Substance-Related Decision-Making Trajectories in Fitness Contexts: A Person-Centered Longitudinal Study.

BACKGROUND: Preventing image- and performance-enhancing drug (IPED) use in recreational fitness settings remains a public health concern. Although theory-informed interventions have been associated with lower risk acceptance and substance-related behavior, heterogeneity in underlying decision-making processes remains insufficiently understood. Identifying distinct psychosocial configurations may inform more differentiated prevention research. METHODS: A three-wave cluster-randomized longitudinal field study was conducted across ten commercial fitness studios in Germany (N = 624 at baseline; n = 602 after data-quality screening; n = 550 with complete three-wave data). Participants had a mean age of 28.4 years (SD = 7.2); 60% identified as male, 39% as female, and 1% as diverse. The intervention comprised three sessions targeting social norms, risk appraisal, and self-regulation. Risk acceptance, protection motivation, intention to remain substance-free, and perceived behavioral control were assessed at baseline (T1), during the intervention period (T2), and follow-up (T3). Latent Transition Analysis (LTA) estimated profile membership and transition probabilities, with intervention condition and baseline psychosocial configuration included as covariates. Cluster-robust standard errors and cluster bootstrapping (B = 1000) were applied. Booster use at follow-up served as an external behavioral criterion. RESULTS: A three-profile solution was retained (BIC = 11,841; entropy = 0.83): High-Protection/Low-Risk (44.7%), Moderate/Ambivalent (38.2%), and High-Risk/Low-Protection (17.1%). Intervention condition was associated with higher estimated odds of the High-Risk/Low-Protection → Moderate/Ambivalent transition (OR = 2.46, 95% CI [1.34, 4.52], p = 0.003), with the association differing between participants with ambivalent and stable protective baseline configurations (interaction p = 0.028). Membership in the combined High-Protection/Low-Risk and Moderate/Ambivalent category was associated with lower odds of booster use than membership in the High-Risk/Low-Protection Profile (OR = 0.42, 95% CI [0.27, 0.65], p < 0.001). CONCLUSIONS: Findings indicate heterogeneity in model-estimated psychosocial configurations and transition probabilities over time. The results support considering psychosocial heterogeneity in the study and evaluation of prevention approaches in recreational fitness settings.