt unable to say whether the claimant should be treated overseas. He said that he could only comment on her clinical priority, which was routine.” No doubt the reason for these conversations was the fact that on 3 December 2002 the claimant’s solicitors had sent a letter before action to the PCT. Proceedings were issued on 12 December 2002 (see further paragraph [31] below).

On 23 December 2002 Mr Alan Dickinson, the Litigation and Controls Assurance Manager of Bedford Hospital NHS Trust, had a meeting with Mr Edge. (He had been asked to assist the PCT with the matter.)

“We discussed the claimant’s treatment. Mr Edge informed me in that meeting that he was not recommending that the claimant have her hip operation abroad.”

On 6 January 2003 the claimant and her daughter travelled to Lille in France, where the claimant was seen at the Polyclinique de la Louvierre by Dr Philippe Delehaye, an orthopaedic surgeon, and Dr Thierry Oszustowicz, a consultant anaesthetist. According to the account subsequently given by Ms Harding in a witness statement dated 20 January 2003, Dr Oszustowicz expressed “great concern” about her mother’s continuing weight loss and worried that if it continued she would not be strong enough for surgery.

“After discussing the matter with Dr Delehaye he indicated that to be on the safe side the operation should be carried out by the middle of March 2003. Any later than that would risk the possibility that my mother would not be able to have the surgery. In the circumstances, it was agreed with the doctors in Lille to pencil in a provisional date for the operation on 11 March 2003. This is of course conditional on the appropriate arrangements being made by the NHS to process the authorisation for funding.”

On 22 January 2003 there was a hearing before Wilson J. Some comment seems to have been made on that occasion that although the skeleton argument lodged on the claimant’s behalf invited the judge to “see … the evidence from Dr Delehaye” no such evidence had been filed. In fact no such evidence has ever been filed. According to Ms Harding’s subsequent witness statement dated 25 March 2003 “after a request to the surgeon to supply information for the High Court, he no longer wished to accept my mother as a patient.”

Following the hearing before Wilson J, and in accordance with his order (see further paragraph [34] below), the PCT reconsidered the claimant’s case. Mrs Watts was re-assessed by Mr Edge on 31 January 2003 and at the same time assessed by Mr Jeremy Sizer, a Consultant Anaesthetist, as to whether or not she was fit for surgery.

Mr Edge wrote the same day (31 January 2003), this time to the claimant’s GP. Since this letter has been subjected to detailed scrutiny and analysis I think I should set it out in full:

“Many thanks for asking me to see this charming lady once again regarding her bilateral painful hips. As you are aware she is involved in negotiations to get her hips done sooner rather than later, possibly at a hospital in Europe. I last saw her on 01.10.02 at The Manor Hospital. At that stage I noted that she had bilateral severe hip pain. She was able to mobilise with difficulty using one or [two] sticks and was battling to mobilise any significant distance.

On questioning her today she says that the pain has now become significantly worse. She does, however, have minimal rest pain when lying in her bed at night. She is able to get down her stairs at home albeit very slowly and mobilises inside her home using 2 sticks all the time. She is now no longer able to mobilise on one stick. The right side is worse than the left.

I note that she is taking Co-dydransol tablets 4 times a day and is unable to take non-steroidals because of her heart condition.

She has no significant pain in her back or the abdomen. She has had no recent chest problems and her pedal oedema remains minimal. I do however note that she has lost one stone in weight since last seen by me 4 months ago. She has had no decrease in her appetite.

On examination today she did battle to walk with a very shuffling gait using 2 walking sticks. She had difficulty in getting on and off the couch and had to lift up her legs using her arms to get onto the couch. Examination of the back was reasonable. The hips were extremely stiff with a range of flexion from 10 to 90 degrees, 20 degrees of adduction on the left and 10 degrees of adduction on the right. There was a jog of rotation. All movements were slow and painful. There was some pedal oedema with good pedal pulses.

X-rays taken today show continued moderate to severe arthritis in both hips with no significant change from the previous films. She had a chest X-ray taken today which has gone for reporting.

She has also been assisted by Jeremy Sizer, Consultant Anaesthetist today. He felt that she was essentially fit for surgery and was likely to remain fit for surgery into the foreseeable provided no new diagnoses were made.

Under the circumstances I feel that this lady probably had deteriorated since last seen. Of course all the patients on my waiting list would have deteriorated in this time but she has perhaps got a little worse than the average patient. I am therefore prepared to re-categorise her as a “soon” case.

You are of course aware that soon has no particular definition. It merely lies somewhere between a routine case and an urgent case. I have several urgent cases on my waiting list which will have to be done prior to Mrs Watts.

Of great concern to me is the loss of one stone in weight over the course of the last 4 months. Patients do not tend to lose weight purely through immobility. She says that she is eating well and this therefore cannot account for her weight loss. I feel it would be prudent to investigate her one stone of weight loss prior to embarking on major reconstructive surgery. I therefore would be most grateful if you could thoroughly investigate her loss of weight (including full blood count, ESR, bone biochemistry and possibly a bone scan) and let me know the results before surgery.”

It will be appreciated that Mr Sizer’s view was less pessimistic than Ms Harding’s report of what Dr Oszustowicz had said.

Ms Laughton saw a copy of Mr Edge’s letter. The same day (31 January 2003) she had a conversation with Mr Edge.

“Mr Edge told me that he considered that a “soon” case should be treated in 3-4 months. This would mean that the claimant would be operated on in April or May 2003. It is Mr Edge’s decision when the claimant has her operation.”

On 3 February 2003 Mr Sizer wrote to Mr Edge, saying that “there is nothing that could be done to reduce her anaesthetic risk, nor indeed does there need to be any action. She is, therefore, fit to proceed to THR. I would be delighted to give an anaesthetic to this lady if it could be arranged.” A copy of Mr Sizer’s letter was of course seen by Ms Laughton.

The PCT’s decision was set out in a letter dated 4 February 2003 from Ms Laughton to the claimant. After summarising Mr Edge’s letter of 31 January 2003, Ms Laughton continued:

“Mr Edge has clarified this with the PCT to mean a wait for surgery of between three and four months. This will mean that Mrs Watts will be listed for surgery in either April or May 2003.

Based on the recent clinical reassessment of Mrs Watts by Mr Edge on 31st January 2003, Bedford PCT has reassessed the application by Mrs Julie Harding on behalf of her mother, Mrs Yvonne Watts, which sought support from Bedford PCT in applying for overseas treatment under form E112.

Having reassessed the new clinical assessment by Mr Edge, the PCT acknowledge that Mrs Watt’s clinical condition has deteriorated enough for Mr Edge to reprioritise Mrs Watts as requiring treatment more quickly than previously thought, within three to four months. This does not alter the PCTs previous determination in this application.

Mrs Watts will be listed for surgery at Bedford hospital for hip replacement surgery between three to four months. This is within the National Plan target for inpatient treatment, which will be a maximum wait of 12 months as from the 1st April 2003. Bedford Hospital achieved the maximum 12 month waiting times target at the end of December 2002.

Therefore, without support for overseas treatment from Mr Edge, Mrs Watts’s Consultant Orthopaedic Surgeon, or from Bedford PCT, (Bedford PCT reiterate previous reasoning for not supporting this application), Bedford PCT is unable to support the overseas treatment application under form E112.”

The claimant was not prepared to wait until April or May 2003. During February 2003 her daughter contacted a specialist health logistics company which arranged for the claimant to have her right hip replaced at the Clinique Sainte Isabelle in Abbeville in France. The claimant and Ms Harding travelled to Abbeville on 1 March 2003. The operation was performed on 7 March 2003. The claimant and her daughter returned to this country on 12 March 2003. The full cost of the operation and hospital stay was a ‘one-off’ fee of about £3,900. According to Ms Harding (neither the PCT nor the Secretary of State has condescended to comment on these figures) the average NHS cost of a similar operation is approximately £4,000, while the cost to the NHS of arranging for such an operation to be carried out in the private sector in this country is somewhat in excess of £6,000.

Before leaving the facts there is one final matter I should mention. As I have said, the claimant did not at any stage make any application to the Department of Health, as opposed to the PCT, whether under Form E112 or otherwise. Ms Harding’s communications with the Department were all with a view to either seeking information or expressing her views. Subsequently, on 4 April 2003, Mr McConn made a witness statement stating that he is the person who takes the decision to grant or refuse applications for treatment abroad under the Form E112 scheme (the Secretary of State having delegated the power to take such decisions to him) and indicating what his decisions would have been had an application been made by the claimant to the Department in December 2002 or in February 2003.

An application in December 2002 or January 2003 would have been refused, according to Mr McConn,

“because there was nothing to distinguish the circumstances of her case from that of others awaiting similar treatment and because Mr Edge had classified [the claimant’s] case as “routine”.”

An application in February 2003 would also have been refused,

“given that Mr Edge and Mr Sizer both considered [the claimant] to be likely to be fit to undergo the required surgery for the foreseeable future. Further, Mr Edge did not consider her case to be “urgent” and because of the “soon” categorisation which he gave her case, she was at that point going to receive her operation in a relatively short period, namely in April or May 2003.”

The claimant’s letter before action was sent to the PCT on 3 December 2002. The PCT’s responses dated 9 and 10 December 2002 not being satisfactory to her, the claimant issued proceedings in the Administrative Court on 12 December 2002. The defendant was the PCT. The Secretary of State for Health and Mr Edge were served as interested parties. (Mr Edge, understandably, has played no part in the proceedings.) The claimant’s Form N461 sought to challenge the PCT’s decision contained in Ms Laughton’s letter of 21 November 2002. The grounds on which judicial review was sought had been settled by Mr Richard Gordon QC and Mr Jeremy Hyam of counsel. The relief sought included (i) a quashing order in relation to that decision, (ii) a mandatory order that the PCT “comply with the law and grant authorisation for the claimant’s treatment overseas under Form E112 forthwith”, alternatively that the PCT forthwith “enter into a direct contract with an overseas hospital for the provision of medical services to the claimant”, (iii) a declaration that the decision was unlawful and in contravention of the claimant’s rights under Article 3 and Article 8 of the European Convention for the Protection of Human Rights and Fundamental Freedoms and (iv) “damages”, including damages under the Human Rights Act 1998.

On 19 December 2002 Lightman J directed the PCT to file an acknowledgement of service no later than 7 January 2003 and directed that the application for permission be heard on notice in court.

The PCT’s acknowledgement of service was filed on 6 January 2003. Attached to it were detailed grounds for opposing the claim settled by Mr Steven Kovats and Ms Nicola Greaney of counsel. The Secretary of State’s acknowledgement of service was filed on 7 January 2003. Attached to it were detailed grounds for contesting the claim settled by Mr David Lloyd Jones QC and Ms Sarah Lee of counsel. At the forefront of that document was the assertion that the application was misconceived, inter alia, in that (a) the PCT was not the appropriate body to authorise treatment under Regulation 1408/71 and (b) no application had been made to the Secretary of State, who was the appropriate person, nor had he taken any decision in the case. The evidence in support of the claim, in the form of a witness statement by Ms Harding, followed on 20 January 2003.

The claimant’s application for permission was heard by Wilson J on 22 January 2003. Wilson J adjourned the application and directed that the application for permission and the substantive hearing be heard together, such hearing to include any challenge to a new decision of the PCT or to a decision of the Secretary of State. The PCT was directed to reach a fresh decision on or before 5 February 2003 – hence the letter from Ms Laughton dated 4 February 2003. The claimant was directed either to withdraw her claim for judicial review or to amend her Form N461 by 12 February 2003.

The claimant served an amended Form N461 on 12 February 2003. She asserted that, insofar as it was or had been appropriate for her to make a separate request for authorisation directly to the Secretary of State, either in November 2002 or following her reassessment by Mr Edge on 31 January 2003, then her amended Form N461 should be treated as such request(s). She went on to say that the amended grounds of judicial review “proceed on the assumption that the [Secretary of State] refuses such request(s) for authorisation for the reasons set out in its “Summary Grounds for Contesting the Claim” dated 7^th January 2003.” The relief sought was amended to include declaratory and mandatory relief against the Secretary of State analogous to that sought against the PCT. In addition the claimant sought “a declaration that [the claimant] is entitled to be indemnified by the [PCT] and/or the [Secretary of State] for the full costs of treatment abroad (including travelling and accommodation costs)” and “restitutionary damages in respect of the full costs of treatment”. The Secretary of State’s factual response to this amended case was set out in the witness statement by Mr McConn to which I have already referred.

The hearing before me began on 2 April 2003 and lasted three days. The claimant was represented by Mr Gordon and Mr Hyam, the PCT by Mr Kovats and Ms Greaney and the Secretary of State by Mr Lloyd Jones and Ms Lee. The evidence consisted of witness statements from Ms Harding dated 20 January 2003 and 25 March 2003, from Ms Laughton dated 7 March 2003, from Mr Dickinson dated 10 March 2003 and from Mr McConn dated 6 March 2003 (two), 31 March 2003 and 4 April 2003. Counsel had prepared very detailed and helpful skeleton arguments for which I am grateful. At the end of the hearing I reserved judgment. It was agreed that I should defer giving judgment until the ECJ had given judgment (then expected on 13 May 2003) in a case in which the Advocate General had given an opinion and the ECJ had heard argument, Muller-Faure v Onderlinge Waarborgmaatschappij OZ Zorgverzekeringen UA, van Riet v Onderlinge Waarborgmaatschappij ZAO Zorgverzekeringen (Case C-385/99). A copy of the judgment was sent to me on 23 May 2003. I was sent further written submissions by Mr Kovats on 13 June 2003 and by Mr Gordon and Mr Lloyd Jones on 16 June 2003. I now (1 October 2003) hand down judgment. I am sorry that it was not possible for me to complete it before the end of term on 31 July 2003.

So much for the facts. I turn to consider the law, first our domestic law.

Mr Gordon accepts, as he has to, that the claimant has in these circumstances no effective remedy under domestic law. The principles are well known and clearly established: see for a general description Kennedy & Grubb, Principles of Medical Law, (ed 1998), paragraph 4.14. I was referred to a number of authorities: R v Secretary of State for Social Services ex p Hincks (1980) 1 BMLR 93, R v Cambridge Health Authority ex p B [1995] 1 WLR 898, R v North Derbyshire Health Authority ex p Fisher (1997) 38 BMLR 76, R v North and East Devon Health Authority ex p Coughlan [2001] QB 213 and R v North West Lancashire Health Authority ex p A [2000] 1 WLR 977. For present purposes I need do no more than quote certain passages from the judgments of Sir Thomas Bingham MR in R v Cambridge Health Authority ex p B [1995] 1 WLR 898 and of Auld and Buxton LJJ in R v North West Lancashire Health Authority ex p A [2000] 1 WLR 977.

