analyzed hazards of smoking initiation, progression to heavy smoking, becoming nicotine dependent, and relapsing from a quit attempt using Cox proportional hazard models. To account for differences in the frequency with which study members attempted cessation, we constructed panel datasets that included one observation per study member per assessment (for the age 18-32 data) and one observation per study member per quit attempt (for the life-history calendar data). We used these panel datasets to analyze the genetic effect on smokers’ risks of cessation failure during ages 18-32 years and on their hazards of relapse during ages 32-38 years. We accounted for non-independence of repeated observations of individuals using generalized estimating equation models of risks and conditional risk-set models of hazards.43,44 We tested whether genetic effects on the mature phenotypes of persistent heavy smoking, nicotine dependence, and relapse were mediated by adolescent developmental phenotypes using the structural equation described by McKinnon and Dwyer and the methods described by Preacher and colleagues.45-47 In order to allow for a single test of mediation, we conducted a principal components analysis48 of the mature phenotypes of persistent heavy smoking (pack years smoked at age 38 years), nicotine dependence (total number of symptoms across all assessments),
