of the main subtypes of diabetes has grown more specific over the years, from ‘early-onset vs. adult-onset’ to ‘insulin-dependent vs non-insulin dependent’ to Type I/Type II, and most recently Type1/Type 2 15;16. Even the most recent definition of Type 2 is recognised as a heterogeneous condition with diverse molecular and environmental pathways 17. In addition, since recruitment of adequate numbers of cases within a highly specific disease definition is often difficult (i.e. requiring multi-centre studies), less specific definitions are frequently introduced to make up sufficient numbers in an attempt to attain a certain level of power. However, in reality a gain of power may not be achieved at all; a reduction in overall power may even be the result, because of increased causal heterogeneity. In practice, the best guideline is to define cases according to a definition that minimises the likely causal heterogeneity based on all existing clinical and biological evidence. For example, in a situation in which a clear, strong, environmental cause is already known for the condition in question, the investigator could limit case and control selection to those unexposed to this cause.
