4â Defectiveness

Applying this terminology, manufacturing defects would be classed as nonstandard products as they would differ from what the producer intended by their failure to comply with the design being used on the production run. In short, rogue products would be classed as non-standard products in this context. Burton J ruled that the primary issue in relation to non-standard products is whether:

the public at large accepted the non-standard nature of the product, i.e., they accept that a proportion of the products is defective … That, as discussed, is not of course the end of it, because the question is of legitimate expectation, and the Court may conclude that the expectation of the public is too high or too low.53

In considering non-standard products, Burton J continued that the avoidability of the harmful characteristic, the impracticality, cost or difficulty of taking precautionary measures and the benefit to society or utility of the product should not be taken into account.54 In respect of the last of these, however, the benefit to society and utility of the product can be considered when the public, with proper knowledge and full information about it, does and ought to accept the risk.55 Assessing the defect in a non-standard product should be relatively straight-forward with the opportunity for the injured user to prove the causal link to the damage and thereby claim compensation.

Applying this alternative terminology, a product with a design defect would nonetheless be a standard product as the product would be as the producer intended, even though the design itself was deficient. Indeed, Burton J acknowledged that if a standard product is unsafe, it is likely to be as the result of either an alleged design error or an allegedly flawed system.56 Assessing defectiveness in standard products is likely to be more problematic, and Burton J opined that the harmful characteristic in the goods must be identified, if necessary with the help of experts. The sole question is the safety of the goods for their foreseeable use, with comparison to other goods on the market being appropriate to identify and compare the relevant features of the products concerned. He included price as a significant factor in legitimate expectation and potentially material when comparing products. However, as with non-standard products, he felt that what the producer could have done differently and whether the producer could or could not have done the same as other producers are not relevant factors in this context.57 Naturally, the inherent risk of taking into account other similar products on the market and comparing the features of those products which the producer could have included in his product is that it would come very close to re-introducing negligence by the back door, as it renders pertinent the actions of the producer by reference to the actions of others. Strict liability is supposed to depend upon the existence of the defect rather than upon whether the producer has failed in some way in allowing the defect to occur.

This definition clearly indicates the adoption of an objective consumer expectation test as the basis underpinning liability. It therefore demands an analysis of who constitutes ‘persons generally’64 for this purpose and how their legitimate expectations are to be assessed. In practice, the court acts as arbiter in analysing the expectations of the public as ‘an informed representative of the public at large’65 and hence decides what persons generally have a right to expect. However, it is possible that the standard adopted by the court does not accord with what either the public generally or an individual subjective consumer might actually have thought or expected. Burton J considered this aspect in A v. National Blood Authority,66 stating that:

The Court decides what the public is entitled to expect … Such objectively assessed legitimate expectation may accord with actual expectation; but it may be more than the public actually expects, thus imposing a higher standard of safety, or it may be less than the public actually expects. Alternatively, the public may have no actual expectation, e.g. in relation to a new product67

The standard of safety is what the public is legitimately entitled to expect. This does not require absolute safety any more than product liability imposes absolute liability as opposed to strict liability for the injuries caused. The level of safety provided by the product will be assessed by reference to all the relevant criteria68 to see whether it satisfies the legitimate expectations of the public, but no more. In A v. National Blood Authority itself, Burton J held that the infected blood was defective as the public had a right to expect that the blood was safe and not infected with Hepatitis C. This was so because there was a known risk that blood used for transfusions might be infected even though there was no test to establish whether any individual bag of blood was so contaminated. It was relevant that the risk was not known to the public generally, although it was recognised within the medical profession, but the knowledge of the ‘learned intermediary’ was not sufficient to avoid liability to the ultimate recipients of the blood.

Q5 Is a consumer expectation test an appropriate test to use when assessing product safety?

