Age-related changes in postural control in older women: transitional tasks in step initiation

There was no group effect on 1^st phase in unperturbed crossing, crossing the obstacle, and step-up and step-down conditions (p > 0.05).

There was a group effect on transit phase in unperturbed crossing, crossing the obstacle, and step-up conditions (Table 2). The greatest differences were observed between group A and C. Group C presented significantly higher values for TT, S1, S2, and DST under unperturbed crossing, crossing the obstacle, and step-up conditions compared to group A. Additionally, group B obtained significantly higher values of TT and S2 relative to group A while crossing the obstacle. Group B also showed significantly higher values of S1 and S2 during the step-up condition compared to group A. The least differences were observed between group B and C. Group B presented significantly lower value only in DST during unperturbed crossing, crossing obstacle compared to group C. There were no significant differences between groups during the step-down condition.

There was a group effect on 3^rd phase in unperturbed crossing, crossing the obstacle, and step-down conditions. The greatest differences were observed between group A and B. Group B presented significantly higher values for v-COP and std-COP under unperturbed crossing and crossing the obstacle conditions compared to group A. During step-down condition group B obtained significantly higher values of v-COP compared to group A. There were no significant differences between groups during the step-up condition.

There was no significant impact of testing conditions on v-COP, and std-COP in all three groups in the first phase (p > 0.05).

There was a significant effect with regard to the testing conditions on TT (Chi2 = 16.74; p < 0.001), S1 (Chi2 = 22.53; p < 0.001), S2 (Chi2 = 11.70; p < 0.008), and DST (Chi2 = 34.74; p< 0.001) in group A. The subjects presented the longest TT, S1, and S2 during the step-down condition. At the same time, the results showed significantly higher values for DST during the step-up condition relative to other conditions (Fig. 3).

There was a significant effect with regard to the testing conditions on TT (Chi2 = 9.54; p < 0.05), S1 (Chi2 = 16.32; p < 0.001), S2 (Chi2 = 11.10; p < 0.05), and DST (Chi2 = 42.31; p < 0.001) in group B. Group B obtained the highest values of TT and S2 during the step-up condition. The same changes were observed when analyzing the DST variable, whereas the highest S1 values were achieved among subjects during the step-down condition (Fig. 3).

There was no significant effect with regard to the testing conditions on all analyzed variables except for DST (Chi2 = 41.04; p < 0.001) in group C. The subjects obtained the longest time for DST during the step-up condition and the shortest time for DST during the step-down condition (Fig. 3).

There was a significant impact of testing conditions on v-COP (Chi2 = 24.06; p < 0.001) and std-COP (Chi2 = 16.26; p = 0.001) in group A. The subjects obtained the highest values of v-COP during step-up condition and the highest values of std-COP during step-down condition (Fig. 4). There was no significant effect of testing conditions on v-COP and std-COP in group B and C (p > 0.05).

The findings in our study comprised no significant difference between all three groups of individuals in vCOP and stdCOP during quiet standing before the transitional task. In spite of previous study, where authors noticed differences between young and older adults in COP during quiet standing [15], our results show that these trials were not enough to detect age-related changes in postural control. However, we have noticed differences between people in middle-age and adults over 60 y/o during quiet standing after completing the transitional task. These results confirmed that dynamic task challenging postural control and change the postural stability. Hence, in our study the older adults presented less stable posture after the movement relative to middle-age women, what indicate that adults over 60 y/o exhibit balance deficits. In addition, our analysis showed a significantly greater velocity and standard deviation of COP displacement in quiet standing after the step-up and step-down trials in middle-age women compared to both older adults' groups. As expected, unperturbed transit and crossing obstacle were the simplest tasks for adults in middle-age. This observation suggests that simple motor task do not hamper postural control in younger women. However, the transitional task was very demanding for both older adults' groups independent of trial condition. Hence, it seems that postural deficit are associated with less efficient postural control during more difficult motor tasks.

Step initiation consists of an anticipatory postural adjustment (APA) and a stepping phase, both of which are impaired in the older adults. Our results demonstrate that the postural preparation time (S1) was longer in adults over 60 y/o and increased by 17% compared to middle-age subjects. These results suggest that APA might be impaired by age-related physiological changes, which are reflected in reduced somatosensory and visual information. In addition, there is evidence that older people present less variability in muscle activity than younger adults during the anticipatory phase [18], which also confirms that the older adults are unable to respond effectively to balance perturbations [19]. Thus, impaired postural preparation comes with a potential for balance loss in the older population.

