Gee & Ors v Depuy International Ltd

Under Part I of the Consumer Protection Act 1987, a product is defective only if its safety falls below the objective level persons generally are entitled to expect. A normal and unavoidable risk inherent in the intended use of a product is not itself a defect. The relevant defect must be identified sufficiently to permit causation to be assessed.

The statutory inquiry requires a flexible assessment of all relevant circumstances. Depending on the product, this may include benefits, avoidability, warnings, regulatory compliance and the presence of a learned intermediary. A claimant alleging an abnormal risk of failure must establish that risk by reliable comparative evidence. On the evidence, the claimants failed to prove that the Pinnacle Ultamet prosthesis carried a materially greater risk of early failure than an appropriate comparator.

The claimants were 312 individuals implanted with Pinnacle Ultamet metal-on-metal hip prostheses manufactured by DePuy International Ltd. They alleged that the prostheses were defective under Part I of the Consumer Protection Act 1987 because they generated metal debris capable of causing an adverse reaction to metal debris and early revision surgery.

The trial concerned a common preliminary issue: whether DePuy was liable, subject to any development risk defence, including issues of causation. The claimants advanced a primary case based on the normal propensity of metal-on-metal hips to generate debris and an alternative case based on an abnormal and materially increased risk of early failure compared with other prostheses. The central questions were the meaning of defect, the legally relevant circumstances, causation, and the reliability of comparative revision statistics.

1. Meaning of defect. Sections 2 and 3 of the Consumer Protection Act 1987, read consistently with Articles 1, 4 and 6 of the Product Liability Directive, impose no-fault liability where a product fails to provide the safety persons generally are entitled to expect. The standard is objective and is assessed by reference to all relevant circumstances, including the time of supply. Information discovered after supply may be considered when deciding whether the product met that standard.
2. The claimant must identify what about the product, its condition, behaviour or risks caused it to fall below the statutory safety standard. The normal propensity of a metal-on-metal hip to shed debris, and the normal possibility that a patient may suffer an adverse reaction, were inherent risks of intended use. They were not defects. The alternative case was legally coherent only if the product had an abnormal risk of early failure compared with an appropriate comparator.
3. Relevant circumstances. The statutory assessment is flexible and fact-sensitive. It is not governed by rigid distinctions between standard and non-standard products. Depending on the circumstances, benefits, avoidability, warnings, regulatory compliance and information supplied to a learned intermediary may be relevant. They must not be used to introduce negligence concepts or to treat compliance as a complete defence.
4. Causation. The appropriate question was whether, on the balance of probabilities, each claimant would have suffered the damage complained of, namely early revision, if the prosthesis had not carried the increased risk said to constitute the defect. The comparison was with the generic incidence of early failure, not merely failure caused by adverse reaction to metal debris. A more-than-double-risk approach was not adopted as a universal bright-line test.
5. Comparative evidence. The appropriate external comparator was an established uncemented metal-on-conventional-polyethylene prosthesis available when the Ultamet entered the market in 2002. The intra-Pinnacle comparison and later products could not determine the entitled expectation of safety. The Court had to assess the reliability of the registry data, including missing data, patient characteristics, activity, asymmetric surveillance, changes in revision practice, media influence and surgeon outliers.
6. The National Joint Registry statistics were insufficiently reliable to establish that the Ultamet had a materially greater ten-year failure risk. The Swedish registry data did not establish such a difference either. The claimants therefore failed to prove that the product was defective or that the alleged increased risk caused the individual revisions.
7. The claimants’ primary case was untenable and their alternative case failed on the evidence. DePuy was not liable under the Act, irrespective of any development risk defence.

First-instance decision. No appellate history is stated in the judgment.