The underlying realities were eloquently described by the Master of the Rolls in R v Cambridge Health Authority ex p B [1995] 1 WLR 898 at p 906C:

“I have no doubt that in a perfect world any treatment which a patient, or a patient’s family, sought would be provided if doctors were willing to give it, no matter how much it cost, particularly when a life was potentially at stake. It would however, in my view, be shutting one’s eyes to the real world if the court were to proceed on the basis that we do live in such a world. It is common knowledge that health authorities of all kinds are constantly pressed to make ends meet. They cannot pay their nurses as much as they would like; they cannot provide all the treatments they would like; they cannot purchase all the extremely expensive medical equipment they would like; they cannot carry out all the research they would like; they cannot build all the hospitals and specialist units they would like. Difficult and agonising judgments have to be made as to how a limited budget is best allocated to the maximum advantage of the maximum number of patients. That is not a judgment which the court can make.”

In R v North West Lancashire Health Authority ex p A [2000] 1 WLR 977 the court was considering an application for judicial review in which the claimants, each of whom suffered from gender identity dysphoria (transsexualism), sought to challenge a health authority’s policy that refused funding for gender reassignment surgery save in rare and exceptional circumstances. Auld LJ said this at p 981D:

“The National Health Service Act 1977, by section 1(1), imposes on the Secretary of State a duty:

“to continue the promotion in England and Wales of a comprehensive health service designed to secure improvement – (a) in the physical and mental health of the people of those countries, and (b) in the prevention, diagnosis and treatment of illness, and for that purpose to provide or secure the effective provision of services in accordance with this Act.”

In section 3, it elaborates on that duty by obliging him:

“to provide … to such extent as he considers necessary to meet all reasonable requirements … (e) such facilities for the prevention of illness, the care of persons suffering from illness and the after-care of persons who have suffered from illness as he considers are appropriate as part of the health service; (f) such other services as are required for the diagnosis and treatment of illness.”

It is important to note the qualifications in those provisions, as this court observed in R v North and East Devon Health Authority ex p Coughlan [2001] QB 213, 230. The first is that section 1(1) does not oblige the Secretary of State to provide a comprehensive health service, but “to continue to promote” such a service. The second and third are that section 3 limits his duty of provision of services “to such extent as he considers necessary to meet all reasonable requirements,” and, in the case of the facilities referred to in section 3(e), to those “he considers are appropriate as part of the health service.”

The Act of 1977 provides, by section 13, that the Secretary of State may direct a regional health authority to exercise those functions and that the authority has a duty to comply with such direction. The Secretary of State, by the National Health Service (Functions of Health Authorities and Administration Arrangements) Regulations 1996 (SI 1996 No 708), has directed that the appellant and other regional health authorities shall exercise the functions in section s 1 and 3 of the Act, thus imposing on them the duty to provide “to such extent as [it] considers necessary to meet all reasonable requirements … such … services as are required for the diagnosis and treatment of illness.” The Act, in section 128, defines the word “illness” as including “mental disorder within the meaning of the Mental Health Act 1983 and any … disability requiring medical … treatment or nursing.” It is common ground for the purpose of this appeal that transsexualism is an illness in the nature of a mental disorder for the purposes of sections 1 and 3.

The qualifications in the statutory duties imposed by the Act of 1977 to which I have referred make plain that it is for the authority to judge what services it should provide, and to what extent, to meet all reasonable requirements for them. In ex p Coughlan the court said as to the originating and corresponding obligations of the Secretary of State, at p 230:

“25.

When exercising his judgment he has to bear in mind the comprehensive service which he is under a duty to promote as set out in section 1. However, as long as he pays due regard to that duty, the fact that the service will not be comprehensive does not mean that he is necessarily contravening either section 1 or section 3. The truth is that, while he has the duty to continue to promote a comprehensive free health service and he must never, in making a decision under section 3, disregard that duty, a comprehensive health service may never, for human, financial and other resource reasons, be achievable. Recent history has demonstrated that the pace of developments as to what is possible by way of medical treatment, coupled with the ever increasing expectations of the public, mean that the resources of the NHS are and are likely to continue, at least in the foreseeable future, to be insufficient to meet demand.

26.

In exercising his judgment the Secretary of State is entitled to take into account the resources available to him and the demands on those resources. In R v Secretary of State for Social Services ex p Hincks (1980) 1 BMLR 93 the Court of Appeal held that section 3(1) of the [National Health Service Act 1977] does not impose an absolute duty to provide the specified services. The Secretary of State is entitled to have regard to the resources made available to him under current government economic policy.””

Auld LJ added this at p 991D:

“As illustrated in the Cambridge Health Authority case [1999] 1 WLR 898 and Coughlan’s case [2001] QB 213, it is an unhappy but unavoidable feature of state funded health care that regional health authorities have to establish certain priorities in funding different treatments from their finite resources. It is natural that each authority, in establishing its own priorities, will give greater priority to life-threatening and other grave illnesses than to others obviously less demanding of medical intervention. The precise allocation and weighting of priorities is clearly a matter of judgment for each authority, keeping well in mind its statutory obligations to meet the reasonable requirements of all those within its area for which it is responsible. It makes sense to have a policy for the purpose – indeed, it might well be irrational not to have one – and it makes sense too that, in settling on such a policy, an authority would normally place treatment of transsexualism lower in its scale of priorities than, say, cancer or heart disease or kidney failure. Authorities might reasonably differ as to precisely where in the scale transsexualism should be placed and as to the criteria for determining the appropriateness and need for treatment of it in individual cases. It is proper for an authority to adopt a general policy for the exercise of such an administrative discretion, to allow for exceptions from it in “exceptional circumstances” and to leave those circumstances undefined: see In re Findlay [1985] AC 318, 335-336, per Lord Scarman. In my view, a policy to place transsexualism low in an order of priorities of illnesses for treatment and to deny it treatment save in exceptional circumstances such as overriding clinical need is not in principle irrational, provided that the policy genuinely recognises the possibility of there being an overriding clinical need and requires each request for treatment to be considered on its individual merits.

However, in establishing priorities – comparing the respective needs of patients suffering from different illnesses and determining the respective strengths of their claims to treatment – it is vital for an authority: (1) accurately to assess the nature and seriousness of each type of illness; (2) to determine the effectiveness of various forms of treatment for it; and (3) to give proper effect to that assessment and that determination in the formulation and individual application of its policy.”

Buxton LJ put the same points very clearly at p 997A:

“A number of propositions are clearly established, mainly by the decision of this court in R v Cambridge Health Authority ex p B [1995] 1 WLR 898. They are: 1. A health authority can legitimately, indeed must, make choices between the various claims on its budget when, as will usually be the case, it does not have sufficient funds to meet all of those claims. 2. In making those decisions the authority can legitimately take into account a wide range of considerations, including the proven success or otherwise of the proposed treatment; the seriousness of the condition that the treatment is intended to relieve; and the cost of that treatment. 3. The court cannot substitute its decision for that of the authority, either in respect of the medical judgments that the authority makes, or in respect of its view of priorities.

I further agree with Mr. Pannick’s submission that it follows from the foregoing propositions that a health authority can in the course of performing these functions determine that it will provide no treatment at all for a particular condition, even if the condition is medically recognised as an illness requiring intervention that is categorised as medical and curative, rather than merely cosmetic or a matter of convenience or lifestyle.

In all of this, the court’s only role is to require that such decisions are taken in accordance with equally well known principles of public law. Those principles include a requirement that the decisions are rationally based upon a proper consideration of the facts. The more important the interest of the citizen that the decision affects, the greater will be the degree of consideration that is required of the decision-maker. A decision that, as is the evidence in this case, seriously affects the citizen's health will require substantial consideration, and be subject to careful scrutiny by the court as to its rationality.”

Judged by these principles it is clear, in my judgment, that the claimant has no remedy in domestic law. I turn to consider human rights law.

The claimant sought to rely upon Articles 3 and 8 of the European Convention for the Protection of Human Rights and Fundamental Freedoms. Article 3 prohibits, without qualification or exception, “torture or … inhuman or degrading treatment”. Article 8, subject to the well-known exceptions in Article 8(2), guarantees the “right to respect for … private and family life”. In relation to Article 3 the argument was founded principally on D v United Kingdom (1997) 24 EHRR 423. In relation to Article 8 the key authority was Botta v Italy (1998) 26 EHRR 241. I was also referred to Bensaid v United Kingdom (2001) 33 EHRR 205 and Pretty v United Kingdom (2002) 35 EHRR 1, [2002] 2 FLR 45.

It so happens that these are topics I have had to consider on a number of occasions in recent months: see Re S (Adult Patient) (Inherent Jurisdiction: Family Life) [2002] EWHC 2278 (Fam), [2003] 1 FLR 292, R (Howard League for Penal Reform) v Secretary of State for the Home Department [2002] EWHC 2497 (Admin), [2003] 1 FLR 484, and R (A, B, X and Y) v East Sussex County Council (No 2) [2003] EWHC 167 (Admin). I do not propose to repeat again the analyses of the relevant Strasbourg jurisprudence that will be found in those cases. I merely make two observations:

Cases such as D v United Kingdom (1997) 24 EHRR 423, Bensaid v United Kingdom (2001] 33 EHRR 205 and Price v United Kingdom (2001) 34 EHRR 1285 show that such obligations may be thrown upon the authorities even where the underlying problem is the claimant’s physical or mental illness. As the Court said in Pretty v United Kingdom (2002) 35 EHRR 1at para [52]:

“Where treatment humiliates or debases an individual showing a lack of respect for, or diminishing, his or her human dignity or arouses feelings of fear, anguish or inferiority capable of breaking an individual’s moral and physical resistance, it may be characterised as degrading and also fall within the prohibition of Article 3. The suffering which flows from naturally occurring illness, physical or mental, may be covered by Article 3, where it is, or risks being, exacerbated by treatment, whether flowing from conditions of detention, expulsion or other measures, for which the authorities can be held responsible.”

Thus in Price v United Kingdom (2001) 34 EHRR 1285 the prison authorities were unable adequately to cope with the special needs of a four-limb-deficient thalidomide victim with numerous heath problems including defective kidneys who had been committed to prison for contempt of court. Holding that there had been a breach of Article 3 the Court said at para [30]:

“There is no evidence in this case of any positive intention to humiliate or debase the applicant. However, the Court considers that to detain a severely disabled person in conditions where she is dangerously cold, risks developing sores because her bed is too hard or unreachable, and is unable to go to the toilet or keep clean without the greatest of difficulty, constitutes degrading treatment contrary to Article 3.”

Now this is all very well but none of it can avail the claimant. Any reliance upon Article 8 in the present circumstances is, in my judgment, precluded by the decision of the Court of Appeal in R v North West Lancashire Health Authority ex p A [2000] 1 WLR 977. As Auld LJ put it at p 995G: “Article 8 imposes no positive obligations to provide treatment”. He went on at p 996C to approve this observation of Hidden J at first instance:

“The Convention does not give the applicants rights to free healthcare in general or to gender reassignment surgery in particular. Even if the applicants had such a right it would be qualified by the authority’s right to determine healthcare priorities in the light of its limited resources.”

Buxton LJ at p 1001E said this:

“ … it is plain that in this case there has occurred no interference with either the applicants’ private life or with their sexuality. The ECHR jurisprudence demonstrates that a state can be guilty of such interference simply by inaction, though the cases in which that has been found do not seem to go beyond an obligation to adopt measures to prevent serious infractions of private or family life by subjects of the state … Such an interference could hardly be founded on a refusal to fund medical treatment.”

Nor in my judgment can the claimant rely upon Article 3. Any reliance upon Article 3 in the present circumstances is also, in my judgment, precluded by the decision of the Court of Appeal in R v North West Lancashire Health Authority ex p A [2000] 1 WLR 977. Auld LJ explained why at p 996A:

“As Mr. Pannick observed, if the applicants have no case under article 8 of failure to respect their private and family life, they could not, a fortiori, establish that they were victims of inhuman or degrading treatment under article 3 since the same essential issues arise: see Olsson v Sweden (1988) 11 EHRR 259, 292, paras 85-87. And, as he also observed, a breach of the article requires “a particular level” of severity which, of course depends on the circumstances of the case. It is plain, in my view, that article 3 was not designed for circumstances of this sort of case where the challenge is as to a health authority’s allocation of finite funds between competing demands.”

Buxton LJ at 1000G said this:

“Article 3 of the ECHR addresses positive conduct by public officials of a high degree of seriousness and opprobrium. It has never been applied to merely policy decisions on the allocation of resources, such as the present case is concerned with. That is clear not only from the terms of article 3 itself, and the lack of any suggestion in any of the authorities that it could apply in a case even remotely like the present, but also from the explanation of the reach of article 3 that has been given by the Convention organs.”

He then referred to passages in Tyrer v United Kingdom (1978) 2 EHRR 1 and East African Asians v United Kingdom (1973) 3 EHRR 76 as illustrating the stress placed on the degree of seriousness of the conduct that article 3 addresses.

Underlying the reasoning of the Court of Appeal in R v North West Lancashire Health Authority ex p A [2000] 1 WLR 977, as it seems to me, is the fundamental point explained by Lord Hoffmann in Matthews v Ministry of Defence [2003] UKHL 4, [2003] 1 AC 1163 at para [26]:

“Human rights are the rights essential to the life and dignity of the individual in a democratic society. The exact limits of such rights are debatable and, although there is not much trace of economic rights in the 50-year-old Convention, I think it is well arguable that human rights include the right to a minimum standard of living, without which many of the other rights would be a mockery. But they certainly do not include the right to a fair distribution of resources or fair treatment in economic terms – in other words, distributive justice. Of course distributive justice is a good thing. But it is not a fundamental human right. No one looking at the legal systems of the member states of the Council of Europe could plausibly say that they treated distributive justice as a fundamental principle to which other considerations of policy or expediency should be subordinated.”

There is in any event another reason why the claim under Article 3 must fail. Article 3 is not engaged unless the “ill-treatment” in question attains a minimum level of severity and involves actual bodily injury or intense physical or mental suffering. However that is not this case. Making every allowance for the constant pain and suffering as described by Mr Edge that the claimant was having to endure – and I do not seek in any way to minimise it – the simple fact in my judgment is that nothing she had to endure was so severe or so humiliating as to engage Article 3.

Accordingly, and despite everything pressed on me by Mr Gordon, this claim cannot succeed insofar as it is founded on human rights law. I come finally, therefore, to the claim, based on European Community law, which lies at the heart of this case.

The claimant, as I have said, founds her claim on two different provisions of European Community law.

First she relies upon certain provisions in the Treaty establishing the European Community. Article 49 (ex Article 59) provides so far as material for present purposes that:

“[R]estrictions on freedom to provide services within the Community shall be prohibited in respect of nationals of Member States who are established in a State of the Community other than that of the person for whom the services are intended.”

Article 50 (ex Article 60) provides so far as material that:

“Services shall be considered to be ‘services’ within the meaning of this Treaty where they are normally provided for remuneration …

‘Services’ shall in particular include … (d) activities of the professions.”

