Product Liability 2020 · 2020. 12. 17. · %JKGH/GFKC1H EGT Fraser Allan CEO Jason Byles Printed...

Product Liability 2020A practical cross-border insight into product liability work

18th Edition

Featuring contributions from:

Addleshaw Goddard LLP

Advokatfirman Glimstedt

Allen & Gledhill LLP

Arnold & Porter

AZB & Partners

Bae, Kim & Lee LLC

Bahas, Gramatidis & Partners

BLD Bach Langheid Dallmayr

Clayton Utz

Faus & Moliner Abogados

Harris Kyriakides LLC

Herbert Smith Freehills LLP

Iwata Godo Law Offices

Kellerhals Carrard

Latham & Watkins LLP

Lee and Li, Attorneys-at-Law

Legaltree

LR LEX Avvocati Associati

Pinheiro Neto Advogados

Rahmat Lim & Partners

Sidley Austin LLP

Soemadipradja & Taher

Squire Patton Boggs

TURUNÇ

Product Liability 202018th Edition

Contributing Editors:

Adela Williams & Tom FoxArnold & Porter

©2020 Global Legal Group Limited. All rights reserved. Unauthorised reproduction by any means, digital or analogue, in whole or in part, is strictly forbidden.

DisclaimerThis publication is for general information purposes only. It does not purport to provide comprehen-sive full legal or other advice. Global Legal Group Ltd. and the contributors accept no responsibility for losses that may arise from reliance upon information contained in this publication. This publication is intended to give an indication of legal issues upon which you may need advice. Full legal advice should be taken from a qualified professional when dealing with specific situations.

ISBN 978-1-83918-045-3ISSN 1740-1887

Published by

59 Tanner StreetLondon SE1 3PLUnited Kingdom+44 207 367 0720 [email protected] www.iclg.com

Group Publisher Rory Smith

Publisher Oliver Smith

Editor Oliver Chang

Senior Editor Sam Friend

Head of Production Suzie Levy

Chief Media Officer Fraser Allan

CEO Jason Byles

Printed by Ashford Colour Press Ltd.

Cover image www.istockphoto.com

Strategic Partners

Table of Contents

Expert Chapters

Q&A Chapters

7

14

U.S. Product Liability Law: Recent Developments, Trends, and Future OutlookDaniel A. Spira & Teri H. Peeples, Sidley Austin LLP

Criminal Liability for Defective ProductsHoward Watson & David Bennett, Herbert Smith Freehills LLP

21 Medicinal Cannabis and CBD Products – Navigating the UK Regulatory LandscapeLouisa Caswell, Mark Chesher & Cécile Burgess, Addleshaw Goddard LLP

26 AustraliaClayton Utz: Colin Loveday & Andrew Morrison

36 BrazilPinheiro Neto Advogados: Sérgio Pinheiro Marçal & Laura Beatriz de Souza Morganti

155 SpainFaus & Moliner Abogados: Xavier Moliner & Juan Martínez

165 SwedenAdvokatfirman Glimstedt: Finn Stenström

1 Update on the EU Product Liability DirectiveAdela Williams & Tom Fox, Arnold & Porter

43 CyprusHarris Kyriakides LLC: Ioanna Leonidou & Charis Andreou

64 FranceSquire Patton Boggs: Carole Sportes & Valérie Ravit

72 GermanyBLD Bach Langheid Dallmayr: Dr. Martin Alexander & Sarah Cloth

79 GreeceBahas, Gramatidis & Partners: Dimitris Emvalomenos

87 IndiaAZB & Partners: Vivek Bajaj, Kaavya Raghavan & Sherien Kaul

96 IndonesiaSoemadipradja & Taher: Romi Emirat, Verry Iskandar, Aris Budi Prasetiyo & Oene Marseille

103 ItalyLR LEX Avvocati Associati: Giuseppe Francesco Bonacci & Edoardo Tosetto

109 JapanIwata Godo Law Offices: Shinya Tago, Landry Guesdon & Kohei Yamada

119 KoreaBae, Kim & Lee LLC: Tony Dongwook Kang & Yongman Bae

127 MalaysiaRahmat Lim & Partners: Karen Foong Yee Ling & Kwong Chiew Ee

136 NetherlandsLegaltree: Antoinette Collignon-Smit Sibinga & Carolien van Weering

172 SwitzerlandKellerhals Carrard: Dr. Claudia Götz Staehelin & Nina Studer

180 TaiwanLee and Li, Attorneys-at-Law: Patrick Marros Chu & David Tien

189 TurkeyTURUNÇ: Naz Esen & Beste Yıldızili Ergül

195 USALatham & Watkins LLP: Gwyn Williams & Christine Rolph

50 England & WalesArnold & Porter: Adela Williams & Tom Fox144 SingaporeAllen & Gledhill LLP: Dr. Stanley Lai, SC & Amanda

Soon

Welcome

EditorialWelcome to the eighteenth edition of International Comparative Legal Guide – Product Liability.

This guide provides corporate counsel and international practitioners with a worldwide legal analysis of the laws and regulations of product liability.

It is divided into two main sections:

Four expert chapters. These chapters are designed to provide readers with an overview of key issues affecting product liability law, with a particular focus in this edition on European and US law.

Country question and answer chapters. These provide a broad overview of common issues in product liability laws and regulations in 21 jurisdictions.

All chapters are written by leading product liability lawyers and industry specialists and we are extremely grateful for their excellent contributions.

Special thanks are reserved for the contributing editors Adela Williams and Tom Fox of Arnold & Porter for their invaluable assistance.

Global Legal Group hopes that you find this guide practical and interesting.

The International Comparative Legal Guide series is also available online at www.iclg.com.

Ian Dodds-SmithPartnerArnold & [email protected]

Welcome

PrefaceI’m delighted to have been asked to introduce the eighteenth edition of International Comparative Legal Guide – Product Liability.

The guide continues to be an ideal reference point with four excellent expert chapters covering significant developments in European and US law. This edition also has a special focus on criminal liability for defective products and UK regulations concerning medicinal cannabis and CBD products.

As always, the bulk of the edition remains the enormously helpful country question and answer section, covering 21 jurisdictions, new to the guide this year being Cyprus, Indonesia, Italy, Malaysia, Sweden and Turkey.

I frequently have cause to make reference to the guide for matters concerning product liability all over the world and will continue to do so as the guide remains a thoroughly informative and comprehensive publication.

Tom SpencerSenior Counsel GlaxoSmithKlineDispute Resolution & Prevention

Chapter 1 1

Update on the EU Product Liability Directive

Arnold & Porter Tom Fox

Adela Williams

Product Liability 2020

formation – ‘New Technologies Formation’ (‘NTF’) – is intended to “assess the implications of emerging digital technologies for the wider liability frameworks at EU and national level”. Both were due to report by ‘mid-2019’.

6. The NTF has already reported and this report is reviewed below. The Product Liability Formation has not yet reported.

NTF Report7. This report titled ‘Liability for Artificial Intelligence and

Other Emerging Digital Technologies’ (‘the AI Report’ or ‘the Report’) was published in August 2019. The meaning of the expression ‘emerging digital technologies’ reflects the Commission Staff Working Document on Liability for emerging digital technologies SWD(2018) 137 final (‘SWD(2018) 137’), which provides no explicit definition, but gives the following examples: “the Internet of Things (IoT), Artificial Intelligence, advanced robotics and autonomous systems.”

8. According to the authors’ description of their remit, the AI Report was meant to look at existing liability regimes and assess whether they are adequate to deal with the emerging realities of new digital technologies. In particular, such liability regimes should be assessed against the criterion of whether they facilitate the uptake of new technologies by fostering investment stability and users’ trust. The AI Report was intended to make recommendations limited to “matters of extracontractual liability, leaving aside in particular corre-sponding (and complementary) rules on safety and technical standards”.

9. Consequently, the AI Report extends beyond the product liability regime covered by the Product Liability Directive. It reviews both this and other liability systems which exist in the EU/EEA Member States and considers how well-adapted they are to dealing with liability issues arising from emerging digital technologies, including AI.

Problems Identified in the AI Report in Relation to EU Liability Systems Generally10. The AI Report considers that the existing liability regimes

in Member States do offer some protection and means of redress to those affected by AI and other new technol-ogies. However, it states that in a number of respects, emerging digital technologies represent a fundamental change. Challenges, in particular in terms of complexity, modification through updates or self-learning during oper-ation, limited predictability, and vulnerability to cyberse-curity threats, mean that compensation for victims may be more difficult to achieve and/or the apportionment of risk between producers/operators of such technologies and the consumers is unfair or imbalanced. The Report therefore recommends that certain adjustments should be made to EU and national liability regimes.

Introduction1. This updating chapter concerns the ongoing review of the

Product Liability Directive 85/374/EEC (‘the Product Liability Directive’), particularly in the context of new technologies and the development of artificial intelligence (‘AI’). It builds on reviews in previous years, and considers the purpose and direction of travel of the ongoing review.

