England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
ADMINISTRATIVE COURT
MR JUSTICE SUPPERSTONE
CO/6789/2013

		Royal Courts of Justice Strand, London, WC2A 2LL
		17/06/2016

Lord Justice Lewison:

Introduction

1. The claimants produce and supply Dolenio, a glucosamine-containing product ("GCP") used in the treatment of osteoarthritis ("OA") with a daily recommended dose of 1500mg. The Medicines and Healthcare Products Regulatory Agency ("MHRA"), which is an agency of the Department of Health, has categorised Dolenio as a "medicinal product" within the meaning of Directive 2001/83/EC ("the Medicinal Products Directive") and, moreover, as a product that can only be supplied on prescription.
2. Glucosamine is a substance which the human body synthesises from glucose. It is a precursor of a component of cartilage, bones, tendons and ligaments. Glucosamine is also found in the shell of crustaceans or can be obtained from wheat or maize. It is the glucosamine from shellfish that is commonly used in most products on the UK market. Chitin, which the human body can break down to form glucosamine, can form part of the normal diet for those who eat lobster bisque, most shellfish soup, shrimp tails (e.g. in sushi) or some oriental snack foods. Chitin is also found in the cell walls of mushrooms, fungi and yeasts. Glucosamine may also be marketed as a food supplement.
3. The claimants' essential complaint is that the MHRA have refused to classify other GCPs, which are identical to Dolenio in terms of their pharmacological composition, their manner of ingestion, their effects on the human body, and the risks associated with ingesting them, as medicinal products. The consequence is that, whereas Dolenio is heavily regulated, their competitors market identical products under a much lighter regulatory regime. The claimants say that the MHRA's decision contravenes EU law. Supperstone J disagreed. His judgment is at [2014] EWHC 1679 (Admin).

Legal framework

4. Article 1.2 of the Medicinal Products Directive contains the following definition:

"Medicinal product :

(a) Any substance or combination of substances presented as having properties for treating or preventing disease in human beings; or

(b) Any substance or combination of substances which may be used in or administered to human beings either with a view to restoring, correcting or modifying physiological functions by exerting a pharmacological, immunological or metabolic action, or to making a medical diagnosis."

5. The first of these limbs is called the "presentational limb", and the second is called the "functional limb". Article 2. 2 provides:

"In cases of doubt, where, taking into account all its characteristics, a product may fall within the definition of a 'medicinal product' and within the definition of a product covered by other Community legislation the provisions of this Directive shall apply."

6. If a product is a medicinal product it may not be placed on the market in a member state unless it has been given a marketing authorisation. There are stringent procedures laid down for obtaining a marketing authorisation; and provisions dealing with the mutual recognition by one member state of a marketing authorisation granted by another. Of particular importance to this case is the method of obtaining a marketing authorisation under article 10a. That provides:

"By way of derogation from Article 8(3)(i), and without prejudice to the law relating to the protection of industrial and commercial property, the applicant shall not be required to provide the results of pre-clinical tests or clinical trials if he can demonstrate that the active substances of the medicinal product have been in well-established medicinal use within the Community for at least ten years, with recognised efficacy and an acceptable level of safety in terms of the conditions set out in Annex I. In that event, the test and trial results shall be replaced by appropriate scientific literature."

7. Under article 70 the competent authorities must classify the medicinal product either as a product subject to medical prescription or as a medicinal product not subject to medical prescription. The choice is governed by article 71.1 which provides:

"Medicinal products shall be subject to medical prescription where they:

— are likely to present a danger either directly or indirectly, even when used correctly, if utilized without medical supervision, or

— are frequently and to a very wide extent used incorrectly, and as a result are likely to present a direct or indirect danger to human health, or

— contain substances or preparations thereof, the activity and/or adverse reactions of which require further investigation, or

— are normally prescribed by a doctor to be administered parenterally."

8. Member states have a duty under article 76.1 to "take all appropriate action to ensure that only medicinal products in respect of which a marketing authorization has been granted in accordance with Community law are distributed on their territory".
9. It is common ground that the Medicinal Products Directive has been correctly transposed into domestic law; and the argument proceeded solely by reference to the Directive itself.
10. Directive 2002/46/EC ("the Food Supplements Directive") regulates food supplements. Article 2 defines a "food supplement" as:

"Foodstuffs the purpose of which is to supplement the normal diet and which are concentrated sources of nutrients or other substances with a nutritional or physiological effect, alone or in combination, marketed in dose form, namely forms such as capsules, pastilles, tablets, pills and other similar forms, sachets of powder, ampoules of liquids, drop dispensing bottles, and other similar forms of liquids and powders designed to be taken in measured small unit quantities."

11. "Nutrients" are in turn defined as "vitamins and minerals". Those vitamins and minerals which may be used in food supplements are listed in the Annexes to the Food Supplements Directive. Glucosamine is not listed (and it is neither a vitamin nor a mineral), so it can only qualify as an "other substance" with a nutritional or physiological effect. Article 6 requires food supplements to bear the name "Food Supplement"; and article 6(2) expressly prohibits labelling or advertising that attributes to food supplements properties "of preventing, treating or curing a human disease". That article also imposes additional labelling requirements.
12. Manufacturers or suppliers of food supplements (and foods) may only make "health claims" if they can be justified. These are defined by Regulation EC/194/2006 ("the Health Claims Regulation") as:

"any claim that states, suggests or implies that a relationship exists between a food category, a food or one of its constituents and health."