Article 48 (ex Article 58), which is applied for this purpose by Article 55 (ex Article 66), provides that:

“Companies or firms formed in accordance with the law of a Member State and having their registered office, central administration or principal place of business within the Community shall, for the purposes of this Chapter, be treated in the same way as natural persons who are nationals of Member States.

‘Companies or firms’ means companies or firms constituted under civil or commercial law, including cooperative societies, and other legal persons governed by public or private law, save for those which are non-profit-making.”

Mr Kovats and Mr Lloyd Jones also direct my attention to Articles 46 and 152. Article 46.1 (ex Article 56.1), which is applied for this purpose by Article 55 (ex Article 66), provides that:

“The provisions of this Chapter and measures taken in pursuance thereof shall not prejudice the applicability of provisions laid down by law, regulation or administrative action providing for special treatment for foreign nationals on grounds of public policy, public security or public health.”

Article 152.5 (ex Article 129.5) provides so far as material that:

“Community action in the field of public health shall fully respect the responsibilities of the Member States for the organisation and delivery of health services and medical care.”

Article 49 on its face is directed to the prohibition of restrictions on the freedom of those who provide services, rather than to the protection of those for whom the services are provided – that is, in the present context, doctors and hospitals rather than their patients. But well-established ECJ jurisprudence, exemplified in this context by Luisi and Carbone v Ministero del Tesoro (Joined Cases 286/82 and 26/83) [1984] ECR 377, shows that Article 49 extends further. As the ECJ said in that case at para [16]:

“ … the freedom to provide services includes the freedom, for the recipients of services, to go to another Member State in order to receive a service there, without being obstructed by restrictions, even in relation to payments and that tourists, persons receiving medical treatment and persons travelling for the purpose of education or business are to be regarded as recipients of services.”

Secondly, the claimant relies upon Article 22 of Council Regulation 1408/71 of 14 June 1971 “on the application of social security schemes to employed persons, to self-employed persons and to members of their families moving within the Community”.

Article 22.1 provides that “An employed or self-employed person who satisfies the conditions of the legislation of the competent State for entitlement to benefits … ” – and it is common ground that the claimant is such a person – “and … (c) who is authorised by the competent institution to go to the territory of another Member State to receive there the treatment appropriate to his condition, shall be entitled … to benefits in kind provided on behalf of the competent institution by the institution of the place of stay or residence in accordance with the provisions of the legislation which it administers, as though he were insured with it”. Article 36(15) provides for reimbursement between the relevant institutions of the two Member States.

Article 22.2 in its original form provided so far as material that:

“The authorisation required under paragraph 1(c) may not be refused where the treatment in question cannot be provided for the person concerned within the territory of the Member State where he resides.”

The meaning of Article 22.2 in its original form was considered by the European Court of Justice in the two Pierik cases. In Bestuur van het Algemeen Ziekenfonds Drenthe-Platteland v Pierik (Pierik I) (Case 117/77) [1978] ECR 825 the ECJ said this (judgment, paras [17], [18], [22]) :

“[17] Whilst the worker’s right to receive benefits in kind provided in another Member State is subject pursuant to the said paragraph to an authorization, the power of the competent institution to refuse the authorization is nevertheless limited by the requirement laid down by the regulation that the worker should be guaranteed the opportunity of receiving treatment appropriate to his state of health provided in any member State, whatever the place of his residence or the Member State to which the social security institution to which he is affiliated belongs.

[18] The provision in the second subparagraph of Article 22(2) that the authorization “may not be refused where the treatment in question cannot be provided for the person concerned within the territory of the Member State in which he resides” means that the authorization may similarly not be refused in cases in which the treatment provided in the Member State of residence is less effective than that which the person concerned can receive in another Member State.

…

[22] The duty laid down in the second subparagraph of Article 22(2) to grant the authorization required under Article 22(1)(c) covers both cases where the treatment provided in another Member State is more effective than that which the person concerned can receive in the Member State where he resides and those in which the treatment in question cannot be provided on the territory of the latter State.”

In Bestuur van het Algemeen Ziekenfonds Drenthe-Platteland v Pierik (Pierik II) (Case 182/78) [1979] ECR 1977 the ECJ elaborated its earlier ruling (judgment, paras [10]–[13]):

“[10] … In setting a limit to the competent institution’s power of decision in this area, paragraph 17 of the said decision implicitly acknowledges that it is for that institution objectively to assess the medical grounds for granting or refusing the authorization required under Article 22(1)(c), having regard inter alia to the state of health of the person concerned, the seriousness of his sickness or disease and the effectiveness of the treatment in question.

[11] When the competent institution, having made use of such power, acknowledges that the treatment in question constitutes an effective treatment of the sickness or disease from which the person concerned suffers, its power of decision is thus bound by the obligation clearly and unequivocally imposed upon it by the second subparagraph of Article 22(2) of the regulation not to refuse in that case the authorization required under Article 22(1)(c).

[12] In fact it emerges from the provisions and the essential aims of Article 22 that it was the intention of the regulation to give medical requirements a decisive role in the decision of the competent institution to grant or refuse the aforesaid authorization by providing generally and unreservedly in the second subparagraph of Article 22(2) that authorization may not be refused “where the treatment in question cannot be provided for the person concerned within the territory of the Member State in which he resides”.

[13] Thus the answer to the second question should be that, when the competent institution acknowledges that treatment in question constitutes a necessary and effective treatment of the sickness or disease from which the person concerned suffers, the conditions for the application of the second subparagraph of Article 22(2) of Regulation No 1408/71 are fulfilled and the competent institution may not in that case refuse the authorization referred to by that provision and required under Article 22(1)(c).”

It was in the light of those decisions that the Council, on the proposal of the Commission, introduced what Advocate General Colomer described in Inizan v Caisse primaire d’assurance maladie des Hauts de Seine (2003) (Case C-56/01) as “une modifaction radicale” (opinion, para [24] – there is as yet no authentic English text).

Article 22.2 was amended by Council Regulation 2793/81 of 17 September 1981 because, as the recitals to the amending Regulation explained:

“the experience gained from implementing Regulations (EEC) No 1408/71 and (EEC) No 574/72 have revealed the need to make some improvements to those Regulations [and] consequently, the discretionary power of an institution of a Member State should be extended in granting or refusing authorization to a worker going to another Member State to receive appropriate medical treatment.”

As Advocate General Colomer put it in Inizan (opinion, para [25]):

“Selon la proposition de la Commission, la pratique avait démontré que l'application de la disposition donnait lieu à certains abus, car l'institution de sécurité sociale d'un État membre pouvait aisément se voir obligée d'accorder l'autorisation à un travailleur, même lorsqu'il n'avait jamais quitté son pays d'origine, dès lors qu'il souhaitait se rendre dans un autre État membre à seule fin d'y suivre un traitement médical qui n'était pas dispensé dans l'État membre d'affiliation. En outre, les difficultés financières des régimes nationaux d'assurance maladie justifiaient l'extension des pouvoirs d'appréciation des organismes de gestion, qui supportaient les frais exposés dans l'autre État, lors de l'octroi de l'autorisation. Elle a donc recommandé de modifier le paragraphe 2, deuxième alinéa, de telle sorte que l'autorisation ne puisse être refusée lorsque les soins figurent parmi les prestations prévues par la législation de l'État de résidence de l'assuré et ne peuvent être dispensés à ce dernier dans le délai normalement nécessaire pour l'obtenir.”

Consequently Article 22.2 was amended to read in material part:

“The authorisation required under paragraph 1(c) may not be refused where the treatment in question is among the benefits provided for by the legislation of the Member State on whose territory the person concerned resides and where he cannot be given such treatment within the time normally necessary for obtaining the treatment in question in the Member State of residence, taking account of his current state of health and the probable course of the disease.”

The inter-relationship between Article 49 of the Treaty and Article 22 of Regulation 1408/71 was considered by Advocate General Tesauro in Kohll v Union des caisses de maladie (Case C-158/96) [1998] ECR I-1931(opinion, paras [11], [31]-[32]):

“[11] … It will likewise be necessary to determine whether Article 22 of the Regulation applies to the case and, if so, whether this fact of itself takes the matter outside the scope of Articles 30 and 59. On this point, I cannot refrain from observing right away that a provision of secondary legislation which is what Article 22 of the Regulation is, cannot in any event be regarded as capable of excluding an examination into whether or not Treaty provisions, such as Articles 30 and 59, have been infringed. I therefore take the view that even if, after examination, the Luxembourg rules are found to come within the scope of the Regulation and to be consistent with it, it does not follow that Articles 30 and 59 are inapplicable to this case.

…

[31] I now turn to the question whether the contested measure is in conformity with Article 22 of the Regulation. It is not in dispute that both measures (Community and national) make the reimbursement of medical expenses incurred in another Member State subject to prior authorisation. Both measures also require, in order for such authorisation to be granted, that the benefits sought by the insured person are among those eligible for reimbursement under the legislation of the Member State in question. I would also recall that Member States are bound to grant authorisation, under Article 22(2), only where the treatment sought cannot be provided within such time as to ensure its effectiveness, thereby leaving all other eventualities to the Member States’ discretion. For their part, the national rules in force in Luxembourg, specifically Article 25 of the UCM Statute, provide that authorisation is to be granted only if the treatment sought is not available in Luxembourg or if the standard of the health-care provided is inadequate for the particular ailment from which the insured person is suffering.

Accordingly, there can be no doubt that the contested rules are consistent with Article 22 of the Regulation. It is quite clear that those rules, at least in terms of their wording, do not go beyond the limits laid down by the latter provision.

[32] That finding, however, as already indicated, does not mean, contrary to the contention of a number of governments during the proceedings, that there is no possibility of the contested rules conflicting with Articles 30 and 59 and that, consequently, there is no need to examine whether the former are compatible with the latter. Furthermore, the arguments put forward in support of this view are, even on the face of it, devoid of substance.”

The ECJ agreed (judgment, paras [22], [25]):

“[22] … the Luxembourg Government submit[s] that Article 22 of Regulation No 1408/71 lays down the principle that prior authorisation is required for any treatment in another Member State. To challenge the national provisions relating to reimbursement of the cost of services obtained abroad amounts to calling into question the validity of the corresponding provision in Regulation No 1408/71. …

[25] It must be stated that the fact that a national measure may be consistent with a provision of secondary legislation, in this case Article 22 of Regulation No 1408/71, does not have the effect of removing that measure from the scope of the provisions of the Treaty.”

The same point emerges from Vanbraekel v Alliance nationale des mutualities chretiennes (Case C-368/98) [2001] ECR I-5363 where the ECJ had to consider (judgment, paras [36]-[37]) whether an obligation by a member state to reimburse a patient might arise under Article 49 in circumstances where it did not under Article 22:

“[36] It should be noted that Article 22 of Regulation No 1408/71 is not intended to regulate, and therefore does not in any way prevent, reimbursement by Member States at the tariffs in force in the competent State, of costs incurred in connection with treatment provided in another Member State (see Case C-158/96 Kohll [1998] ECR I-1931, paragraph 27) where the legislation of the Member State in which the person concerned is insured makes provision for such reimbursement and the tariffs applied under that legislation are more beneficial than those applied by the Member State in which the treatment was provided.

[37] Although Article 22 of Regulation No 1408/71 does not have the effect of preventing extra reimbursement, additional to that resulting from the application of the system of the Member State where the treatment was provided, when the system applied in the Member State in which the person concerned is insured is more beneficial, that provision does not have the further effect of requiring such additional reimbursement. Consequently, it is necessary to consider whether such an obligation might arise under Article 59 of the Treaty.”

In the event the ECJ held (see below) that there was such an obligation.

Most recently the relationship between Article 49 and Article 22 has been explained by Advocate General Colomer in Inizan (opinion, paras [28]-[31]):

“[28] La situation des patients est différente selon qu'ils font usage de la procédure prévue à l'article 22 du règlement n° 1408/71 ou invoquent directement l'article 49 CE.

[29] L'article 22 du règlement n° 1408/71 régit exclusivement les relations entre les institutions de sécurité sociale des États membres. …

[30] Au contraire, l'article 49 CE permet à tous les ressortissants des États membres établis dans la Communauté de solliciter le remboursement, selon le barème de l'État d'affiliation, des frais médicaux exposés dans un autre État membre sans qu'ils aient à disposer d'une autorisation. …

[31] Ces deux dispositions visant des hypothèses différentes et leur application conduisant à des résultats distincts, il est difficile d'admettre qu'elles sont incompatibles. Les citoyens dont le risque de maladie n'est pas couvert par l'un des régimes légaux nationaux de sécurité sociale ou qui ont souscrit une assurance privée ne se voient pas empêchés de se rendre dans les autres États membres pour y bénéficier de prestations médicales. Ceux qui sont assurés par l'un de ces régimes peuvent choisir de faire usage de la procédure prévue à l'article 22, paragraphe 1, point c), du règlement n° 1408/71 ou, dans les limites établies par la jurisprudence, d'invoquer l'article 49 CE.”

From this I derive two propositions of importance for present purposes:

I shall deal first with the claim based on Article 49 of the Treaty. The Secretary of State denies that Article 49 applies at all. His position is very clear: “The only mechanism by which an NHS patient as such may be entitled to seek treatment in another member state is pursuant to Council Regulation 1408/71”. The PCT on this issue adopts a position of disengaged neutrality.

To put the arguments into context I need first to indicate the nature of the claims put forward in the various cases before the ECJ to which I have been referred and then to summarise the different types of social security and health care schemes that are to be found in the various member states of the Community.

The cases before the ECJ to which I have been referred on this point are, in chronological order, Kohll v Union des caisses de maladie (Case C-158/96) [1998] ECR I-1931, Vanbraekel v Alliance nationale des mutualities chretiennes (Case C-368/98) [2001] ECR I-5363, Geraets-Smits v Stichtung Ziekenfonds VGZ, Peerbooms v Stichtung CZ Groep Zorgverzekeringen (Case C-157/99) [2002] QB 409 and Muller-Fauré v Onderlinge Waarborgmaatschappij OZ Zorgverzekeringen UA, van Riet v Onderlinge Waarborgmaatschappij ZAO Zorgverzekeringen (2002) (Case C-385/99). I have also been referred to the French text (there is, as yet, no authentic English text) of the Opinion of Advocate General Colomer in Inizan v Caisse primaire d’assurance maladie des Hauts de Seine (2003) (Case C-56/01), a case in which the ECJ has not yet given judgment.

The claims in each of these cases can be summarised as follows:

As I understand it (see the Opinion of Advocate General Colomer in Geraets-Smits, para [46]) the statutory social security and health care systems in force in the member states of the Community fall into three broad categories:

The system in the Netherlands has three components based on three different Laws, the ZFW, the AWBZ and the WTZ. They were described by the ECJ in Geraets-Smits as follows (judgment, para [3]):

“Both the ZFW and the AWBZ establish a system of benefits in kind under which an insured person is entitled not to reimbursement of costs incurred for medical treatment but to free treatment. Both Laws are based on a system of agreements made between sickness funds and providers of health care. The WTZ, on the other hand, establishes a system under which insured persons are reimbursed costs and is not based on a system of agreements.”