While section 3(2) of the 1987 Act raises the issue of the time of production when assessing defectiveness so as to put the incident into context, time plays no part in the expectation of persons generally that the product be safe. Thus, the court in Abouzaid v. Mothercare (UK) Ltd69 held that defectiveness

64It is noteworthy that s.3 refers only to the level of safety expected by persons generally and does not expressly refer to particular categories of users, such as children or the elderly, who may legitimately expect higher levels of safety. Contrast the requirements for a safe product under the General Product Safety Regulations 2005, reg. 2, where categories of consumers at risk are dealt with explicitly.

65A v. National Blood Authority [2001] 3 All ER 289, para. 31(vii).

66Ibid.â 67â Ibid. para. 31(vii).

68 Consumer Protection Act 1987, s.3(2).â 69â [2000] All ER (D) 2436.

in a product is to be judged by the expectations of the public rather than by the passage of time. The plaintiff, aged twelve, was helping to fix a ‘Cosytoes’ fleecelined sleeping bag to a pushchair used for his younger brother. The Cosytoes was manufactured by the defendant and was affixed using two Âelasticated straps, which went around the back of the pushchair and fastened with a buckle. While attaching the Cosytoes, one of the straps sprang from the plaintiff’s grasp and the buckle on the end of the strap hit him in the left eye, resulting in permanent sight damage. The defendant tried to claim a development risk defence, which the court rejected, deciding that the lack of any previous similar accidents was not sufficient to satisfy the defence.70 The judge, finding for the plaintiff, opined that the inherent risk in the product arose from the propensity of elastic to spring back. If that propensity constituted a defect in 1999, when the product was considered by a safety expert prior to the court hearing, it was equally a defect in 1990 when the accident occurred. The Court of Appeal upheld this decision and rejected an appeal by the defendant.

In assessing defectiveness in any individual item, section 3(2) of the 1987 Act provides that all the circumstances must be taken into account and proceeds to provide a non-exhaustive list of three specific factors that are relevant. In this respect, it echoes the thrust of the corresponding provision of the Product Liability Directive,71 although the 1987 Act goes into greater detail, outlining the three criteria thus:

(a)the manner in which, and purposes for which, the product has been marketed, its get-up, the use of any mark in relation to the product and any instructions for, or warnings with respect to, doing or refraining from doing, anything with or in relation to the product;

(b)what might reasonably be expected to be done with or in relation to the product; and

(c)the time when the product was supplied by its producer to another.

The first criterion deals with the presentation of the product and the purpose for which it has been put into the market. It expressly refers to instructions and warnings, reflective of the fact that this criterion deals primarily with duty to warn defects. The get-up or presentation of the product is most unlikely to impact upon either a manufacturing defect or a design defect but may be the major factor in a duty to warn defect. Section 3(2)(a) encompasses a few related issues, the first of which is the manner in which, and purposes for which, the product has been marketed. This allows the court to consider issues such as whether the product is aimed at any particular section of the community such as children or the infirm, and whether any advertising accompanying the product is geared specifically to appeal to that group. This would also be relevant in later consideration of whether instructions and warnings were appropriate

70The development risk defence under s.4(1)(e) will be discussed in greater detail below.

71Product Liability Directive, Art. 6.

313 4â Defectiveness

to foreseeable product users. The marketing of the product might also indicate whether the product is aimed only at experienced users of such products or the public in general, which would again impact on the instructions and warnings provided with the product.

