Turning difficulty is common among people with Parkinson disease (PD) in the advanced stages. People with PD turn slowly with multiple small steps1 and often demonstrate freezing of gait (FOG).2 The small steps and FOG during turning further increase the risk of imbalance and falls.3 Therefore, since turning is an important component of functional upright mobility, it is important to reduce turning difficulty in people with PD for fall prevention.
Previous studies have shown that people with PD, especially those having FOG and turning difficulty, demonstrated greater gait variability and asymmetry than healthy adults during straight walking.4–6 Yogev et al4,5 found that people with PD had greater stride time variability and asymmetric leg swing times than the healthy controls, and a dual task (ie, walking with number subtracting) increased such variability and asymmetry more remarkable in people with PD than in the healthy controls. These findings suggest that the ability to generate a stable gait is less automatic and becomes more attention demanding in people with PD.6
Compared with walking in a straight line, asymmetric tasks such as turning involve greater challenge on bilateral limb coordination. As a result, the arrhythmic, asymmetric gait could be exacerbated during turning and even induce gait disturbances including small, shuffling steps and FOG. Indeed, empirical findings have shown that people with PD demonstrate a remarkable reduction in cadence during turning,1 and most gait disturbances in PD take place during turning.2 Therefore, it is possible that the turning difficulty is associated with gait variability and asymmetry.
The cognitive movement strategies are a set of compensatory skills to improve mobility function in people with PD.7,8 The principle of the cognitive movement strategies is to decompose a complex motor task into discrete steps, so people with PD can consciously plan and rehearse the task before execution.7,8 In addition, the cognitive movement strategies discourage dual tasking such as talking while walking, so people with PD can focus solely on a single motor task.7,8 The clock-turn strategy is an example of the cognitive movement strategies for turning in a narrow space.7 People with PD are taught to imagine a virtual clock, divide the whole turning angle into discrete parts, and complete each part with a step-to gait pattern. For example, a 180° turn is divided into 2 parts: first from 12 to 3 (or 9) o'clock and then from 3 (or 9) to 6 o'clock (Figure 1). It is claimed that the clock-turn strategy emphasizes attentional focus on decomposing turning into discrete step-to cycles and thus helps maintain a stable step pattern during turning.7
Despite having been documented in literature,7 the effectiveness and robustness of the clock-turn strategy have never been examined. Therefore, this study aimed to investigate the immediate effects of the clock-turn strategy on the pattern of turning steps, turning performance, and FOG during a narrow turning. We hypothesized that people with PD using the clock-turn strategy have lower step time variability and asymmetry than those who are not using the strategy, thus show better turning performance and less FOG. We further hypothesized that the aforementioned benefits of the clock-turn strategy would be compromised under dual task due to the competition for attentional resources between the motor (ie, the strategy) and cognitive tasks.
We recruited people with idiopathic PD from the neurology clinic of National Taiwan University Hospital. The inclusion criteria were as follows: (1) aged older than 50 years with intact cognitive function (Saint Louis University Mental Status Examination score ≥25)9; (2) able to walk without an assistive device for at least 30 m during the medication OFF period (8-12 hours after the last medication intake); and (3) having frequent FOG 1 week before the study (FOG Questionnaire score ≥8).10 People with PD having untreated musculoskeletal (eg, knee osteoarthritis) or cardiopulmonary diseases (eg, chronic obstructive pulmonary disease) that might affect walking were excluded. This study was approved by the research Ethics Committee of National Taiwan University Hospital, and written informed consent was obtained from all participants prior to the study.
This study focused on the immediate effects of the clock-turn strategy; therefore, we employed a cross-sectional, 2-group design, with one group using the clock-turn strategy and the other group not using the strategy. We conducted this study in the OFF period of PD medication, during which turning difficulty was more prominent, to emphasize the possible effects of the clock-turn strategy.
