Running towards injury? A prospective investigation of factors associated with running injuries

Abstract

Background Given the high incidence and heavy burden of running related injuries, large-scale, prospective multifactorial investigations examining potential risk factors are warranted. This study aimed to identify factors associated with running related injuries and to evaluate their potential in injury screening. Study design Prospective cohort study. Materials and methods Two hundred and seventy-four recreational runners were recruited. Clinical measures (strength, range of motion, foot position), injury and training history (via questionnaire), impact loading (via accelerometery) and running technique measures were collected at baseline. Runners were tracked for injury for one year via fortnightly check-ins. A binary logistic regression, (injury versus no injury), was performed for each variable univariably, and then adjusting for age, sex and mileage. A multivariable regression was also performed to evaluate the model’s discriminative ability. Results Of the 225 runners included in the final analysis 52% experienced a running related injury. Injury history in the past year, less navicular drop, and measures of running technique (knee, hip, and pelvis kinematics) were associated with increased odds of injury (p < .05). The multivariable logistic regression model was statistically significant, χ2(11) = 56.45, p < .001, correctly classifying 74% of cases with a sensitivity and specificity of 72% and 76%, respectively. The area under the receiver operating characteristic curve was 0.79 (CI95% = 0.73–0.85), demonstrating acceptable discriminative ability. Conclusions This study found a number of clinical and running technique factors to be associated with prospective running related injuries among recreational runners. With the exception of injury history, the factors identified as being significantly associated with injury may be modifiable and therefore, could form the basis of interventions. Range of motion, spatiotemporal parameters and strength measures were not associated with injury and thus their utilisation in injury prevention practices should be reconsidered.