EDITOR, - In their audit of patients with breast cancer treated in the Thames regions A M Chouillet and colleagues show and comment on wide variations in the treatment delivered in early 1990 to women of the same age and with disease of the same stage.1 Reasons for these variations are not hard to identify. Women are referred to many general surgical clinics, which lack the range of skill required for managing women with breast problems. Multidisciplinary breast clinics achieve a measure of uniformity and quality by working according to agreed protocols, whose implementation can be audited prospectively.

Referral to specialised breast clinics will not eliminate variation in treatment altogether because of differences of opinion between experts. Differences in treatment practices reflecting clinical uncertainty should act as a spur to participation in large randomised trials rather than to pressure for consensus guidelines. Where the adjuvant systemic treatment of early breast cancer is concerned, a worldwide meta- analysis carried out in 1992 showed conclusively that tamoxifen, chemotherapy, and ovarian suppression each reduce 10 year mortality in premenopausal and perimenopausal women.2 About 5-10 extra women with early breast cancer are alive at 10 years for every 100 women treated with any of these modalities. Tamoxifen and chemotherapy are each effective in postmenopausal women. Tamoxifen has a relatively favourable toxicity profile and is simple to administer. Five years' treatment at 20 mg a day costs just under pounds sterling200. It should therefore be considered for most women with early breast cancer, regardless of age. The optimal duration of tamoxifen treatment is being tested in the Cancer Research Campaign's trial and the “adjuvant tamoxifen treatment - offer more?” (aTTom) trial, and participation in these should be encouraged.

The overview carried out in 1992 also raised the possibility of appreciable further reductions in …