Efficacy of Tandem Gait to Identify Impaired Postural Control after Concussion

The purpose was to evaluate tandem gait (TG), Balance Error Scoring System (BESS), and modified Balance Error Scoring System (mBESS) performance acutely after concussion in collegiate student-athletes. In addition, we sought to evaluate the psychometric properties of TG, including minimal detectable change (MDC), sensitivity, and specificity.

Seventy-six National Collegiate Athletic Association student-athletes performed TG and BESS tests: 38 acutely after concussion and 38 controls. Participants were tested at baseline (time 1) and again acutely after concussion, or the following year for controls (time 2). Ten controls, tested simultaneously by two researchers, established a TG interrater minimal detectable change. A 2 × 2 mixed-design ANOVA compared each outcome variable. An receiver operating characteristic curve analysis was used to evaluate sensitivity, specificity, and area under the curve (AUC).

There was a significant interaction (F = 8.757, P = 0.004) for TG whereby the concussion group was slower after concussion (10.59 ± 1.53 vs 11.80 ± 2.67 s), whereas there was no difference for controls (10.13 ± 1.72 vs 9.93 ± 1.85 s). There was no significant interaction for BESS (F = 0.235, P = 0.630) or mBESS (F = 0.007, P = 0.935). TG had a sensitivity of 0.632, a specificity of 0.605, and an AUC of 0.704. BESS had a sensitivity of 0.447, a specificity of 0.500, and an AUC of 0.508. mBESS had a sensitivity of 0.474, a specificity of 0.632, and an AUC of 0.535.

Participants completed TG significantly slower after concussion, whereas no change across time was detected for controls. In contrast, BESS and mBESS performances were similar at both testing times in both groups. Our AUC analysis was acceptable for TG, but a failure for both BESS and mBESS; thus, TG may be a useful alternative for clinicians conducting postconcussion postural control assessments.