The validity of epidemiological and clinical observations is paramount, not only for scientific advancement but also in evidence-based practice. Concern regarding validity with respect to the design and conduct of analytic epidemiological studies is often under-appreciated, partly as a consequence of our focus on randomized experimental designs as the highest standard of ‘proof’ in clinical science. We review the design, conduct and interpretation of rigorous analytic epidemiological study designs, with specific reference to periodontology. We give special emphasis to intrinsic validity and the use of focused aims to re-frame the perspective on the strength of evidence in reviews of the literature. Specifically, we draw on the periodontal research literature to provide an overview of the appropriate design and conduct of cohort studies, including randomized designs, case–control and cross-sectional studies. The concepts of selection bias, information bias and confounding are explored for each study design, both in general terms and with respect to a critical review of the literature on the epidemiology of periodontal diseases.