Friendship Networks and Emergence of Substance Use

DESCRIPTION (provided by applicant): The proposed renewal of the PROSPER Peers Project will improve efforts to prevent adolescent substance use and young adult sexual risk behaviors for HIV/AIDS by using the remarkable peer network data of the PROSPER prevention trial to advance knowledge of network influences during an age span that is largely unaddressed by prevention research and programming. To date we have studied friendship networks and the emergence of substance use among over 10,000 students across 5 waves of data for 6th-9th grade. In this continuation, we propose coding and analyzing 3 additional waves of peer network data through 12th grade and merging these data with outcome data at age 19 to 24, producing the largest and longest-term dataset on friendship networks and substance use in the world. The first aim of the project is to examine the contributions of friendship networks t substance use throughout late adolescence, adding a period of rapid growth in the rate and seriousness of risky behavior. The additional years of data will enable us to move beyond the most common substances (alcohol and tobacco) to use bringing greater harmful consequences for health and life prospects. The longer time span also will permit us to break new ground by examining the relation of trajectories of positions in friendship networks (e.g., popularity, isolation) to varying trajectories of use (e.g., earlier versus later onset; maintenance; escalatio). Our second aim is to expand the scope of our investigation to include the emerging domains of romantic relationships and depression to generate knowledge about how preventive interventions can best combat their contributions to substance use. We will examine romantic partnerships as bridges that may facilitate diffusion of substance use across distant pockets of peer networks. Depression also becomes considerably more prevalent during the high school years, and when combined with substance use, it foretells more problematic substance use. Our network data enable us to test the three proposed sources of depression's association with substance use using considerably more rigorous methods than prior work. Our third aim is to examine how adolescent friendships and substance use jointly contribute to sexual risk behavior that carries a high risk of HIV/AIDS, during the peak age of acquisition of HIV/AIDS, which is 20-24 years old. Both peers and substance use are associated with risky sexual behavior in adolescence, but little is known about their connection to sexual risk behavior at the key later ages. The final aim of our project is to test how friendship networks contribute to long-term impacts of prevention programs. Our study has the unique advantage of being embedded in the PROSPER community-level randomized trial of adolescent substance use prevention that has been shown to reduce substance use through at least 12th grade, and our prior work demonstrated that PROSPER succeeded in altering peer networks to reduce the potential for peer influence toward substance use. The additional years of data will enable us to assess the durability of program network effects and their role in mediating long-term effects.