13. Claims of this kind are distinguished from medicinal claims, i.e. claims regarding the treatment or prevention of disease. The Health Claims Regulation also lays down a procedure for authorising particular health claims. The details need not concern us. But the upshot is that claims that glucosamine helps maintain healthy joints or that it helps maintain healthy and flexible joints have not been permitted to be made since 2 January 2014. The reason for this is that there was no scientific evidence to validate such claims. I will return to this point later.

The marketing authorisation

14. The claimants first obtained their marketing authorisation under the Medicinal Products Directive in Denmark. Their product has glucosamine sulphate as its active ingredient. The application was opposed by the MHRA which alleged that there were serious public health concerns; and which also required the application to be an application for a prescription only medicinal product, and for any marketing authorisation to be limited to treatment of OA in the knee (as opposed to any other joint). One must infer that the reason why the MHRA required that the product be classified as a prescription only medicinal product was because of one of the factors identified in article 71 of the Medicinal Products Directive, otherwise such a requirement would have been irrational. The modifications proposed by the MHRA found their way into the marketing authorisation granted by the Danish authorities. Through a mutual recognition procedure the claimants then obtained a marketing authorisation in the UK in June 2009. That authorisation required the production of a patient information leaflet, which contains the statement:

"Dolenio tablet is a medical product used for the relief of symptoms in mild to moderate osteoarthritis of the knee."

15. The leaflet also has a number of warnings. They include instructions not to take Dolenio if allergic to shellfish or taking warfarin; and to consult a doctor if suffering from diabetes, cardiovascular disease, asthma or reduced kidney or liver function. The leaflet also warns of possible side effects.

The position of the MHRA

16. Mr Carter, a Senior Pharmaceutical Regulatory Adviser to the MHRA, explained the MHRA's approach. He says that the MRHA considers products on a case by case basis. Where a product is not associated with a medicinal claim the MHRA considers the product under the functional limb of the definition. Such a product may be associated with a health claim which is usually expressed in terms of maintenance of existing good health rather than targeting a disease. The MHRA publish guidance on how they approach that task. Paragraph 29 of MHRA Guidance Note No 8 says that factors particularly relevant to the functional limb are:

"the pharmacological, immunological or metabolic properties of the ingredient(s) and any significant effect(s) the product will have on physiological function in humans, …

• the composition of the product;

• the manner in which the product is used;

• the product promotional literature, including testimonials and any literature issued by a third party on behalf of the person who places the product on the market;

• the familiarity of the product to consumers and the extent of its distribution in the UK;

• the product form, (capsule, tablet, etc.) and the way it is to be used;

• the presence of essentially similar licensed, registered or exempt medicines on the UK market;

• the risks which use of the product may pose."

17. Mr Carter stresses that this is not a "tick box" exercise and may lead to two apparently similar products having different classifications. He also points out that different member states have come to different conclusions on the question whether GCPs are medicinal products. In one member state (Slovenia) they are always medicinal products; in four they are not unless accompanied by a medical claim (The Netherlands, Cyprus, Malta and Belgium) and in yet others whether or not they are classified as medicinal products depends on the recommended dosage or daily amount. In each of the latter cases a GCP with a recommended dosage or daily amount of 1500 mg would be classified as a medicinal product (and in some of those cases the applicable threshold is significantly less).
18. Mr Carter describes the legal test that the MHRA applies under the functional limb:

"The law is clear that for a product to be classified as medicinal under the functional limb it has to be used as intended and has to be capable of appreciably restoring, correcting or modifying physiological functions in human beings." (Emphasis in original)

19. He adds that simply because a product has been found to be efficacious with regard to some medical conditions, it does not follow that it is efficacious with regard to others. A product may, therefore, receive a marketing authorisation for a narrow and particular medical condition, but it does not follow that the product necessarily satisfies the functional limb in relation to other conditions, even if they are similar. The existence of a significant pharmacological or metabolic effect on the body is not the only criterion that justifies a classification of a product as a medicinal product: it must be considered in the round. Thus the MHRA will consider the familiarity that consumers have with the product as a non-medicinal substance and the manner in which it is actually used. These factors assume particular significance in the case of substances that have both a medicinal and a non-medicinal use. Examples of such products are: salt, ascorbic acid, alcohol, caffeine and even water. The less potent the ingredients of a product, the greater the importance given to other factors, such as consumer familiarity and manner of use; particularly in the case of products such as GCPs which have a long history of use as food supplements.
20. GCPs have been a very popular food supplement and are one of the top selling products in health food shops. There are approximately 800 GCPs on the UK market. The MHRA has taken action against GCPs which claim to treat or prevent a variety of joint pains and conditions, including problems with the knee, but have not routinely classified them as medicinal products. The factors particularly relevant to the MHRA's conclusions are the following:

i) The clinical evidence supplied to obtain a marketing authorisation is for one very specific indication. If the consumer does not have mild to moderate OA of the knee, there is a lack of evidence that a GCP is a medicine by function.

ii) GCPs have not been shown to be a potent product in terms of their effect on the human body; nor have they been found to carry any significant health risks.

iii) There has been a large and well-established market for GCPs dating from long before any GCP received a marketing authorisation.

iv) Apart from cases where medicinal claims are made (which would prompt the MHRA to take action) GCPs are presented and marketed for use as food supplements rather than for medicinal purposes.

v) The averagely well-informed consumer is aware of the myriad of products which are for the maintenance of normal functioning of the body. Glucosamine falls into that category and it is reasonable that a healthy individual would wish to take it to ensure that they maximise the production of cartilage as they age.

vi) Although there is evidence that non-medicinal GCPs are used by pharmacists to fulfil doctors' prescriptions, that is only a small fraction of the total market for GCPs: approximately one fifth.

