Sport participation as a mediating factor between substance use and self-harming behaviour among Australian adolescents: a prospective cohort study

Technical Summary: Sport Participation as a Mediating Factor Between Substance Use and Self-Harming Behaviour

Problem Statement
Adolescence is a critical developmental period characterized by increased risk-taking behaviors, including substance use, self-harm, and suicidality. These behaviors are frequently interrelated, with substance use identified as a primary risk factor for self-harm and suicidal behaviors in Australian adolescents. While sports participation is generally viewed as a protective factor for mental health, existing evidence regarding its specific role in mitigating the harmful effects of substance exposure is mixed. Furthermore, the literature lacks clarity on whether the type of sports participation—specifically distinguishing between individual and team sports—differentiates its protective capacity against self-harm and suicidality in the context of substance use. This study addresses the gap in understanding whether sports participation acts as a mediating factor in the relationship between substance use and self-harming behaviors among Australian adolescents.

Methodology
This study employed a secondary analysis of data from the Growing Up in Australia: The Longitudinal Study of Australian Children (LSAC), specifically utilizing Waves 6, 7, and 8. The analytic sample consisted of 2,040 adolescents (aged 14–19 years) from the Kindergarten (K) cohort who provided complete data across all variables of interest.

• Variables:

  • Independent Variables (X): Exposure to four substance categories: alcohol, tobacco smoking, marijuana, and other drugs (including sniffing). These were measured via self-report over the preceding 12 months.
  • Dependent Variables (Y): Self-harm (Y1) and Suicidality (Y2). Self-harm was defined as intentional self-injury in the past 12 months. Suicidality was a composite variable derived from serious consideration of suicide and the formulation of a suicide plan.
  • Mediating Variables (M): Sports participation, categorized into Team Sports (e.g., football, cricket) and Individual Sports (e.g., swimming, tennis, martial arts). Participation was recorded as "Yes" if the adolescent engaged in the activity at any point during Waves 6–8.
  • Covariates: Age, sex, area of residence (metropolitan, regional, remote), and socioeconomic position (based on the Index of Relative Socioeconomic Advantage and Disadvantage).

• Analytical Approach:
  The study utilized Structural Equation Modelling (SEM) based on a modified Baron and Kenny approach to test mediation effects. The analysis followed a three-step process:

  1. Step 1: Assessment of statistical significance for Path A (Substance → Sport), Path B (Sport → Outcome), and Path C (Substance → Outcome).
  2. Step 2: If Path A and Path B were significant, the Sobel test was conducted to determine the significance of the indirect effect.
  3. Step 3: Interpretation of mediation (partial, complete, or none) based on the significance of the Sobel test and the direct effect (Path C).
     All models were adjusted for the specified covariates.

Key Results
Descriptive statistics indicated that 62.8% of the sample participated in team sports and 44.3% in individual sports. Substance use prevalence was 58.5% for alcohol, 41.1% for smoking, 16.8% for marijuana, and 2.8% for other drugs. Self-harm and suicidality were reported by 18.5% and 23.2% of the sample, respectively.

• Team Sports Mediation:
  The analysis revealed that team sports participation significantly mediated the association between substance use (specifically alcohol and tobacco) and both self-harm and suicidality.

  • Path A: Alcohol and tobacco use were significantly associated with team sports participation.
  • Path B: Team sports participation was significantly negatively associated with self-harm and suicidality.
  • Path C: Substance use was significantly associated with the outcomes.
  • Mediation Effect: The Sobel test confirmed significant indirect effects for alcohol and tobacco, indicating that team sports partially mediate the relationship between these substances and self-harm/suicidality.
  • Non-Results: No mediating effect was found for marijuana or other drugs, as the association between these substances and team sports (Path A) was not statistically significant.

• Individual Sports Mediation:
  Individual sports participation did not demonstrate a mediating effect on the relationship between any substance use and self-harm or suicidality. Path A (Substance → Individual Sports) was not statistically significant for any substance category, preventing the analysis from proceeding to the Sobel test for mediation.

Key Contributions
The study provides empirical evidence distinguishing the protective mechanisms of team versus individual sports in the context of adolescent risk behaviors. It contributes to the literature by:

1. Identifying that the mediating role of sports participation is specific to team-based activities rather than individual pursuits.
2. Demonstrating that this protective mediation is specific to alcohol and tobacco use, rather than illicit substances like marijuana or other drugs.
3. Utilizing a large, nationally representative longitudinal dataset to model complex mediation pathways between substance exposure, sports engagement, and severe mental health outcomes.

Significance and Claims
The authors conclude that team-based sports participation may play a protective role against self-harm and suicidality among adolescents exposed to alcohol and tobacco use. The paper posits that the social aspects of team sports—such as peer connection and social cohesion—are likely the mechanisms driving this protective effect, rather than the physical activity itself.

The study highlights the importance of integrating team sports into mental health promotion and suicide prevention strategies. It suggests that the decline in sports participation observed in recent years could exacerbate risks associated with substance use and self-harm. However, the authors maintain a modest tone regarding the complexity of these relationships, noting that cultural norms (e.g., the link between team sports and alcohol in Australia) and potential selection biases (healthier adolescents being more likely to participate in sports) must be considered. The paper does not claim that sports participation eliminates risk but rather that it serves as a significant mediating factor that can mitigate the association between specific substance exposures and self-harming behaviors.