Finally, this sample was ascertained to detect genetic factors contributing to nicotine (rather than alcohol) dependence. Given the co-occurrence of drinking and smoking, we may hesitate to generalize these findings due to the potential for alcohol use in this sample greater than that in the whole population. The prevalence of alcohol dependence in this sample is similar to the general population prevalence, however, which along with the fact that we adjusted for nicotine dependence, increases our confidence that our results are unbiased by ascertainment.