Participants with PD were asked to omit their morning dose of medication and participated in the study procedures around 9:00 to 12:00 AM, 12 to 15 hours after the last medication intake (OFF period).11 The baseline demographics (ie, age, sex, Hoehn-Yahr scale, PD duration), FOG symptom (ie, FOG Questionnaire), and walking function (ie, Timed 10-Meter Walk Test) were collected. The Hoehn-Yahr scale is a 5-point system to grade the progress of symptoms of PD, with a high score indicating severe motor symptoms.12 The FOG Questionnaire is a 6-item self-report tool to grade the frequency and duration of FOG during activities including walking, gait initiation, and turning in the recent week.10 The score on the FOG Questionnaire ranges from 0 to 24, with a high score indicating great FOG severity. The FOG Questionnaire has shown high test-retest reliability (intraclass correlation coefficient [ICC] >0.8) and strong correlation with the items of Unified Parkinson Disease Rating Scale, which monitors mobility function (r >0.8).10 The Timed 10-Meter Walk Test measures the preferred walking speed by recording the time required to traverse the intermediate 6 m of a 10-m walkway, excluding the acceleration (first 2 m) and deceleration (last 2 m) phases.13 The 10-Meter Walk Test has shown high test-retest reliability (ICC = 0.96) in people with PD.13
Participants were assigned to the clock-turn or usual-turn group by a dynamic randomization algorithm using age, Hoehn-Yahr scale, FOG Questionnaire, and Timed 10-Meter Walk Test as the stratification factors14 to match the FOG symptom and walking function between the groups. The adaptive randomization algorithm took into account the stratification factors of a newly entered participant and the previously enrolled participants and then assigned the new participant to the group that resulted in minimum imbalance between groups. For example, in a study with n groups, the algorithm first assigns the new participant to group 1 to calculate the imbalance score 1 (SCORE1) and then repeats the process for y times (SCORE1, SCORE2, ... SCOREy); the new participant will be assigned to the group that results in the minimum imbalance score. After group allocation, 39 infrared-reflexive markers were placed upon anatomical landmarks according to the Vicon plug-in-gait model.15
The Timed Up and Go (TUG) test was used to investigate the effect of the clock-turn strategy on the pattern of turning steps, turning performance, and FOG. Participants performing the TUG test stood up from a chair, walked 3 m, turned 180° around, and returned to the seat, walking in their comfortable speed.16 To increase the task difficulty, the 180° turn was confined within a narrow space (58 × 46 cm) marked at the end of walkway (Figure 1). This narrow turning specification replicated the scenario that the clock-turn strategy aimed to address. To investigate whether the effect of the clock-turn strategy was compromised by performing a cognitive task, the TUG test was conducted with (dual-task TUG) and without (single-task TUG) concurrently preforming a cognitive task. The cognitive task was an odd/even number classification test,17 given at the 180° turn during the TUG test via a wireless earphone. Participants were instructed that they might hear a verbal question during the TUG test, which contained 2 numbers (random from 11 to 99, exclude 20, 30, 40, ... 90). They should answer “Yes” if the numbers were either both odd (eg, 21 and 43) or both even (eg, 26 and 68), otherwise answer “No.” Participants were asked to respond as fast as possible without interrupting their stepping. Our pilot study showed that this number classification test was challenging enough to disturb gait stability but not sufficiently challenging to stop the performance of stepping.
Participants engaged in a 15-minute learning session to practice the number classification test (5 minutes) and the TUG test (10 minutes) after marker placement. There were 5 practice trials (3 single-task TUG tests, 2 dual-task TUG tests) at the end of the learning session to ensure participants were familiar with the tasks. In the experimental trials, each participant completed 3 valid single-task TUG tests and 3 valid dual-task TUG tests (random in sequence). The single-task and dual-task TUG tests were performed in random order to eliminate confounding by order (ie, where one type of trials is better than the other because it is conducted first). A 1- to 5-minute rest period was given between trials, depending on participant needs. The experimental trials were considered valid if the participants completed the TUG test (and cognitive task, if required) without losing balance and if the 180° turn took place within the 58 × 46-cm space.
The clock-turn group performed the 180° TUG turn using the clock-turn strategy (Figure 1). At the beginning, the clock-turn group watched a video demonstrating a TUG test with the clock-turn strategy, and a verbal instruction was given as follows:
This clock-turn strategy aims to reduce gait disturbances when turning in a narrow space such as bathroom or corridor. Image a virtual clock when reaching the spot of turning, and your forward, backward, rightward, and leftward are 12, 9, 3, and 6 o'clock, respectively. Now divide your turning arc in to a 12 → 3 → 6 (ie, turning right) or 12 → 9→ 6 (ie, turning left) o'clock sequence with 3 step-to cycles. Try to intentionally lift your turning steps high, like a marching solider.
