The association between type 2 diabetes and some types of cancer is well established and the complexity of this association is receiving increasing recognition. Many factors influence the risk of both diabetes and cancer including age, sex, ethnicity, socioeconomic status, obesity/insulin resistance, diet (including alcohol intake), physical activity levels and smoking history. The presence of diabetes may influence the uptake of cancer screening. Using an experimental approach to investigate the effect of better control of diabetes is helpful in assessing whether hyperglycaemia has a causal relationship with cancer. Diabetes treatments may influence the risk of cancer independently of their effect on glycaemia and complicate investigation of the association between diabetes and cancer. This article discusses the complexity involved in attempting to clarify the factors that contribute to the associations between diabetes, hyperglycaemia, diabetes treatment and cancer.