Although rs16969968 is the most strongly associated SNP genome-wide, multiple associations in the region form haplotypes associated with varying risk of smoking behavior.10,16–19 Furthermore, although this article references rs16969968 as the likely causal association, more than half of the studies used the SNP rs1051730 to tag rs16969968, and there are multiple other SNPs spanning the CHRNA5-CHRNA3-CHRNB4 cluster that are indistinguishable from rs16969968 in populations of European descent. Therefore, the SNP analysis is a simplification of the true genetic model and does not represent the full complexity of the relationship of age at onset with the association between genotype and smoking in this region. In addition, further study is required to better understand other modifiers of the association between CPD and rs16969968. For example, parental monitoring and peer smoking have been shown to modify this genetic association.39,40