The claims in Geraets-Smits, Peerbooms, Muller-Faure and van Riet all related to the ZFW. The operation of the ZFW is described in some detail in the Opinion of Advocate General Colomer in Geraets-Smits, starting at paras [40]-[41]:

“[40] … the Netherlands has organised a compulsory sickness insurance scheme which covers all persons whose income does not exceed a certain amount and which is managed by sickness funds with separate legal personality. It is financed from the contributions paid by insured persons and employers, and an annual payment made by the state, from the public purse, to the general sickness insurance fund. The sickness insurance funds are responsible for concluding with medical practitioners and specialist institutions agreements for the provision of health care to the persons registered with them.

[41] Unlike the situation in Kohll with respect to the social security scheme in Luxembourg … health care under the Netherlands compulsory sickness scheme is free for insured persons. (It would be true to say that it is free in virtually all cases, even though certain benefits may be subject to the payment of a contribution by the person concerned. The legislation does provide, however, that the persons receiving the benefits do not all have to make the same contribution.) In order to obtain the health-care benefits they require, however, insured persons must use one of the medical practitioners or health-care institutions with whom or which their fund has concluded an agreement because, if they decide to use non-contracted providers, they are required to pay any costs they incur, without entitlement to reimbursement.”

Having then described in some detail the funding arrangements as between the sickness funds and the contracting doctors and hospitals, the Advocate General continued at para [46]:

“Under that system of compulsory sickness insurance, the funds operate by concluding with health-care institutions and independent medical practitioners agreements in which they determine in advance the extent and quality of the benefits to be provided, and the financial contribution the fund will make, which, for practitioners, consists in the payment of a fixed flat-rate amount, and, for each hospital, in the payment of an attendance charge, which is intended to finance the institution rather than to cover the real cost of hospital accommodation.

Viewed from that angle, it is very much like the systems operated in certain member states where the social security institutions have their own resources and staff which they engage directly – such staff have the status of civil servants or quasi civil servants whose relationship with the social security institutions is governed in certain member states by public law – for a preset number of hours and a given salary. To my mind, it is clearly different from other systems, such as that in Luxembourg, the subject matter of Kohll v Union des Caisses de Maladie (Case C-158/96) [1998] ECR I-1931, which I have already described (and the French and Belgian systems operate according to the same principles as that in Luxembourg). Under the Luxembourg system, as indeed the court confirmed, the relationship between the insured person and the practitioner is characterised by a provision of “services” within the meaning of article 60 of the Treaty (that also applies to health care provided in hospital, since, although treatment is paid for by the sickness fund rather than the insured person, there is a charge for each medical procedure based on the cost of the benefits involved), but I am convinced that there is no such provision of “services” under the system I am considering here, since the element of remuneration required by article 60 of the Treaty is lacking.”

From the outset various member states, amongst whom the United Kingdom has been prominent, have sought to deploy a succession of arguments designed to demonstrate that Article 49 does not apply in situations such as the one I am concerned with. Thus far, all such arguments have been rejected by the ECJ.

In Kohll various governments, including the United Kingdom, sought to argue (see the judgment, para [16]) that “the rules at issue in the main proceedings do not fall within the scope of the Community provisions on freedom to provide services, in that they concern social security, and so should be examined solely from the point of view of Article 22 of Regulation No 1408/71.” That argument was rejected by the ECJ:

“[20] The Court has held that the special nature of certain services does not remove them from the ambit of the fundamental principle of freedom of movement (Case 279/80 Webb [1981] ECR 3305, paragraph 10).

[21] Consequently, the fact that the national rules at issue in the main proceedings fall within the sphere of social security cannot exclude the application of Articles 59 and 60 of the Treaty.”

In Vanbraekel a number of governments, including as I understand it the United Kingdom, sought to argue (see the judgment, para [39]) that “hospital services cannot constitute an economic activity for the purposes of Article 60 of the EC Treaty (now Article 50 EC).” That argument was likewise rejected by the ECJ:

“[41] It is settled case-law that medical activities fall within the scope of Article 60 of the Treaty, there being no need to distinguish in that regard between care provided in a hospital environment and care provided outside such an environment (see Joined Cases 286/82 and 26/83 Luisi and Carbone [1984] ECR 377, paragraph 16, Case C-159/90 Society for the Protection of Unborn Children Ireland [1991] ECR I-4685, paragraph 18, and Kohll, paragraphs 29 and 51).

[42] It is also settled case-law that the special nature of certain services does not remove them from the ambit of the fundamental principle of freedom of movement (Case 279/80 Webb [1981] ECR 3305, paragraph 10, and Kohll paragraph 20), so that the fact that the national rules at issue in the main proceedings are social security rules cannot exclude application of Articles 59 and 60 of the Treaty (Kohll, paragraph 21).

[43] … the hospital services at issue in the main proceedings fall within the scope of freedom to provide services … ”

In Geraets-Smits and Peerbooms the arguments against the applicability of Article 49 were more complex. The Advocate General summarised them as follows (opinion, para [32]):

“ … the member states appear to fall into two major groups in terms of their views. Those in the first group, namely Belgium, France and Austria, consider that the health care provided under a public social security scheme is a service within the meaning of article 60 of the Treaty. Those in the second group, however, which comprises Germany, Ireland, the Netherlands, the United Kingdom, Denmark, Sweden, Finland and Iceland, take the view that health care under a social security system organised in the form of benefits in kind does not constitute a service within the meaning of article 60. Portugal and Norway have expressed no views on this point. The division of opinion among the member states ends there, since they all agree that the requirement of prior authorisation from the institution of the state of insurance in order to receive care in another member state, although constituting a barrier to freedom to provide services, is justified.”

The various arguments were described as follows by the ECJ (judgment, paras [48]-[51]):

“[48] A number of the governments which have submitted written observations to the court have argued that hospital services cannot constitute an economic activity within the meaning of article 60 of the Treaty, particularly when they are provided in kind and free of charge under the relevant sickness insurance scheme.

[49] Relying in particular on Belgian State v Humbel (Case 263/86) [1988] ECR 5365, 5388, paras 17-19, and Society for the Protection of Unborn Children Ireland Ltd v Grogan (Case C-159/90) [1991] ECR I-4685, 4739, para 18, they argue, in particular, that there is no remuneration within the meaning of article 60 of the Treaty where the patient receives care in a hospital infrastructure without having to pay for it himself or where all or part of the amount he pays is reimbursed to him.

[50] Some of those governments also maintain that it follows from Gravier v City of Liège (Case 293/83) [1985] ECR 593 and Wirth v Landeshauptstadt Hannover (Case C-109/92) [1993] ECR I-6447, 6469, para 17, that a further condition to be satisfied before a service can constitute an economic activity within the meaning of article 60 of the Treaty is that the person providing the service must do so with a view to making a profit.

[51] The German Government considers that the structural principles governing the provision of medical care are inherent in the organisation of the social security systems and do not come within the sphere of the fundamental economic freedoms guaranteed by the EC Treaty, since the persons concerned are unable to decide for themselves the content, type and extent of a service and the price they will pay.”

On this issue the Advocate General and the ECJ took different views. The Advocate General accepted the view put forward by, amongst others, the Netherlands and the United Kingdom. Having drawn the distinction between the Luxembourg scheme and the Dutch scheme in the passage in his opinion to which I have already referred, the Advocate General continued in relation to the Dutch scheme (opinion, para [42]):

“That fundamental difference causes me to question whether the treatment provided by medical practitioners and health-care institutions in such circumstances may be regarded as a service within the meaning of article 60 of the Treaty, in view of the fact that the person for whom the service is provided does not receive it in return for remuneration.”

He continued (opinion, para [46]):

“Under the Luxembourg system, as indeed the court confirmed, the relationship between the insured person and the practitioner is characterised by a provision of “services” within the meaning of article 60 of the Treaty (that also applies to health care provided in hospital, since, although treatment is paid for by the sickness fund rather than the insured person, there is a charge for each medical procedure based on the cost of the benefits involved), but I am convinced that there is no such provision of “services” under the system I am considering here, since the element of remuneration required by article 60 of the Treaty is lacking.”

He concluded (opinion, para [49]):

“In the light of the characteristics of the Netherlands compulsory sickness insurance scheme which I have described, I take the view that the health-care benefits in kind which it provides to insured persons lack the element of remuneration and are not therefore services within the meaning of article 60 of the EC Treaty.”

The ECJ disagreed. Having rehearsed the various arguments as I have already set them out, it continued (judgment, paras [52]-[59]):

“[52] None of those arguments can be upheld.

[53] It is settled case law that medical activities fall within the scope of article 60 of the Treaty, there being no need to distinguish in that regard between care provided in a hospital environment and care provided outside such an environment: see Luisi v Ministero del Tesoro (Joined Cases 286/82 and 26/83) [1984] ECR 377, 403, para 16; Society for the Protection of Unborn Children v Grogan [1991] ECR I-4685, 4739, para 18, concerning advertising for clinics involved in the deliberate termination of pregnancies, and Kohll [1998] ECR I-1931, 1945, 1950, paras 29 and 51.

[54] It is also settled case law that the special nature of certain services does not remove them from the ambit of the fundamental principle of freedom of movement (Criminal proceedings against Webb (Case 279/80) [1981] ECR 3305, 3323, para 10, and Kohll, para 20), so that the fact that the national rules at issue in the main proceedings are social security rules cannot exclude application of articles 59 and 60 of the Treaty (Kohll, para 21).

[55] With regard more particularly to the argument that hospital services provided in the context of a sickness insurance scheme providing benefits in kind, such as that governed by the ZFW, should not be classified as services within the meaning of article 60 of the Treaty, it should be noted that, far from falling under such a scheme, the medical treatment at issue in the main proceedings, which was provided in member states other than those in which the persons concerned were insured, did lead to the establishments providing the treatment being paid directly by the patients. It must be accepted that a medical service provided in one member state and paid for by the patient should not cease to fall within the scope of the freedom to provide services guaranteed by the Treaty merely because reimbursement of the costs of the treatment involved is applied for under another member state’s sickness insurance legislation which is essentially of the type which provides for benefits in kind.

[56] Furthermore, the fact that hospital medical treatment is financed directly by the sickness insurance funds on the basis of agreements and preset scales of fees is not in any event such as to remove such treatment from the sphere of services within the meaning of article 60 of the Treaty.

[57] First, it should be borne in mind that article 60 of the Treaty does not require that the service be paid for by those for whom it is performed: Bond van Adverteerders v Netherlands State (Case 352/85) [1988] ECR 2085, 2131, para 16, and Deliège v Ligue francophone de judo et disciplines associées ASBL (Joined Cases C-51/96 and 191/97) [2000] ECR I-2549, 2616, para 56.

[58] Secondly, article 60 of the Treaty states that it applies to services normally provided for remuneration and it has been held that, for the purposes of that provision, the essential characteristic of remuneration lies in the fact that it constitutes consideration for the service in question: Belgian State v Humbel (Case 263/86) [1988] ECR 5365, 5388, para 17. In the present cases, the payments made by the sickness insurance funds under the contractual arrangements provided for by the ZFW, albeit set at a flat rate, are indeed the consideration for the hospital services and unquestionably represent remuneration for the hospital which receives them and which is engaged in an activity of an economic character.

[59] … the provisions of services at issue in the main proceedings do fall within the scope of the freedom to provide services within the meaning of articles 59 and 60 of the Treaty … ”

It will be appreciated that the most important part of this ruling for present purposes is in that part of para [55] where the ECJ said (and I repeat the crucial words):

“It must be accepted that a medical service provided in one member state and paid for by the patient should not cease to fall within the scope of the freedom to provide services guaranteed by the Treaty merely because reimbursement of the costs of the treatment involved is applied for under another member state’s sickness insurance legislation which is essentially of the type which provides for benefits in kind.”

The reference there to “legislation which is essentially of the type which provides for benefits in kind” is, of course, primarily to the Dutch ZFW scheme. But it is obvious that the decision of the ECJ in Geraets-Smits and Peerbooms poses certain difficulties for the Secretary of State and not surprisingly Mr Lloyd Jones was at pains to try and minimise its impact. He questioned some of the ECJ’s reasoning, suggesting that the ECJ had in fact given no reasons at all to support either its bald rejection (in para [52]) of the argument summarised in para [50] or, more importantly, its crucial holding in para [55]. He pointed out that the ECJ appeared not to have addressed the implications of the specific reference to non–profit–making bodies in Article 48 nor to have addressed the implications of Articles 46.1 and 152.5. He sought to suggest that the analysis in paras [55]–[58], even if correctly applicable to the ZFW scheme in the Netherlands, was inapt if sought to be applied to the NHS.

Mr Lloyd Jones’s points, when reduced to essentials, really came to this: First, that the NHS, in contrast to the ZFW scheme, is not a sickness insurance fund. It is itself a provider of health care and, in the overwhelming majority of cases, delivers that health care itself direct to its patients. Only exceptionally, and for a comparatively small number of patients, does it discharge its responsibilities by purchasing hospital services from other providers, either within the United Kingdom (from the private health sector) or abroad. Second, and a related point, that even though under both the NHS and the ZFW scheme treatment is free at the point of delivery to the patient, the doctors are remunerated differently, the Dutch doctors being reimbursed preset scale fees by the state whilst in the United Kingdom hospital doctors are salaried employees of the state. That may all be so, but I find it hard to imagine that the ECJ could have thought that the applicability of Article 49 might turn on such seemingly technical and almost marginal points.

Be that as it may, I have a more fundamental difficulty with Mr Lloyd Jones’s approach. It is not, as it seems to me, for the courts of a member state to dissect and deconstruct the considered decision of the ECJ on an issue going to the applicability of a provision as fundamental as Article 49, let alone to treat a decision of the ECJ on such a point as having been (as Mr Lloyd Jones seemed at times almost to be hinting with his submissions based on Article 48) decided per incuriam. Subject only to any subsequent decisions of the ECJ it is my obligation to apply what seem to me to be the quite clear and explicit statements of principle to be found in the relevant parts of the ECJ’s judgment in Geraets-Smits and Peerbooms and, in particular, the crucial statement of principle in para [55].

Mr Lloyd Jones’s arguments have in any event been overtaken by the subsequent decision of the ECJ in Muller-Fauré and van Riet, where issues very similar to those previously considered in Geraets-Smits and Peerbooms were raised by a number of member states, including the Netherlands and the United Kingdom. So far as concerned the Netherlands (judgment, para [31]):

“the referring court draws attention to the characteristics of the Netherlands sickness insurance scheme. In essence, unlike ‘reimbursement’ schemes, the scheme guarantees that benefits in kind will be provided. In the submission of the defendants in the main actions, the financial balance of the scheme could be jeopardised if it were possible for insured persons to obtain reimbursement, without prior authorisation, of the costs of care provided in another Member State. The national court refers in that regard to national measures taken to control the costs of hospital care, in particular the rules laid down … concerning the planning and geographical distribution of care, and those … limiting reimbursement to care provided by authorised hospitals.”

