Using complex systems to understand running-related injuries.

This study aims to investigate anthropometric and training characteristics associated with injuries in runners using a complex approach. We conducted a cross-sectional study with 97 Brazilian runners of both sexes (age: 39.1 ± 11.5 years; body mass index: 24.2 ± 4.2 kg/m2). A web survey was used for data collection, including age, sex, body weight, body height, training volume/frequency, session duration, running pace and injury information (yes or no; the number of injuries). The partial correlation parameter was used for network estimation, considering both sexes. For women, injury occurrence was positively and strongly associated with training frequency (r = 0.57) and the number of competitions (r = 0.64), while for men, a negative and strong association was found with age (r=-0.69), and positive associations were observed with BMI (r = 0.89) and training volume (r = 0.82). Network centrality showed that training session duration and the number of competitions/year presented a higher expected influence value for women (-1.47) and men (-1.31). Using a non-linear approach with Brazilian non-professional runners, training characteristics had a stronger influence on running-related injuries.