2. The Product Liability Directive was adopted only after extensive negotiation and debate. It is characterised by an explicit balance between consumer and producer inter-ests (a ‘fair apportionment of risk’). It retains features which show clearly that, while it is intended to be fully harmonised in respect of the matters which it covers, some aspects of product liability are left to the legal systems of the Member States and it co-exists with some other liability regimes (e.g. contract and negligence) in Member States.

3. Reviews of the Product Liability Directive have been noto-riously slow and difficult, but have all broadly concluded that the Product Liability Directive is suitably flexible and still fit for purpose. That position is allegedly challenged by new technologies. Past reviews have concluded, and chapters in previous versions of this guide have argued, that significant changes are unnecessary. This position is called into question by the Commission’s Fifth Report and, more recently, by the report from one of the Expert Working Groups tasked by the Commission with consid-ering changes to the Product Liability Directive.

The Commission’s Fifth Report4. This report, which covered the period 2011–2015, followed

an extensive consultation exercise and was submitted in 2018. Discussed extensively in previous versions of this guide, the Fifth Report identified challenges, and reflected the concerns of the stakeholders consulted, some of whom felt that new technology made the case for changes to the Product Liability Directive. The Fifth Report did not go so far in its conclusions, but made the case for there to at least be clarification of the status of software in the context of the Product Liability Directive, and perhaps also clarification, via examples, of the application of the Product Liability Directive to emerging technologies such as complex networked products and/or products involving AI.

The Expert Working Group: Two Formations5. In March 2018, after the European Parliament adopted a

resolution on Civil Law Rules on Robotics, the Commission established an Expert Working Group in two formations: the first formation – ‘Product Liability Formation’ – is intended to assist the Commission in drawing up guid-ance on the Product Liability Directive; and the second

© Published and reproduced with kind permission by Global Legal Group Ltd, London

2 Update on the EU Product Liability Directive


Fault-based liability and burden of proof

16. The AI Report states that “in the traditional model, the proper functioning of the tool and the expected outcome of the operation are known and easy to establish”. Emerging digital technologies by contrast are fast-developing, opaque, and may have a degree of autonomy and limited predictability. All these challenge the ability to establish the standard of care and breach necessary to prove liability.

17. The AI Report states that “AI systems cannot all be measured according to duties of care designed for human conduct”. Which prompts the question: how should liability be assessed in relation to some complex systems? The AI Report correctly notes that products tend to be governed by standards in Europe, and suggests that these might form the basis for the development of a ‘duty of care’ standard against which the performance of such systems might be measured, to inform decisions on liability in a tort context. However, it observes that such standards will take some time to develop, so will not immediately be available as a yardstick. This is probably correct, but it is already the case that tort law is informed by, rather than reliant upon, standards. For existing products and services, regulations or other harmonised standards may set out the essential safety requirements; a failure to meet such essential safety requirements may then provide evidence of negligence (or of defect in a product liability case). Standards also evolve with technological developments, so they are not fixed.

Strict liability and product liability

18. The AI Report discusses strict liability generally, noting that tort systems in Europe include a variety of mecha-nisms. While acknowledging that the application of existing strict liability and product liability systems to emerging digital technologies raises fewer difficulties than the application of fault-based systems, the AI Report concludes that certain problems remain.

19. The authors acknowledge the risk that the imposition of strict liability may inhibit innovation, but appear to consider this risk to be exaggerated compared to the rela-tively higher business uncertainty where the question of liability is unresolved and left to be dealt with under general tort systems. Ultimately, the AI Report favours strict liability for new technologies, recognising that meas-ures such as the inclusion of caps on liability may promote certainty and make liability risks more insurable.

20. The Report notes that the Product Liability Directive has been an important part of the European consumer protec-tion system for over 30 years. However, as acknowl-edged in the (Fifth) Report from the Commission to the European Parliament on the functioning of the Product Liability Directive, submitted in 2018, some concepts underpinning the Directive require updating in order to address the potential risks of emerging digital technolo-gies. In particular, the concepts of ‘product’ and ‘defect’ may not be readily translated to emerging digital technol-ogies and the focus of the Directive on the product at the date it was put into circulation does not reflect the reality of many emerging technologies which involve frequent updates and may evolve based on data inputs and interac-tions with the environment. These issues are considered in more detail in the context of the recommendations of the AI Report, below.

Complexity

11. The AI Report suggests that liability systems that have existed to date are not well-suited to dealing with complexity. Where there are multiple factors at play, and links in the chain of causation are under the control of the defendant manufacturer or service provider, the Report states that it is already difficult for claimants to estab-lish liability. Where a product or service involves AI and machine learning, changes made on the basis of post-mar-keting data inputs will make the position more compli-cated and require expert evidence and analysis.

12. The AI Report acknowledges that many existing products which do not involve new technologies are complex, and also accepts that: “Problems of uncertain causation are of course not new to European legal systems.” However, it suggests that increasing complexity of new technologies (and their inter-action with, and alteration via data) will mean that it is harder than ever for claimants to discharge the burden of proof.

13. The Report states that the ‘classic’ response of European tort systems, when faced with the impossibility of deter-mining causation between competing claims, is to find no one liable (as the victim’s evidence fails to reach the threshold to prove liability by a single cause) or, the majority approach, all potential actors are jointly and severally liable. The authors of the Report comment that proportional liability for potential tortfeasors, based on the likelihood of their having caused the harm, is a more ‘modern’ approach that is applied in some cases. As the AI Report acknowledges, European legal systems are well-experienced in handling complex questions of causa-tion and have been creative in developing responses to situations where victims might fail to establish liability. Aside from findings of joint and several liability, legal systems have developed other mechanisms to allow claim-ants to cut through in cases of uncertainty and complexity, including theories of material contribution and rebuttable presumptions. Complexity and asymmetry of informa-tion is also nothing new. It may be addressed by features of national litigation systems such as expert evidence and disclosure obligations.

14. The AI Report uses the example of an autonomous car crashing into a tree to illustrate the challenges associated with emerging digital technologies and outlines the poten-tial difficulty in establishing whether this was caused, for example, by a flaw in the design of the car, by the car receiving incorrect data or incorrectly processing correct data, or by the failure of a software update. In such an example, plainly the car is not meant to crash, so an error has occurred or some event has occurred which was not foreseen when the car or its data systems were designed or updated. The investigation of what occurred in the given example would, as indicated in the Report, undoubtedly, be complex. However, it is not clear that emerging digital technologies involve such a major qualitative change in complexity that existing liability systems cannot accom-modate them.

15. This may be the case in the future, but in the meantime, at least on the basis of this Report, there seems to be little evidence that high-tech businesses are evading responsi-bility because the complexity of their products/services makes it too hard to establish liability. As the digital tech-nologies change and become more complex, so too do the resources and technologies which enable the functioning of such technologies to be understood.


3Arnold & Porter


applicable to emerging digital technologies. The operator may be the manufacturer of the technology or another party, but the key thing is that they are identifiable as the party that exerts control over the risks posed by the operation of the product in terms of a range of activities from activating the system to determining the output or result. Economic benefit may also be relevant to determining who should be viewed as the oper-ator, but is unlikely to be the sole decisive factor. One or more parties may be classed as ‘operator’, but the AI Report envis-ages that only the operator with the most control over the “risks of the operation” should be subject to strict liability. Since the AI Report’s recommendations include that producer liability for defective products should remain, the intention appears to be that this new form of liability should exist in addition to rather than as a replacement for other forms of liability. If these recommendations are introduced, a claimant injured by an emerging digital technology will be potentially able to pursue claims on alternative bases in negligence, contract, against the producer in a product liability claim or on a strict liability basis against the operator.

25. The basis upon which emerging technologies would or would not be categorised as products or systems that are subject to this new form of strict liability is unclear. Any legislation introducing such liability would need to define the category. There would need to be some clear way of identifying and treating specific new technologies as a separate category subject to strict operator liability, so that it is clear in principle why they should be treated differ-ently to other complex products and systems which remain subject only to existing forms of liability.

26. The AI Report in some respects builds upon the more expan-sive work of the Commission in SWD(2018) 137 on digital liability. The annex to SWD(2018) 137 discusses the charac-teristics of emerging digital technologies (Internet of Things and AI) as including high complexity and autonomous behaviour, being data-driven, and ‘openness’, meaning that the digital dimension encompassing the tangible and intan-gible (software and data) can interact to create new tangible products or deliver services. SWD(2018) 137 describes with approval the introduction of strict ‘operator’ liability rules to cover the use of drones and it seems a small step to extend such rules to autonomous vehicles and other AI technolo-gies. However, an important difficulty is likely to be the challenge in differentiating emerging digital technologies subject to any new rules on liability from other products.

Strict product liability

27. The AI Report supports the application of strict product liability regimes to AI and emerging digital technologies.

28. The authors suggest that the Product Liability Directive already achieves some of the elements viewed as necessary in terms of a liability regime applicable to emerging digital technologies. However, as indicated above, the Report identifies a number of areas where it concludes that the Product Liability Directive is problematic, notably: a. The definitions of the key concepts of ‘product’,

‘producer’ and ‘defect’.b. The absence of a duty to monitor products once placed

on the market.c. The existence of a development risks defence.d. The type of damage that can be claimed.

