There are two judgments on BAILLI this morning in the Glaxosmithkline case where the judge has resisted the claimants’ attempts to widen the scope of their case beyond the pleaded case and the issues set out in a Group Litigation Order. The case is interesting in respect of the ingenious attempts by the claimants to argue new matters. Also note the claimants’ complaints that the defendant had not applied to strike the action out…

“The Claimants having unequivocally pinned their colours to the mast, the Defendant was entitled to lock horns with the claim as pleaded against it and assert that, as a matter of law, the Claimants’ approach to defect was flawed”


The claimants bring an action based on their use of the drug Seroaxat.   The proceedings were started in 2007. The were stayed in 2001. In hearings in 2016 and 2017 it was determined that the stay should be lifted “but only on the basis that the Claimants’ case should remain as pleaded at the date of the vacated trial.”


In the first hearing, Bailey & Ors v Glaxosmithkline (UK) Ltd [2019] EWHC 337 (QB),   prior to the trial starting, the claimant were trying to argue it could raise “freestanding defect” as an an issue at the trial, based on symptoms caused when patients ceased taking the drug.  Mrs Justice Lambert rejected the claimants’ argument.

    1. I agree with Mr Gibson that the freestanding defect is not in scope for the following reasons:
a. the high point of the Claimant’s case is paragraph 12.1 of the Particulars of Claim. That paragraph refers to the adverse effects on discontinuance making it “more” difficult for the user to withdraw from usage of the drug; even within this assertion is an implicit comparison as to the ease of discontinuance with other drugs.
b. I also note the way in which the case is pleaded at paragraph 5.1 where the Claimants set out their generic case on defect:

The Claimants contend that:

5.1 the product was defective as defined in the Directive and the Act because the safety of the Product was not such as persons generally were entitled to expect in that the capacity of the Product to cause adverse effects consequent upon or following discontinuance (withdrawal) was such as to prevent or make more difficult the ability of users to discontinue, withdraw from or remain free from taking the product to an extent greater than other SSRIs”

c. The GLO makes no reference to the freestanding defect now alleged. Critically, the GLO reflected the agreed list of issues of fact and law at the time of the adjourned trial before Mackay J.
d. Foskett J set out the Claimants’ case on defect in clear terms: in the February 2016 ruling at [5] and [7]. At [58] of the same judgment he records that Professor Hotopf (an expert instructed by the Claimant) “explained to the new legal team that his changed position still supported the main thrust of the case that the adverse effects of the drug were “worst in class.”
e. Foskett J, in his judgment of March 2017 (and perhaps because of his sense that there was some attempt by the Claimant to expand their case) repeated at [11] the essential nature of the case advanced on behalf of the Claimants’ and expressed his belief that his summary was “common ground” At [20] he observed: “since close of pleadings it has been plain that the Claimant’s case proceeds onlyon the basis of what can be described as a worst in class for discontinuation symptoms for SSRIs allegation and the associated allegation of a failure to warn that Seroxat was “worst in class” in this respect“. It does not seem to me that Foskett J could have been any clearer as to his understanding of the Claimants’ case on defect.
  1. My analysis of the issues raised in the pleadings accords with that of Foskett J. Further, Foskett J was not exercising a purely case management role at the hearings in February 2016 and March 2017, at least not in the usual sense. The issue which Foskett J was grappling with was whether the case should be revived at all in 2015, after a 5 year hiatus. He was required to make his ruling in the face of powerful arguments deployed by the Defendant that he should not do so. He sought to strike a fair balance between the parties. The balance came down in favour of permitting the claim to proceed, but only on the basis of the case as advanced by the Claimants in 2010 when it was stayed. It seems to me that in those circumstances, I should in any event be very slow to disturb his decision and expand the scope of the trial. I also add that I am informed by Mr Gibson that if the scope of the proceedings were to be expanded, the Defendant would be prejudiced as there would be insufficient time to embark upon examining and meeting a broader case than that pleaded. I accept this.
  2. My ruling is therefore that the Claimant’s case on defect is restricted to that reflected in the GLO and does not include the freestanding case based upon the incidence, severity and duration of adverse events upon withdrawal from the drug.


The second judgment, Bailey & Ors v Glaxosmithkline (UK) Ltd [2019] EWHC 1167 (QB), and is a judgment made in the context of an ongoing trial,  again the judge rejected an attempt by the claimants to introduce new matters into the action where these had not been pleaded.