Those arguments are set out in greater detail in passages (judgment, paras [47]-[50]) which there is no need for me to quote.

In the light of this (see the judgment, para [34]) “the referring court informed the Court that it was maintaining the reference since Smits and Peerbooms did not specifically deal with the attributes of the Netherlands sickness insurance scheme, which is a benefits-in-kind scheme based on agreements.”

The United Kingdom submitted (see judgment, paras [55]-[56]) that:

“[55] … if insured persons were entitled to go to a Member State other than that in which they are insured in order to receive treatment there, there would be adverse consequences for the setting of priorities for medical treatment and the management of waiting lists, which are significant aspects of the organisation of sickness insurance. In that regard, the United Kingdom Government points out that the finite financial resources allocated to the National Health Service (‘the NHS’)” – the French text of the judgment here uses the words “allouées au National Health Service (service national de santé, ci–après le ‘NHS’)” – “are managed by local health authorities which establish timetables based on clinical judgments and medically determined priorities for different treatments. Patients do not have the right to demand a certain timetable for their hospital treatment. It follows that if patients could shorten their waiting time by obtaining, without prior authorisation, medical treatment in other Member States for which the competent fund was none the less obliged to assume the cost, the financial balance of the system would be threatened and the resources available for more urgent treatment would be severely depleted, thereby placing at risk its ability to provide adequate levels of health care.

[56] The United Kingdom Government adds that if hospital services were to be liberalised, its own hospitals would be unable to predict either the loss of demand that would follow from recourse being had to hospital treatment in other Member States or the increase in demand that would follow from persons insured in those other States being able to seek hospital treatment in the United Kingdom. Those effects of liberalisation would not necessarily offset each other and the impact would be different for every hospital in the United Kingdom.”

Furthermore, the United Kingdom (see the judgment, para [59]):

“points to the specific characteristics of the NHS” – in the French text, “particularités du NHS” – “and asks the Court to uphold the principle that health care provided under such a national sickness insurance scheme does not fall within the scope of Article 60 of the Treaty and that the NHS,” – in French, “le NHS” – “which is a non-profit-making body, is not a service provider for the purposes of the Treaty.”

The ECJ responded to this as follows:

“[38] It should be borne in mind, as a preliminary point, that it is settled case-law that medical activities fall within the scope of Article 60 of the Treaty, there being no need to distinguish in that regard between care provided in a hospital environment and care provided outside such an environment (see, most recently, Smits and Peerbooms, paragraph 53).

[39] The Court also found, in paragraphs 54 and 55 of Smits and Peerbooms, that the fact that the applicable rules are social security rules and, more specifically, provide, as regards sickness insurance, for benefits in kind rather than reimbursement does not mean that the medical treatment in question falls outside the scope of the freedom to provide services guaranteed by the EC Treaty. Indeed, in the disputes before the national court, the treatment provided in a Member State other than that in which the persons concerned were insured resulted in direct payment by the patient to the doctor providing the service or the establishment in which the care was provided. …

[103] … as has already been made clear in paragraph 39 above, a medical service does not cease to be a provision of services because it is paid for by a national health service” – in the French text, “par un service national de santé” – “or by a system providing benefits in kind. The Court has, in particular, held that a medical service provided in one Member State and paid for by the patient cannot cease to fall within the scope of the freedom to provide services guaranteed by the Treaty merely because reimbursement of the costs of the treatment involved is applied for under another Member State’s sickness insurance legislation which is essentially of the type which provides for benefits in kind (Smits and Peerbooms, paragraph 55). The requirement for prior authorisation where a person is subsequently to be reimbursed for the costs of that treatment is precisely what constitutes, as has already been stated in paragraph 44 above, the barrier to freedom to provide services, that is to say, to a patient’s ability to go to the medical service provider of his choice in a Member State other than that of affiliation. There is thus no need, from the perspective of freedom to provide services, to draw a distinction by reference to whether the patient pays the costs incurred and subsequently applies for reimbursement thereof or whether the sickness fund or the national budget pays the provider directly.”

That last passage is important. It shows that the entire debate in relation to Article 49 is not about the right to travel abroad for treatment – for which, of course, one needs no more than a passport – but about the right to be reimbursed for the cost of treatment obtained abroad. And, moreover, it shows that in this context it matters not whether the cost of such treatment is to be reimbursed to the patient or paid direct to the foreign service provider. Nor does it matter if the reimbursement comes from a sickness fund or fromthe national budget.

The ECJ’s rejection of the arguments addressed to it by the Netherlands and the United Kingdom is also clear in that part of the answer which it gave to the question referred to it where it said (see judgment, para [109]):

“ … Articles 59 and 60 of the Treaty … preclude … legislation in so far as it makes the assumption of the costs of non-hospital care provided in another Member State by a person or establishment with whom or which the insured person’s sickness fund has not concluded an agreement conditional upon prior authorisation by the fund, even when the national legislation concerned sets up a system of benefits in kind under which insured persons are entitled not to reimbursement of costs incurred for medical treatment, but to the treatment itself which is provided free of charge.”

In my judgment, whatever ambiguity in relation to the position of the NHS there may still have been following the decision of the ECJ in Geraets-Smits and Peerbooms – and I should make clear that I do not accept that there was in fact any ambiguity – has now been set to rest. It is quite clear that in para [39] of its judgment in Muller-Fauré and van Riet the ECJ was reiterating in relation to the Dutch ZFW scheme the view it had earlier expressed in para [55] of its judgment in Geraets-Smits and Peerbooms. It is also, in my judgment, quite clear that the ECJ rejected the attempt by the United Kingdom to draw some distinction between the NHS and the Dutch ZFW scheme and, more importantly, that it rejected the attempt by the United Kingdom to argue that the provision of health care under the NHS does not fall within the scope of Articles 49 and 50.

It is quite clear (if not from the English text then certainly from the French text) that when in para [103] of its judgment in Muller-Fauré and van Riet the ECJ referred to “a national health service” (in French, “un service national de santé”) it was using that expression both in the generic sense and specifically as referring to the NHS: compare the passage in para [55] where the ECJ referred to “the finite financial resources allocated to the National Health Service (‘the NHS’)” (in French, “ … allouées au National Health Service (service national de santé, ci–après le ‘NHS’)”. In my judgment the fundamental principles articulated by the ECJ in the passages in paras [103] and [109] of its judgment in Muller-Fauré and van Riet that I have set out above were intended to be, and are, as applicable in the case of the NHS as in the case of the Dutch ZFW – a conclusion which, as it seems to me, is only reinforced by the ECJ’s important, and plainly intentional, reference in para [109] to it being immaterial whether reimbursement comes from a “sickness fund or the national budget” (emphasis added).

The Secretary of State is wrong. Article 49 applies in principle to a case such as this. The medical services provided to the claimant in France and paid for by her there cannot cease to fall within the scope of the freedom to provide services guaranteed by the Treaty – do not fall outside the ambit of Article 49 – merely because reimbursement of the costs of the treatment involved is applied for under the NHS, any more than the medical services provided to the various claimants in Geraets-Smits, Peerbooms, Muller-Fauré and van Riet could fall outside the ambit of Article 49 merely because reimbursement of the costs of the treatment involved in those cases was applied for under the Dutch ZFW scheme. Article 49 does not cease to apply merely because reimbursement is applied for under a scheme which, as in the case of both the NHS and the ZFW, essentially provides for benefits in kind.

In the light of all this the ECJ learning on what I have referred to as the applicability of Article 49 can I think be summarised for present purposes as follows:

Despite the applicability of Article 49, various member states, amongst whom the United Kingdom has again been prominent, have from the outset sought to argue that national systems which make reimbursement subject to prior authorisation and other restrictions are not a barrier to and do not restrict freedom to provide services or, alternatively, that if they do such restrictions are nonetheless justified. In this connection member states have pointed to Article 46 (ex Article 56).

The ECJ has consistently rejected the first argument. It has been consistently sceptical about the second: it has never been prepared to go as far as the member states would wish though it has not – yet – been willing to go as far in the other direction as some would wish.

The ECJ addressed the first argument in Kohll as follows:

“[31] Mr Kohll and the Commission submit that the fact that reimbursement of the cost of medical services, in accordance with the legislation of the State of insurance, is subject to prior authorisation by the institution of that State where the services are provided in another Member State constitutes a restriction on freedom to provide services within the meaning of Articles 59 and 60 of the Treaty.

[32] The Member States which have submitted observations consider, on the contrary, that the rules at issue do not have as their purpose or effect to restrict freedom to provide services, but merely lay down the conditions for the reimbursement of medical expenses.

[33] It should be noted that, according to the Court’s case-law, Article 59 of the Treaty precludes the application of any national rules which have the effect of making the provision of services between Member States more difficult than the provision of services purely within one Member State (Case C-381/93 Commission v France [1994] ECR I-5145, paragraph 17).

[34] While the national rules at issue in the main proceedings do not deprive insured persons of the possibility of approaching a provider of services established in another Member State, they do nevertheless make reimbursement of the costs incurred in that Member State subject to prior authorisation, and deny such reimbursement to insured persons who have not obtained that authorisation. Costs incurred in the State of insurance are not, however, subject to that authorisation.

[35] Consequently, such rules deter insured persons from approaching providers of medical services established in another Member State and constitute, for them and their patients, a barrier to freedom to provide services (see Joined Cases 286/82 and 26/83 Luisi and Carbone v Ministero del Tesoro [1984] ECR 377, paragraph 16, and Case C-204/90 Bachmann v Belgium [1992] ECR I-249, paragraph 31).

[36] The Court must therefore examine whether a measure of the kind at issue in this case may be objectively justified.”

The ECJ has never resiled from that position. In Vanbraekel it said this:

“[43] Since the hospital services at issue in the main proceedings fall within the scope of freedom to provide services, it is necessary to go on to consider whether the fact that national legislation does not guarantee a person covered by its social insurance scheme who has been authorised to receive hospital treatment in another Member State in accordance with Article 22(1)(c) of Regulation No 1408/71 a level of payment equivalent to that to which he would have been entitled if he had received hospital treatment in the Member State in which he was insured entails a restriction of freedom to provide services within the meaning of Article 59 of the Treaty.

[44] In that regard, it is settled case-law that Article 59 of the Treaty precludes the application of any national rules which have the effect of making the provision of services between Member States more difficult than the provision of services purely within one Member State (Case C-381/93 Commission v France [1994] ECR I-5145, paragraph 17, and Kohll, paragraph 33).

[45] In the present case, there is no doubt that the fact that a person has a lower level of cover when he receives hospital treatment in another Member State than when he undergoes the same treatment in the Member State in which he is insured may deter, or even prevent, that person from applying to providers of medical services established in other Member States and constitutes, both for insured persons and for service providers, a barrier to freedom to provide services (see, by analogy, Luisi and Carbone, paragraph 16, Case C-204/90 Bachmann v Belgium [1992] ECR I-249, paragraph 31, and Kohll, paragraph 35).

[46] Consequently, it is necessary to examine whether the fact that the national legislation of a Member State does not guarantee a person insured in that State at least an equally advantageous level of cover when hospital services are provided in another Member State could be objectively justified.”

It said precisely the same in Geraets-Smits (judgment, paras [59]-[69], esp paras [61], [69]). It maintained the same position in Muller-Faure (judgment, paras [40]-[44]).

It is clear, in particular from paras [34]–[35] of its judgment in Kohll and para [46] of its judgment in Vanbraekel, that in the eyes of the ECJ the essential vice which in this context implicates Article 49 is the operation by a member state of a system under which the reimbursement of the cost of treatment obtained in another member state is denied to those who have not obtained prior authorisation whilst at the same time the cost of treatment in the ‘home’ member state is not subject to such authorisation; or, putting the same point in different words, where the system operated by a member state does not guarantee at least an equally advantageous level of cover where treatment is obtained in another member state. Such differential financial treatment, in the view of the ECJ, deters patients from approaching providers of medical services in another member state and thus creates, both for the foreign doctor and for his potential patient, a barrier to freedom to provide services.

So a national system which, as in the case of the United Kingdom, makes reimbursement subject to prior authorisation and other restrictions thereby creates a barrier to and restricts freedom to provide services, in a manner which requires to be justified if there is not to be a breach of Article 49. I turn therefore to the question of justification.

In Kohll both Luxembourg and a number of other member states sought to justify the requirement of prior authorisation on various grounds summarised by the ECJ as follows:

“[37] … Governments of the Member States which have submitted observations submit that freedom to provide services is not absolute and that reasons connected with the control of health expenditure must be taken into consideration. The requirement of prior authorisation constitutes the only effective and least restrictive means of controlling expenditure on health and balancing the budget of the social security system.

[38] According to … the Luxembourg Government and the Commission, the risk of upsetting the financial balance of the social security scheme, which aims to ensure a balanced medical and hospital service available to all its insured, constitutes an overriding reason in the general interest capable of justifying restrictions on freedom to provide services. …

[43] The Luxembourg Government also relies on grounds based on the protection of public health, arguing, first, that the rules at issue are necessary to guarantee the quality of medical services, which in the case of persons going to another Member State can be ascertained only at the time of the request for authorisation, and, second, that the Luxembourg sickness insurance system aims to provide a balanced medical and hospital service open to all insured persons.”

The ECJ rejected all these arguments:

[41] It must be recalled that aims of a purely economic nature cannot justify a barrier to the fundamental principle of freedom to provide services (see, to that effect, Case C-398/95 SETTG v Ypourgos Ergasias [1997] ECR I-3091, paragraph 23). However, it cannot be excluded that the risk of seriously undermining the financial balance of the social security system may constitute an overriding reason in the general interest capable of justifying a barrier of that kind.

[42] But, contrary to the submissions of … the Luxembourg Government, it is clear that reimbursement of the costs of dental treatment provided in other Member States in accordance with the tariff of the State of insurance has no significant effect on the financing of the social security system.”

Recognising (para [45]) that “under Articles 56 and 66 of the EC Treaty Member States may limit freedom to provide services on grounds of public health” the ECJ continued:

[46] However, that does not permit them to exclude the public health sector, as a sector of economic activity and from the point of view of freedom to provide services, from the application of the fundamental principle of freedom of movement (see Case 131/85 Gül v Regierungspräsident Düsseldorf [1986] ECR 1573, paragraph 17). …

[48] … doctors and dentists established in other Member States must be afforded all guarantees equivalent to those accorded to doctors and dentists established on national territory, for the purposes of freedom to provide services.

[49] Consequently, rules such as those applicable in the main proceedings cannot be justified on grounds of public health in order to protect the quality of medical services provided in other Member States.

[50] As to the objective of maintaining a balanced medical and hospital service open to all, that objective, although intrinsically linked to the method of financing the social security system, may also fall within the derogations on grounds of public health under Article 56 of the Treaty, in so far as it contributes to the attainment of a high level of health protection.

