This is a revised application to support secondary analyses of 5 data sets collected to test Problem Behavior Theory (PBT) as well as a sixth data set, the National Longitudinal Study of Adolescent Health (Add Health). The goal of the proposed analyses is to establish empirically the stages of adolescent involvement with alcohol and to investigate antecedent psychosocial risk factors for transitions between the resultant stages. Stages of alcohol involvement in adolescence, such as experimental drinking or heavy drinking, have been arbitrarily defined rather than empirically determined. The variety of operational definitions of each stage has led to confusion in the field. These 6 data sets employed a consistent set of alcohol behavior questions assessing ever use, frequency of drinking, usual intake, high-volume drinking, drunkenness, and negative consequences of drinking. Proposed analyses address the following 8 aims.
Aim 1 : Latent Class Analysis (LCA) will be employed to determine the empirical latent classes of adolescent involvement with alcohol in the most recent data set.
Aim 2 : Resultant latent classes of alcohol involvement will be cross-validated using the other PBT data sets collected earlier.
Aim 3 : Gender, age, and ethnic-racial differences in the structure of the latent classes of adolescent alcohol involvement will be examined using data from the five PBT data sets.
Aim 4 : Two 4-wave PBT data sets collected 20 years apart will be examined to describe the natural history of adolescent involvement in these stages.
Aim 5 : The two 4-wave PBT data sets will be further analyzed using Latent Transition Analysis (LTA) to investigate the antecedent risk factors for transitions from one stage to another.
Aim 6 : Gender differences in the antecedent risk factors for these stages will be examined in the two longitudinal PBT data sets.
Aim 7 : Data from the national Add Health study will be used to further test the replicability of the classes and stages of change in adolescent involvement with alcohol.
Aim 8 : Add Health data will be used to investigate racial/ethnic differences in these classes and in the risk factors for transitions among the stages. Establishing the stages of adolescent alcohol involvement and the risk factors for transitions among these stages should facilitate the design of more effective prevention and intervention efforts in this population.
| Donovan, John E; Chung, Tammy (2015) Progressive elaboration and cross-validation of a latent class typology of adolescent alcohol involvement in a national sample. J Stud Alcohol Drugs 76:419-29 |
| Donovan, John E; Molina, Brooke S G (2013) Types of alcohol use experience from childhood through adolescence. J Adolesc Health 53:453-9 |