    1. The Claimants’ written opening was served on 5 April 2019. It contained the following at paragraphs 57, 58 and 59: “it is not and has never been the Claimants’ case that a product can be shown to be defective within the CPA 1987 merely by identifying one negative and/or undesirable aspect of it whilst ignoring any advantages. It is a reductio ad absurdum on the part of the Defendant that fails to recognise the case being advanced by the Claimants.” The submission went on “what is being advanced herein is indeed a holistic approach, namely that whatever benefits asserted by the Defendant for this product in these proceedings, they are outweighed by the risks and problems associated with DS, having regard to inter alia the existence of equally efficacious products which do not have those risks/problems. In simple terms, if all the SSRIs are equally efficacious (as is agreed), why would any patient take the one which is “worst in class” for DS?
    2. Correspondence from the Defendant followed on, predictably quickly, from service of the Claimants’ opening. The Court was reminded by the Defendant that there had been a slew of rulings from the Court in which the Claimants’ case had been defined; that the case was about whether Seroxat was “worst in class” for discontinuation symptoms and that, in both the ruling of Foskett J of March 2017 and my ruling of February 2019, the relative benefits of Seroxat whether in comparison with other SSRIs or otherwise had been determined to not form part of either parties’ case. The Defendant stated that it was not open to the Claimants to shift the goal posts and now maintain that their case on defect was “holistic” and that the relative benefits or risks of Seroxat (other than in respect of symptoms on discontinuation) should be taken into account by the Court when considering the circumstances relevant to safety: this was not their pleaded case on defect which focussed only upon the worst in class aspect.
    3. The forensic effect of the point being made by the Claimants in their written opening emerged clearly only during the course of Ms Perry’s oral opening submissions. Her submission was that, when the Court was considering whether Seroxat was unsafe (by reference to what persons generally are entitled to expect) the Court must “start with a level playing field.” In other words, the Court should assume that Seroxat has no particular benefits over and above other comparator drugs. Ms Perry submitted that “if you start with products that all, basically, do the same thing, and no one of those products is said to be particularly better than the others. You, therefore, are starting at a position where – and the medical lecture support this – the only thing that somebody prescribing this drug should or needs to be concerned about is what are the downsides? What are the discontinuation effects? What are the discontinuation symptoms? Because if, ultimately, you’ve got five drugs all doing the same thing, it must follow that if they’re all going to achieve the same result, but one of them is going to cause far more problems than the others, it isn’t just a case of being worst in class“. Therefore, I should approach the question of whether the drug is not safe in all of the circumstances because there are “no benefits” associated with the drug but “just burdens.
    4. Ms Perry submits that her case is consistent with the pleadings. She argues that too much emphasis has been placed upon the response to the Request for Further Information (question 6) which, to adopt the language of Foskett J, appeared to disavow consideration of the relative benefits of the drug when considering whether it was an unsafe and therefore defective product. That response she submits should be viewed in conjunction with the response to question 7 which contained a denial that the product had any greater efficacy or other substantial benefit than other drugs in the class. Furthermore, she submits, the Defendant has admitted that Seroxat has no particular advantage so far as its efficacy is concerned. Ms Perry also submitted that it was up to the Defendant to plead the asserted relative benefits of the drug. It has not done so. In the absence of a pleaded case, the Defendant must be taken to have, in effect, conceded that Seroxat has no relative benefits. It has raised the issue of the lawfulness of the Claimants’ approach to defect and yet not made an application for summary judgment or to strike out the Claimants’ case. It should have done so. She submits that as the evidence has emerged, the experts support the case that in fact Seroxat has no particular benefits or indications over and above other drugs of a similar nature. As such, the Court should proceed on the basis of the facts as they have emerged irrespective of the state of the pleadings. Finally, she submits that, to the extent that Foskett J analysed the Claimants’ case, he did not drill down into the detail of the pleadings; at the hearing before him which led to the March 2017 ruling she had been unable to attend and the Claimants had been represented by Mr Lambert who had limited knowledge of the case having only been instructed for a short period of time.
    5. The Defendant submits that this approach represents a very significant and fundamental expansion on the Claimants’ pleaded case. The Claimants’ pleaded case on defect expressly and unambiguously excludes consideration of any possible relative benefits associated with Seroxat compared with other drugs in the comparator class and the case on defect focusses only upon one product characteristic, namely the alleged greater incidence, severity and duration of discontinuation symptoms. This is the case which the Defendant equally unambiguously challenged in its defence as a matter of law and on the facts. Foskett J correctly analysed the Claimants’ case and the defence. In March 2017 Foskett J noted Mr Lambert’s submissions on relative risks and benefits and in his ruling tackled the question of whether, on the pleadings, relative risks and benefits formed part of the circumstances relevant to the Claimants’ case on safety and defect. Foskett J made it plain that they did not do so and that Mr Lambert’s attempt to expand the case was impermissible