In spite of the foregoing considerations, our secondary hypothesis that sport participation characteristics would be related to injuries was only partly confirmed, in that none of those variables tested in our regression analysis were significant. One reason could be that our approach is rather global, in the sense that sport participation characteristics were determined as average values over the period of interest, while sports injuries may be more related to the changes of weekly load and behavior. Indeed, short-term changes of sport participation characteristics did reveal a higher weekly intensities, as well as a trend for an increased number of weekly sessions prior to injury (Table 2).