In this article, we propose a general structure for designing a research protocol of an observational epidemiological study. We start by highlighting the importance of the research protocol, namely in accounting for some bias and guaranteeing methodologic rigor and study reproductability. Next, we reflect on some of the essential elements of a research protocol no matter its objective. We further present some specific issues to be included according to the type of study: cross-sectional, case-control and cohort.