The purpose for which the product has been marketed delineates the intended use of the product but this will not necessarily negate the liability of the producer if an injury is caused when the product is used for a non-intended purpose, but one that would constitute a reasonably foreseeable misuse. However, much might depend on the warnings provided and the previous known history of the product. As Professor Miller points out,72 Thalidomide, which has a notorious reputation given its impact on the foetus if used during pregnancy, would not be classed as defective if marketed as, and used solely as, a treatment for leprosy, for which it has proved very effective. This would, of course, depend on it not being prescribed for use by pregnant women. In fact, Thalidomide has received approval by the Food and Drug Administration (FDA) for use in the United States in the treatment of leprosy, all the more remarkable given that the FDA refused to license it originally and thereby avoided the tragic injuries to unborn foetuses that were suffered by children in countries where the drug was used during pregnancy.73

The get-up of the product and the use of any mark in relation to it may also contribute to the perception of how safe the product is likely to be. The use of marks would include things such as the BSI Kitemark, which provides a clear statement of compliance of the relevant aspect of the product with safety requirements. However, the inclusion of a mark should not be interpreted as meaning that the product is safe in every regard. Any safety standard must be applied and interpreted in the light of what it seeks to control. Thus, for example, compliance with a BSI Standard that governs the flammability of a product does not address the strength of the item or the constituent substances from which it is made. Further, compliance with a voluntary standard does not necessarily mean that the product is safe, as a standard may not reflect the highest level of safety possible at the time of the incident as opposed to the level acceptable at the time when the standard was published. It follows that a product that complies with a voluntary standard could still be defective for the purposes of the 1987 Act and compliance with that standard would not be a defence. It may, demonstrate the intention of the producer to supply a quality product, but will not protect him from the consequences of supplying a defective item to an injured user.

A similar issue to the voluntary use of a quality mark such as the Kitemark is the compliance of the product with a mandatory standard. Again, the standard will not necessarily reflect the highest level of safety achievable at the time,

72Miller and Goldberg, above n. 10, para. 10.55.

73The FDA official who refused to license Thalidomide originally was subsequently awarded a Congressional medal.

(d)â Need for causation

While strict liability has altered both the basis of liability and the range of potential defendants, it has not affected the requirement that the claimant proves the causal link between the defective product and the damage caused. Indeed, the Product Liability Directive expressly states that the injured person is ‘required to prove the damage, the defect and the causal relationship between defect and damage’.86 Producers can still rely on the defence that, while the claimant used their product, it was not their product that caused the damage and that some other factor is responsible for the injuries sustained by the claimant. There is no reversal of proof and thus the burden lies on the claimant to prove the causal link rather than on the defendant to disprove it. Causation may be very difficult to prove, particularly in cases involving pharmaceuticals and medical products, leading to suggestions that such injuries should be dealt with via a no-fault compensation scheme, such as that in New Zealand, or through an industrybased scheme like the German Pharmapool. Given that most, if not all, pharmaceuticals have side-effects and are designed to impact physiologically upon the user, it is very difficult to establish causation in any individual case. Each person has a different metabolism, which, combined with differences of illhealth, age, sex, weight, etc., can lead to differing reactions to the same product. The net result is that causation may be difficult to prove unless there is a proven link between the pharmaceutical in question and the type of injury suffered by the claimant, as occurred in the Thalidomide situation.

It must also be acknowledged that the fact that the claimant can prove a causal link does not automatically mean that the product is defective. It may be that the claimant is an idiosyncratic user whose reaction to the product is abnormal. Thus, for example, nuts are not rendered defective because a person who is allergic to nuts becomes ill after eating them. In respect of pharmaceuticals, a risk-benefit analysis comes into play and the incidence of a relatively minor side-effect would not make a product defective if the benefit it bestows is to save the life of the user. Causation was specifically addressed in X v. Schering,87 a case in which the user suffered a thrombosis allegedly caused by the use of a third generation contraceptive pill. The plaintiff, the lead claimant in a group action, had used the contraceptive Femodene, a third generation COC, which was manufactured by the defendant. The plaintiff had suffered a deep vein thrombosis while others in the group litigation had suffered various cardiovascular disorders, which, in some instances, had proved fatal. The claimant alleged that the third generation COCs carried more than twice the risk of the second generation equivalent, a fact she needed to prove to establish causation. In giving judgment for the defendant, Mackay J held

86 Product Liability Directive, Art. 4.â 87â (2003) 70 BMLR 88.