Prevalence and predictive factors of musculoskeletal injuries in triathletes: a cross-sectional study

Triathlon combines swimming, cycling, and running over varying distances and has grown in popularity in recent years. Despite this, few studies investigate the prevalence and risk factors of injuries. This study aimed to analyze injury prevalence and distribution across body regions in Brazilian triathletes, and to identify predictive factors associated with injury development. A cross-sectional study was conducted with 758 Brazilian triathletes through an online form. Data collected included sociodemographic and anthropometric information, sports participation duration, injuries in the past 12 months, and number of competitions. Descriptive statistics were applied, with group comparisons using t-test, Mann-Whitney, or Chi-square tests. Binary logistic regression identified injury predictors, with variables (p < 0.20) entered into multiple regression via forward selection. Model fit was assessed using R² Cox & Snell, R² Nagelkerke, and ROC curve analysis. Of the 758 triathletes, 614 (81%) were male. More than half (n = 425) reported an injury in the past 12 months, most frequently in the lower limbs, particularly the knee (16.88%). Logistic regression revealed associations between injury occurrence and receiving prevention guidance from a physician (OR = 0.512; 95% CI: 0.320–0.818; p = 0.005), physiotherapist (OR = 0.249; 95% CI: 0.178–0.348; p < 0.001), and physical education professional (OR = 0.338; 95% CI: 0.244–0.467; p < 0.001). Participation in Ironman 70.3 distance in the last year increased injury likelihood (OR = 1.018; 95% CI: 1.216–2.339; p = 0.002). The ROC curve demonstrated good discriminative capacity (AUC = 0.759; 95% CI: 0.724–0.793; p = 0.001). Lower limb injuries were the most prevalent among Brazilian triathletes. Injury prevention guidance from physicians, physiotherapists, and physical education professionals was associated with reduced risk, whereas competing in Ironman 70.3 distance increased injury likelihood. The predictive model showed good accuracy.