Mendelian randomization is the term applied to the random assortment of alleles at the time of gamete formation, a process which results in the population distributions of genetic variants which are generally independent of the behavioural and environmental factors which typically confound epidemiologically derived associations between the putative risk factors and disease. In some of the circumstances taking the advantage of Mendelian randomization might provide an epidemiological study with properties of a randomized comparison, for instance, a randomized clinical trial. In particular, the use of the Mendelian randomization might allow causal inferences to be drawn from epidemiological studies, typically using the instrumental variable estimator.