Definitions29. The authors of the Report note that the Product Liability

Directive was designed in the context of traditional, physical products, which may not easily be applied to emerging digital

Recommendations of the AI Report21. The AI Report concludes that digitisation results in some

fundamental changes to the environment, some of which impact liability systems. While the authors conclude that liability laws may cope, in principle, with new tech-nologies, they suggest that there are some fundamental changes which require the introduction of “certain adjust-ments” to EU and national liability regimes. They recog-nise however that, given the diversity of emerging digital technologies and the correspondingly diverse range of risks that may be associated with these, it is impossible to propose a single solution capable of addressing all issues.

22. The AI Report therefore makes certain key recommenda-tions in relation to operator and producer liability including:a. Systems of fault-based liability and strict liability

should continue to co-exist, for harm caused by emerging digital technologies; contractual claims may also exist in some scenarios.

b. For the purposes of liability, it is not necessary to give autonomous systems legal personality.

c. Strict liability should apply to producers and also to a new category of legal person who is liable for the harm caused by an AI or autonomous system, namely the ‘operator’.

d. Liability for harm caused by defective products should apply regardless of whether the product is digital or physical.

e. Operators should be subject to “an adapted range of duties of care” with regard to choosing the right system, and monitoring and maintaining the system.

f. Producers, whether they are also operators or not, should be required to design their products in such a way as to enable operators to discharge their duties.

g. Emerging digital technologies should be designed so that the device/system’s activity is logged and absence of logging information should lead to a reversal of the burden of proof.

h. Where it is difficult to establish defect in relation to a particular technology, this should be made easier, e.g. by reversing the burden of proof.

i. There should also be a reversal of the burden of proof where there is a breach of relevant safety rules (including those relating to cybersecurity) aimed at preventing the type of harm which has eventuated.

j. Where there is particular difficulty in establishing causa-tion, this should, in certain circumstances, be made easier.

k. If it is established that the product/service has caused harm, but fault still needs to be proved, then the burden of proving fault should be reversed if the cost/difficulty would be ‘disproportionate’ for the party suffering loss.

l. Existing defences should be reconsidered and possibly removed, including, in the product liability context, the development risks defence.

m. The AI Report suggests that compulsory insurance may be suitable for some emerging technologies, but acknowledges there may be difficulties with this.

n. The definition of damage should be expanded to include harm to, or loss of, data.

o. Comparable risks should be addressed by similar liability regimes and differences between liability regimes should be eliminated.

23. Some of these recommendations are discussed further in the following paragraphs.

Proposed strict ‘operator liability’

24. The AI Report introduces the concept of operator liability




Fault liability and duty to monitor

34. The AI Report recognises that existing liability systems often impose such duties, and suggests that supply and use of emerging digital technologies should result in specific duties of care for the purposes of potential fault liability. The authors of the Report recommend that operators of such technologies should be under specific duties to choose the correct system for the right task and skills, to monitor the system and to maintain the system. In addi-tion, the Report proposes that producers should be under duties to design, describe and market products so as to allow operators to comply with the duties described above and should also be under a duty to monitor such products after putting them into circulation.

Logging by design

35. An important recommendation of the AI Report, and plainly seen as necessary in the context of highly complex products where the existence and causation of errors is unlikely to be transparent, is the introduction of a duty, imposed on producers, to equip such technologies with the means of recording information about its operation (described as logging by design). Any data logging would have to be consistent with requirements for protection of personal data and rules concerning the protection of trade secrets.

36. The Report suggests that, where logged information is not available or is not provided to the victim, there should be a rebuttable presumption that the aspect of liability to be established by the missing information is fulfilled. If, as a result of this presumption, the operator is required to compensate the victim, the operator would have a right of recourse against the producer for failing to equip the tech-nology with logging facilities.

Easing of burden of proof in relation to causation

37. EU case law currently provides no support for reversing the burden of proof in product liability matters. In Case C-621/15, N. W and Others v Sanofi Pasteur MSD SNC and Others, the court held that the Product Liability Directive permitted factual findings under “national evidentiary rules” which lead to a finding of liability on the basis of evidence that is “serious, specific and consistent” even if it is supported by clear scientific evidence. However, it cautioned that: “National courts must…ensure that their specific application of those evidentiary rules does not result in the burden of proof introduced by Article 4 being disregarded or the effectiveness of the system of liability introduced by that directive being undermined.” In other words, there is considerable leeway for national evidentiary rules to ease the burden of proof on the claimant, but there cannot be a formal reversal of the burden of proof.

38. As the AI Report notes, the standard of proof varies between national legal systems, although national courts have, in all jurisdictions considered, found ways to alle-viate the burden of proving causation, where the claim-ant’s position is deemed to be weaker than in typical cases. The Report therefore recognises that introducing meas-ures which cut across national law provisions and impose, for example, a reversal of the burden of proof would (aside from affecting the existing product liability regime) conse-quently also have variable impacts and risk interfering with national procedure rules.

technologies, which frequently involve a constant interac-tion between products and services with no clear delinea-tion between the two elements. Furthermore, it has long been recognised that the status of software as a ‘product’ within the scope of the Directive is uncertain, particularly given differences between embedded and non-embedded software, including over-the-air software updates and other data feeds. The Report therefore describes the current definition of ‘product’ under the Directive as “an inade-quate match for the potential risks of emerging digital technologies” citing a general statement to that effect in the Commission’s Fifth Report and suggests that strict liability should apply to defective products and their components, whether these take a tangible or a digital form.

30. The Report also suggests that current definitions of ‘producer’ may not reflect the reality of emerging digital technologies where many different actors may contribute to design, func-tioning and use. Particular difficulties may emerge where updates or data feeds are provided from outside the EEA and there is no traditional producer or importer within the EEA for the purposes of claims by the victim.

31. In terms of ‘defect’, the Report suggests that the intercon-nectivity of products and systems involved in emerging digital technologies may make it difficult to identify defects. Further, the authors point out the difficulties associated with the application of general principles of defect to situations such as an unpredictable deviation from the decision-making path of an autonomous system with self-learning capabilities. Finally, the Report states that the complexity inherent in many emerging digital technology systems creates problems for consumers seeking to discover and prove the existence of a defect.

Other key elements32. The AI Report refers to the fact that the Product Liability

Directive focuses on the time when the product is placed into circulation and suggests that this excludes the possi-bility of claims arising from subsequent additions or modi-fications by the producer. To the extent that any update or upgrade itself constitutes a product with a new time of placing into circulation, this concern does not appear to require any change to the existing regime. The authors of the Report suggest that strict liability imposed on the producer should be supplemented by fault-based liability for failure to warn (see below).

33. The development risks defence remains a controversial aspect of the product liability regime and is raised at every review of the Product Liability Directive. In the context of emerging digital technologies, the AI Report notes that a defect in digital content of a product may result from the impact of the environment in which the product operates or from the product’s evolution, for which the manufac-turer only created a general framework, but did not design in detail. The Report suggests that, in view of the need to share benefits and risks efficiently and fairly, “the development risks defence, which allows the producer to avoid liability for unfore-seeable defects, should not be available in cases where it was predict-able that unforeseen developments might occur”. The Report does not analyse the relevance of the development risks defence in the context of supporting innovation and such a change in the general approach would involve re-opening the settlement of ‘fair apportionment of risk’ agreed when the Product Liability Directive was adopted. This seems likely to require greater consideration of specific unfairness asso-ciated with the current regime in the context of emerging technologies and the importance of the development risks defence in allowing access to technological advances.


5Arnold & Porter


Conclusions43. The AI Report constitutes an interesting review of the

possible shortcomings of existing liability systems and makes proposals for revision. However, as recognised in the Report, national liability systems have largely proven adaptable and adequate to the challenges of developing technology and it seems likely that further research or experience is needed in order to understand the extent to which change is in fact needed.

44. The Product Liability Directive is not the sole focus of the review, and consequently is addressed with rather less detail than in SWD(2018) 137 on which the review builds. While the AI Report identifies some issues, it does not make the case for a wholesale replacement of the product liability regime. It remains to be seen what guidance or legisla-tive recommendations emerge from the Product Liability Formation of the Expert Working Group, although clar-ification of the definitions of product and property, to clarify the status of software and data, seems likely to be recommended.

45. While the EU has considerable influence, it is not certain that, given the relative openness and global connectedness of Member States’ digital economies, the same degree of control can be easily exercised over the digital economy as for physical products placed on the market. It is likely that existing large technology providers will already have an EU presence and thus be amenable to regulation to some degree. However, it may be possible for other emerging global providers to make their emerging digital technol-ogies available online only without subjecting themselves to EU regulation or systems of redress. This is perhaps not a large risk with, e.g., US technology companies which are based in a jurisdiction with well-established redress mechanisms, but it is imaginable that future digital tech-nologies may originate in and be sold online directly from other jurisdictions with regimes much less accessible to an injured EU claimant.

