It is also possible that the effect of rs6265 reflects Mendelian randomization, in which the association between genotype and an outcome (i.e., cannabis use) relates to individuals being “randomized” at birth to a higher likelihood of environmental exposure (i.e., smoking initiation; Smith and Ebrahim, 2003). If carriers of the G (Val) allele of rs6265 are more likely to initiate cigarette smoking, and cigarette smoking, in turn, increases the likelihood of cannabis initiation via environmental pathways, then the association between rs6265 and cannabis use may reflect an environmental process rather than evidence for genetic overlap. Due to the rather general biological role of rs6265, these processes are difficult to disentangle.