[51] Article 56 of the Treaty permits Member States to restrict the freedom to provide medical and hospital services in so far as the maintenance of a treatment facility or medical service on national territory is essential for the public health and even the survival of the population (see, with respect to public security within the meaning of Article 36 of the Treaty, Case 72/83 Campus Oil v Minister for Industry and Energy [1984] ECR 2727, paragraphs 33 to 36).”

The ECJ said much the same in Vanbraekel (judgment, paras [47]-[49]), in Geraets-Smits (judgment, paras [72]-[74]) and in Muller-Fauré (judgment, paras [67], [72]-[73]).

In Kohll itself these arguments failed. As the ECJ concluded (judgment, paras [52]-[53]):

“[52] However, neither [the Luxembourg authority] nor the Governments of the Member States which have submitted observations have shown that the rules at issue were necessary to provide a balanced medical and hospital service accessible to all. None of those who have submitted observations has argued that the rules were indispensable for the maintenance of an essential treatment facility or medical service on national territory.

[53] The conclusion must therefore be drawn that the rules at issue in the main proceedings are not justified on grounds of public health.”

These issues were at the heart of the debate in Geraets-Smits where the scene was set by the ECJ (judgment, para [60]) as follows:

“It is necessary to determine whether there is a restriction on freedom to provide services within the meaning of article 59 of the Treaty where the costs of treatment provided in a hospital in another member state is assumed under the sickness insurance scheme only on condition that the person receiving the treatment obtains prior authorisation, which is granted only if the treatment concerned is covered by the sickness insurance scheme of the member state in which the patient is insured, which requires that the treatment be “normal within the professional circles concerned”, and where the insured person's sickness fund has decided that his medical treatment requires that he be treated in the hospital establishment concerned, presupposing that adequate timely treatment cannot be provided by a contracted care provider in the member state in which the patient is insured.”

I am not concerned in the present case with any condition that the treatment be “normal”, so I can in large measure pass over that part of the ECJ’s judgment directed to that issue.

Having held (judgment, paras [61]-[69]) that these restrictions imposed a barrier to freedom to provide services, the ECJ continued:

“[70] Consequently, it is necessary to examine whether, in so far as they concern medical services provided within a hospital infrastructure, such as those at issue in the main proceedings, such rules can be objectively justified.

[71] In that regard, it is first necessary to determine whether there are overriding reasons which can be accepted as justifying barriers to freedom to provide medical services supplied in the context of a hospital infrastructure, then to determine whether the prior authorisation principle is justifiable in the light of such overriding needs and last to consider whether the conditions governing the grant of prior authorisation can themselves be justified.”

Having referred (judgment, paras [72]-[74]) to the principles laid down in Kohll paras [41], [50] and [51], the ECJ continued:

“[75] It is therefore necessary to determine whether the national rules at issue in the main proceedings can actually be justified in the light of such overriding reasons and, in such a case, in accordance with settled case law, to make sure that they do not exceed what is objectively necessary for that purpose and that the same result cannot be achieved by less restrictive rules: Commission of the European Communities v Federal Republic of Germany (Case 205/84) [1986] ECR 3755, 3802-3803, paras 27 and 29; Commission of the European Communities v Italian Republic (Case C-180/89) [1991] ECR I-709, 722-723, paras 17 and 18, and Ramrath v Ministre de la Justice (Case C-106/91) [1992] ECR I-3351, 3384-3385, paras 30 and 31.”

That was reiterated by the ECJ in Muller-Fauré (judgment, para [68]). The same principle had previously been expressed by the Court of Appeal in R v Human Fertilisation and Embryology Authority ex p Blood [1999] Fam 151 at p 183D as follows:

“the fact that there is interference with the freedom to provide services does not mean that article 59 is infringed. It means no more than the second stage has been reached and the interference has to be justified in accordance with the well established principles if it is not to contravene article 59. Those principles are correctly summarised … in the case of an administrative decision as being that the decision must be non-discriminatory, it must be justified by some imperative requirement in the general interest, it must be suitable for securing the attainments of the objects which it pursues and it must not go beyond what it is necessary to attain that objective.”

I return to Geraets-Smits. The ECJ then turned to consider the prior authorisation requirement:

“[76] As regards the prior authorisation requirement to which the ZFW subjects the assumption of the costs of treatment provided in another member state by a non-contracted care provider, the court accepts, as all the governments which have submitted observations have argued, that, by comparison with medical services provided by practitioners in their surgeries or at the patient's home, medical services provided in a hospital take place within an infrastructure with, undoubtedly, certain very distinct characteristics. It is thus well known that the number of hospitals, their geographical distribution, the mode of their organisation and the equipment with which they are provided, and even the nature of the medical services which they are able to offer, are all matters for which planning must be possible.

[77] As may be seen, in particular, from the contracting system involved in the main proceedings, this kind of planning therefore broadly meets a variety of concerns.

[78] For one thing, it seeks to achieve the aim of ensuring that there is sufficient and permanent access to a balanced range of high-quality hospital treatment in the state concerned.

[79] For another thing, it assists in meeting a desire to control costs and to prevent, as far as possible, any wastage of financial, technical and human resources. Such wastage is all the more damaging because it is generally recognised that the hospital care sector generates considerable costs and must satisfy increasing needs, while the financial resources which may be made available for health care are not unlimited, whatever the mode of funding applied.

[80] From both those perspectives, a requirement that the assumption of costs, under a national social security system, of hospital treatment provided in another member state must be subject to prior authorisation appears to be a measure which is both necessary and reasonable.

[81] Looking at the system set up by the ZFW, it is clear that, if insured persons were at liberty, regardless of the circumstances, to use the services of hospitals with which their sickness insurance fund had no contractual arrangements, whether they were situated in the Netherlands or in another member state, all the planning which goes into the contractual system in an effort to guarantee a rationalised, stable, balanced and accessible supply of hospital services would be jeopardised at a stroke.

[82] Although, for the considerations set out above, Community law does not in principle preclude a system of prior authorisation, the conditions attached to the grant of such authorisation must none the less be justified with regard to the overriding considerations examined and must satisfy the requirement of proportionality referred to in paragraph 75 above.”

The ECJ then turned to consider the condition requiring the treatment to be necessary:

“[99] Under the rules at issue in the main proceedings, the grant of authorisation allowing assumption of the costs of a medical service provided abroad is subject to a second condition, namely that it be proved that the insured person's medical treatment requires that service. …

[101] As regards the provision of hospital treatment outside the Netherlands, the national court states, however, that in practice that condition often appears to be interpreted as meaning that the provision of such treatment is not to be authorised unless it appears that appropriate treatment cannot be provided without undue delay in the Netherlands. No distinction is therefore drawn in this respect between whether the treatment could be provided by a contracted establishment or by a non-contracted establishment.”

The heart of the ECJ’s reasoning, and that part of its judgment which in the final analysis is of most importance in the present case, follows:

“[103] In view of what is stated in paragraph 90 above, it can be concluded that the condition concerning the necessity of the treatment, laid down by the rules at issue in the main proceedings, can be justified under article 59 of the Treaty, provided that the condition is construed to the effect that authorisation to receive treatment in another member state may be refused on that ground only if the same or equally effective treatment can be obtained without undue delay from an establishment with which the insured person's sickness insurance fund has contractual arrangements.

[104] Furthermore, in order to determine whether equally effective treatment can be obtained without undue delay from an establishment having contractual arrangements with the insured person's fund, the national authorities are required to have regard to all the circumstances of each specific case and to take due account not only of the patient’s medical condition at the time when authorisation is sought but also of his past record.

[105] Such a condition can allow an adequate, balanced and permanent supply of high-quality hospital treatment to be maintained on the national territory and the financial stability of the sickness insurance system to be assured.

[106] Were large numbers of insured persons to decide to be treated in other member states even when the hospitals having contractual arrangements with their sickness insurance funds offered adequate identical or equivalent treatment, the consequent outflow of patients would be liable to put at risk the very principle of having contractual arrangements with hospitals and, consequently, undermine all the planning and rationalisation carried out in this vital sector in an effort to avoid the phenomena of hospital overcapacity, imbalance in the supply of hospital medical care and logistical and financial wastage.

[107] However, once it is clear that treatment covered by the national insurance system cannot be provided by a contracted establishment, it is not acceptable that national hospitals not having any contractual arrangements with the insured person's sickness insurance fund be given priority over hospitals in other member states. Once such treatment was ex hypothesi provided outside the planning framework established by the ZFW, such priority would exceed what was necessary for meeting the overriding requirements referred to in paragraph 105 above.

[108] In view of all the foregoing considerations, the answer to be given to the national court must be that articles 59 and 60 of the Treaty do not preclude legislation of a member state, such as that at issue in the main proceedings, which makes the assumption of the costs of treatment provided in a hospital located in another member state subject to prior authorisation from the insured person's sickness insurance fund and the grant of such authorisation subject to the condition that (i) the treatment must be regarded as “normal in the professional circles concerned”, a criterion also applied in determining whether hospital treatment provided on national territory is covered, and (ii) the insured person’s medical treatment must require that treatment. However, that applies only in so far as (a) the requirement that the treatment must be regarded as “normal” is construed to the effect that authorisation cannot be refused on that ground where it appears that the treatment concerned is sufficiently tried and tested by international medical science, and (b) authorisation can be refused on the ground of lack of medical necessity only if the same or equally effective treatment can be obtained without undue delay at an establishment having a contractual arrangement with the insured person’s sickness insurance fund.”

This last passage again goes to show that the debate in relation to Article 49 is not about the right to travel abroad for treatment but about the right to be reimbursed for the cost of treatment obtained abroad.

In the context of the dispute with which I am concerned the fundamental proposition is obviously that contained in para [103] and in the last few lines of para [108], namely (putting the point in a domestic context), that prior authorisation for treatment by a NHS patient in another member state at the expense of the NHS can be refused on the ground of lack of medical necessity only if the same or equally effective treatment can be obtained without undue delay at an NHS establishment. So where, as in the present case, there is no suggestion that the same or equally effective treatment cannot be obtained both in the United Kingdom and in France, the crucial issue is whether that treatment can be provided by the NHS “without undue delay”.

The ECJ returned to consider this issue in Muller–Faure, where the referring court asked the ECJ (see judgment, paras [34]-[35]) to explain the import of paragraph [103] of its judgment in Geraets-Smits. More specifically, the referring court sought guidance on “what is meant by ‘without undue delay’ and, in particular, whether that condition must be assessed on a strictly medical basis, regardless of the waiting time for the treatment sought.”

The Netherlands submitted (see judgment, para [51]) that:

“the mere fact of a person being on a waiting list does not mean that such treatment is not available. If it were to adopt a different interpretation, the Court would significantly extend the conditions in which benefits are awarded, which are a matter of national competence. Moreover, it would cast uncertainty over all attempted planning and rationalisation in the health-care sector aimed at avoiding over-capacity, supply-side imbalance, wastage and loss.”

The United Kingdom submitted (see judgment, para [58]) that

“To compel the competent authorities to authorise treatment abroad in circumstances other than where there is a delay beyond the normal waiting time and to pass the cost on to the NHS would have damaging consequences for its management and financial viability.”

The Commission of the European Communities (see judgment, para [64]) submitted that “only the patient’s medical condition should be taken into account”, seeking to rely for that purpose upon what the ECJ had said in paragraph [104] in Geraets-Smits. That approach was adopted by Advocate General Colomer. Having commented with reference to what the ECJ had said in Geraets-Smits at paragraph 104 that “Comme on peut le voir, la Cour n'a donc évoqué aucune raison qui ne soit médicale” (opinion, paragraph [66] – there is, as yet, no authentic English text: I have been shown the French and Spanish versions and quote from the former), he continued (opinion, para [67]):

“Selon moi, il convient de préciser à la juridiction nationale que la condition «en temps opportun» («tijdig») doit être appréciée d'un point de vue strictement médical, indépendamment de la durée de la période d'attente nécessaire pour obtenir le traitement souhaité.”

Addressing the main issue, the ECJ said (judgment, para [69]):

“In this instance the arguments put forward to justify the requirement for prior authorisation seek to establish that, if it were open to patients to get treatment in a Member State other than that in which they are insured, without prior authorisation to that effect, the competent State could no longer guarantee that in its territory there would be a high-quality, balanced medical and hospital service open to all and hence a high level of public health protection.”

It continued (judgment, para [74]):

“It is self-evident that assuming the cost of one isolated case of treatment, carried out in a Member State other than that in which a particular person is insured with a sickness fund, can never make any significant impact on the financing of the social security system. Thus an overall approach must necessarily be adopted in relation to the consequences of freedom to provide health-related services.”

Having in paragraphs [76]-[84] rehearsed its earlier findings in Geraets-Smits at paragraphs [76]-[82], which I have already set out, the ECJ then restated in slightly different language (judgment paras [89]-[90]) the principles it had earlier laid down in Geraets-Smits at paras [103]-[104]):

“[89] The condition concerning the necessity of the treatment, laid down by the legislation at issue in the main proceedings, can be justified under Article 59 of the Treaty, provided that the condition is construed to the effect that authorisation to receive treatment in another Member State may be refused on that ground only if treatment which is the same or equally effective for the patient can be obtained without undue delay from an establishment with which the insured person’s sickness insurance fund has an agreement (Smits and Peerbooms, paragraph 103).

[90] In order to determine whether treatment which is equally effective for the patient can be obtained without undue delay in an establishment having an agreement with the insured person's fund, the national authorities are required to have regard to all the circumstances of each specific case and to take due account not only of the patient's medical condition at the time when authorisation is sought and, where appropriate, of the degree of pain or the nature of the patient’s disability which might, for example, make it impossible or extremely difficult for him to carry out a professional activity, but also of his medical history (see, to that effect, Smits and Peerbooms, paragraph 104).”

The ECJ then added these crucial observations (judgment, para [92]):

“ … a refusal to grant prior authorisation which is based not on fear of wastage resulting from hospital overcapacity but solely on the ground that there are waiting lists on national territory for the hospital treatment concerned, without account being taken of the specific circumstances attaching to the patient's medical condition, cannot amount to a properly justified restriction on freedom to provide services. It is not clear from the arguments submitted to the Court that such waiting times are necessary, apart from considerations of a purely economic nature which cannot as such justify a restriction on the fundamental principle of freedom to provide services, for the purpose of safeguarding the protection of public health. On the contrary, a waiting time which is too long or abnormal would be more likely to restrict access to balanced, high-quality hospital care.”

The ECJ concluded (judgment, para [109]) that:

“Articles 59 and 60 of the Treaty must be interpreted as not precluding legislation of a Member State, such as that at issue in the main proceedings, which (i) makes the assumption of the costs of hospital care provided in a Member State other than that in which the insured person’s sickness fund is established, by a provider with which that fund has not concluded an agreement, conditional upon prior authorisation by the fund and (ii) makes the grant of that authorisation subject to the condition that such action is necessary for the insured person’s health care. However, authorisation may be refused on that ground only if treatment which is the same or equally effective for the patient can be obtained without undue delay in an establishment which has concluded an agreement with the fund.”

