Prof David Coggon, Professor of Occupational and Environmental Medicine, Southampton University, said:

"As the authors recognise, in a study of this sort positive associations often occur simply by chance. They carry little weight in the absence of stronger supportive evidence from other research."

Prof Anthony Dayan, Toxicologist, said:

"The conclusions are presented too strongly in relation to the weight of evidence. The odds ratios almost always include ‘1’ and so are far from definitive. The grouping of solvents into simple classes is a way of increasing numbers in each class to a level at which statistical analysis is feasible but it is much harder to justify in terms of pathogenetic mechanisms as the ‘reactive metabolites’ involved differ greatly in nature and properties. The strength of the association with ‘fibres’ is very odd indeed and although the attempt is made to link that to dyes and flame retardants it remains a very peculiar feature indeed. I wonder about suitability of controls – largest number had genito-urinary tract neoplasms. If that included ovarian and/or uterine tumours then there might have been serious confounding with other hormonal effects. I am not clear whether they have dealt properly with the questions of smoking and oral contraceptives, both known associates of breast cancers."