The clock-turn group first practiced the strategy with a standing turn (ie, turning in place) and then practiced the strategy with a walking turn. In the 5 practice trials, successful learning was defined as the clock-turning strategy being employed at least 4 of 5 practice trials. In the event that a participant in the clock-turn group failed the aforementioned criterion, the participant performed addition practice trials until 4 successful trials were achieved out of the latest 5 attempts (ie, until the participant met the 4/5 criterion).
Participants in the usual-turn group, on the contrary, performed the 180° turn of the TUG test in their usual manner. Participants in the usual-turn group was instructed to perform the 180° turn in their own way but not step outside the specified 58 × 46-cm space. The usual-turn group practiced the TUG test freely during the learning session and completed 5 practice trials before the experimental trials.
An 8-camera 3-dimensional motion capture system (Bonita cameras, Vicon Motion Systems, Oxford, United Kingdom) with a sampling rate 120 Hz was used to track the infrared-reflexive markers placed on participants' anatomical landmarks. Participants were also videotaped by a 30-Hz frame rate digital video recorder (MV600i; Canon Inc, Tokyo, Japan), and the clips were reviewed offline by 3 experienced physical therapists to identify FOG episodes during experimental trials. A FOG episode was defined as “a brief, episodic absence or marked reduction of forward progression of the feet despite the intention to walk.”2p.872 Cohen's unweighted κ showed that the 3 therapists had good to excellent agreement (κ = 0.76-0.96).
The dependent variables in this study were grouped into 3 categories: pattern of turning steps, turning performance, and FOG severity (definitions are provided in Table 1). Values for participants in the same task type (single-task or dual-task TUG test) were averaged.
The pattern of turning steps was evaluated by foot clearance, step time variability, and step time asymmetry. Foot clearance was used to examine the height to which participants lifted their feet when stepping through the turns. Step time variability measures the temporal rhythmicity of turning steps. Step time asymmetry quantifies the temporal unevenness between the bilateral step times on a scale from 0 (perfect symmetry) to infinity (extreme asymmetry).
Turning performance was evaluated by turning time and number of turning steps. A short turning time and fewer turning steps indicate an efficient turning. FOG severity was evaluated by the number of FOG episodes and FOG duration, which were based on the video review. FOG episodes not within the turning were excluded. A low FOG number and short FOG duration indicate a less severe FOG.
Sample size calculation was based on the reported minimal detectable change of the TUG test in people with PD.18 Therefore, we assumed that a mean difference in the TUG test between the clock-turn and usual-turn groups would be 3.5 ± 2.9 seconds. Under this assumption, 13 participants per group would be required to reach a power of 0.8 at an α level of .05.
Seven 2-way mixed analyses of variance (ANOVAs) were employed to investigate the effect of group (clock-turn and usual-turn) and task (single task and dual task) on step pattern (3 variables: foot clearance, step time variability, and step time asymmetry), turning performance (2 variables: turning time and number of turning steps), and FOG severity (2 variables: FOG number and FOG duration). Both FOG number and duration were submitted to a rank transformation prior to ANOVA due to the violation of normal distribution.19 Tukey's HSD test was used for post hoc comparisons if significant group × task interaction was identified. Statistical tests were conducted by the SPSS software (version 16.0; SPSS, Inc, Chicago, Illinois), and the significance level was set to P < .05. The significance level of the ANOVA was not subject to the Bonferroni adjustment to avoid a surge risk of type I error.20
Twenty-five people with PD were enrolled during the approved study period (December 2012 to September 2013). The participants were allocated to either the clock-turn group (n = 12) or the usual-turn group (n = 13), and their baseline characteristics are given in Table 2. Statistics showed that the 2 groups were equivalent in demographics, PD duration, FOG symptom, and walking function.