21. The MHRA say that their multi-factorial approach is firmly based on the case law of the CJEU; so it is to that that I now turn.

The case law

22. The first of the leading cases is (Joined cases C-211/03, C-299/03 and C-316/03 to C-318/03) HLH Warenvertriebs GmbH v Germany [2005] ECR I-5141 ("Warenvertrieb" (which, thanks to McCombe LJ, I now know has no "s" if used as a free-standing word)). The case concerned a number of products which were intended to be marketed as food supplements. However, the German national authorities decided that they needed marketing authorisations as medicinal products. The products in question were a variety of high dosage vitamins, bioflavonoids, and bacterial cultures. At [51] the court considered the application of the functional limb of the definition. They said:

"…for the purposes of determining whether a product comes within the definition of a medicinal product 'by function' within the meaning of [the Medicinal Products Directive], the national authorities, acting under the supervision of the courts, must proceed on a case-by-case basis, taking account of all the characteristics of the product, in particular its composition, its pharmacological properties, to the extent to which they can be established in the present state of scientific knowledge, the manner in which it is used, the extent of its distribution, its familiarity to consumers and the risks which its use may entail."

23. Having said that the court went on to consider pharmacological and risk factors before concluding at [54]:

"… that the pharmacological properties of a product are the factor on the basis of which the authorities of the Member States must ascertain, in the light of the potential capacities of the product, whether it may, for the purposes of the second subparagraph of Article 1(2) of [the Medicinal Products Directive], be administered to human beings with a view to making a medical diagnosis or to restoring, correcting or modifying physiological functions in human beings. The risk that the use of a product may entail for health is an autonomous factor that must also be taken into consideration by the competent national authorities in the context of the classification of the product as a medicinal product."

24. The court also recognised in that case that a product classified as a foodstuff in one member state could be classified as a medicinal product in a different member state; and vice versa.
25. Commission of the European Communities v Federal Republic of Germany [2008] 1 CMLR 36 ("Garlic Capsules") concerned garlic capsules which the German national authorities had classified as a medicinal product. The Commission challenged that classification. At [55] the court repeated what had been said in Warenvertrieb at [51] as being the applicable test. It referred to the pharmacological properties of the product (as in Warenvertrieb at [54]) but added at [60]:

"… that criterion must not lead to the classification as a medicinal product by function of substances which, while having an effect on the human body, do not significantly affect the metabolism and thus do not strictly modify the way in which it functions."

26. At [61] the court said:

"… the definition of medicinal product by function is designed to cover products whose pharmacological properties have been scientifically observed and which are genuinely designed to make a medical diagnosis or to restore, correct or modify physiological functions."

27. Thus in order to fall within the definition of a medicinal product by function "it is not sufficient that product has properties beneficial to health in general, but it must strictly speaking have the function of treating or preventing disease" see [64]. This contrasted with the definition of a medicinal product by presentation, which is intended to protect consumers from products which do not have the effectiveness which they are entitled to expect.
28. In (Case C-140/07) Hecht-Pharma GmbH v Staatliches Gewerbeaufsichtsamt Lüneburg [2009] 2 CMLR 23 ("Hecht") the court considered a product manufactured from fermented rice and sold as a food supplement. The national authorities prohibited sales on the ground that it was a medicinal product and did not have a marketing authorisation. By the time of the decision in Hecht, the Medicinal Products Directive had been amended and one question for the court was whether the amendment had cast doubt on the previous interpretation of the definition of medicinal product. The court answered that question in the negative. It said at [37]:

"Article 1(2)(b) of [the Medicinal Products] Directive must be interpreted as meaning that the characteristics of the manner in which a product is used, the extent of its distribution, its familiarity to consumers and the risks which its use may entail are still relevant to determining whether that product falls within the definition of a medicinal product by function."

29. One other important principle in the application of the Medicines Directive is proportionality. The need for a marketing authorisation would potentially cut across the EU pillar of the free movement of goods. As the court explained in Garlic Capsules at [71]:

"… the Community provisions relating to medicinal products must ensure, in addition to the protection of human health, the free movement of goods, so that the interpretation of the provisions of [the Medicinal Products Directive] in general, and the definition of medicinal products in particular, cannot result in obstacles to the free movement of goods which are entirely disproportionate to the pursued aim of protecting health."

30. In principle the application of the multi-factorial approach mandated by EU law requires a judgment which is to be made by the national authorities: in the case of the UK by the MHRA (acting as the delegate of the Secretary of State for Health). Since the application of the multi-factorial approach requires a judgment to be made, the court will not interfere with the judgment of the responsible decision maker except on the traditional grounds for judicial review: R v Medicines Control Agency, Ex p Pharma Nord (UK) Ltd [1998] 3 CMLR 109 at [41].
31. At the heart of the claimants' case is the perceived inconsistency, within the same member state, between classifying Dolenio as a medicinal product and at the same time refusing to classify as medicinal products which have the same pharmacological properties, the same dosage and are ingested in the same way. While inconsistency between different member states may be tolerated because EU law has not been completely harmonised in this respect, inconsistency in the same member state should not be.
32. This argument relies heavily on article 2.2 of the Medicinal Products Directive, and the decision of the CJEU in (Case C-109/12) Laboratoires Lyocentre v Lääkealan turvallisuus- ja kehittämiskeskus and Sosiaali- ja terveysalan lupa- ja valvontavirasto [2013] ECR I-0000 ("Lyocentre").
33. Article 2.2 was inserted into the Medicinal Products Directive by amendment contained in Directive 2004/27/EC. Recital (7) explained the purpose of the amendment:

