Abstract
Background:
Following concussion, individuals often adopt a conservative gait pattern, characterized by slower, more cautious movements. However, standard clinical concussion assessments do not detect subtle gait alterations. Although quantitative gait analysis shows promise for identifying latent impairments, especially under dual-task conditions which require simultaneous cognitive/motor task execution, traditional approaches (e.g., motion capture) are costly and laboratory-bound. Smartphones contain inertial sensors and offer portable, cost-effective, valid alternatives for gait analysis. Yet, their utility to detect post-concussion gait alterations remains unexplored. We investigated spatiotemporal gait parameters among adolescents with and without concussion using a novel, validated smartphone application under single- and dual-task conditions.
Hypothesis:
Adolescents with recent concussions will exhibit more conservative spatiotemporal gait parameters (lower steps/cadence, shorter step length/time, slower gait speed, and lower step length symmetry), on average, compared to uninjured controls. Group differences will be greater under dual-task conditions.
Methods:
In a cross-sectional, case-control study, adolescents with and without recent concussion (<21 days post-injury) completed demographic/medical history questionnaires and standardized gait assessments. Gait assessments involved 1-minute of continuous walking along a 10-meter path at self-selected speeds under single- and dual-task conditions. Dual-task walking included a continuous verbal fluency task. A smartphone affixed to the lower back recorded accelerometer data to calculate spatiotemporal gait parameters (steps, cadence, step length, step time, gait speed, and step length symmetry). Between-group comparisons were performed using independent samples t-tests and Cohen’s d effect sizes. Multivariable linear regression models evaluated associations between group and gait parameters, adjusting for age, sex, concussion history, and musculoskeletal injury history.
Results:
We enrolled 179 participants: 124 with recent concussion (15.8±1.5 years; 51% female; 13.5±4.7 days post-injury) and 55 uninjured controls (15.9±1.5 years; 69% female). Compared to controls, participants with concussion took fewer steps and walked slower under both single- and dual-task conditions (Figure – A, E). During dual-task walking, they demonstrated lower cadence and longer step time (Figure – B, D). After covariate adjustment, group differences remained for single- and dual-task step length and dual-task gait speed; no other parameters were associated with group (Table).
Conclusion:
Smartphone-based gait analysis identified univariable spatiotemporal differences between adolescents with and without concussion. Although these differences were altered in adjusted models, highlighting concussion complexity, our findings support inclusion of smartphone-based quantitative gait analysis in multimodal concussion assessment. Future research should investigate longitudinal applications of smartphone-assessed gait analysis to characterize concussion recovery and inform return-to-activity decision-making.
