A Gateway That Never Was

This brief communication responds to the article by Rajabi et al., recently published in Behavior Genetics. To test the hypothesis of cigarette smoking as a "gateway" for subsequent opium use and contrast it with the common liability model, Mendelian randomization analysis was applied to data obtained from an Iranian sample, using CHRNA3 rs1051730 as an instrumental variable. It is doubtful, however, if the assumptions of instrumental variable analysis hold in this case. The authors misstate both the gateway hypothesis and the common liability model. The article has many other deficiencies that diminish the veracity of its categorical conclusions that accept the causal interpretation of the "gateway hypothesis" and reject the common liability model, with which the data are fully consistent.