"Particularly as a result of scientific and technical progress, the definitions and scope of [the Medicinal Products] Directive should be clarified in order to achieve high standards for the quality, safety and efficacy of medicinal products for human use. In order to take account both of the emergence of new therapies and of the growing number of so-called "borderline" products between the medicinal product sector and other sectors, the definition of "medicinal product" should be modified so as to avoid any doubt as to the applicable legislation when a product, whilst fully falling within the definition of a medicinal product, may also fall within the definition of other regulated products. …With the same objective of clarifying situations, where a given product comes under the definition of a medicinal product but could also fall within the definition of other regulated products, it is necessary, in case of doubt and in order to ensure legal certainty, to state explicitly which provisions have to be complied with. Where a product comes clearly under the definition of other product categories, in particular food, food supplements, medical devices, biocides or cosmetics, this Directive should not apply..."

34. The first of the cases in the CJEU to consider article 2.2 was Hecht. As mentioned, the national authorities classified the fermented rice product as a medicinal product. However they did so even though its pharmacological effect had not been demonstrated, relying on the fact that it probably had a pharmacological effect, thus bringing the "avoidance of doubt" principle in article 2.2 into play. That was the issue raised before the CJEU. It was in that connection that the court said:

"[24] It is clear from recital 7 in the preamble to Directive 2004/27 that art. 2(2) was inserted into [the Medicinal Products] Directive in order to make clear that when a product falls within both the definition of a medicinal product and that of other regulated products, it must be made subject to the provisions of [the Medicinal Products] Directive. Thus, art. 2(2) of [the Medicinal Products] Directive starts from the premise that the product concerned satisfies the conditions for classification as a medicinal product (see, to that effect, HLH Warenvertriebs and Orthica at [43] and [44]).

[25] It should be borne in mind in that regard that, contrary to the definition of medicinal product by presentation, the broad interpretation of which is intended to protect consumers from products which do not have the effectiveness which they are entitled to expect, the definition of medicinal product by function is designed to cover products the pharmacological properties of which have been scientifically observed and which are genuinely designed to make a medical diagnosis or to restore, correct or modify physiological functions (Commission of the European Communities v Germany (C-319/05) [2008] 1 C.M.L.R. 36 at [61]).

[26] Thus, [the Medicinal Products] Directive does not apply to a product in respect of which it has not been established that it is a medicinal product within the meaning of art.1(2)(b) of that directive, that is to say, a product in respect of which it has not been scientifically established that it is capable of restoring, correcting or modifying physiological functions by exerting a pharmacological, immunological or metabolic action, or that it may be used to make a medical diagnosis.

[27] That interpretation is corroborated by the case law to the effect that the interpretation of the provisions of [the Medicinal Products] Directive —which is intended, in addition to protecting human health, to safeguard the free movement of goods within the Community—cannot result in obstacles to the free movement of goods which are entirely disproportionate to the pursued aim of protecting health (see, to that effect, Commission of the European Communities v Germany (C-319/05) at [62] and [71])."

35. I agree with Mr De La Mare QC for the claimants that the CJEU was concentrating on the scientifically established properties of the product in question. Although it is true that in deciding whether a product is or is not a medicinal product the decision maker must apply the Warenvertrieb criteria, scientifically established pharmacological properties are a necessary condition to qualification as a medicinal product. This, in my judgment, is borne out by the court's description of the aim of the definition by function at [25], which is to capture "products the pharmacological properties of which have been scientifically observed and which are genuinely designed to make a medical diagnosis or to restore, correct or modify physiological functions"; and also by the court's statement at [26] again referring to a product which has not been scientifically established to be capable of producing the required effect. It is also consistent with what the CJEU said in Warenvertrieb at [54], namely that the pharmacological properties of a product are "the" factor on the basis of which the national authorities must decide whether it can be administered to human beings for the purposes envisaged by the Medicinal Products Directive.
36. Lyocentre was the first case about consistency of treatment within a single member state. It concerned a vaginal capsule containing live lactobacilli intended to restore balance to the vagina's normal protective bacterial flora. It was marketed in Finland (and other member states) as a 'medical device or accessory.' However, in 2008 the national authorities decided to classify another similar product as a medicinal product because it achieved its principal effect through metabolic and pharmacological action; with the consequence that they decided to reclassify Lyocentre's product as a medicinal product. The sale and advertising of Lyocentre's product therefore required marketing authorisation as a medicinal product, which had not been the case before. Lyocentre challenged that decision, and the Finnish courts referred a number of questions to the CJEU. The first question concerned differential treatment of the same product in different member states; and the CJEU reaffirmed that differential treatment of that kind was not contrary to EU law. The second question was procedural and need not detain us. However, the third question is of importance because it concerns consistency of treatment within a single member state. The CJEU formulated it at [57] in this way:

"… the referring court seeks to ascertain, in essence, whether, within the same Member State, a product which, while not identical to another product classified as a medicinal product, none the less has in common with it an identical substance and the same mode of action, may be marketed as a medical device in accordance with [the Medical Devices Directive]."

