Somatosensory perception sensitivity in voluntary postural sway movements: Age, gender and sway effect magnitudes

1) to develop a reliable device for assessing somatosensory perception sensitivity in voluntary postural sway movement, specifically a sway discrimination apparatus (SwayDA) for testing voluntary lateral sway discrimination sensitivity (VLSDS); 2) to explore the relationship between mobility performance and VLSDS in older adults, and 3) to determine the effects of age, gender and sway magnitude on VLSDS.

First, eighteen healthy young adults (8 males, 10 females, age ranging from 22 to 70) were recruited for a test-retest reliability study. During the SwayDA test, the participants were asked to discriminate between four possible medial-lateral sway extents when moving away from neutral standing. For Objective 2, twenty-five older participants (9 males, 16 females, mean age 70.1) undertook both the SwayDA and the mobility tests. The mobility testing battery consisted of single task and cognitive dual task timed-up and-go tests, and the 5 times sit-to-stand test. Pearson's correlation was calculated between SwayDA scores and mobility performance. For Objective 3, 20 community-dwelling adults over 65 years old (10 males, 10 females, mean age 71.3) and 20 young volunteers (10 males, 10 females, mean age 23.6) were recruited to study the effects of age, gender and sway magnitude on VLSDS. To obtain a bias-free measure of VLSDS, the probability of correct response was considered as the true-positive judgment, while the probability of incorrect response was considered as false-positive judgment, and these were cumulated across the response values. A receiver operating characteristic (ROC) curve was then generated and the Area Under the ROC Curve (AUC) was used to measure VLSDS.

There was no significant difference in AUC scores between Day 1 and Day 8 (p > 0.05). ICC(3,1) reliability indices were 0.750 for sway to the left and 0.879 for sway to the right. Pearson's correlation revealed a significant correlation between the SwayDA sores and timed-up-and-go (TUG), cognitive dual task TUG, 5 times sit-to-stand test (r = -0.456, -0.522, and - 0.416 respectively, all p < 0.05). Factorial ANOVA showed age and gender main effects (F = 8.144, p < 0.01, and F = 8.806, p < 0.01, respectively), suggesting older adults and females had worse VLSDS. In addition, a significant difference was found between the young and older participants in the inner range of VLSDS (t = -2.875, p < 0.017), indicating that the decline of somatosensory perception of postural sway in older people may be magnitude-specific, and greatest for small deviations from upright stance.

The SwayDA has good to excellent test-retest reliability. The finding that VLSDS score was significantly correlated with mobility performance in older adults highlights the importance of somatosensory perception in postural control. More importantly, the significantly worse VLSDS in older people observed in the inner lateral movement range may represent a unique characteristic of neuromuscular degeneration associated with aging, which should be monitored and addressed in rehabilitation programs.