All 12 participants in the clock-turn group learned the clock-turn strategy and passed the 4/5 criterion before taking the experimental trials. Four of them had 1, 2, 2, and 4 additional practice trials, respectively, to meet the criterion. The pattern of turning steps was evaluated for the foot clearance, step time variability, and step time asymmetry (Table 3). In the foot clearance, all group main effect, task main effect, and group × task interaction were significant. Inspection of the interaction profile plots indicated that dual task reduced the foot clearance in the clock-turn group whereas the usual-turn group was unaffected. The clock-turn group overall had greater foot clearance than the usual-turn group. In the step time variability, only group main effect and task main effect were significant, indicating that the clock-turn group had lower step time variability than the usual-turn group, and both groups had higher step time variability under the dual-task TUG test condition compared with the single-task TUG test. In the step time asymmetry, only the group main effect and task main effect were significant, indicating the clock-turn group had lower step time asymmetry than the usual-turn group, and both groups had higher step time asymmetry under the dual-task TUG test than under the single-task TUG test.
Turning performance was evaluated by the turning time and number of turning steps (Table 3). In the turning time, only the group main effect was significant, indicating that the clock-turn group took less time turning than the usual-turn group. In number of turning steps, on the contrary, all the group main effect, task main effect, and group × task interaction were not significant.
FOG severity was evaluated by FOG number and FOG duration (Table 3). A significant group main effect was found in both FOG number and FOG duration, indicating that the clock-turn group had lower FOG severity than the usual-turn group during the 180° turn of the TUG test. The task main effect and group × task interaction, on the contrary, were not significant in both FOG number and FOG duration.
Finally, we also recorded the time to perform the TUG test. The clock-turn group took 19.67 ± 3.55 and 21.20 ± 5.48 seconds to complete the single-task TUG and dual-task TUG tests, respectively. The usual-turn group, on the contrary, took 23.37 ± 7.94 and 24.17 ± 6.63 seconds to complete the single-task TUG and dual-task TUG tests, respectively. Statistic showed that all group main effect (F1,46 = 3.64, P = .063), task main effect (F1,46 = 0.44, P = 0.511), and group × task interaction (F1,46 = 0.05, P = 0.834) were not significant.
The clock-turn strategy is a cognitive movement strategy for turning in a narrow space,7 such as bathroom or corridor. We used the TUG test, which includes a 180° turn, to replicate the scenario of turning in a narrow space. Furthermore, we increased the task difficulty by specifying a narrow space (58 × 46 cm) for the 180° turn of the TUG test. Our results showed that the group that executed the TUG test using a clock-turn strategy for the turn had lower step time variability, lower step time asymmetry, faster turns, and reduced FOG severity compared with the usual-turn group. In both groups, step time variability and step time asymmetry were increased under the dual-task condition, but dual tasking did not significantly affect turning performance and FOG severity.
Effect of Clock-Turn Strategy
In this study, there were 2 features that were emphasized to ensure that participants learned the clock-turn strategy during practice: one was employment of the step-to pattern, and the other was encouraging a substantial increase in foot clearance during turning. As shown in Table 3, the clock-turn group had higher foot clearance than the usual-turn group, confirming that the strategy was employed. Our results suggested that the clock-turn strategy can be learned and employed during the OFF period. The clock-turn group had lower step time variability and step time asymmetry than the usual-turn group, suggesting that the clock-turn strategy improves gait rhythmicity and temporal symmetry during turning. People with PD are known to have difficulty performing complex sequential motor tasks.21,22 It is possible that the clock-turn strategy decomposed turning into several step-to cycles to downgrade the complexity of walking with changing direction. By adhering to the step-to pattern, the clock-turn strategy avoids the increase of step time variability and asymmetry during turning. It is also possible that the clock-turn strategy facilitated an attentional focus on “lifting foot high,” thus helped maintain a stable step pattern during turning. Yogev et al4,5 found that the ability to generate a rhythmic and symmetric gait became less automatic in people with PD. Therefore, a compensatory skill such as the clock-turn strategy that facilitates the attentional focus on movement rhythm/amplitude might avoid gait disturbances. Indeed, previous studies have shown that the attentional focus on “taking big steps” could reduce gait variability and shuffling steps in people with PD.23–25 Our results indicate that the advantages of attentional control that have been observed during walking in a straight line also apply to turning.
