One of the main problems of clinical research lies in the difficulty in selecting the best design to give an answer to the problem of research - which can trigger in vain efforts to address and resolve the problem. The selection of the best design basically depends on the research question, the objectives and the assumption to contrast. We present the key aspects related with the kind of question that respond, bias, sample size and analysis plan should be considered a priori in the selection of cross-sectional design.