46. Ultimately, it is likely to prove to be the case that either AI/new technologies are substantially analogous to other new technologies that are adequately managed by existing regu-latory and liability systems, or they represent such a major change that they will alter life fundamentally and bring with them large, unknowable, future risks and challenges. If the former is true, then it is likely that the risks can be managed through relatively minor adaptations to existing systems such as the regime governed by the Product Liability Directive. If the latter is true, then arguably the use and introduction of the new technologies ought to be closely monitored and regulated to manage exposure to those large, unknowable risks. A balanced approach to regulation may be necessary however, as the imposi-tion of excessively tight controls may permit technolog-ical advancement at a faster rate in other global regions and leave EU businesses and consumers in a position where they have to accommodate a series of new tech ‘utilities’ which they need or want, but which are geographically and/or for reasons of scale substantially outside the regu-latory control of the EU/Member States.

39. However in view of the Report’s conclusion that victims may be in a weaker position to establish causation arising from liability in the context of emerging digital technolo-gies than in other cases, where the events leading to the harm may be more readily analysed, it recommends allevia-tion of the burden of proving causation in light of the chal-lenges of emerging digital technologies where a balance of various factors warrants such action. The factors proposed by the authors of the Report in this context are: the likelihood that the technology at least contributed to the harm; the likelihood that the harm was either caused by the technology or by some other cause within its sphere; the risk of a known defect within the technology (even if its actual causal impact is unclear); and the availability/non-availability and comprehensibility of information regarding processes within the technology that may have contributed to the cause.

Compulsory insurance

40. The AI Report suggests that compulsory insurance may be appropriate for some emerging technologies, but recog-nises that there are difficulties, including that insurance may not be commercially available where, as is more likely to be the case with new technologies, it is not readily possible for insurers to price the risks. Further, if the cost of insurance is too high, it may, by being passed on to the consumer of the technology, make the technology more expensive and discourage innovation. In a global marketplace, extra costs for compulsory liability insurance may put technology that is developed with such consider-ations in mind at a competitive disadvantage relative to technology developed in jurisdictions where there is no requirement for insurance.

41. The AI Report recommends caution and careful analysis of whether compulsory insurance is really needed. It notes that, in particular where true strict liability applies, risks can be rendered insurable by capping them. A cap on risks, whether in terms of the amount of compensation payable and/or, via limitation, on the time for bringing a claim, can make it possible to insure. Without capping, new tech-nologies, particularly ones like AI which involve machine learning and therefore perhaps a lower degree of transpar-ency and predictability, may represent open-ended risks that are hard to price, meaning that insurance, if available at all, may be prohibitively expensive.

42. Aside from motor vehicle third party risk insurance, there is no EU-wide compulsory insurance applicable to other complex and innovative products, such as new medicines and medical devices, though insurance may be required for producers in some jurisdictions (e.g. for medical device manufacturers in France). While there have been recom-mendations for compulsory insurance of some prod-ucts, and legislation proposed or introduced at national level (e.g. for self-driving cars), it is not certain on the basis of the AI Report’s analysis that there yet exists a clearly-defined category of emerging digital technologies in respect of which it would be appropriate to introduce compulsory insurance.




Adela Williams is a partner in the London office of Arnold & Porter, specialising in product liability litigation (unitary actions and group litiga-tion), principally involving life sciences clients and including claims involving unlicensed medical products in the research context as well as marketed products. Such litigation has often involved co-ordinating proceedings within the EU and advising on forum and other jurisdictional issues. Past cases include the fetal anticonvulsant litigation and the successful defence of group litigation involving more than 100 claims relating to the “third generation” oral contraceptive pill on behalf of two of the defendant manufacturers. She also advises clients in relation to the regulation of medicinal products, medical devices, foods and cosmetics in the EU and acts on their behalf in litigation arising from the decisions of regulatory bodies.She is an Assistant Coroner.

Arnold & PorterTower 42, 25 Old Broad StreetLondon EC2N 1HQUnited Kingdom

Tel: +44 20 7786 6115Fax: +44 20 7786 6299Email: [email protected]: www.arnoldporter.com

Arnold & Porter is an international law firm with over 1,000 attorneys in 16 offices in the US, London, Brussels, Frankfurt, Shanghai and Seoul. With 40 partners and counsel specialising in product liability matters, the firm is one of the most experienced firms internationally, providing clients with an integrated product liability service on a transatlantic basis.The European product liability group is a recognised leader in the UK and Europe, with comprehensive experience in handling the defence of claims. Its lawyers have been at the forefront of “group action” litigation, with expe-rience derived from the successful defence of many major multi-claimant cases that have been brought in the UK and elsewhere in the EU over the last 30 years. In the US, the firm has acted both as national counsel for companies and as trial counsel in cases involving personal injury and prop-erty damage claims.

Tom Fox is a counsel in the London office of Arnold & Porter, whose practice focuses on litigation and general product safety regulatory work. His main litigation practice concerns the defence of product liability claims on behalf of medical device and pharmaceutical companies. He also has considerable experience of commercial litigation and personal injury. He also has experience in bringing judicial review actions based on public and administrative law on behalf of pharmaceutical companies, both at the Court of Justice of the European Union and in national courts. He advises on general product safety and regulatory issues such as conformity marking, labelling, and compliance with standards in relation to chemicals and a range of consumer products including electrical and electronic goods, clothing, cosmetics and toys.

Arnold & PorterTower 42, 25 Old Broad StreetLondon EC2N 1HQUnited Kingdom

Tel: +44 20 7786 6187Fax: +44 20 7786 6299Email: [email protected]: www.arnoldporter.com

Please contact Ian Dodds-Smith, Dr. Adela Williams or Tom Fox in the London Office for UK or EU product liability enquiries, and Anand Agneshwar in the New York Office for US enquiries.

www.arnoldporter.com


Chapter 2 7

U.S. Product Liability Law: Recent Developments, Trends, and Future Outlook

Sidley Austin LLP Teri H. Peeples

Daniel A. Spira


found that a corporation is typically only “at home” in its “place of incorporation and principal place of business”. 134 S. Ct. 746, 760 (2014). Only in an “exceptional” case will “a corporation’s operations in a forum other than its formal place of incorpora-tion or principal place of business … be so substantial and of such a nature as to render the corporation at home in that state”. Id. at 761-62 & n.19 (declining to find general jurisdiction over Daimler in California despite its subsidiary’s sales of $4.6 billion in the state) (citing Burger King Corp. v. Rudzewicz, 471 U.S. 462, 472 (1985)).

In BNSF Ry. Co. v. Tyrrell, the Supreme Court further described the narrow circumstances under which general jurisdiction over a defendant may be found. 137 S. Ct. 1549, 1554, 1559 (2017). The Court held that there was no general personal jurisdiction over BNSF in Montana, when BNSF was not incorporated in Montana and did not maintain its principal place of business there, even though the company had over 2,000 miles of rail-road track and more than 2,000 employees in the state. Id. at 1559. These contacts were not substantial enough for general jurisdiction, because they represented only a small portion (less than 10 per cent) of BNSF’s total presence in the United States. Id. at 1554, 1559.

These cases are of significance to defendants in product liability litigation, where hundreds or thousands of plaintiffs may individually sue the same defendant manufacturer over the same product or device. Narrowing the venues in which these defendants are considered “at home” can critically curtail plaintiffs’ efforts to “forum shop” in seemingly favourable juris-dictions. In a recent attempt to dull the impact of Daimler and BNSF, plaintiffs have argued that defendants “consented” to jurisdiction anywhere they are registered to do business. See, e.g., Bors v. Johnson & Johnson, 208 F. Supp. 3d 647 (E.D. Pa. 2016) (“[B]ecause [defendant] was authorized to do business in Pennsylvania, it was subject to the exercise of personal jurisdic-tion by Pennsylvania courts”); see also Williams v. Takeda Pharms. America, Inc., No. 18-4774, 2019 WL 2615947, at *3 (E.D. Pa. June 26, 2019) (“Without any further analysis from the Supreme Court or the Third Circuit, a foreign corporation’s registration to do business in Pennsylvania establishes consent to personal jurisdiction”). Several courts, however, have recently declined to embrace this argument. See, e.g., Waite v. All Acquisition Corp., 901 F.3d 1307, 1318-22 (11th Cir. 2018) (registration to do business in Florida did not subject the corporate defendant in a product liability suit to jurisdiction in the state), cert. denied, 139 S.Ct. 1384 (2019); Sullivan v. A.W. Chesterton, Inc. (In re Asbestos Prod. Liab. Litig.), 384 F. Supp. 3d 532, 545 (E.D. Pa. 2019) (“Pa. Statutory scheme requiring foreign corporations to register to do busi-ness and, therefore, to consent to general personal jurisdiction in Pennsylvania, offends the Due Process Clause and is uncon-stitutional”); Aspen Am. Ins. Co. v. Interstate Warehousing, Inc., 90 N.E.3d 440, at 447-48 (Ill. 2017) (“that a foreign corporation has

IntroductionU.S. product liability law has continued to evolve over the past year. Courts have arguably narrowed the scope of federal personal jurisdiction, and have continued to interpret the appli-cability of federal preemption to tort claims addressing phar-maceutical products. Additionally, given the prevalence of multi-district litigation (“MDL”) that continues to encompass a substantial portion of civil cases in federal court – especially those involving product liability claims – problems pertaining to MDLs have garnered significant attention. Proposed MDL rule changes, if implemented, would likely alter the product liability landscape. Finally, as the design, manufacturing, and function of traditional products evolve through the advent of new tech-nologies, courts, regulators, practitioners, and companies are being forced to re-examine, develop, and adapt product liability law.