That, it will be noted, is not materially different from what the ECJ had earlier said in Geraets-Smits (see paragraph [130] above). But in two important respects the ECJ’s judgment in Muller–Fauré goes beyond its earlier judgment in Geraets–Smits. The first is in explicit adoption of what might be thought a middle course, rejecting, as it seems to me, both the submission of the European Commission, supported by the Advocate General, that the only matter to be taken into account is the patient’s medical condition, and the opposing submission of the Netherlands and the United Kingdom that in effect the normal domestic waiting time should be treated as determinative. The second, and ultimately more important for present purposes, is in its analysis (in para [92]) of the significance – if any – of national waiting lists.

I do not propose to attempt what would in any event be unnecessary and wholly inappropriate, a summary in my own words of the ECJ’s jurisprudence. But it may be helpful at this stage to identify some of the key points which emerge from the judgments in Kohll, Muller–Faure and Geraets–Smits, particularly as they bear upon the case I am considering:

So far so good. But how do national waiting times fit into this analysis? It is at this point that the judgment of the ECJ in Muller–Fauré is, as it seems to me, so important. I do not think that I am placing undue weight on the language adopted by the ECJ if I assert that there are four crucial propositions to be found in para [92]:

The ECJ’s rejection of the proposition that waiting times and waiting lists are significant in assessing whether there is likely to be “undue delay” – “It is not clear … that such waiting times are necessary … for the purpose of safeguarding the protection of public health. On the contrary, a waiting time which is too long … would be more likely to restrict access to balanced, high–quality hospital care” – seems to me, if I may say so, to be entirely unsurprising and entirely consistent with the ECJ’s whole approach to the issue.

The ECJ recognises the need for the United Kingdom to provide and maintain an adequate, balanced and permanent supply of high–quality medical and hospital services accessible to all through the NHS, and the need to avoid the risk of seriously undermining the financial balance of the NHS. And it recognises the risk to the NHS were there to be logistical or financial wastage resulting from hospital overcapacity caused by the outflow from the NHS of large numbers of NHS patients going abroad to be treated. But as is obvious (and the point is reinforced by the fact, deposed to by Mr McConn in one of his witness statements, that maximum waiting times are planned to fall by progressive reductions over a period of five years from 2003 to 2008), waiting times at present in the NHS are not determined simply by reference to the medical needs of patients and the evident needs of a complex organisation to make the best and most efficient use of its human and non–human resources.

The difficulties facing the NHS result not from overcapacity but rather from its present undercapacity, which provides patients like the claimant with an understandable medical incentive to go abroad for treatment. Why, after all, does someone in the circumstances in which the claimant found herself want to go abroad for treatment which it is reasonable to believe, other things being equal, she would much rather have in this country? It was surely because she felt she had to wait too long for an operation in circumstances where, as Mr Edge readily acknowledged on 28 October 2002, she had experienced “severe deterioration in her function”, had to use two sticks to mobilise and “is in constant pain”.

Moreover, the idea that, if given the chance, so many NHS patients will go abroad to be treated that the NHS will face the spectre of overcapacity, with resultant logistical or financial wastage, seems to me to be wholly unproven by anything that I have seen, either in the evidence adduced in the present case or in anything hitherto brought to the attention of the ECJ. It is a speculation unnourished by common sense. There is nothing to suggest that huge numbers of NHS patients will, if only given the chance, be clamouring to travel, whether by Eurostar or cross–channel ferry, for treatment in France, Belgium, the Netherlands, Germany or any other member state. And in any event common sense suggests that those who do so make that choice only because of the delays which they currently experience in the NHS. Surely, as the NHS moves from a condition of undercapacity to one of what one might call adequate or balanced capacity – and as in consequence NHS waiting times fall – the medical incentives to travel abroad for treatment will correspondingly reduce. And surely at, if not before, the point at which the NHS might otherwise be faced with overcapacity as a result of this flood of patients travelling abroad for treatment, the flood will be reduced to at most a trickle. For at the point where the NHS has adequate or balanced capacity, but not yet overcapacity, any medical incentive to travel abroad for treatment will simply evaporate, for there will then no longer be the delays which may at present induce patients like the claimant to go abroad for treatment. And in any event, whatever patients may or may not wish to do in those circumstances, there will be no problem for the NHS because by that point – indeed somewhat before that point is reached – it will be impossible for patients to demonstrate that they are faced with the “undue delay” which alone entitles them to reimbursement under Article 49.

Plainly, if patients in the claimant’s position are entitled to be reimbursed for the cost of treatment obtained abroad, the short–term financial costs of the NHS will be increased – though by how much I do not know – for the NHS will be having to fund such treatment in addition to all the other treatment which it is able to provide using its own resources. (In the long run, of course, there may be no additional cost, for the additional expense incurred in paying the claimant for the cost of treatment obtained by her in France in the financial year 2002/2003 will be offset by the fact that she will not require NHS treatment in (say) the financial year 2003/2004.) But that, as it seems to me, is an example of what the ECJ referred to as a “consideration of a purely economic nature” which, as it said, “cannot as such justify a restriction on the fundamental principle of freedom to provide services”.

Consideration of NHS waiting times and waiting lists is not irrelevant when it comes to an assessment of whether a patient is faced with “undue delay”, for, as the ECJ has made clear, one has to have regard to “all the circumstances”. Moreover, a waiting time which, even if some might think it unacceptably long, is at least tolerated by the medical profession may properly be thought to throw at least some light on what constitutes “undue delay” even if one assesses that from an exclusively medical point of view. But in my judgment, and for all the reasons I have given, although the waiting time applicable in any particular case is a relevant matter to be considered it cannot be determinative. And in many – indeed probably in most – cases it is, for the reasons I have sought to explain, unlikely to be even a significant matter when assessing whether or not a patient is faced with “undue delay”.

Before leaving this aspect of the case there is one final matter to which I must draw attention. In Geraets–Smits the ECJ said (judgment, para [90]):

“It likewise follows from settled case law that a scheme of prior authorisation cannot legitimise discretionary decisions taken by the national authorities which are liable to negate the effectiveness of provisions of Community law, in particular those relating to a fundamental freedom such as that at issue in the main proceedings: see, to that effect, Criminal proceedings against Bordessa (Joined Cases C-358 and 416/93) [1995] ECR I-361, 385, para 25; Criminal proceedings against Sanz de Lera (Joined Cases C-163, 165 and 250/94) [1995] ECR I-4821, 4837-4838, paras 23-28, and Asociación Profesional de Empresas Navieras de Líneas Regulares (Analir) v Administración General del Estado (Case C-205/99) [2001] ECR I-1271, 1313, para 37. Therefore, in order for a prior administrative authorisation scheme to be justified even though it derogates from such a fundamental freedom, it must, in any event, be based on objective, non-discriminatory criteria which are known in advance, in such a way as to circumscribe the exercise of the national authorities' discretion, so that it is not used arbitrarily: Analir, para 38. Such a prior administrative authorisation scheme must likewise be based on a procedural system which is easily accessible and capable of ensuring that a request for authorisation will be dealt with objectively and impartially within a reasonable time and refusals to grant authorisation must also be capable of being challenged in judicial or quasi-judicial proceedings.”

The ECJ repeated this in Muller–Fauré (judgment, para [85]).

So any administrative scheme which impinges upon the fundamental freedom protected by Article 49 must satisfy three further criteria:

So much for the jurisprudence of the ECJ in relation to Article 49. I turn to consider the claimant’s case under Article 49 – and first, her case in relation to the decision contained in the PCT’s letter dated 21 November 2002.

I can take this quite shortly. I start with Mr Edge’s professional assessment as set out in his letter to Ms Laughton dated 28 October 2002. I repeat the salient passages. He described the claimant as suffering:

“severe bilateral hip pain. She has experienced severe deterioration in her function over the course of the last three months. She now has to use two walking sticks to mobilise.

Clinical examination and X-ray examination show her to have severe arthritis of both hips.

This lady requires bilateral total hip replacements. She is battling tremendously with her mobility and is in constant pain.

… this lady’s function is severely hampered by her arthritis”.

Mr Edge’s clinical assessment of the time that the claimant would have to wait for the necessary hip surgery in her local hospital is perhaps captured by his comment that “Unfortunately my waiting list is approximately one year” (emphasis added).

Now I have to say that it seems to me almost self–evident (unless one treats the waiting list time as being if not determinative then at least of preponderant weight) that to delay in treating a patient in this condition for a year is, on any sensible view, to subject her to what is manifestly an “undue” delay. The claimant, be it noted, is not merely a patient with a treatable condition. She is a patient with a treatable condition which, whilst it remains untreated, subjects her to “constant pain” and “severely hamper[s]” her function and mobility. Surely a patient in the claimant’s condition is the very kind of patient the ECJ had in mind when, in para [90] of its judgment in Muller–Fauré, it abjured the national authorities:

“to have regard to all the circumstances of each specific case and to take due account not only of the patient’s medical condition at the time when authorisation is sought and, where appropriate, of the degree of pain or the nature of the patient’s disability which might, for example, make it impossible or extremely difficult for him to carry out a professional activity”.

In my judgment, any national authority properly directing itself in accordance with the principles laid down by the ECJ, in particular in Geraets–Smits and Muller–Fauré, would have been bound to conclude in October–November 2002 that the anticipated delay of approximately one year was on any view “undue”, and thus such as to trigger the claimant’s right under Article 49 to reimbursement of the cost of obtaining more timely treatment in another member state.

The heart of the decision letter dated 21 November 2002 is to be found in the following passages:

“Mr Edge’s letter dated 1st October 2002 does not indicate that your mother requires treatment outside of the draft NHS guidelines for overseas treatment which states that overseas treatment can only be supported “if treatment cannot be provided within the time normally necessary for obtaining the treatment in question taking account of the patient’s current state of health and the probable course of disease”.

The treatment your mother requires “can be provided locally within the time normally necessary for obtaining the treatment in question taking into account the patient’s current state of health”.

… “authorisation can be refused on the ground of lack of medical necessity only if the same or equally effective treatment can be obtained without undue delay at an establishment having a contractual arrangement with the insured person’s sickness insurance fund”.

Bedford PCT interpret ‘undue delay’ as meaning, within the Governments NHS Plan targets”.

It is quite clear that this decision was based on the assumption that “undue delay” was to be understood by reference to the Article 22 criterion of “the time normally necessary for obtaining the treatment in question … taking account of [the patient’s] current state of health and the probable course of the disease”. That in my judgment was a plain error of law. The test of “undue delay” for the purposes of Article 49 is not the same test as that applicable under Article 22. Moreover, it is equally plain that for the purposes of the decision “undue delay” was treated as being determined by reference to waiting list times. That also, in my judgment, was a plain error of law.

The PCT points out that Mr Edge in his letter of 28 October 2002 conspicuously refrained from recommending either that the claimant be treated abroad or that she be treated as a priority case and, further, that in December 2002 he made it clear to Mr Dickinson that he was not recommending that she be treated abroad. This, says Mr Kovats, is fatal to the claimant’s case. I entirely disagree.

The relevant authorities are, of course, entitled to proceed in relation to the medical aspects of the case on the basis of the professional views and opinions expressed by the treating consultant or other medical practitioner. Thus the PCT was entirely justified in accepting Mr Edge’s view as to whether the claimant needed treatment. But that cannot help the PCT here, for his opinion plainly was that she did need treatment. The PCT would likewise have been entirely justified in acting on the basis of any views Mr Edge might have expressed – in fact he never did – as to whether from the medical point of view the treatment available for the claimant’s condition in the United Kingdom was, as the case might be, either more or less suitable than the corresponding treatment available, for example, in France. What, however, in my judgment the PCT or other determining authority cannot do – and this is really what Mr Kovats’ submission comes to, however attractively it is put – is to shelter behind a consultant’s view as to whether or not a patient should be treated abroad when that view is based (as Mr Edge’s view here, insofar as he expressed one, was based) as much on his understanding of complex legal issues and of the Department of Health’s policy as on any purely clinical assessment. In fact to be fair to Mr Edge he appears to have been scrupulous in not going outside his purely clinical competence. What he said to Ms Laughton in December 2002 was “that he felt unable to say whether the claimant should be treated overseas … he could only comment on her clinical priority, which was routine.” All he said to Mr Dickinson, later the same month, was that he was “not recommending that the claimant have her hip operation abroad”. That was a factually accurate statement, but it has to be borne in mind that it was not a statement that he was recommending that she not have her operation abroad, and it certainly contained nothing to absolve the relevant decision makers from properly applying their minds to the relevant questions which required to be considered under Article 49.

The PCT, in my judgment, did not address itself to the correct principles to be found in Geraets–Smits and Muller–Fauré. And nor, for that matter, would Mr McConn have done (see paragraph [29] above). They are perhaps not to be blamed for their erroneous approach. Muller–Fauré had not at that time yet been decided and, more fundamentally perhaps, both the PCT and Mr McConn were approaching the case on the basis that the matter was governed by Article 22 and not by Article 49.

Nonetheless, if matters had not moved on thereafter, the claimant would in my judgment have been entitled to appropriate relief in relation to a decision which, for the reasons I have given, was erroneous in law and, moreover, arrived at in circumstances where, as I have already said, any national authority properly directing itself in accordance with the principles laid down by the ECJ would have been bound to conclude in October–November 2002 that the anticipated delay of approximately one year was on any view “undue”, and thus such as to trigger the claimant’s right under Article 49 to reimbursement of the cost of obtaining more timely treatment in another member state. But the fact is that matters did move on and any relief which the claimant might otherwise have been entitled to has, as it seems to me, been overtaken by subsequent events.

Much energy was devoted by the Secretary of State and a certain amount of time was taken up with the argument that the PCT is not the appropriate body for authorising and indeed has no power to authorise treatment sought in another member state; that this function is performed by the Secretary of State (acting through officials of the Department of Health) and has not been delegated to PCTs; and that the Secretary of State, acting through the Department of Health, has not been asked to take a decision on the claimant’s request, nor has he done so. The point is made that although NHS patients seeking treatment abroad “should normally first seek the opinion of their NHS consultant and home PCT”, a patient “may approach the Department of Health directly”. It is also said by the Secretary of State that the PCT’s letter dated 21 November 2002 did not purport to be, nor did it constitute, a decision to refuse a request for treatment abroad or to refuse to authorise the issue of a Form E112: “it merely stated that the [PCT] was unable to support the request”.

I do not propose to take up time debating at any length issues which in the event do not arise for decision and which are unlikely to arise again in future (see further below). I pause only to observe that:

I turn therefore to consider those subsequent events and the claimant’s case in relation to the decision contained in the PCT’s letter dated 4 February 2003.

