Background: Causality has to be assessed by randomised controlled trials. However, since these are not always ethically or technically possible, other study designs such as cohort studies, case-control and cross-sectional analytic studies may be used.
Methods: In this review, we focus on the strengths and limitations of epidemiological studies whose aim is to identify the causal relationship between study factor and outcome (cohort studies, case-control studies, cross-sectional analytic studies). We also introduce the absolute and non-absolute parameters useful in determining the causal relationship in epidemiological studies.
Results: To establish a causal relationship between a factor and an outcome, exposure must precede the outcome, the association must not be due to chance, bias, confounding or misclassification. It is important when reading this kind of research papers that all these factors are properly considered. Studies in which strong associations are documented, whereby a dose-effect relationship exists, and those which are consistent with the results of previous epidemiological studies and document biologically plausible associations are generally stronger evidence than those in which these criteria are not found.
Conclusions: The design and conduct of valid cohort, case-control and other type of studies, when randomised trials are unfeasible or unethical to answer causality and intervention questions, is a challenge for the nephrology community, presently lacking valid clinical evidence and adequate answers to many questions.