This chapter summarises and provides updates on each of the following topics pertaining to U.S. product liability law:■ PersonalJurisdiction;■ FederalPreemption;■ MDLs;and■ EmergingAreasofProductLiabilityLaw.

Personal JurisdictionThe constitutional requirement of personal jurisdiction protects defendants from being sued in jurisdictions in which they do not have certain minimum contacts. See Int’l Shoe Co. v. Wash., 326 U.S. 310, 316 (1945). For a court to exercise jurisdiction, the defendant must also have purposefully availed itself of the priv-ileges of conducting activities within the forum. See Walden v. Fiore, 571 S. Ct. 1115, 1122 (2014).

There are two different types of personal jurisdiction, without either of which a case must be dismissed. “General jurisdic-tion” exists when a defendant has such substantial contacts with a forum that it is essentially “at home” in the forum and can be subject to any claim there, regardless of whether the lawsuit relates to the forum. See Goodyear Dunlop Tires Operations, S.A. v. Brown, 564 U.S. 915, 924-25 (2011). By contrast, “specific juris-diction” can be invoked even when a defendant has few contacts with the forum, as long as the claim arises directly out of those contacts. Id. at 414 & n.8.

In recent years, case law has narrowed the scope of general and specific jurisdiction, with significant implications for product liability suits and other cases.

General Jurisdiction

The U.S. Supreme Court clarified the scope of general jurisdic-tion in Daimler Ag v. Bauman, a landmark case in which the Court


8 U.S. Product Liability Law


Liab. Litig., No. 09-2047, 2017 WL 5971622, at *12 (E.D. La. Nov. 30, 2017) (“[BMS] was not a class action …. [T]his factor materially distinguishes this action from Bristol-Myers because in class actions, the citizenship of the unnamed plaintiffs is not taken into account for personal jurisdiction purposes”) (internal quotations omitted); Cabrera v. Bayer Healthcare, LLC, No. CV 17-08525, 2019 WL 1146828, at *7-8 (C.D. Cal. Mar. 6, 2019) (discussing split of authority and holding that “decisions concluding that Bristol-Myers does not apply in the class action context are more persuasive”).

BMS may also serve to limit specific jurisdiction where plain-tiffs’ alleged injuries only occurred in the jurisdiction as a result of a product traveling there through the “stream of commerce”. Under the “stream of commerce” theory, plaintiffs allege that minimum contact with the forum is established merely because defendant’s product ultimately reaches a state and the product allegedly causes injury in that state. Recently, the Supreme Court granted certiorari regarding reliance on a stream of commerce theory to establish specific jurisdiction in automobile cases. See Bandemer v. Ford Motor Co., 931 N.W.2d 744 (Minn. 2019), cert. granted, No. 19-369, 2020 WL 254152 ( Jan. 17, 2020) and Ford Motor Co. v. Montana Eighth Judicial Dist. Court, 443 P.3d 407, cert. granted, No. 19-368, 2020 WL 254155 ( Jan. 17, 2020). Because the automobiles at issue in those cases were not purchased, manu-factured or designed by defendants in the state where the inju-ries occurred, the question for the Court is whether the defen-dant nevertheless has sufficient contacts with the forum under a “stream of commerce” theory to confer specific jurisdiction over the matter. The Supreme Court’s approach to this question could have a significant impact on the jurisdictions available for plaintiffs to bring suits against broad classes of manufacturers whose products are placed in the stream of commerce and ulti-mately are widely distributed by third parties.

Federal Preemption Where state law conflicts with federal law, state law is preempted under the Supremacy Clause of the U.S. Constitution. U.S. Const. art. VI, cl. 2. Preemption may be expressed in an explicit provision of federal law or implied in the structure and scope of the federal regulatory scheme. See, e.g., Buckman Co. v. Plaintiffs’ Legal Comm., 531 U.S. 341 (2001). Preemption with respect to pharmaceutical products and medical devices approved by the U.S. Food and Drug Administration (“FDA”) has received considerable attention from courts in recent years and remains one of the most hotly contested issues in product liability cases.

Pharmaceutical “Impossibility” Preemption

In PLIVA, Inc. v. Mensing, 564 U.S. 604 (2011), a case with signif-icant implications for pharmaceutical preemption, plaintiffs alleged that manufacturers of a generic drug failed to adequately warn of the risk of a neurological disorder, as required under state law. The manufacturers argued that the claims were preempted because federal law requires generic medications to carry warn-ings identical to their brand-name equivalents, making compli-ance with both federal law and the alleged duty under state law impossible (a recognised ground for implied preemption). Id. at 610; see also id. at 617-18. (“Where state and federal law directly conflict, state law must give way”, and “state and federal law conflict where it is impossible for a private party to comply with both state and federal requirements”.)

Plaintiffs argued that compliance was not impossible, because the manufacturers could have independently modified the warnings by: (1) using FDA’s changes-being-effected (“CBE”) process; (2) sending Dear Doctor letters to physicians; or (3) proposing stronger warnings to the FDA. Id. at 614-16. The

registered to do business under the Act does not mean that the corporation has thereby consented to general jurisdiction over all causes of action, including those that are completely unre-lated to the corporation’s activities in Illinois”). This theory will likely see continued litigation across the country as litigants attempt to test the limitations put in place by the Supreme Court.

Specific Jurisdiction

The U.S. Supreme Court’s 2017 decision in Bristol-Myers Squibb Co. v. Superior Court of Calif. (“BMS”) significantly limited the forums in which specific jurisdiction may be invoked to those connected to a particular plaintiff’s claims. BMS, a pharma-ceutical company, was sued in a product liability lawsuit in California state court by a group of plaintiffs, 80 per cent of whom were out-of-state residents. 137 S. Ct. 1773, 1778 (2017). Where plaintiffs lived, purchased, or were prescribed the drug in California, the court had specific jurisdiction over BMS for inju-ries that arose out of BMS’s conduct in that state, and plaintiffs were permitted to bring an action against BMS in California. Id. at 1779. But non-resident plaintiffs whose claims had no relation to California were not allowed to piggyback onto the specific jurisdiction over the resident plaintiffs’ claims and assert their own claims against BMS in California. Id. at 1782.

Courts across the country have consistently applied BMS to limit forum shopping by out-of-state plaintiffs asserting product liability claims. See, e.g., Jordan v. Bayer Corp., 4:17-cv-00865, 2018 WL 837700, at *4 (E.D. Mo. Feb. 13, 2018) (granting motion to dismiss because non-Missouri plaintiffs’ “allegations are simply too attenuated to serve as a basis for specific personal jurisdic-tion over Bayer”); Hinton v. Bayer Corp., No. 4:16-cv-1679, 2018 WL 3725776, at *4 (E.D. Mo. July 27, 2018) (similar); Timpone v. Ethicon, No. 4:19 CV 821 RWS, 2019 WL 2525780, at *2 (E.D. Mo. June 19, 2019) (also similar); In re Zostavax (Zoster Vaccine Live) Prods. Liab. Litig., 358 F. Supp. 3d 418, 423-424 (E.D. Pa. Jan. 7, 2019) (Defendant’s in-state activities not involving the alleged injury-causing product were not jurisdictional contacts); Campbell v. Acme Insulations, Inc., 2018 IL App (1st) 173051, 105 N.E.3d 984, 995 (2018) (In BMS “the United States Supreme Court rejected the notion that specific jurisdiction could be asserted under a ‘sliding scale’ theory”; rather, “for purposes of specific personal jurisdiction, there must be a connection between the forum and the specific claims at issue”) (internal quotations omitted).

In the context of class actions, some courts have similarly recognised that BMS requires dismissal of non-resident putative class members’ claims that have no connection to the forum. E.g., Mussat v. IQVIA Inc., No. 17 C 8841, 2018 WL 5311903, at *5 (N.D. Ill. Oct. 26, 2018) (“Following the Supreme Court’s lead in [BMS] and applying its core reasoning here, due process … requires a connection between the forum and the specific claims at issue. This recognition bars nationwide class actions in fora where the defendant is not subject to general jurisdiction. Whether it be an individual, mass, or class action, the defend-ant’s rights should remain constant”); DeBernardis v. NBTY, Inc., No. 17 C 6125, 2018 WL 461228, at *2 (N.D. Ill. Jan 18, 2018) (“The Court believes that it is more likely than not … that the courts will apply [BMS] to outlaw nationwide class actions in a for[u]m, such as in this case, where there is no general juris-diction over the Defendants”); Chavira v. OS Restaurant Services, LLC, No. 18-cv-10029-ADB, 2019 WL 4769101, at *6 (D. Mass. Sept. 30, 2019) (motion to strike non-resident class action allega-tions was granted, noting that BMS applies to all cases, including class actions). Other courts have maintained that BMS is inap-plicable to proposed class members’ claims, resulting in a split of authority. See, e.g., In re: Chinese-Manufactured Drywall Prod.


