People With MS May Have Lower Cancer Risk

Dear Sir

It is surprising that Dr Chaudhuri suggests that unequal numbers in the two cohorts represents a potential flaw in this study. In epidemiological research the majority of observational studies use groups of unequal size and this is not generally regarded as a limitation. If the two groups are equivalent in terms of the study base (the population they are drawn from) with appropriately standardised duration of follow-up, then there is no reason to suspect that differences in group size will influence the magnitude of association estimates. Inadequate statistical power can be a concern in epidemiological research, but here the number of subjects is large, with sufficient power to detect even modest differences between the cohorts. The constancy in findings across various cancer types lends additional assurance that the results are not simply due to chance associations with rare events.

Yours faithfully

Scott M Montgomery (Prof.)

Professor Scott Montgomery 11:45AM April 01, 2009