In its fundamental approach the PCT’s second letter differed not at all from its earlier letter:

“without support for overseas treatment from Mr Edge, Mrs Watts’s Consultant Orthopaedic Surgeon, or from Bedford PCT, (Bedford PCT reiterate previous reasoning for not supporting this application), Bedford PCT is unable to support the overseas treatment application under form E112.”

Insofar as there was error of law in the letter dated 21 November 2002 the same error of law, unsurprisingly, pervades the letter dated 4 February 2003. But this, in my judgment, is to ignore the underlying realities of the situation.

The fact – and it is a fact – is that the PCT’s stance had been radically and fundamentally transformed as a result of receiving Mr Edge’s letter dated 31 January 2003. The heart of Mr Edge’s letter is to be found in the following passages:

“the pain has now become significantly worse. … She is now no longer able to mobilise on one stick … she did battle to walk with a very shuffling gait using 2 walking sticks. She had difficulty in getting on and off the couch and had to lift up her legs using her arms to get onto the couch. … Under the circumstances I feel that this lady probably had deteriorated since last seen. Of course all the patients on my waiting list would have deteriorated in this time but she has perhaps got a little worse than the average patient. I am therefore prepared to re-categorise her as a “soon” case.”

It is clear, as Mr Kovats submits, that the PCT reconsidered the facts – including the medical facts – of the claimant’s case in the light of Mr Edge’s letter and the subsequent conversation which he had with Ms Laughton. Acting in accordance with Mr Edge’s opinion the PCT recategorised the claimant as a “soon” case, which meant that she would have surgery in either April or May 2003:

“Mr Edge has clarified this with the PCT to mean a wait for surgery of between three and four months. This will mean that Mrs Watts will be listed for surgery in either April or May 2003. … Having reassessed the new clinical assessment by Mr Edge, the PCT acknowledge that Mrs Watt’s clinical condition has deteriorated enough for Mr Edge to reprioritise Mrs Watts as requiring treatment more quickly than previously thought, within three to four months. … Mrs Watts will be listed for surgery at Bedford hospital for hip replacement surgery between three to four months.”

The position therefore was this: As at 4 February 2003 the PCT was committed to providing the claimant with surgery in April or May 2003, but, as I have said, she was not prepared to wait. She arranged to be treated in France, the surgery in the event being performed on 7 March 2003.

Mr McConn (see paragraph [30] above) rightly describes this as “a relatively short period”, a view with which, in all the circumstances, I entirely agree. Making every allowance for the pain and suffering to which by then, if not previously, she was subject, the delay which the claimant faced on 4 February 2003 – whether for this purpose one treats the relevant period as being that from 4 February 2003 until the date of the proposed operation in April or May 2003 or that from 7 March 2003 until April or May 2003 – simply cannot in my judgment be described as “undue”. Yes, the claimant was still facing delay. But delay alone is not sufficient to trigger rights under Article 49: the delay has to be “undue”, having regard, of course, to all the circumstances, including in particular in a case such as this the constant pain the claimant was suffering and the extent of the disability resulting from her condition so long as it remained untreated. However, one must not forget that amongst the circumstances that the relevant authority was entitled to take and no doubt would have taken into account – just as the PCT actually did, and just as Mr McConn says that he would have done – was Mr Edge’s assessment of the claimant’s case as lying “somewhere between a routine case and an urgent case” and his observation that he had “several” urgent cases on his list that had to be done prior to the claimant. Article 49 does not impose impossible burdens on member states. It does not guarantee instantaneous treatment or treatment within the kind of timescale that might be appropriate in an ideal world.

In my judgment, in this case, having regard to all the circumstances, the period of delay which was tolerable before it reached the level of what is “undue” was a period very much less than the year with which the claimant was originally faced but, on the other hand, a period significantly (though probably not substantially) greater than the period of delay until April or May 2003 with which she was faced on 4 February 2003. On that simple ground, as it seems to me, the claimant’s case breaks down. Whether or not the PCT misdirected itself in law – and in my judgment it did – the claimant has wholly failed to establish, as she must if her claim under Article 49 is to succeed, that she was faced with the prospect of “undue delay” on 4 February 2003, using that phrase in the sense in which it is here to be used. In my judgment she was not. As I have said, the delay which she faced on 4 February 2003 simply cannot, in my judgment, be described as “undue”.

It follows that the claimant’s case fails insofar as it is based on Article 49. She has succeeded in demonstrating that the Secretary of State’s understanding of the law is wrong, and that the law is, in effect, as she has contended for. But her claim nonetheless fails on the facts.

I turn to the claimant’s alternative case based on Article 22. I can take this very much more shortly.

The amendment of Article 22 in 1981 was, indeed, as Advocate General Colomer described it (see paragraph [68] above), “une modification radicale”. It was, as it seems to me, so radical that little if any useful guidance is now to obtained from the earlier decisions of the ECJ in Pierik I and Pierik II. Attention has now to focus on the amended text, which provides in material part, and the language is clear enough, that:

“authorisation … may not be refused where … the person concerned … cannot be given … treatment within the time normally necessary for obtaining the treatment in question in the Member State of residence, taking account of his current state of health and the probable course of the disease.”

It will be recalled (see paragraph [72] above) that in Kohll Advocate General Tesauro said of Article 22 (opinion, para [31]) that

“Member States are bound to grant authorisation … only where the treatment cannot be provided within such time as to ensure its effectiveness, thereby leaving all other eventualities to the Member States’ discretion” (emphasis added).

I respectfully agree. The national authority’s otherwise unfettered discretion is limited only in this one respect. And the test the patient has to surmount before bringing herself within Article 22 is, in my judgment, a significantly higher test in the present type of case than that required to bring herself within Article 49. Waiting lists and waiting times, although relevant under Article 49 will typically, as we have seen, have little if any significance in that context. In contrast, they are plainly of central significance in the context of Article 22 which, although it requires one to take account of the patient’s current state of health and the probable course of the disease, primarily directs attention to “the time normally necessary for obtaining the treatment in question in the Member State of residence”.

In Inizan, Advocate General Colomer having referred (opinion, para [42]) to what the ECJ had said in Geraets-Smits at paragraph [104] in relation to what he called “une condition très similaire requise par la législation néerlandaise” concluded (opinion para [46]) that:

“aux fins d'apprécier si un traitement peut être dispensé dans le délai normalement nécessaire pour l'obtenir dans l'État membre de résidence, les autorités nationales doivent prendre en compte l'ensemble des circonstances de l'espèce, la situation médicale du patient ainsi que ses antécédents”.

I have no quarrel with any of that, but it does not mean that one imports into Article 22 the Article 49 concept of “undue delay”, let alone that one imports into the jurisprudence on Article 22 the whole of the jurisprudence on Article 49 to be found, for example, in Geraets–Smits and Muller–Fauré. Article 22 and Article 49 are separate provisions. They do not stand or fall together. The test under each is different.

The Secretary of State and the PCT submit that in coming to the decision embodied in its letter dated 21 November 2002 the PCT directed itself correctly as to the law in its consideration of Article 22. The express reference in Article 22 to “the time normally necessary for obtaining the treatment in question in the Member State of residence” makes clear, it is said, that the authorities of member states are entitled to take account of normal waiting times, even though they are at the same time entitled, indeed obliged, to have due regard to the particular circumstances of the individual patient’s case and to the priority of the patient compared with other cases. The United Kingdom, it is said, is entitled to refuse authorisation under Article 22 notwithstanding the existence of a waiting time for the required treatment, provided that the length of the waiting time reflects appropriate clinical priority for patients in need of such treatment and provided that any decision refusing authorisation takes account of the medical condition and needs of the particular patient concerned.

In practice, as I understand it, the “normal” waiting time for the purposes of Article 22 is determined by the Secretary of State by reference to the normal local waiting time for the procedure in question, in the case of a patient in the same clinical condition and with the same prognosis as the applicant – in practice, the actual waiting time for that procedure in the case of a patient in the same clinical condition and with the same prognosis as the applicant on the list of the consultant who will undertake the procedure, but ignoring any exceptional events such as the exceptional non–availability of medical staff or hospital facilities.

The Secretary of State and the PCT point to the opinions expressed by Mr Edge in his letter to Ms Laughton dated 28 October 2002 and in his conversations with Ms Laughton and Mr Dickinson in December 2002. They submit that in all the circumstances the PCT was entitled to conclude that the treatment requested could be provided within the time normally necessary for obtaining the treatment in question in the United Kingdom taking account of the patient’s current state of health from the probable course of the disease and to decline to support the application.

I agree. Despite everything pressed on me by Mr Gordon I am unable to detect, in relation to Article 22, any error of law or any error of approach in the letter dated 21 November 2002.

If the claimant’s case under Article 22 fails in relation to the decision letter dated 21 November 2002 it must, in the circumstances, equally fail in relation to the later decision letter. Indeed, this is hardly surprising given that in relation to the later letter even her claim under Article 49 has failed.

I conclude therefore that the claimant has failed to make out any part of her case under Article 22.

In the light of the conclusions to which I have come, the issue of reimbursement does not arise. It is, however, convenient that I say something on a matter that was the subject of some helpful submissions.

The principles governing the patient’s right to reimbursement were considered by the ECJ in Vanbraekel (judgment, para [53]):

“ … Article 22(1)(c) and (i) of Regulation No 1408/71 is to be interpreted as meaning that, when an insured person has been authorised by the competent institution to go to another Member State for treatment, the institution of the place where the treatment is provided is required to provide him with benefits in kind in accordance with the rules on assumption of the costs of health care which the latter administers, as if the person concerned were registered with it.

Where the request of an insured person for authorisation on the basis of Article 22(1)(c) of that regulation has been refused by the competent institution and it is subsequently established that such refusal was unfounded, the person concerned is entitled to be reimbursed directly by the competent institution by an amount equivalent to that which would have been borne by the institution of the place of treatment under the rules laid down by the legislation applied by the latter institution if authorisation had been properly granted in the first place.

As Article 22 of that regulation is not intended to regulate any reimbursement at the tariffs in force in the Member State of registration, it does not have the effect of preventing or prescribing payment by that State of additional reimbursement covering the difference between the system of cover laid down by the legislation of that State and the system applied by the Member State of treatment, where the former is more advantageous than the latter and such reimbursement is provided for by the legislation of the Member State of registration.

Article 59 of the EC Treaty is to be interpreted as meaning that, if the reimbursement of costs incurred on hospital services provided in a Member State of stay, calculated under the rules in force in that State, is less than the amount which application of the legislation in force in the Member State of registration would afford to a person receiving hospital treatment in that State, additional reimbursement covering that difference must be granted to the insured person by the competent institution.”

It will be seen that different principles apply under Article 22 and Article 49.

Reimbursement under Article 22 is calculated in accordance with the legislation of the member state where the treatment is performed, in the present case France. The evidence adduced by the Secretary of State indicates that the most the claimant would be entitled to on this basis is a percentage – 75% I am told – of the cost of the operation, being the equivalent of the amount that would be funded under French legislation, which requires the patient to pay a proportion of the cost (known as a co–payment) of in–patient hospital treatment.

Article 22 would not in these circumstances entitle the patient to reimbursement of any travel and accommodation costs. That is the assertion of both the Secretary of State and the PCT and, as Mr Kovats correctly points out, there is no evidence before me that the French system includes such costs. I do not decide the point. I merely observe that if in some future case it is to be said that the right of reimbursement extends to travel and accommodation costs it would seem to me to be for the claimant to establish that fact – and strictly speaking, being a matter of foreign law it is a question of fact – and to do so by adducing proper evidence of the relevant foreign law.

Reimbursement under Article 49 is calculated on a different basis, namely, if more generous, by reference to the legislation in force in the member state of residence. Since hospital treatment in the United Kingdom is free at the point of delivery this would seem to mean that in a case such as this reimbursement under Article 49 would not be limited to the 75% payable under Article 22 but would extend to the full cost of the treatment.

Whether reimbursement under Article 49 would extend to travel and accommodation costs will, it would seem, depend upon the extent to which (if at all) payment of such costs is provided for under the legislation governing the NHS. That is a complex topic, only briefly touched on in the course of argument, upon which I propose therefore to express no views whatever.

Although the claimant has succeeded in significant measure in her arguments on the law, having demonstrated that the Secretary of State is wrong both in his view that Article 49 is not applicable in a case such as this and also in his over–restricted view as to the ambit of any rights conferred on NHS patients by Article 49, she has nonetheless failed on the facts. Her claims to substantive relief must therefore be dismissed. I shall hear further argument, if desired, on the question of whether it would be appropriate for me to grant any, and if so what, declaratory relief.

There is one final matter I must mention. I referred above (see paragraph [166]) to the unhelpful and somewhat opaque terms in which the Department of Health’s literature – and in particular ‘Health advice for travellers’ – is drafted. This problem is exemplified by the seeming conflict between the statement in ‘Health advice for travellers’ that the patient should apply to the PCT (which will then transmit the application to the Department) and the Secretary of State’s complaint in the present case that no application was ever made to the Department.

True it is, as Mr Lloyd Jones points out, that ‘Health advice for travellers’ states that the patient “requires authorisation from the Department of Health” and that paragraph 1.15 of the draft ‘Guidance’ document states that “Final approval of the Department of Health … needs to be obtained”. Accepting all that, any ordinary reader of ‘Health advice for travellers’ would, as it seems to me, clearly understand that the appropriate procedure is for the patient to apply to the PCT, which will then forward the application on to the Department. Indeed, as Mr Gordon submits, there is, so far as one can see, no document published by the Department which states explicitly (or even, he says, implicitly) that an application either can or ought to be made direct to the Department.

Mr Lloyd Jones accepts that ‘Health advice for travellers’ does not expressly state how an application should be made if, as here, a PCT, having declined to support an application, does not transmit it to the Department. He submits that an applicant “could reasonably be expected to understand that he or she should raise the matter with the … Department”. Mr Kovats, for his part, goes so far as to assert that it is “obvious” that in these circumstances the applicant can approach the Department directly. I have no wish to rub salt in the wound, but this appears to have been very far indeed from obvious to the official in the Department who on 3 December 2002 gave Ms Harding what is now conceded to have been the wholly erroneous advice that “there is no system of appeal against any decision that the PCT may take.”

It is not a happy or reassuring picture. There is a basic lack of clarity in the Department’s published materials as to the procedures an applicant should adopt. The procedural system as described in those materials seem to me to fall short of the requirement, identified by the ECJ in Geraets–Smits at para [90] and reiterated in Muller–Fauré at para [85], of a “procedural system which is easily accessible”.

I understand that, irrespective of the outcome of these proceedings, the Department had already decided to revisit its published materials with a view to spelling out more clearly the right of an applicant to approach the Department directly. That is welcome news. No doubt the Department has also taken in hand the need for internal arrangements adequate to ensure that, in future, approaches to it by persons in Ms Harding’s position are directed to and handled by officials in the correct office and not, as apparently happened in her case, by an official in the wrong office with no responsibility for E112 matters.