Background and purpose: Understanding the individual risk factors for injury in soccer is paramount to develop preventive measures. Therefore a multivariate model is needed to examine the contribution of the various factors in injury etiology and to explore their interrelationship. The purpose of the study was to assess whether different factors (anthropometrics, flexibility, strength and endurance, balance and proprioception, lower limb function test) could be identified as risk factors for injuries in a multivariate model.

Relevance: Physiotherapists would be measuring intrinsic risk factors of athletes to identify their predisposition to injury.

Participants: 502 male soccer players from the highest division in Rwanda were followed up.

Methods: During the 2014-2015 preseason period, we examined the following factors: age, experience, body mass, stature, lower limb flexibility, calf endurance, dynamic balance, hop and lower limb function test. The technical and medical staff prospectively recorded exposure and injury data respectively for the period of two seasons.

Analysis: Binary Logistic Regression and Chi-squared Automatic Interaction Detection (CHAID) were performed to assess the impact of variables, identified as significant in the univariate analyses, on the likelihood of sustaining an injury.

Results: Players that scored 11cm or less on the Sit and Reach test were at higher risk of sustaining overall injuries (odds ratio [OR] = 0.95; 95%CI: 0.92-0.96; P<0.001).  A timed hop test of more than 2.5 seconds constituted a risk for overall injuries (OR = 2.23; 95%CI: 1.09-4.57; P=0.028). Lower limb injury was also associated with a smaller range in the Sit and Reach test and those with a range of 2cm or less were particularly at risk (OR = 0.95; 95%CI: 0.93-0.98; P=0.001). Duration of more than one year playing in the current team was also associated with a greater incidence of lower limb injury (OR = 1.25; 95%CI: 1.01-1.56; P=0.042). For thigh injuries, the significant risk factors were decreased flexibility of back structures and the hamstring (OR = 0.94; 95%CI: 0.90-0.98; P=0.006) and quadriceps (OR = 0.92; 95%CI: 0.89-0.96; P0.001). For ankle injuries, the predictor risk factors were poor balance (OR = 0.94; 95%CI: (0.91-0.97); P<0.001).

Conclusions: The flexibility of the back structures and the hamstring muscles as well as balance and co-ordination emerged as being significant predictors of an increased risk of injury among first division soccer players in Rwanda.

Implications: Assessment of intrinsic risk factors may help identify those at risk for lower extremity injuries to inform preventive interventions.