9Sidley Austin LLP


ensuring that courts will continue to grapple with the “impos-sibility” of drug manufacturers complying with both state law duties to warn and federal limitations on their ability to do so.

Several noteworthy opinions following Mensing and Bartlett have also addressed whether state law design defect claims are preempted, where defendants could not, pursuant to federal law, unilaterally make “major changes” to pharmaceutical drug designs without FDA approval. See, e.g., Yates v. Ortho-McNeil-Janssen Pharms., Inc., 808 F.3d 281, 298-300 (6th Cir. 2015) (claims were preempted because “once a drug, whether generic or brand-name, is approved [by the FDA], the manufacturer is prohibited [by federal law] from making any major changes to the qualitative or quantitative formulation of the drug product …”, and plain-tiff’s additional argument that the defendant could have utilised a different design “in the first instance” before obtaining FDA approval was similarly preempted, in part because “Defendants could not have complied with whatever pre-approval duty might exist without ultimately seeking the FDA’s approval prior to marketing [the product], and certainly prior to [plaintiff’s] use of the drug”); Gustavsen v. Alcon Labs., Inc., 903 F.3d 1, 10 (1st Cir. 2018) (“federal law preempts plaintiffs’ [design defect] cause of action because defendants cannot lawfully make such a [design] change without prior FDA approval”); Drescher, 2020 WL 699878, at *8 (“Because it would constitute a major change under the [federal] regulations, Defendants were not able to alter their [pharmaceutical formulations] without obtaining prior FDA approval”, and plaintiff’s design defect claim was there-fore preempted); Thomas v. Bracco Diagnostics Inc., 3:19-cv-00493, 2020 WL 1016273, at *9 (W.D. La. Feb. 27, 2020) (Defendant “could not have independently adopted an alternative design for [its pharmaceutical product] after the FDA approved the drug because federal drug regulations prohibit [defendant] from altering its composition. Accordingly, any state require-ment that [defendant] should have adopted an alternative design after FDA approval is preempted.”). But see, e.g., Guidry v. Janssen Pharms., Inc., 206 F. Supp. 3d 1187, 1206-08 (E.D. La. 2016) (rejecting “the Sixth Circuit’s reasoning in Yates concerning preemption in the pre-FDA approval context”, because “[f ]ederal law does not prevent a drug manufacturer from complying with this state-imposed duty [to consider feasible, alternative designs] before seeking FDA approval”); Paulsen v. Abbott Labs., 368 F. Supp. 3d 1152, 1173 (N.D. Ill. 2019) (whether design defect claims are preempted under Bartlett is a question that “has divided federal courts across the country”).

Buckman Preemption

In Buckman Co. v. Plaintiffs’ Legal Comm., the U.S. Supreme Court addressed claims that plaintiffs suffered injuries from the use of orthopaedic bone screws, and that the manufacturer of the device and its consultant “made fraudulent representations to the [FDA] in the course of obtaining approval to market the screws”. 531 U.S. 341, 343 (2001). The Court held that “plain-tiffs’ state-law fraud-on-the-FDA claims conflict with, and are therefore impliedly preempted by, federal law. The conflict stems from the fact that the federal statutory scheme amply empowers the FDA to punish and deter fraud against the Administration, and that this authority is used by the Administration to achieve a somewhat delicate balance of statutory objectives. The balance sought by the Administration can be skewed by allowing fraud-on-the-FDA claims under state tort law”. Id. at 348. Almost 20 years later, courts continue to apply Buckman with varying inter-pretations and results.

For instance, in 2013, the Ninth Circuit held that federal law did not impliedly preempt Arizona state law failure-to-warn claims predicated on a medical device manufacturer’s alleged failure to

Court rejected the first two arguments, noting that a generic drug maker cannot unilaterally strengthen its labelling through the CBE process or disseminate a Dear Doctor letter. Id. at 614-15. As to the third basis, the Court held that it was not clear that the FDA would have permitted a new warning. Id. at 619-20. Plaintiffs’ claims were therefore preempted: “[W]hen a party cannot satisfy its state duties without the Federal Government’s special permission and assistance, which is dependent on the exercise of judgment by a federal agency, that party cannot inde-pendently satisfy those state duties for pre-emption purposes.” Id. at 623–24.1

In Mut. Pharm. Co. v. Bartlett, 570 U.S. 472 (2013), the Supreme Court found that state design-defect claims against the manu-facturer of a generic drug were impliedly preempted. Similar to Mensing, a defendant manufacturer argued in Bartlett that it was impossible to comply with both its alleged state law duty to strengthen the warnings for its drug, and its federal law duty not to alter its approved labelling. Although the Court of Appeals held that plaintiff’s claims were not preempted, as defendant “could escape the impossibility of complying with both its federal- and state-law duties by choosing not to make [the drug] at all”, the Supreme Court rejected that reasoning, holding that “an actor seeking to satisfy both his federal- and state-law obli-gations is not required to cease acting altogether in order to avoid liability”. Id. at 488.

Notwithstanding the availability of the CBE process to manufacturers of brand-name pharmaceuticals (unlike the generic drug manufacturers in Mensing and Bartlett), brand-name manu-facturers have likewise succeeded in arguing that state law failure-to-warn claims are subject to “impossibility preemption”. Specifically, whereas it was “impossible” for the generic drug manufacturer in Mensing to comply with state law warning duties because federal law does not allow generic manufacturers to invoke the CBE process to supplement drug warnings, so too is it “impossible” for brand-name manufacturers to comply with state law warning duties where plaintiffs have not identified “newly required information” of the type that is required before federal law permits such manufacturers to invoke the CBE process. See, e.g., Sabol v. Bayer Healthcare Pharm, Inc., No. 18 Civ. 11169, 2020 WL 705170, *14 (S.D.N.Y. Feb. 12, 2020) (“Plaintiff does not sufficiently plead that . . . [defendant] had reasonable evidence of a causal association between [the pharmaceutical at issue] and a clinically significant adverse reaction in patients with normal kidney function. Thus, [plaintiff ] does not plead facts showing that [defendant] had or should have had newly acquired infor-mation permitting it to unilaterally add her claimed warning under the CBE regulations. Consequently, [plaintiff’s] claims against [defendant] are preempted”); Drescher v. Bracco Diagnostics Inc., No. CV-19-00096, 2020 WL 699878, at *4-6 (D. Ariz. Jan. 31, 2020) (similar); Pradaxa Cases, No. CJC-16-004863, 2019 WL 6043513, at *2 (Cal. Super. Ct. Nov. 8, 2019) (granting summary judgment for defendant on claim that warning label was inade-quate, because “between the inception of Pradaxa in the U.S.” and the “date of [plaintiff’s injury], there was no relevant ‘newly acquired information’ about Pradaxa that, under federal law, would permit [defendant] to change the Pradaxa label”). But see, e.g., Holley v. Gilead Sciences, Inc., 379 F. Supp. 3d 809, 826, 829 (N.D. Cal. May 10, 2019) (dismissing certain failure to warn theories for failure to sufficiently plead “newly acquired infor-mation”, but finding impossibility preemption inapplicable to other failure to warn theories alleging that defendant “should have submitted different warnings” before FDA approval of a drug, because there is “no federal law that would prevent a drug manufacturer from submitting a different warning label to the FDA prior to initial approval”). In Merck Sharp v. Dohme Corp., 139 S. Ct. 1668, 1676 (2019), the Supreme Court reaffirmed that “a judge, not the jury, must decide the pre-emption question”,




manufacturers has also received considerable attention in recent years. In 2008, the U.S. Supreme Court held that claims against manufacturers of Class III pre-market approved (“PMA”) devices are expressly preempted to the extent they would impose requirements “different from, or in addition to the requirements imposed by federal law”. Riegel v. Medtronic, Inc., 552 U.S. 312, 330 (2008) (internal quotation omitted). This ruling shields manufacturers from tort liability for most claims related to PMA devices, which are subject to the most rigorous FDA review. Id. at 318-20. Riegel left open, however, the possibility that plaintiffs could plead viable “parallel” state law claims, where PMA-approved medical devices deviate from federally-imposed, device-specific requirements, in violation of both federal and state law. Id. at 330.