and inconsistent with the basis upon which he was prepared to allow the claim to continue. I made a similar point in February 2019. Neither the Foskett rulings nor my ruling were appealed. If I or Foskett J had misinterpreted the pleadings, then the Claimants should have appealed. The Defendant has not, by deciding not to advance a positive case on benefits, conceded that no benefits exist. The Defendant has in its pleading simply responded to the claim as pleaded. The Defendant submits that the Claimants’ case as set out in the opening represents a significant expansion of the case which, if permitted to be ventilated at trial, would result in unfairness to the Defendant: the disclosure exercise and the assembling of the expert evidence have all proceeded on the basis that the Claimants case is limited to the “worst in class” for discontinuation symptoms allegation. The experts have not been instructed to consider the relative benefits of Seroxat, precisely because this topic is not in issue in the case.
My Decision and Reasons
    1. I have already informed the parties of my decision which is that, when considering whether the safety of the drug, Seroxat, is such as persons generally are entitled to expect, the Claimants are not entitled on their pleadings to submit that the drug has no relative benefits and that so far as relative risk and benefits are concerned, there is a “level playing field.”
    2. I made this decision for the following reasons:
a. The case now advanced in the Claimants’ opening is not consistent with the pleadings. The defect alleged in the Particulars of Claim is limited to the greater capacity of the drug to cause discontinuation symptoms (in respect of incidence, frequency and severity) compared with other similar drugs. Whilst appreciating that Ms Perry does not approve of the label “worst in class” as a shorthand for her case, I, like Foskett J, find it to be an accurate characterisation of the pleaded case. The case on defect does not include an holistic assessment of relative risks and benefits across the drugs in the group: the response which was given to question 6 in the Request for Further Information confirmed that benefits of the drug against other SSRIs for a particular claimant were not material or to be taken into account. As I noted in my ruling of February 2019, the Claimants were here stating that their case on defect was independent of and irrespective of any relative benefits which the drug might have. The fact that in a subsequent response, the Claimants asserted that (to the extent that it was relevant) the drug did not have any potential relative benefits, does not undermine the Claimants’ case that relative benefits were not material to the consideration of defect. The effect of the case now advanced by the Claimants is that the Court is being invited to take into account, when considering the safety of the drug, the relative benefits of Seroxat compared with other comparator drugs; and to take them into account on the basis that the drug has no such relative benefits. This is not the pleaded case.
b. The Claimants having unequivocally pinned their colours to the mast, the Defendant was entitled to lock horns with the claim as pleaded against it and assert that, as a matter of law, the Claimants’ approach to defect was flawed. Again, as I noted in February 2019, the Defendant was under no obligation to advance a positive case as to any relative benefits which the drug might possess. The Defendant could have “further and alternatively” set out its stall as to any benefits or indications for use which Seroxat had in comparison with, say, citalopram or other drugs of a similar pharmacological make up. But it chose not to do so. The fact that it did not do so is not a concession that no such benefits existed. It might be said that the Defendant was, by taking the strategic approach that it did and not pleading the range of indications for Seroxat in comparison with other SSRIs, taking something of a risk; I am not determining the merits of the Defendant’s legal challenge in this ruling which may or may not have merit. But it was an approach that the Defendant was entitled to take on the pleadings.
c. The only limited concession which the Defendant has made in the pleadings is that, so far as efficacy of the drug is relevant there is no basis for distinguishing Seroxat from other SSRIs (see paragraph 49(c) of the Particulars of Claim and the Response to the Request for Further Information on that paragraph). At times, during Ms Perry’s submissions, it appeared that she was reading this statement as a concession that Seroxat had no relative benefits, rather than the statement being limited to a reference to therapeutic efficacy. I do not read the statement as amounting to anything more than an admission by the Defendant that, so far as the drug acts as a treatment for its licensed indications (e.g anxiety or depression), there is nothing to choose between it and other SSRIs. Nor do I find that on any plain reading of the statement in paragraph 49(c) “efficacy” could sensibly be said to be referring to anything other than therapeutic efficacy.
d. Just pausing here and standing back from the Claimants’ submission that the Defendant has, by its failure to set out its stall in respect of the range of relative benefits associated with the drug, conceded that no such benefits exist, it is clear to me that the submission is wrong. One prong of the twin attack (that is, the legal challenge) by the Defendant in its Defence is that the Claimants have failed to approach defect lawfully by failing to adopt an holistic assessment of relevant circumstances which impact upon the safety of the product and that such an assessment would or should include assessment of relative risks and benefits across the drugs class. Against this background it would be surprising if in fact the Defendant had conceded that no such benefits exist. I find that the Defendant has not done so.
