Inflection points in longitudinal models: Tracking recovery and return to play following concussion

Area of Science:

• Sports Medicine
• Biostatistics
• Neurology

Background:

• Longitudinal studies tracking concussion recovery often overlook the return to play (RTP) process.
• Omitting RTP as a factor can negatively impact statistical model fit and interpretation.
• Standard statistical designs may fail to capture crucial recovery changes around RTP.

Purpose of the Study:

• To compare two linear mixed-model designs for analyzing concussion recovery data.
• To demonstrate the impact of incorporating an RTP inflection point into statistical models.
• To evaluate how different statistical approaches affect the interpretation of recovery findings.

Main Methods:

• Two linear mixed-model designs were applied: a standard between-group longitudinal (GROUP) analysis and an RTP analysis incorporating an inflection point.
• Instrumented balance data from 23 concussed athletes and 25 controls over 8 weeks post-concussion were analyzed.
• Outcome measures included total sway area and range of mediolateral acceleration.

Main Results:

• The standard GROUP design yielded no significant findings for either outcome measure.
• The RTP analysis revealed significant effects of time, RTP change, and group interactions for both total sway area and range of mediolateral acceleration.
• The RTP model demonstrated a significantly better fit to the data compared to the GROUP model (likelihood ratio tests, P ≤ 0.027).

Conclusions:

• Incorporating an inflection point for RTP in statistical designs enhances understanding of recovery around critical clinical time points.
• The choice of statistical model significantly influences concussion recovery study interpretations.
• This study advocates for advanced statistical methods that account for clinical milestones like RTP in longitudinal data analysis.