Numerous courts have rejected plaintiffs’ attempts to circum-vent Riegel. See, e.g., Lawrence v. Medtronic, 791 F. App’x 679, 680 (9th Cir. 2020) (affirming dismissal of state law claims, where plain-tiff “identifies no federal authority that requires medical-device manufacturers to send removed medical devices anywhere for evaluation”, such that “a tort claim premised on such a course of conduct would impose a requirement that is ‘different from’ and ‘in addition to’ what is required under federal law”); id. (affirming dismissal of “product-defect theories of recovery”, because plaintiff “fail[s] to identify any specific federal require-ment that was violated or the specific nature of the [product’s] purported defects”); Green v. Medtronic, Inc., No. 1:19-cv-3242, 2019 WL 7631397, at *4 (N.D. Ga. Dec. 31, 2019) (dismissing manu-facturing defect claim as expressly preempted, because “Plaintiff does not identify the specific federal regulations or statutes that the Defendants purportedly violated”); Walls v. Medtronic, Inc., No. 19-3690, 2019 WL 6839942, at *4 (E.D. Pa. Dec. 16, 2019) (dismissing claims that the PMA medical device was defective, lacked adequate instructions and warnings, and was negligently designed, because plaintiffs “have not identified any violations of specific FDA regulations”). Other courts, however, have diverged from these analyses and rejected arguments that claims regarding PMA devices are expressly preempted under Riegel. See, e.g., Bull v. St. Jude Med., Inc., No. 17-1141, 2018 WL 3397544, at *8 (E.D. Pa. July 12, 2018) (“Plaintiff’s state law failure to warn claim identified a state duty to warn physicians of risks inherent in its medical devices that is parallel to St. Jude’s duty to comply with MDR reporting requirements …. As such, it parallels these federal requirements, and is not expressly pre-empted”).

MDL Trends The nature of product liability litigation can lead to a substantial volume of individual claims in different courts, each alleging, for instance, similar injuries arising from exposure to the same pharmaceutical product or medical device. Handling these cases on an individual basis can become unwieldy and expensive. As a result, plaintiffs and/or defendants may support the centralisa-tion or coordination of litigation before one judge in one court. 28 U.S.C. § 1407 provides one mechanism for doing so, allowing lawsuits “involving one or more common questions of fact” to be coordinated or consolidated in one federal district court for pre-trial proceedings, called an MDL. The general purposes of an MDL are “to avoid duplication of discovery, to prevent inconsistent pretrial rulings, and to conserve the resources of the parties, their counsel and the judiciary”. U.S. Judicial Panel on Multidistrict Litigation, Overview of Panel, https://www.jpml.uscourts.gov/overview-panel-0.

As of year-end 2019, there were 190 active MDLs. See https://www.jpml.uscourts.gov/sites/jpml/files/JPML_Calendar_Year_Statistics-2019_1.pdf. By some measures, MDLs now contain more than half of all civil cases pending in federal court, see https://www.law360.com/classaction/articles/1138928/mdls-surge-to-majority-of-entire-federal-civil-caseload, although

“report to the FDA any complaints about the product’s perfor-mance”. Stengel v. Medtronic Inc., 704 F.3d 1224, 1232 (9th Cir. 2013) (en banc). The court distinguished Buckman on the ground that the plaintiff’s “claim specifically alleges, as a violation of Arizona law, a failure to warn the FDA”, id. at 1233, whereas “the plain-tiffs in Buckman alleged no state-law claim and were concerned exclusively with alleged fraud on the FDA that had occurred as part of that approval process”. Id. at 1230. In other words, unlike in Buckman, the plaintiff in Stengel asserted “a state-law duty that paralleled a federal-law duty …” Id. at 1232. See also, e.g., Bledsoe v. Medtronic, Inc., No. 2:18-cv-133, 2020 WL 43107, at *7 n.4 (N.D. Ind. Jan. 3, 2020) (narrowly interpreting Buckman as only barring claims that assert “fraud on the agency”).

However, the Supreme Court of Arizona unanimously rejected the Ninth Circuit’s interpretation of Arizona law. Conklin v. Medtronic, Inc., 431 P.3d 571 (Ariz. 2018). Holding that failure to warn claims against the manufacturer of a medical device were impliedly preempted, the court explained that Stengel “was based on the unsupported premises that Arizona law contemplates a warning to a third party such as the FDA”, whereas “established law does not recognise a claim merely for failing to provide some-thing like adverse event reports … to a government agency that has no obligation to relay the information to the patient”. Id. at 579 (internal quotations omitted). “Absent an independent state law duty to submit adverse event reports to the FDA, [plaintiff’s] failure-to-warn claim, at bottom, is an attempt to enforce a federal law requirement”, and is therefore preempted. Id. at 578; see also, e.g., McNeil-Williams v. DePuy Orthopaedics, Inc., 384 F. Supp. 3d 570, 576-77 (E.D.N.C. 2019) (suggesting that North Carolina simi-larly “does not recognize an independent state law duty to make adverse event reports to the FDA”). The Ninth Circuit’s decision also departed from the holding of some other courts that allega-tions that a manufacturer “failed to provide the FDA with suffi-cient information and did not timely file adverse event reports, as required by federal regulations”, are “foreclosed by [the FDCA] as construed in Buckman”. In re Medtronic, Inc., Sprint Fidelis Leads Prods. Liab. Litig., 623 F.3d 1200, 1205–06 (8th Cir. 2010); see also, e.g., Tinkler v. Mentor Worldwide, LLC, 1:19-cv-23373, 2019 WL 7291239, at *4 (S.D. Fla. Dec. 30, 2019) (“Buckman’s holding—i.e., that fraud-on-the-FDA claims are impliedly preempted—extends to failure-to-warn claims where the plaintiff alleges the device’s warnings were inadequate because the defendant-manufacturer failed to provide sufficient information to the FDA”).

Courts have relied on Buckman to also hold that other tort claims against pharmaceutical and medical device manufacturers are impliedly preempted. See, e.g., Markland v. Insys Therapeutics, Inc., 758 F. App’x 777, 778-80 (11th Cir. 2018) (affirming dismissal of state law claims that the pharmaceutical manufacturer “engaged in a ‘fraudulent’ and ‘unlawful’ marketing scheme to push doctors to prescribe [the drug] ‘off label’”, because plaintiff “has not pointed to any traditional state-law duty owed by [defendant] to [plaintiff ] that was breached by the company’s marketing of [the product] for off-label use. It is only because of the FDCA and FDA enforce-ment decisions that the promotion of off-label uses is prohibited”, and, therefore, “[a]s with the Buckman plaintiffs, [plaintiff ] seeks to enforce a duty that ‘exists solely by virtue of the FDCA’”); Green v. Medtronic, Inc., 2019 WL 7631397, at *5 (N.D. Ga. Dec. 31, 2019) (“a claim for negligence per se is impliedly preempted insofar as it is premised on breaches of duties created by the FDCA”, because under Buckman “there is no private right of action for violations of the FDCA”) (internal quotations omitted).

Express Preemption of Claims Against Manufacturers of Certain Medical Devices

The express preemption of claims against medical device


11Sidley Austin LLP


potentially impacting the safety and effectiveness of the device. Threats and vulnerabilities cannot be eliminated, therefore, reducing cybersecurity risks is especially challenging. The heath care environment is complex, and manufacturers, hospitals, and facilities must work together to manage cybersecurity risks”. U.S. FDA, “Cybersecurity”, available at https://www.fda.gov/medical-devices/digital-health/cybersecurity.

In January 2019, the U.S. Consumer Product Safety Commission (“CPSC”) also released “A Framework of Safety for the Internet of Things”, intended “to provide an over-view of technology-neutral best practices to ensure consumer product safety in the design and deployment of devices, soft-ware and systems used with Internet-connected consumer prod-ucts”. U.S. CPSC, “A Framework of Safety for Internet of Things: Considerations for Consumer Product Safety”, avail-able at https://www.cpsc.gov/s3fs-public/A_Framework_for_Safety_Across_the_Internet_of_Things_1-31-2019_0.pdf?1KJ.t4Tn04v9OtEBr2s0wyLAP.KsuuQ3. The Commission advised that “[p]roduct designers need to be aware of the capabilities of every component of their final product and predict unin-tentional uses and intentional misuses that could lead to fore-seeable hazardous conditions. These best practices will help ensure that devices and components of devices are designed to prohibit unsafe system actions, command and control critical safety functions, and signal precursors to hazardous events as reliably as possible over the expected lifespan of the system.” Id.; see also CPSC, “Status Report on the Internet of Things (IoT) and Consumer Product Safety”, Sept. 25, 2019, avail-able at https://www.cpsc.gov/s3fs-public/Status-Report-to-the-Commission-on-the-Internet-of-Things-and-Consumer-Product-Safety.pdf?6sv9HwTXKHrkdmAyAkQ0_TsKCkpl1lR2.

