This chapter describes in a simple way the most important issues of epidemiological design, with emphasis on studies on metabolic polymorphisms. Different options are offered to the researcher who approaches a molecular epidemiology study. Case-control studies and cohort studies have different and sometimes complementary advantages and disadvantages. The sources of bias and the choice of controls, with specific problems in the context of investigations on metabolic polymorphisms, are key issues. Confounding is an issue that needs further clarification in the field, partly because of the lack of complete biological knowledge on the complex relationships between exposure, markers of susceptibility, and cancer. Metabolic polymorphisms are correctly interpreted as effect modifiers of the exposure-disease relationship. This interpretation implies that studies should be planned in order to have sufficient statistical power and that an interactive term should be modelled in the statistical analysis.