e. Foskett J analysed the pleadings with care. He had to do so because the only basis upon which he was prepared to permit the action to continue was that the action was confined to the pleaded claim as at 2010/2011. I have set out the relevant excerpts from his judgments above and I do not accept that he, for whatever reason, failed to drill down into the issues in the case. Nor could he have been clearer in his analysis. His analysis identified correctly the nature of the Claimants’ case. His understanding, as he recorded it, was that his analysis was common ground between the parties. In March 2017, he was faced with submissions by Mr Lambert that “negative risk benefit” was a relevant factor in assessing whether discontinuation problems were a defect (as part of all the relevant circumstances) and he rejected the submission that this was part of the Claimants’ case (“there has been a hint in some of Mr Lambert’s submissions that there is now a desire to engage, at least to some extent, in a risk/benefit analysis, something which had previously been expressly disavowed… it is now too late to do so”). It seems to me that the way in which the claim is now being advanced by the Claimants flies in the face of that ruling. There was no appeal from Foskett J’s ruling.
f. The Claimants could have, in response to the Defence, sought to amend their pleadings and plead, even as an alternative case, that an holistic assessment including consideration of relative benefits and risks formed part of the circumstances which impacted upon safety and so defect. This did not happen. Mr Kent has submitted that it would have been impossible for the Claimants to have done this. He submits that the Claimants could not have, proleptically, set out to deny each and every possible relative benefit which might be asserted by the Defendant. I do not understand this point. On the basis of such expert evidence as was available to them, the Claimants could have pleaded as part of an holistic case on defect that there were no benefits or indications for prescribing Seroxat to any patient given its “worst in class” status on discontinuance. It did not do so. Although Mr Kent correctly submits that the absence of such relative “substantial benefit” finds its way into the response to the Request for Further Information (at question 7), it is clear from the response (“Strictly, given the answer to 6, an answer is not required“) that the assertion did not form part of the Claimants’ case on defect and that the absence of substantial benefits was not therefore relevant.
g. Nor do I accept that the fact that there is or may be expert evidence to be deployed at trial which supports the Claimants’ case that there is a level playing field across all the drugs in the appropriate class is relevant. This is, in any event, disputed by Mr Gibson. Further, I am told by him, and accept, that there is expert evidence which could have been deployed which would support the existence of certain benefits and indications associated with Seroxat which would make it the drug of choice for some patients. The fact however is that given that the relative benefits of the drug were not in scope the experts have not examined the topic. Further, it would be too late to do so now. One point of agreement (and the only point of agreement it seems) between Ms Perry and Mr Gibson is that neither are ready or able to embark upon an examination of the particular relative benefits of Seroxat in this trial.
  1. There is no application before me to amend the Particulars of Claim. Had one been made, I would have refused permission. It is now too late. If this claim is to proceed then it must do so on the basis of the current pleadings as analysed by Foskett J in 2016 and 2017. I accept that it would cause unfairness to the Defendant if the case were permitted to go forward on the understanding/assumption/inference that when considering whether Seroxat is defective the drug has no relative benefits compared with other SSRIs (or others in the appropriate comparator class).
  2. I make two final observations. The first addresses the Claimants’ recurring complaint that the Defendant, having challenged the lawfulness of the approach taken to defect in the Particulars of Claim, should have launched an application for summary judgment or for a strike out. Foskett J was clear that he could as part of his case management tools have listed the issue for a preliminary hearing. He decided against this course. I too resisted the Claimants’ invitation in February to rule on the point in the absence of an application by the Defendant. The Defendant has made clear in the past that it wishes the litigation to be determined, once and for all, on all issues. However, whatever, may be the reason for the absence of an application to strike out the claim, it does not seem to me to be relevant to the issue before me now, which is the scope of the claim on the pleadings. I make no ruling on the merits of the Defence case on the point. I am not invited to do so, given that this ruling on scope is a necessary prequel to any further legal argument. That said, it would be wrong for me not to acknowledge the force of the Defence submission in the light of the recent judgments of the High Court in Wilkes v Depuy International Limited [2016] EWHC 3096 (QB) and Gee v Depuy International Limited[2018] EWHC 1208 (QB) which have, in both cases, underscored the relevance of relative risks and benefits of a medicinal product when considering safety. I emphasise however that I have not heard full argument on the point.
  3. Second, it seems to me that the issue which is addressed in this ruling (and which has taken considerable court time to ventilate) has already been covered, centrally, by Foskett J in March 2017 and also by myself in my ruling in February 2019. As I have already said, there has been no appeal from either of those rulings. What the Claimants have sought to do by opening the case in the way they have, is to seek to justify the limited approach (said to be flawed by the Defendant) on defect on the basis of an asserted concession by the Defendant that if a wider risk/benefits analysis were to be undertaken it would reveal a level playing field across the class of drugs. This case simply does not square with the Claimants’ pleaded claim nor with Foskett J’s analysis, nor mine. If either Foskett J or I were thought to be wrong in our analysis, then the proper course would have been to have appealed the relevant rulings. It is now too late to do so.