Lawsuits have similarly emerged pertaining to IoT devices. For instance, several class action lawsuits consolidated in the Central District of California were recently brought against Ring LLC and Amazon.com, Inc., regarding the companies’ “Wi-FI cameras” that are “designed to be strategically placed throughout a prop-erty, enabling authorized users to see covered areas in high defi-nition and to communicate directly with occupants via a two-way speaker-microphone system”. Orange v. Ring LLC, et al., No. 2:19-cv-10899 (C.D. Cal. Dec. 26, 2019), Compl. (Dkt. No. 1) ¶ 6. Plaintiffs – asserting negligence, invasion of privacy, breach of implied contract, breach of implied warranty, and unjust enrich-ment claims on behalf of themselves and putative class members – allege that “[l]ax security standards and protocols render [defend-ants’] camera systems vulnerable to cyber-attack. Indeed, over the past several months numerous Ring customers reported that their camera systems had been hacked by malicious third parties who gained access to the video and two-way speaker-microphone system which they used to invade the privacy of customers’ homes and terrorize unsuspecting occupants, many of whom are chil-dren.” Id. ¶ 7; see also id. ¶ 28 (describing alleged “hacking inci-dents”). This litigation remains in its early stages, but the manner in which litigants and courts handle these and similar cases have the potential to impact future cases involving a broad spectrum of IoT products. Indeed, given the quantity and increasing breadth of IoT devices on the market and their rapid development, regu-latory challenges and consumer claims involving IoT devices will almost certainly continue to grow and evolve, causing a re-ex-amination of security, privacy, and traditional notions of product liability law.

Autonomous Vehicles

Autonomous vehicles, also known as driverless cars, have the potential to reduce traffic, increase safety, lower energy usage, and generate free time. Nevertheless, they also may lead to new

the number of MDL dockets has been declining. See Meeting of the Advisory Committee on Civil Rules, 207 (April 2–3, 2019), available at https://www.uscourts.gov/sites/default/files/2019-04_civil_rules_agenda_book.pdf. Of the 190 MDLs pending in 2019, 65 of them are classified as “Products Liability” litigation, the largest of any category. (See https://www.jpml.uscourts.gov/sites/jpml/files/JPML_Calendar_Year_Statistics-2019_1.pdf.)

In recent years, defence and plaintiff practitioners have argued that several aspects of MDLs and the manner in which they are litigated undermine their effectiveness for resolving their substantial inventory of cases. Accordingly, in November 2017, the Advisory Committee on Civil Rules formed an MDL subcommittee to consider rules to address commonly raised concerns with MDLs, including, among others, (1) their tendency to attract (and failure to weed out) meritless claims, (2) infrequent appellate review of pre-trial decisions, and (3) high pressure to engage in bellwether trials. See https://www.uscourts.gov/sites/default/files/2017-11-CivilRulesAgendaBook_0.pdf.

Industry groups have proposed several Federal Rule revisions to the Committee, including: requiring claimants to disclose preliminary evidence showing the cause and nature of the injury alleged; allowing mid-case appellate review of decisions on topics that have litigation-wide implications, such as preemption or expert testimony; requiring disclosure of plaintiffs’ outside sources of funding; barring use of bellwether trials without party consent; and increasing pleading standards to discourage meritless claims. See id. at 469-542.

In October 2019, the Committee reported that it “has continued to explore and gather information about the issues”; however, “it remains unclear whether rule changes would be helpful”. Meeting of the Advisory Committee on Civil Rules, 189-190 (October 29, 2019), available at https://www.uscourts.gov/sites/default/files/2019-10_civil_rules_agenda_book.pdf. The Committee continues to highlight particular areas of focus for potential rule changes, such as interlocutory appellate review of preemption and Daubert rulings, and MDL transferee judge review of proposed settlements. Id. The discussion of whether to limit claims through the use of Plaintiff Fact Sheets has evolved to the proposal for a “census” of cases in an MDL that would permit an initial assessment of plaintiffs’ claims. Id. at 193. One MDL judge has plans to pilot the “census” in a pending litiga-tion. Id. Although new rules are not imminent, MDLs continue to have an immense impact on product manufacturers and other defendants faced with these centralised cases.

Emerging Areas of Product Liability Law As technologies advance and change at an accelerating pace, new products entering the market are likely to become the subject of future product liability litigation. Among the numerous advancements that continue to create unique product liability challenges are the “internet of things” (“IoT”) and autonomous vehicles.

IoT

The expansive development of internet-connected consumer products, also known as IoT devices, have prompted consumer privacy and product liability concerns arising out of alleged manufacturing and design defects.

For instance, the FDA now recognises that “[m]edical devices are increasingly connected to the Internet, hospital networks, and other medical devices to provide features that improve health care and increase the ability of health care providers to treat patients. These same features also increase the risk of potential cybersecurity threats. Medical devices, like other computer systems, can be vulnerable to security breaches,




Note This chapter has been prepared for informational purposes only and does not constitute legal advice. This information is not intended to create, and the receipt of it does not constitute, a lawyer-client relationship. Readers should not act upon this without seeking advice from professional advisers. The content therein does not reflect the views of the firm.

Endnote1. Although the FDA proposed a rule in 2013 that would have

permitted generic drug manufacturers, like brand-name manufacturers, to change their warning labels through the CBE process, the rule was withdrawn in December 2018. See FDA Statement (Dec. 13, 2018), available at https://www.fda.gov/news-events/press-announcements/statement-fda-commissioner-scott-gottlieb-md-and-director-fdas-center-drug-evaluation-and-research (explaining that the “the new policy would have resulted in labels for the same drug that varied between different generic manufacturers”, which “could have led to consumer and provider confusion”).

liability risks. In fact, high-profile product liability lawsuits have already emerged against the manufacturers of self-driving vehicles. E.g., Huang v. Tesla Inc., No. 19-cv-346663 (Cal. Super. Ct. Apr. 26, 2019) (alleging negligence/wrongful death, strict liability, and other claims based on an auto-accident and death that resulted when Tesla’s allegedly defective Autopilot feature caused a car to accelerate into a concrete median).

In addition, developing regulations continue to address stand-ards surrounding autonomous vehicles, with most states consid-ering or enacting legislation. See https://www.ncsl.org/research/transportation/autonomous-vehicles-self-driving-vehicles-en-acted-legislation.aspx. The U.S. Department of Transportation has also released a series of principles for “Automated Vehicles”, on which it has invited public comment. See U.S. Dept. of Transportation, “Automated Vehicles 4.0”, available at https://www.transportation.gov/av/4. State regulations and federal guidelines remain in their infancy, and will undoubtedly continue to evolve with this developing technology.


13Sidley Austin LLP


Daniel A. Spira, a litigation lawyer in Sidley’s Chicago office, focuses his practice on the defence of product liability cases and consumer class actions. In both state and federal trial and appellate courts, Dan has represented clients in a wide array of industries, including phar-maceuticals, medical devices, nutritional supplements, food and beverages, automotives, construction products and consumer goods. In addition to first-chair jury trial experience in federal court, Dan has extensive experience managing multi-district litigation, writing and arguing successful motions at various stages of litigation, taking and defending depositions, working with medical and economic expert witnesses and managing complex discovery. Dan’s track record of securing favourable judgments for his clients spans every phase of litigation, including motions to dismiss, summary judgment, class certification, trial and appeal. Dan has also advised clients through pre-litigation risk assessments, as well as regarding regulatory reporting obligations to the Consumer Product Safety Commission (CPSC).

Sidley Austin LLPOne South DearbornChicago, IL 60603USA

Tel: +1 312 853 7000Email: [email protected]: www.sidley.com

Sidley Austin LLP is a premier law firm with a practice highly attuned to the ever-changing international landscape. The firm has built a reputation for being an adviser for global business, with more than 2,000 lawyers in 20 offices worldwide. Sidley successfully represents clients on complex and multidisciplinary transactional, regulatory and litigation matters spanning a range of legal areas. From its offices in the commercial, financial and regula-tory centres of the world, Sidley harnesses its knowledge to provide thoughtful advice for the myriad legal and business challenges that clients face.

www.sidley.com

Teri H. Peeples practises in the Product Liability and Mass Torts group in the firm’s New York office. Her practice focuses on representing clients in the life sciences industry, including pharmaceutical and medical device companies, in federal and state courts, as well as in multi-district litigation and class actions. Teri advises clients at all stages of the litigation process, but maintains an emphasis on the scien-tific and medical issues involved in these types of matters. This includes causation and mechanism of action, safety-related allegations and the pre-clinical and clinical development of the products at issue. Teri is critically involved in the development of the defence strategy, iden-tification and development of expert witnesses, preparation of, taking and defending fact and expert witness depositions, arguing discovery motions in state and federal court and day-to-day case management. Teri also counsels multinational pharmaceutical and medical device companies in comprehensive FCPA risk reviews and conducts internal FCPA investigations.

Sidley Austin LLP787 Seventh Avenue New York, NY 10019USA

Tel: +1 212 839 5300Email: [email protected]: www.sidley.com


Chapter 314

Criminal Liability for Defective Products

Herbert Smith Freehills LLP David Bennett

Howard Watson


to EU-level regulation. These a

Date post:	30-Jan-2021
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times

Product Liability 2020 · 2020. 12. 17. · %JKGH/GFKC1H EGT Fraser Allan CEO Jason Byles Printed...

Documents