Effects of balance and gait training with augmented cues and feedback on falls in people with Parkinson's disease

Abstract

Background: Fall prevention is essential for people with Parkinson's disease (PD) because of the high incidence of falls in that population. Falls result in devastating outcomes including physical injuries such as bone fractures and negative psychological effects such as fear of falling, and these could lead to self-imposed functional restrictions, deconditioning and early institutionalisation. Few studies have examined the effects of exercise training on fall prevention among those with PD, especially over the long-term. We designed a balance and gait training programme with augmented cues and feedback. The primary objective was to compare the effects of the balance and gait training programme on preventing falls with strength training programme in short-term and long-term over a period of 12 months after training completion, in people with PD.The secondary objective was to compare the effects of the balance and gait training programme on enhancing postural stability and alleviating fear of falling in short-term and long-term over a period of 12 months after training completion,in people with PD. We hypothesized that the innovative balance and gait training is better than strength training on decreasing fall rate and number of fallers,improving their postural stability,and alleviating their fear of falling over both the short and long terms.Method: This was a randomized, controlled experiment with the tester blinded to the group assignment of participants.Fifty-one eligible participants were randomly allocated to an experimental group receiving balance and gait training with augmented cues and feedback (the BAL group, N=26) or an active control group which received lower limb strengthening exercises (the CON group, N=25). The training in each group lasted for 12 weeks, with 4 weeks at a university followed by 4 weeks of self-administered training at home and another 4 weeks at the university.At the university,the balance and gait training was augmented with audio-visual cues and feedback on each participant's performance in the BAL group,whilst the CON participants received strength training with gym-based equipments. At home, participants in the BAL and CON groups performed balance and gait exercises and strengthening exercises respectively. All of the participants were followed up for 12 months after the treatment ended. The primary outcomes were fall rate and the number of fallers. The secondary outcomes were time to first fall,single-leg-stance (SLS) time, limits of stability(LOS),the latency and amplitude of postural response,spatio-temporal gait characteristics, and Activities-specific Balance Confidence (ABC) scores. The participants were assessed before the intervention (PRE),immediately after training (POST),then 3 months (FU3m) and 12 months after the treatment ended (FU12m). All of the tests were conducted during the participants' "on" medication state. The data were analyzed based on the intention-to-treat principle. Negative binomial regression models and Pearson χ2 tests were used to compare the between-group differences in the fall rate and the ratio of fallers to non-fallers respectively. Two-way repeated-measure analyses of variance were used to analyze the effects on the secondary outcomes with the normally distributed data. For the data not normally distributed,Wilcoxon tests and Mann-Whitney U tests were applied.Multivariate analysis of variance was used to investigate the between-group differences in the changes for all secondary outcomes from the baseline values. The significance level was set at 5%. The Bonferroni correction was used in multiple comparisons. Results: Forty-five participants who completed the 12-week training were included in the data analysis. Participants in two groups demonstrated comparable characteristics in demographics,fall history,balance and gait outcomes at baseline. For fall prevention,the BAL group exhibited lower fall rates (IRR=0.111 at POST, 0.188 at FU3m, 0.407 at FU12m, p<0.05 at POST and FU3m, p=0.057 at FU12m) and had significantly fewer fallers (RR=0.190 at POST, 0.285 at FU3m, and 0.483 at FU12m, p<0.05) than the CON group. For short-term effects on postural stability and fear of falling at POST,the BAL group individuals had significantly increased their SLS time, endpoint excursion of LOS,gait velocity and stride length, reduced the latency and amplitude of their postural response (all p<0.017),and marginally increased the ABC score (p=0.025). The CON participants only significantly increased their endpoint excursion of LOS, gait velocity and cadence (all p<0.017).When the changes from baseline were compared,the BAL group showed significantly more reduction in the latency of postural response (by 7% vs. 1%, p<0.05), more increase in the stride length (by 13% vs. 3%, p<0.05) and marginal more increase in the SLS time than the CON group (by 44% vs. 20%, p=0.064). For the long-term effects on postural stability and fear of falling at FU3m and FU12m, the BAL group had sustained all the improvements achieved after training (p<0.017)except the improved latency of postural response at FU3m and FU12m,and the amplitude of postural response and the endpoint excursion of LOS at FU12m.In addition, their ABC score had further increased to reach a significant level at FU3m and FU12m (p<0.017). The CON group only maintained their increase in gait velocity and cadence at FU3m and FU12m (p<0.017).When compared the changes from the baseline at FU3m with the CON group, the BAL group demonstrated significantly greater improvement in the SLS time (by 45% vs. 7%), movement velocity of LOS(by 17% vs. -6%), latency of postural response (by 5% vs. 2%) and stride length (by 9% vs. 3%) than the CON group (all p<0.05). At FU12m, the BAL group demonstrated significantly greater improvement from the baseline than the CON group in the stride length (by 12% vs. 2%),latency of postural response (by 6% vs.3%) and ABC score (by 6% vs. -1%) (all p<0.05).Conclusion: This is the first study to report that 12-week balance and gait training with augmented cues and feedback reduces falls,enhances postural stability and alleviates their fear of falling over a period of at least 15 months. The positive findings support the clinical use of balance and gait training with augmented cues and feedback in fall prevention among people with PD.