Breast cancer rates in populations of single women.

The well known associations of breast cancer with fertility patterns and diet are interdependent and it is difficult to estimate the extent to which breast cancer is related to diet. This was attempted by analysing breast cancer rates in populations of single (never married) women for which the contribution of childbearing would be small. Age specific breast cancer rates for single women showed the same variation by country, social class, urban-rural area and with time, as did the corresponding rates for married women, suggesting that common or related factors determined breast cancer rates in single and married women. Also, dietary correlations of breast cancer rates at 55-64 years, around 1960, were not sifnificantly different for single women and the general female population. This supported the view that the dietary associations with breast cancer, observed in larger studies of general female populations, did not arise indirectly from an association with childbearing rates. It was pointed our that the positive association of breast cancer with sugar, observed for single and for all women, was accopanied by a negative association with starch. These opposite associations with two forms of varbohydrate seemed inconsistent on general nutritional grounds and could be explained as arising indirectly to the association of breast cancer with affluence. Otherwise, it would seem necessary to establish a nutritional difference between starch and sugar, which could reasonably influence breast cancer rates, before the association was accepted as indicating cause.