The epidemiology of diabetes and cancer.

The literature on cancer occurrence in persons with diabetes has almost invariably been concerned with relative measures. In this paper, we briefly review this, but the aim is to quantify the absolute occurrence of diabetes and cancer in the population in order to give a fuller picture, which also includes the competing mortality risk. Overall, we find that some 35 % of the population will have a diagnosis of diabetes in their lifetime, 44 % a diagnosis of cancer, and about 15 % will have both diagnoses. The impact of differing mortality between persons with and without diabetes is illustrated by the fact that a person without diabetes at age 50 has a smaller lifetime risk of cancer than a person aged 50 with diabetes. Thus, the differences in cancer occurrence between persons with and without diabetes are of quantitatively smaller importance than the differences in mortality.