DESCRIPTION (provided by applicant): Prenatal exposure to cigarettes represents a potent risk factor for adverse behavioral outcomes including conduct problems and youth smoking. Existing research compellingly demonstrates that these increased behavioral risks are (a) robust to obvious confounding; (b) present across diverse contexts and; (c) consistent with evidence from basic science. Despite the robustness of these findings, considerable gaps remain in our understanding of mechanisms that link exposure to adverse behavioral outcomes. These gaps reflect fundamental methodologic limitations of studies to date and the need for theoretically-driven model development and testing that examines the role of characteristics of mothers who smoke during pregnancy in pathways from exposure to problem behavior in youth. The proposed study is designed to examine three questions fundamental to exploration of potential mechanisms and pathways that link prenatal cigarette exposure to later adverse behaviors: (1) What are the salient parameters of exposure that affect risk? (2) What is the specific nature of the problem behavior associated with prenatal exposure to cigarettes? and (3) How do the family contexts of prenatally exposed youth differ from those of non-exposed youth? The proposed research takes advantage of a unique opportunity to conduct a follow-up of the adolescent offspring of the Maternal-Infant Smoking Study of East Boston (MISSEB), a population-based cohort study with repeated, prospective biochemical and selfreported measurement of prenatal cigarette smoke exposure. Specific aims of the proposal are to: (1) examine the effects of variation in duration, timing and intensity of prenatal exposure on the development of youth conduct problems and smoking; (2) delineate specific types and patterns of problem behavioir" associated with prenatal exposure (including examination of sex differences) and; (3) characterize the family context of exposed youth and explore its potential contribution to pathways to problembehavior. 260 girls and boys (mean age=l 3) and their mothers will be studied prospectively over 4 years. The study will use existing MISSEB data on prenatal exposure and collect new data on problem behavior and family context from the MISSEB cohort at adolescence. Prenatal smoking patterns will be used as predictors of youth problem behavior. Multi-level modeling techniques will be used including survival analyses and longitudinal regression techniques. The family contexts of youth whose mothers smoked during pregnancy will be compared to those of youth whose mothers smoked but not during pregnancy and youth whose mothers are nonsmokers. The potential moderational role of family factors will be explored.