37. They answered that question as follows:

"[58] To the extent that another product has several of the significant characteristics set out in Article 1(2)(b) of [the Medicinal Products Directive], namely, where it has one of the same substances and the same mode of action as the product classified as a medicinal product, the former should, in principle, also be classified and marketed as a medicinal product. That being the case, it is for the referring court to verify, on a case-by-case basis, as referred to in paragraph [42] above, that another characteristic that is specific to that product and relevant for the purposes of Article 1(2)(a) of [the Medical Devices Directive] does not preclude the product from being classified as a medicinal product and marketed as such.

[59]      It should, moreover, be recalled that Article 2(2) of [Medicinal Products] Directive states that, in cases of doubt, where, taking into account all of its characteristics, a product may fall within the definition of a 'medicinal product' and within the definition of a product covered by other Union legislation, it must be classified as a medicinal product.

[60]      In the light of the foregoing considerations, the answer to the third question is that, within the same Member State, a product, which, while not identical to another product classified as a medicinal product, none the less has in common with it an identical substance and the same mode of action, cannot, in principle, be marketed as a medical device in accordance with [the Medical Devices Directive], unless as a result of another characteristic that is specific to that product and relevant for the purposes of Article 1(2)(a) of the [Medical Devices Directive], it must be classified and marketed as a medical device, which is a matter for the referring court to verify."

38. One other feature of the CJEU's case law that calls for comment is its frequent references to assessment of a product on a case by case basis: see for example (Case 227/82) Officier van Justitie v van Bennekom [1985] 2 CMLR 692 at [29]; Warenvertrieb at [51]; Garlic Capsules at [65].
39. In (Case C-387/99) Re Vitamin Supplements: Commission of the European Communities v Germany [2006] 3 CMLR 16 the court was concerned with a rule of thumb by which Germany classified as medicinal products preparations containing water soluble vitamins containing more than three times the amount of the daily recommended dose. This was called "the triple amount rule". Having reiterated the Warenvertrieb criteria, including the requirement to work on a case by case basis, the court went on to say:

"[59] In this case, it must be stated that the German practice applies a general rule, applicable without distinction to all vitamin preparations regardless of the vitamin in their composition, which classifies them as medicinal products where they contain more than three times the recommended daily amount.

[60] That practice does not therefore make a distinction in relation to the different vitamins in the preparations examined, even though it is common ground that no vitamin has the same effects on health in general, and, in particular, no vitamin has the same degree of potential harmfulness. As it is applicable without distinction, the triple amount rule can therefore have the effect of classifying certain vitamin preparations as medicinal products even though they are not capable of "restoring, correcting or modifying human physiological functions"."

Discussion

40. The parties differ fundamentally on what the principle in Vitamin Supplements entails. Does it, as Mr De La Mare argues, require the MHRA to investigate GCPs brand by brand; or does it, as Mr Henshaw QC argues for the MHRA, enable the MHRA to treat GCPs generically? Or does it require something between the two?
41. One further matter of debate between the parties was whether the decision of the CJEU in Lyocentre on what have been called "borderline products" was confined to the borderline between medicinal products and medical devices, or whether it is equally applicable to products on the borderline between medicinal products and food supplements. The particular context which the CJEU considered was the borderline between medicinal products and medical devices; but their observations at [59] were not so limited and article 2.2 itself is general in its reference to "other Community legislation".
42. That, said, the borderline between different classifications is not necessarily the same in all cases. It is necessary to inquire what is it that causes a product to fall one side of the border or the other. The particular border under consideration in Lyocentre was that between a medicinal product and a medical device. Article 1.2 (a) defines "medical device" as follows:

"'medical device' means any instrument, apparatus, appliance, material or other article, whether used alone or in combination, including the software necessary for its proper application intended by the manufacturer to be used for human beings for the purpose of:

- diagnosis, prevention, monitoring, treatment or alleviation of disease,

- diagnosis, monitoring, treatment, alleviation of or compensation for an injury or handicap,

- investigation, replacement or modification of the anatomy or of a physiological process,

- control of conception,

and which does not achieve its principal intended action in or on the human body by pharmacological, immunological or metabolic means, but which may be assisted in its function by such means."

43. Article 1.5 (b) provides that the Medical Devices Directive does not apply to medicinal products covered by the Medicinal Products Directive. To complete the picture we must recall that article 2.2 of the Medicinal Products Directive provides that where a product falls both within the scope of the Medicinal Products Directive and also within the scope of other Community legislation, the Medicinal Products Directive takes priority. The borderline between the Medicinal Products Directive and the Medical Devices Directive is the means by which the device in question achieves its intended function. Is it by pharmacological, immunological or metabolic means; or is it by something else? It is, therefore, a critical part of the definition to consider the means by which the device achieves its intended function.
44. It is for that reason that the CJEU singled out for particular mention "the same mode of action as the product classified as a medicinal product", because that is the precise borderline between the two categories. Accordingly, in my judgment although the principle in article 2.2 of the Medicinal Products Directive is a general one, the particular test for distinguishing products that the CJEU laid down in Lycocentre was specific to the border between medicinal products and medical devices.
45. Mr De La Mare accepted (a) that the test laid down in Warenvertrieb was the applicable test for determining on which side of the border a particular product lay and (b) that article 2.2 of the Medicinal Products Directive could not bring within the scope of that Directive a product which had been shown not to satisfy the definition of a medicinal product. However he argued that the MHRA had not properly applied either the Warenvertrieb test or article 2.2. They had also ignored relevant factors and taken into account irrelevant factors, and thus had made an error of law.
46. In my judgment the general point made by the CJEU in Lyocentre was that where a product has "several of the significant characteristics" of a medicinal product, it should in principle be classified as a medicinal product unless it has "another characteristic specific to that product" and relevant to the test. It should be noted that the court did not require that the product in question should have all the characteristics of the product classified as a medicinal product: only that it should have several of them.
47. It is therefore necessary to consider which (if any) of the significant characteristics of Dolenio are shared by unlicensed GCPs. The significant characteristics are those laid down by Warenvertrieb, namely:

"its composition, its pharmacological properties, to the extent to which they can be established in the present state of scientific knowledge, the manner in which it is used, the extent of its distribution, its familiarity to consumers and the risks which its use may entail."

