Assessing the Influence of Healthy Leisure On South African Adolescent Substance Use Trajectories: A Growth Mixture Modeling Approach

Adolescent substance use has been associated with long- and short-term risk outcomes including risky
sexual behaviors, school dropout, future substance abuse, violence, and suicide. Leisure time provides
a context where adolescents are presented with multi-faceted opportunities to engage in risky and
healthy behaviors. Leisure research suggests that adolescents who have access to and can make use of
structured recreation opportunities are at lower risk for substance use problems. Adolescents living
in contexts with few leisure/recreation opportunities and where adolescents have large amounts
of discretionary time may be at increasingly high risk for substance use. In this paper we examine
how leisure experiences, specifically healthy leisure, may influence adolescent patterns of substance
use. Making use of a growth mixture modeling framework and rich longitudinal data obtained from
a sample of at-risk South African youth, we (1) identify distinctive developmental patterns of South
African adolescent substance use and (2) examine if and how those trajectories may be influenced by
adolescents’ ability to make healthy use of their leisure time.

The current study analyzed data from South African adolescents serving as control participants in the
HealthWise effectiveness trial where respondents had between 5 and 8 waves of data collected every
6 months between 8th and 11th grade (N=3,654). Associations were examined between each of two
healthy leisure factors, including adolescents’ initial efficacy for healthy leisure (α=0.83; i.e., I feel good
about myself in my free time) and healthy leisure planning/decision-making (α=0.85; i.e., I know how to
plan my free time activities), and multiple trajectories of substance use (measured as a composite index
of alcohol, tobacco, marijuana, methamphetamines, and inhalants).

Latent class growth analyses revealed the presence of three stable and three increasing trajectory
groups: Abstainers, Consistent Experimenters, Consistent Users, Escalators, Early Escalators, and
Late Escalators. In line with theoretical prediction, adolescents with initial higher efficacy for healthy
leisure were more likely to be in the Consistent Experimenters (OR=1.38), Abstainers (OR=1.21), or
Late Escalators (OR=1.25) than in the Early Escalators class. There were no class differences for healthy
leisure planning/decision-making. Findings help to better understand differing trajectories of substance
use within this population, support the need to intervene in early adolescence to prevent substance
use behaviors, and emphasize the importance of addressing leisure skills central to healthy leisure
engagement, especially in developing contexts such as South Africa.