Identifying Risk Profiles for Substance Use and Comorbid Behavior

点击翻译按钮获取中文摘要

DESCRIPTION (provided by investigator): Substance use and comorbid health-related behaviors among youth and young adults have a high cost to society. Although considerable progress has been made in developing effective intervention programs for prevention, their effectiveness depends to a large extent on a clear understanding of multiple risks as they coexist. Prevention scientists everywhere have access to rich sources of data on early and concurrent risk factors for developmental outcomes. Many of these data sets have been previously analyzed with the objective of examining risk factors, and in fact much progress has been made in the identification of developmental models that describe how individual and multiple risks contribute to disorders such as substance abuse. However, it is now possible to move beyond the traditional approaches to modeling risk that have been used previously in order to examine the complex interplay among risk factors at multiple levels. Intervention scientists stand to gain a powerful new understanding of risk by moving from a traditional 'risk factors' approach to a novel 'risk profiles' approach. The proposed research employs a relatively new and underutilized person-centered statistical technique, latent class analysis with covariates, to (a) identify nuanced multilevel risk profiles in existing empirical data and (b) establish the ability of the risk profiles to predict future problem behavior. The data sets to be analyzed are from two community-based and two national longitudinal studies and contain rich data on a variety of risk factors as well as substance use and comorbid behaviors. Gender and ethnic group differences will be explored in the prevalence of risk profiles and the link between risk profile membership and later health-related outcomes. The proposed research will fill an important gap in current knowledge about the interplay of multiple risks at multiple levels, thereby helping intervention scientists to develop more effective programs and to target those programs more effectively. A series of articles will be submitted to peer-review journals, and a project Web site will be added to the Methodology Center Web site at Penn State, where free SAS software for latent class modeling currently is available for download. The project site will provide information for intervention scientists on how to use latent class analysis to uncover risk profiles in their own data. Substance use and comorbid health-related behaviors among youth and young adults have a high cost to society. Although considerable progress has been made in developing effective intervention programs for prevention, their effectiveness depends to a large extent on a clear understanding of multiple risks as they coexist. The proposed research will fill an important gap in current knowledge about the interplay of multiple risks at multiple levels, thereby helping intervention scientists to develop more effective programs and to target those programs more effectively.