Progress in our ability to evaluate the interaction between genetic effects and environmental factors in understanding risk for nicotine dependence has been significantly impeded by inadequate single study sample sizes and meta analyses that are necessarily limited to the estimation of effect sizes for research questions that have been previously addressed within single studies. Fortunately, recent methodological innovations now permit the combination of data sets that differ in terms of study design, populations, and measures. This integrative data analysis (IDA) allows us to pool the raw data from multiple studies, allowing us to test new hypotheses and results in more powerful, more comprehensive, and more rigorous studies than we achieve through analysis of single studies or by using traditional meta analytic techniques. The present research will combine both an innovative approach (IDA) and innovative statistical methods (moderated nonlinear factor analysis; MNLFA and time-varying effects models; TVEM), to evaluate the association between genetic variants in the CHRNA5/CHRNA3/CHRNB4 and CHRNA6/CHRNB3 regions and nicotine dependence symptoms across levels of smoking exposure. Taking advantage of existing resources, we will pool extant data from the Social Emotional Contexts of Adolescent Smoking Project (SECASP), The National Longitudinal Study of Adolescent Health (AddHealth), the National Youth Survey Family Study (NYSFS) and the Study of Addiction: Genetics and Environment (SAGE) to evaluate changes in the association between SNPs in the acetylcholine receptor gene clusters and an empirically harmonized nicotine dependence score across levels of smoking exposure and to estimate the additional contribution of timing of smoking exposure in explaining the association between individual genetic variants and nicotine dependence symptoms. The innovative methods of data integration with MNLFA will allow us to begin to explore exposure varying effects in an extremely cost-effective manner in that we will leverage previous investment in tracking, contacting, assessing, and collecting DNA from 4 large studies assessing adolescents and adults. The empirically based nicotine dependence symptom score derived from the cross-study MNLFA model will provide a much more sensitive measure of the nicotine dependence phenotype than traditional methods of scoring within study, and one that is uniquely harmonized to account for between study differences.
The present research will combine both an innovative approach (IDA) and innovative statistical methods (moderated nonlinear factor analysis; MNLFA and time-varying effects models; TVEM), to evaluate the association between variants in the CHRNA5/CHRNA3/CHRNB4 and CHRNA6/CHRNB3 gene clusters and nicotine dependence symptoms across levels of smoking exposure. Resulting evidence will provide guidance as to the feasibility of using this approach to guide personalized interventions in specific environments (e.g. smoking exposure and/or timing of exposure) based on a person?s genetic background. Given that tobacco use is the single most preventable cause of death in the world and is associated with increased risk of several cancers, this study has the potential to have an important impact on public health.
