An SPSS mixed model analysis was performed on the clusters of brain structures showing significant (FWE corrected) differences in our whole brain VBM analysis. The volumes of these brain structures were extracted using MarsBaR and used in SPSS based mixed model analysis to provide statistical control for participants from families with multiple siblings using a family ID variable. Using the same covariates used in the VBM analysis, we investigated the effect of familial risk and the effect of prenatal alcohol exposure on gray matter volume. HR subjects were compared to LR subjects, and within the HR group, those with prenatal alcohol exposure were compared with HR subjects who had no or low prenatal alcohol exposure. The cumulative estimates of personal use of alcohol, drugs and cigarettes available from longitudinal data collection were used as quantitative variables in the SPSS analyses.
