Few studies have examined patterns of marijuana use from adolescence to young adulthood, particularly among African Americans. African American high school students consistently exhibit higher rates of marijuana use than Whites, and in the past decade rates of abuse and dependence have increased among young adult African Americans, but not their White counterparts. Limited research suggests marijuana use continues well into young adulthood for African Americans with greater long-term consequences. There is also evidence that African Americans initiate marijuana use prior to cigarettes in contrast to the 'gateway hypothesis';a recent study suggests this pattern of initiation may be associated with poorer physical environments, greater exposure to drugs and more neglectful parents. Previous work suggests that differential rates and patterns of marijuana use in African Americans are attributable, in part, to differences in the importance of various ris factors. Risk factors for marijuana use, however, have not been studied extensively in African Americans, and therefore, it is unclear if the risk factors found to be important for Whites can generalize to African Americans. Further, neighborhood context which is particularly relevant for low-income urban youth who are persistently exposed to drug activity, disorder and violence, represents an important but equally understudied risk factor. The goal of this application is to conduct a series of new analyses of longitudinal data from the 2nd generation Johns Hopkins Prevention Intervention Research Center (JHPIRC) trial to examine the role of individual, social and neighborhood level factors on: (1) the process underlying developmental transitions in marijuana use from earliest opportunities to use the drug down through the later stages of drug dependence in a sample of African American youth followed from 1st grade to age 25, and (2) the clustering of marijuana use behaviors in the urban neighborhoods where they live. Our goals will be accomplished using two innovative statistical techniques~ latent transition analysis (LTA) for studying empirically-derived stages of marijuana use and alternating logistic regressions (ALR) for estimating and modeling the magnitude of clustering in neighborhoods. This is one of the few studies that follows African Americans from childhood to young adulthood and includes extensive measures of individual, family and peer factors including annual self-reported measures of the neighborhood environment and more recently, an innovative field-rater assessment of the urban, primarily economically-disadvantaged, neighborhoods where they live Findings from this innovative and cost-effective project could have a significant impact on the development of culturally-appropriate, community-wide prevention programs targeted at urban-dwelling African Americans for whom effective programs are urgently needed.
African Americans exhibit higher rates of marijuana use than Whites and their rates of abuse and dependence are increasing. By using innovative statistical techniques and measures of the social and neighborhood environment, new understandings of the course of marijuana use and the influence of factors that are more salient in this population will lead to the development of culturally appropriate prevention programs targeted at urban-dwelling African Americans for whom effective programs are needed.