Risk Of Running-related Injury Associated With Center Of Mass Acceleration Complexity

Abstract

Dynamical systems theory suggests that examining the complexity of biological signals may be more sensitive in differentiating between groups of varying health status, including predicting who may develop a running-related overuse injury (RROI). This theory has yet to be tested in prospective running studies. PURPOSE: To evaluate if changes in complexity, quantified by mean control entropy, of center of mass (COM) resultant acceleration during running is associatedwith RROI development. METHODS: 30 collegiate runners wore an ActiGraph GTX3+ during ‘easy’ training runs throughout a cross-country season. Clinician-diagnosed RROI were reported via an online survey. Complexity was calculated on the resultant acceleration time series from each run. Cox proportional hazards analysis assessed injury risk as a function of the mean complexity for each runner across the season and the change in complexity per week. RESULTS: Seven runners sustained an injury. Across all 30 participants over the season, the mean ± 1SD complexity was 0.623 ± 0.086 units. The mean change in complexity over the course of the cross country season was -0.0014 ± 0.0043 units per week. Although not statistically significant, each 0.1-unit increase in complexity was associated with a 1.65-fold increase in injury rate (95% CI 0.70-3.89, p=0.30) and a 0.01-unit increase in complexity per week was associated with a 1.50-fold increase in injury rate (95% CI 0.06-35.5, p = 0.80). CONCLUSION: These preliminary findings suggest that COM complexity could at least contribute to RROI detection strategies in cross-country runners. Given these preliminary results, additional prospective studies with larger sample sizes are necessary to further assess relationships between baseline complexity and changes in complexity during running training that can be monitored with wearable technology.