Smear
Sir: Writing about breast screening and cervical smears Dr Sharon McCullough ('Useless smear campaign', 11 February) suggests that assessing their value is 'vir- tually impossible, given the present know- ledge of the natural history of breast and cervical malignancy'. This is surely wrong. The proof of the pudding is in the eating. In other words, forget 'natural history' and medical theories, just compare the death rate in those having smears (or any other intervention) with the death rate in those not having them. This is the only way to show if lives are really being saved or not.
The only snag is that, for the comparison to be reliable, the two groups need to be alike in every way except the screening or smearing. This is why 'randomised' com- parisons are usually needed. Without them there is a great danger of coming to false conclusions.
Thurstan B. Brewin
Marie Curie Cancer Care, 28 Belgrave Square, London SW1
Previous page