48. Let me take these in turn, although in a slightly different order. First, the composition of the products. The active ingredient in Dolenio is glucosamine sulphate. The active ingredient in many unlicensed GCPs is also glucosamine sulphate. Those GCPs share the significant characteristic of composition. There are other GCPs whose active ingredient is a different salt (e.g. glucosamine hydrochloride) and which, therefore do not share that significant characteristic. Second, the pharmacological properties of the products. The scientific evidence establishes that glucosamine sulphate, at a daily dosage of 1500 mg, has a beneficial effect for the relief of symptoms in mild to moderate OA of the knee. This is true of all glucosamine sulphate, both Dolenio and those unlicensed GCPs whose active ingredient is glucosamine sulphate. By contrast there is no scientific evidence that other glucosamine salts have any effect, either medicinal or indeed in terms of maintaining health. That is why health claims have been outlawed. Nor, indeed, is there any scientific evidence that glucosamine sulphate has any beneficial effect on persons who are not suffering from OA in the knee. Third, the risks. Since the active ingredient of Dolenio and unlicensed GCPs whose active ingredient is glucosamine sulphate have the same pharmacological effect, the risks associated with taking them (whatever they may be) are the same. They therefore share that significant characteristic.
49. Fourth, the manner in which the products are used. Dolenio and unlicensed GCPs are ingested orally, and in many cases unlicensed GCPs whose active ingredient is glucosamine sulphate carry a recommended dose of 1500 mg per day. Accordingly, to the extent that the manner of use refers to how the products are used, they share that characteristic. Thus far, it is fair to say that Dolenio and those unlicensed GCPs whose active ingredient is glucosamine sulphate with a recommended daily dose of 1500 mg do share "several" significant characteristics.
50. However, the manner of use also encompasses why they are used. In that respect, there is some dispute so it is necessary to examine some of the evidence.
51. The first GCP to receive a marketing authorisation in the United Kingdom did so in 2007. Mr Aurora's evidence, which is not challenged in this respect, is that in 2008 about 610,000 prescriptions on the NHS were made out for GCPs, in 2009 about 870,000, and in 2010 about 880,000. It may well be inferred, in the absence of any contrary evidence, that medical practitioners would not prescribe GCPs unless they were thought to have some benefit, either medicinal or related to health. This level of prescription declined dramatically in 2011 as a result of changes in prescribing practice. In 2011 the number of prescriptions on the NHS for GCPs fell to about 310,000. Of that number 89% were filled by the dispensing of unlicensed GCPs, rather than by those with marketing authorisations. In the absence of any contrary evidence it seems reasonable to infer that those whose prescriptions were dispensed by unlicensed GCPs were taking them for the same reasons as those whose prescriptions were dispensed by authorised GCPs. To that extent, therefore, they shared that part of the significant characteristic of the manner in which the products are used. This accounts for some 17 per cent of revenue from GCPs.
52. In addition to the evidence of GCPs being prescribed by medical practitioners there is a substantial body of evidence that the public self-medicate by buying unlicensed GCPs over the counter. I quote some examples from reputable sources:

"Oral glucosamine sulphate 1,500 mg daily probably provides modest symptom relief in patients with osteoarthritis of the knee and its efficacy appeals similar to that of NSAIDs. … On current evidence it seems reasonable to suggest glucosamine sulphate 1,500 mg daily as a treatment option for patients with knee osteoarthritis." (Drugs and Therapeutics Bulletin)

"Many people with osteoarthritis take over-the-counter neutriceutical products and may benefit from clear, evidence-based information. … The GDC felt that it would be beneficial to advise people who wanted to trial over-the-counter glucosamine that the only potential benefits identified in early research are purely related to the reduction of pain (to some people, and only to a mild or modest degree) with glucosamine sulfate 1500 mg daily. They could also benefit from advice on how to perform their own trial of therapy, that is, to evaluate their pain before starting glucosamine and to ensure they review the benefits of glucosamine after three months." (NICE guidelines)

"Glucosamine has attained great popularity as a nutritional supplement, primarily for osteoarthritis….

The fact remains that glucosamine has become popular with patients. This is, to no small degree, due to the encouragement of the medical profession who, when it was first marketed, advocated it as a low-risk alternative to non-steroidal anti-inflammatory drug." (Patient.co.uk)

"Many people take glucosamine sulphate tablets with or without chondroitin for osteoarthritis." (Arthritis Research UK)

"Nevertheless, patients might wish to purchase and try glucosamine sulphate and the evidence suggests that this is a reasonable strategy." (Drugs and Therapeutics Bulletin)