In this study, all parameters (TT, S1, S2, DST) increased with age. Also, in previous studies [17] all these parameters were higher in Parkinson’s Disease participants, who present well documented severe balance and gait deficits [20, 21]. In addition, the longer S1, the longer the time of TUG performance. Moreover, the longer the S1, the less points patients gained in BBS, Tinetti test or FRT, which is associated with increased risk of falling and impaired stability. Similar relationships can be noted in TT, which also shows numerous negative correlations with the FRT, Tinetti and BBS [16]. Therefore, we assumed that higher values of the measured parameters indicate impaired postural control during the performance of a transitional task. The transit time was longer by 30% in subjects over 70 y/o, which lead to prolonged DST, S1, and S2. In previous studies, the authors analyzed various gait parameters in older adults, mostly gait speed, step length, and stride length, and rarely double-support time [22, 23]. Some authors [24] observed that adults aged 65–79 y/o present a 20% slower gait speed relative to young adults (20–25 y/o). The decrease in gait speed may reflect a protective adaptation to a perceived threat to stability, as the center of mass must be accelerated from a stationary state, and the relatively small base of support in the first step [24]. Although gait speed has been identified as an important predictor of the onset of immobility and balance disorders among older population [25‐27] our results support the idea that double-support time might also be considered as a valuable marker of age-related changes in locomotion. Moreover, there is convincing evidence that DST is highly correlated with gait speed [28]. In our study, DST was longer by 11% among adults 60-70 y/o, but in people over 70 y/o, it was longer by almost 50% compared to middle-age subjects. Since decreased gait speed predicts balance deficits in the older adults, we assumed that a longer DST, which is associated with slower gait speed, indicates impaired postural control during gait among the older women.

Additionally, we have noticed a longer regained stability time (S2) in adults over 60 y/o with respect to middle-age participants. The S2 time increased by 18% among adults 60-70 y/o, but in participants over 70 y/o, it increased by 32%. Gait termination changes the gait patterns and thereby threatens the stability of the older adults [29], and the older people also generated less braking force than the middle-aged group [30]; therefore, it takes more time to regain a stable posture after movement. In addition, there is evidence that older women exhibit motor decline, which is due by the lower muscle strength relative to younger women. Hence, the older group exhibited longer balance recovery times compared to the younger women [14, 31]. Our results support these findings; we have noticed that adults over 70 y/o present difficulties with regaining stability after a motor task.

In addition, all of the measurements changed across the different conditions. Our results demonstrate that independent of age, TT, S1, and S2 increased while crossing an obstacle. An explanation of these findings consists of the fact that stepping over obstacles increases gait challenges at every age, even in the middle-age population [32, 33]; however, this motor task is still more demanding for the older adults compared to young adults [34]. Previous studies reported that the older people used a more conservative strategy for crossing obstacles relative to young adults, including a slower crossing speed and higher foot clearance while crossing over obstacles [11, 35]. Our study supports these findings, as adults over 60 y/o present a longer TT; in other words, they needed more time for negotiating an obstacle. In addition, TT increased in middle-age adults, which corresponds with recent studies [36]. However, the changes in obstacle crossing in the middle-aged group were observed only in the most challenging tasks (obstacle height 26cm) [26].

Additionally, we noticed that TT increased during step-up and step-down conditions. There is evidence that ascending and descending stairs is a hazardous activity of daily life for adults over 60 y/o [12]. However, our findings show that the postural control alteration while negotiating stairs already occurs in middle-age. The preparatory time and the regained stability time increased in adults over 50 y/o, which is a very important sign, because it may mark the onset of age-related gait changes. Moreover, we observed several changes in DST. Independent of age, we noticed longer DST in the step-up condition compared to flat crossing, which may be an indication of balance disorders imposed on the older women during stair climbing. Prolonged DST provides evidence that older adults need to spend more time on double limb support before transitioning to a single support phase while stepping up. These changes are also already apparent in adults 50-60 y/o, further evidence proving the early signs of age-related gait changes.

In addition, during step initiation before descending stairs, the older adults present decreased stability relative to younger participants [37]. Bosse et al. [37] claimed that the older people generated less braking forces while descending stairs, thus they sway forward like a pendulum instead of controlling the movement of center of mass; this offers evidence that the older adults may not be able to effectively reduce their body sway before the initiation of stepping down. In our case, the most surprising issue was the decreased DST in every age group of adults during the step-down condition. On the basis of the above literature, we assumed that adults with impaired postural control are not able to control the forward COP movement, therefore they shorten the double support phase during the descending step task. Moreover, in the women over 70 y/o every task was difficult; we have not noticed differences between all conditions. Therefore, we assumed that the adults in the 70-80 y/o category would present advanced postural control impairments during transitional task.

A limitation of the study was that our procedure included one step, while in other studies the subjects usually performed a few steps. This one step did not reflect exactly the same conditions as during normal daily life situations. However, our procedure is the simplest and less complicated compared to standard measures of gait initiation. Furthermore, in our study, we investigated only three ranges of age; therefore, we recommend that future studies should include more age groups to better assess the onset of age-related gait changes.