The purpose of this study was to assess gait initiation (GI) performance longitudinally across clinical concussion recovery milestones through return to participation (RTP). We recruited 54 collegiate student-athletes, 27 with concussions and 27 matched controls (15 Female and 12 Male per group). Participants performed 5 trials of GI at baseline and again at five post-concussion clinical milestones: 1) Acute, the day clinical tests achieved baseline values on the 2) Balance Error Scoring System (BESS), 3) ImPACT, 4) Asymptomatic, and 5) RTP day. GI performance on six outcome measures (A/P and M/L Center of Pressure displacements and velocities during the anticipatory postural adjustment (APA) phase and initial step length and velocity) with repeated measures mixed model and pairwise post-hoc. A reliable change index (RCI) was calculated and post-concussion participant’s performance was compared to the RCI at milestones. There were significant interactions for APA posterior and lateral displacement, APA posterior velocity, step length, and step velocity. The post-hoc tests identified significant deficits across clinical milestones and at RTP for APA posterior and lateral displacement, step length, and step velocity. There were no post-hoc differences for any outcome measure in the control group. At RTP, 85.2 – 88.9% of concussion participants had at least one outcome measure which exceeded the 80% or 95% RCI. The primary finding of this study was persistent impairments in dynamic postural control suggesting ongoing neurophysiological impairment despite clinical recovery. These results suggest that collegiate student-athletes may be RTP prior to neurophysiological recovery and potentially exposing themselves to elevated risk of recurrent concussion or subsequent musculoskeletal injury.