53. The statement by NICE is of particular importance for two reasons. First, it was combined with NICE's guidelines to the effect that it was not cost effective to prescribe glucosamine on the NHS. This may well be the reason for the dramatic decline in the number of prescriptions for GCPs from 880,000 to 310,000. Second, the NICE guidelines contain, as Mr De La Mare rightly submitted, clear encouragement to self-medication through the purchase of GCPs containing glucosamine sulphate over the counter. The only GCPs that could be purchased in that way, are, by definition, unlicensed GCPs because all authorised GCPs are available on prescription only.
54. Because there has been no investigation into the extent of self-medication, it is not possible to say with any confidence what is the extent of the market for self-medication. However, as Mr De La Mare pointed out, since the only scientifically established benefit of GCPs is that glucosamine sulphate has a beneficial effect on OA in the knee, while other GCPs have no beneficial effect on any part of the human body, it seems to be an obvious inference that at least some of those who buy GCPs over the counter do so in order to self-medicate. One might also infer that at least some of the half million patients who used to be prescribed GCPs before 2011 transferred to self-medication once the prescription rate collapsed.
55. Why consumers buy GCPs over the counter cannot be divorced from another significant Warenvertrieb characteristic, namely the familiarity of the product to consumers. This seems to me to be largely concerned with public perception. I have already referred to reputable sources of advice encouraging self-medication. I found persuasive Mr De La Mare's submission that when unlicensed GCPs are dispensed against prescriptions that is likely to normalise the use of such GCPs as medicines; and that submission also finds support from the quotation from Patient.co.uk. To this must be added the fact that some UK retailers have made medicinal claims for GCPs (until stopped by the MHRA); that such claims are also made by companies based off-shore against whom the MHRA cannot take enforcement action, and that such claims are also made on the internet. The making of such claims is likely to have an effect on public perception. On top of that, one must not forget that market authorisations have been obtained on the basis of well-established medical use. For example the Public Assessment Report for an authorised GCP produced by the Swedish authorities stated:

"Glucosamine was mainly introduced on the world-wide market as a food-supplement but with the aim to improve symptoms in patients with osteoarthritis or joint pain or function. These products gained very high sales figures with a huge interest from patients, due to limitations in the existing alternatives to relieve the often debilitating pain in osteoarthritis."

56. The claimant's own application stated that glucosamine sulphate is registered as a medicinal product in 13 member states, and also in Russia. Mr Henshaw said that medicinal use in those countries was not evidence of medicinal use in the UK. In the narrow sense that is no doubt true, but the body of evidence I have described, does, in my judgment, call for investigation of the true extent of medicinal use of glucosamine sulphate in the UK in the absence of any reason why UK consumers are likely to behave any differently from consumers in other member states.
57. One might ask at this stage: why has there been no investigation of the extent of self-medication? The answer appears from the MHRA's letter of 6 March 2013 in response to the claimants' solicitors' letter of 6 December 2012 in which the latter asserted that GCPs were positioned, perceived and used as medicines. The MHRA's response was:

"It is the understanding of the MHRA that, supported by information from the relevant UK Trade Associations, many GCP have been sold simply as food supplements for over 20 years in the UK."

58. In my judgment this was a wholly inadequate response. First, the question is not whether "many" GCPs have been sold as food supplements. The question is whether a significant proportion (and if so what proportion) have been sold as medicinal products. Second, the question is not whether they have been "sold as" food supplements, but how (and why) they have been used. The mere fact that they have been sold as food supplements does not exclude the possibility (for which there is substantial evidence) that they have in fact been used to relieve the symptoms of OA. The MHRA returned to the point in their letter of 18 April 2013 in which they said:

"Moreover, it is the understanding of the MHRA that many GCP have been sold simply as food supplements for over 20 years in the UK. This is, in our view, the predominant perception of these products, rather than their being generally seen or used as medicines."

59. This response suffers from the same flaw in reasoning, namely that it equates how a product is sold with how it is used. In the same letter the MHRA said that consideration of the functional limb of the definition of medicinal product "requires consideration of the manner of use of the relevant product in general." It is, however, clear from that letter that what the MHRA regards as the "relevant product" is the whole class of GCPs. It does not even divide the class into GCPs whose active ingredient is glucosamine sulphate and those whose active ingredient is a different glucosamine salt. As Mr Carter made clear in his second witness statement the MHRA does not review individual brands except by focussing on their presentation and claims made (explicitly or impliedly) for those brands.
60. Two further questions arise at this stage. First, is the relevant question whether the use of GCPs for medicinal purposes is the predominant use, or does it suffice that there is use for that purpose by a substantial cohort of consumers? Second, is the MHRA's decision not to investigate susceptible to challenge? Neither of these questions is answered by current EU jurisprudence.
61. Mr Henshaw emphasised the principle of proportionality which applies to decisions to classify products as medicinal products. There is a tension between the protection of the free movement of goods on the one hand, and risks to public health on the other. The potency of the product (and the risks that taking it entails) must be weighed against the proportion of the product which is in fact used by consumers as a medicinal product. If there is some evidence that a small number of consumers have used a low potency product medicinally it would be disproportionate to classify all similar products as medicinal products. In the abstract, I agree. But the difficulty is that the MHRA has simply not investigated what proportion of unlicensed GCPs are used as medicinal products; and there is a cogent body of evidence that suggests that self-medication is widespread. Nor is there an explicit link in the correspondence or the evidence adduced on behalf of the MHRA that because of the lack of potency of glucosamine sulphate it is acceptable that a greater proportion of consumers use it to self-medicate.
62. Mr Henshaw relied on R (London Criminal Courts Solicitors' Association) v Lord Chancellor [2015] EWHC 295 (Admin); [2015] ACD 95 in which Laws LJ considered the duty to inquire placed upon a decision maker. He referred to the well-known principle that a decision maker must take reasonable steps to acquaint himself with the relevant information. He quoted from an earlier judgment of his own in the following terms:

"where a statute conferring discretionary power provides no lexicon of the matters to be treated as relevant by the decision-maker, then it is for the decision-maker and not the court to conclude what is relevant subject only to Wednesbury review. By extension it gives authority also for a different but closely related proposition, namely that it is for the decision-maker and not the court, subject again to Wednesbury review, to decide upon the manner and intensity of enquiry to be undertaken into any relevant factor accepted or demonstrated as such."

