Background: Obese Latino adolescents are at increased risk for stress and depressive symptomology, but interventions to target both physiological and mental health outcomes are scarce. This pilot randomized controlled trial assessed feasibility and preliminary efficacy for a home-based strength training (HBST) intervention on stress-related and mental health outcomes in obese Latino adolescent males. Methods: Fifty-two adolescents were randomized to HBST (n = 26) or control conditions (n = 26). Participants randomized to HBST completed a 16-week progressive resistance training intervention performed twice per week at home on non-consecutive days. Primary and secondary outcomes were assessed at baseline and immediately post-intervention and included measures of upper- and lower-body strength (1RM), body mass index (BMI), BMI percentile, BMI Z-score, salivary cortisol, depressive symptoms (CES-D), and perceived stress (PSS-14). Results are presented using completer-only analyses (n = 25) and mixed-design ANOVA models. An ANCOVA sensitivity analysis was conducted for depressive symptoms due to baseline imbalance, including baseline CES-D as a covariate in the model. Results: Recruitment goals were met, but retention was lower than expected (48% overall; HBST = 31%, control = 54%). Analyses revealed a significant Time × Group interaction for salivary cortisol (F(1, 20) = 5.70, p = 0.027, ηp2 = 0.222), such that cortisol decreased over time in HBST participants and increased in control participants. While all strength and anthropometric outcomes improved descriptively from baseline to follow-up in the intervention condition, no significant interactions were present between groups. Depressive symptoms also decreased descriptively in HBST participants, but this effect was no longer significant after adjusting for baseline CES-D using ANCOVA (F(1, 19) = 0.002, p = 0.968). There were no significant findings for perceived stress. Conclusions: Differential effects were observed on salivary cortisol, suggesting HBST may be feasible in obese Latino adolescents. However, results should be interpreted with caution given baseline imbalance, small sample size, high attrition, and limitations with measuring cortisol at one time point without adjustment for time of day or key psychosocial and physiological confounders. All psychological and anthropometric outcomes were exploratory and non-significant after adjustment. A larger, multisite trial using baseline-adjusted analytic procedures, repeated physiological sampling, objective measures of adherence, and extended follow-up is needed to determine whether HBST produces meaningful effects that are sustained over time.
Kelly et al. (2026) studied this question.