PI Practitioner – each issue a particular topic is highlighted, citing some of the useful cases and other materials in that area


Using epidemiological evidence to prove individual causation


• In Loveday v Renton [1990] 1 MLR 117 the plaintiff claimed that the pertussis vaccine caused brain damage. The court considered a preliminary question, namely ‘can or could pertussis vaccine used in the UK…cause permanent brain damage or death in young children.

The centrepiece of the plaintiff’s evidence was an epidemiological study, the NCES. At Chapter 11 of his judgment (p.161 et seq.), Stuart-Smith LJ held that ‘where given effects…occur with or without the pertussis vaccination, it is only possible to assess whether the vaccine is a cause, or more precisely a risk factor, when the background incidence of the disease is taken into account. The question therefore is, does the effect occur more often after the pertussis vaccination than could be expected by chance? It is the function of epidemiology to answer this question’.

The judge gave weight to the study (see p.181) but concluded that it did not support the plaintiff. This finding contributed to the claim being dismissed at a preliminary stage.

• Reay v British Nuclear Fuels plc [1994] 5 Med LR 1 is an attempt to establish the correct approach for a court to adopt to epidemiological evidence, including use of the Bradford-Hill Criteria (at p.13). These criteria are commonly used in epidemiology to prove causality, i.e. the relationship between an incidence and a consequence.

On the facts, the Claimant failed to establish a sufficiently strong causal relationship between his leukaemia and the incident (a fire at Sellafield nuclear plant). This failure was partly attributable to methodological deficiencies in the main epidemiological study relied upon; and to the application of the Bradford-Hill criteria (see conclusions pp.48 – 49).

• In McTear v Imperial Tobacco Ltd [2005] 2 SC 1 (OH) the question was whether epidemiological data could be used to draw conclusions about the cause of a disease in any individual (in the context of a widow bringing an action for the death of her husband from lung cancer). The Court found at paragraph [6.180] that it could not: ‘epidemiological evidence cannot be used to make statements about individual causation’.

• Novartis Grimsby Ltd v John Cookson [2007] EWCA Civ 1261 was a case concerning bladder cancer, in which there was a dispute between a clinician and an epidemiologist about whether the Claimant’s cancer had been caused primarily by smoking or by exposure to industrial dyes (see paragraph [52]).

The clinician’s view was accepted by the Recorder at first instance. The Recorder’s decision to rely upon the clinician’s evidence was upheld on appeal. The epidemiologist had not appreciated all of the facts but refused to revise his view when these facts were brought to his attention by counsel (see paragraph [53]).

November 4, 2011 · Editorial Team · Comments Closed
Posted in: Cases