Health: Second Opinion

BEING struck by lightning is one of the few accidents for which the victim cannot easily find someone to blame - and possibly sue. It seems to be human nature to look for some explanation for personal disasters, but scientific methods of analysis are being improved to distinguish cause from coincidence.

Around one in 20 babies has some defect at birth. Most of these are minor. One in 100 has a heart defect, usually requiring treatment. One in 300 has a cleft lip or palate. One in 1,000 has a major defect affecting an arm or a leg. Genetic factors play a part in causing some defects, but in most cases there is no identifiable cause.

Thirty years ago the thalidomide tragedy - in which thousands of babies around the world were born with severe abnormalities after their mothers had taken the drug in early pregnancy - alerted people to the possibility that some birth defects might be caused by chemicals acting on the developing embryo. In fact, very few substances other than thalidomide have been proved to affect foetal development.

Suspicion of possible effects on the foetus has fallen on many other drugs, however, and nowadays pregnant women are prescribed as few medicines as possible.

Ten years ago a drug widely used to treat vomiting in pregnancy, known as Bendectin in the US and Debendox in Britain, was withdrawn by its manufacturers in the face of thousands of legal actions from women who believed that they had given birth to babies with limb defects because they had taken the drug. The manufacturers disputed this. More than 33 million women had taken the drug during pregnancy; in some communities it had been prescribed for one-third of all pregnant women. Inevitably, therefore, one-third of women in those communities whose children had birth defects of any kind would have taken the drug and many blamed it.

Slowly the legal system has ground its way through the actions. In the US 38 cases (representing large numbers of plaintiffs) had come to trial by the end of 1993; the manufacturers won in 36 of these and lost in two. Those two, and six of the defence wins, have gone to appeal.

The Supreme Court's judgement in one of these cases, Daubert v Merrill Dow Pharmaceuticals, may have a wide impact on the interpretation of expert evidence in all medical cases. The court has decided that in disputes between expert witnesses the greatest weight should be given to evidence that is based on 'peer review and publication of scientific data and conclusions'. What this means is that the court will be most likely to believe an explanation of causation if that explanation has passed the scrutiny of the panels of experts and statisticians (mostly unpaid) used by top-ranking scientific journals such as the New England Journal of Medicine. Publication in such a journal will be strong evidence of 'general acceptance by the scientific community'. The court will rely on that consensus more than on any expert's individual opinion, no matter how eminent or persuasive he or she might be.