63. He continued at [35]:

"… where there is more than one reasonable view of an issue which a decision-maker must consider – what he should treat as relevant, or what he should enquire into – the court should not interfere with the view of it which the decision-maker takes."

64. In this case although the Medicinal Products Directives provides no lexicon of the matters to be treated as relevant, that deficiency is supplied by the CJEU. The matters to be treated as relevant are those contained in the Warenvertrieb criteria. What, then, of the decision of the MHRA not to investigate the manner in which glucosamine sulphate is used, except at a high level of generality? As Laws LJ said, such a decision is capable of challenge on Wednesbury grounds. For the reasons I have explained, I consider that the MHRA not only asked itself the wrong question, but also closed its mind to the cogent body of evidence suggesting that self-medication by the use of glucosamine sulphate was widespread. In addition Mr Carter's evidence in his first witness statement was that:

"The less potent the ingredients of a product, the greater the importance which factors other than effect, such as consumer familiarity and manner of use, are likely to have. Thus those factors assume greater importance in relation to less potent products which have other long-established uses as foods or food supplements."

65. The logic of this way of putting the point is that in the case of a product like glucosamine sulphate which is said not to be a potent one, it becomes all the more important to investigate its manner of use. But that, as it seems to me, is the opposite of what the MHRA have done.
66. It follows, in my judgment, that the MHRA's decision is capable of challenge on Wednesbury grounds; and that the challenge succeeds. In short, orally ingested GCPs whose active ingredient is glucosamine sulphate and which carry a recommended daily dose of 1500 mg share several significant characteristics with a product classified in the UK as a medicinal product; and in accordance with article 2.2 of the Medicinal Products Directive must be classified in the same way unless they have another significant characteristic which takes them outside the definition of "medicinal product". That characteristic may well be the manner in which they are used, but it has not so far been demonstrated.
67. It is convenient at this point to return to the question: does a "case by case" assessment require the MHRA to investigate GCPs brand by brand, or does it entitle the MHRA to deal with GCPs generically? It seems to me to be clear that, on the facts, the MHRA has dealt with them generically. In my judgment that was an error of law. It puts into the assessment products which do not share significant characteristics of Dolenio, and in particular its active ingredient (glucosamine sulphate) or its recommended daily dosage. This, in my judgment, is directly contradicted by Vitamin Supplements, because in the case of salts other than glucosamine sulphate, according to the scientific evidence, they are not capable of restoring, correcting or modifying human physiological functions; and glucosamine sulphate itself does not have the required effect except at a daily dose of 1500 mg.
68. On the other hand, in Vitamin Supplements what the court envisaged would comply with EU law was described as follows:

"[81] A less restrictive measure would be to fix, for each vitamin or group of vitamins on the basis of its pharmacological properties, a threshold value above which preparations containing one of those vitamins are subject, under national law, to the rules governing medicinal products, while below that value those preparations would obtain a simple product authorisation.

[82] It is true that evaluation by the competent German authorities of the pharmacological properties of each vitamin or group of vitamins for the purposes of classification of vitamin preparations may correctly lead to the same result as the triple amount rule in some cases."

69. There is no suggestion here of a requirement to evaluate vitamin preparations brand by brand. Rather the court envisaged an evaluation vitamin by vitamin, and the basis of assessment was the pharmacological properties of each. Mr De La Mare argued that the inclusion in the Warenvertrieb criteria of both familiarity to consumers and also extent of distribution could vary from brand to brand; and that accordingly the assessment on a case by case basis could not be carried out except on a brand by brand basis. Although I see the force of the point, the fact is that in Vitamins Supplements the CJEU specifically stated at [57] that the Warenvertrieb criteria applied to the assessment; and I do not consider that its conclusion at [81] and [82] can be taken to have intended any departure from those criteria. Moreover the court specifically envisaged that a threshold value for an individual vitamin could be set based on its pharmacological properties. There was no suggestion that consumer familiarity or extent of distribution would make a difference to that threshold value. I would not, therefore, go as far as Mr De La Mare. I consider that the limit of the required investigation is an investigation into the manner of use of orally ingested GCPs whose active ingredient is glucosamine sulphate and which have a recommended daily dose of 1500 mg.
70. There is one other point to mention. The stabilising agent in Dolenio is sodium chloride, whereas other GCPs containing glucosamine sulphate use potassium chloride as their stabilising agent. The effect of this is said to be that the amount of pure glucosamine delivered by an equivalent daily dose is lower. This is not a point that the MHRA has so far made, but it may well be relevant to the question whether GCPs other than those with marketing authorisations have a significant characteristic which takes them outside the definition of "medicinal product". That will be for the MHRA to decide.

Result

71. For the reasons I have given I would allow the appeal, quash the decision of the MHRA and remit the question whether to classify some or all GCPs as medicinal products for further consideration by the MHRA in the light of this judgment.

Lord Justice McCombe:

72. I agree.

Lord Justice Longmore:

73. I also agree.