In addition to FTQ/FTND measures, the subjects were assessed for abuse and dependence of alcohol and illicit substances using DSM-IV criteria. To evaluate whether the nAChR genes were associated with the abuse/dependence symptoms of these other substances, we queried our twin databases and selected categories with reasonable frequencies (>25%) in our combined VAANX-ND and VAFTND samples. We found that only the symptom counts of alcohol and cannabis abuse/dependence had sufficient frequencies to permit useful analysis. Therefore, we used the symptom counts of alcohol and cannabis abuse/dependence (SymAlcAD and SymCanAD, respectively) as phenotypes instead of a dichotomized affected/unaffected category. In parallel to FTND scores where only regular smokers were assessed, we used a “regular drinker” definition (those who used five or more drinks in a month) as the inclusion criterion for SymAlcAD. Similarly, a “regular cannabis user” definition (those who used cannabis five or more times in a month) was used as inclusion criterion for SymCanAD. The rationale for using these definitions was that we wanted to use only those subjects with sufficient exposures to drinking and cannabis use. In the