We found that the clock-turn group turned faster and had lower FOG severity than the usual-turn group, which may be associated with the reduced step time variability and asymmetry. One reason that people with PD turn slowly is that their cadence is substantially reduced,1 perhaps as the result of unsmooth transition from walking to turning. What is worse, people with PD may spend extra time struggling to reinitiate gait when FOG takes place. FOG has a close relationship with an unstable gait pattern, and it has been proposed that FOG might be due to a failed attempt to maintain a rhythmic, symmetric gait.26 This assumption is supported by the finding that FOG frequently follows a substantial increase of gait variability.27,28 By attentional focus on turning decomposition, the clock-turn strategy might reduce step time variability and asymmetry, thus helping people with PD turn faster and avoid FOG.
In number of turning steps, we found no significant difference between the clock-turn and usual-turn groups. It turned out that the clock-turn group took about 12 steps to turn 180° rather than the ideal 6 steps (ie, 3 step-to cycles). We speculated that the symptoms such as fatigue and weakness during the OFF period may contribute to the unexpected extra turning steps. Most of our participants reported fatigue during trials due to their medication dosing being overdue. This may have prevented participants in the clock-turn group from lifting their feet high and pivoting 90° at the same time. Instead, participants reduced the pivoting angle to ensure foot clearance, thus took extra steps to complete turning. The result may suggest that the cognitive movement strategy was compromised by the participants being in the medication OFF phase.
Effect of Dual Task
We found that the clock-turn group had lower foot clearance under the dual-task TUG test condition compared with the single-task TUG test, whereas the usual-turn group had no difference in foot clearance between the dual task and single-task conditions. It appears that dual tasking disturbed participants with PD conducting the clock-turn strategy, so the foot clearance was no longer maintained. Lohnes and Earhart29 reported that the increase of walking speed achieved by focusing on “taking big steps” was compromised by dual task. Our results were in line with Lohnes and Earhart29 and support the hypothesis that the clock-turn strategy would compete with cognitive task for attentional resources under the dual-task condition.
The step time variability and asymmetry significantly increased under the dual-task condition in both groups, indicating that dual task is detrimental to walking regardless the application of the clock-turn strategy. This is in line with the results reported by Yogev et al,4,5 who showed that the step time variability and step time asymmetry substantially increased under dual-task walking in people with PD. On the contrary, we found that the effect of dual tasking on turning time and number of turning steps was not significant. Previous studies have shown that step time variability and asymmetry are sensitive indicators of motor alteration in the early-stage PD, even before the presence of notable gait disturbances.30,31 Therefore, we speculate that the step time variability and asymmetry are more sensitive to the effect of dual task than is turning performance.
To our surprise, FOG number and FOG duration were not affected by dual tasking. Similar results were reported by Spildooren et al,11 who showed that dual task did not significantly increase the occurrence of FOG during 180° turning. The authors speculated that turning is a more important trigger of FOG than dual task. Our results supported the notion of Spildooren et al11 that dual task adds no extra risk of FOG during 180° turning. Future studies of the clock-turn strategy may benefit from adding dual tasking as a training condition.32
The first limitation of this study is that participants in the clock-turn group did not perform the single-task and dual-task TUG tests in their usual manner. We did not ask the clock-turn group to conduct trials both in the usual manner and using the clock-turn strategy because we found that participants had difficulty tolerating high numbers of repetitions (at least 12 trials) during their OFF period in the pilot study. Despite the limitation, the results of the FOG Questionnaire and the Timed 10-Meter Walk test showed that the 2 groups were matched at baseline; thus, the differences in outcome measures seem to be associated with employing of the clock-turn strategy.
The second limitation is that we did not measure the Unified Parkinson's Disease Rating Scale Motor score to confirm that the 2 groups were comparable in the level of motor deficits during their OFF period. We chose, instead, to use the Timed 10-Meter Walk test to ensure that the 2 groups were matched in walking function during the OFF period. A third limitation is that the cross-sectional design of this study merely addressed the immediate effect of the clock-turn strategy. Future studies are suggested to test whether people with PD can retain this skill and adopt it as part of their routine daily activities. Finally, this study enrolled 25 participants rather than the preset 26 during the approved study period. However, the effect of this unmet sample size might be negligible considering the differences identified between the clock-turn and usual-turn groups.
The clock-turn strategy appears to help people with PD turn faster and avoid FOG, perhaps by recruiting attentional control to reduce step time variability and asymmetry. In addition, dual task increases step time variability and asymmetry during turning in people with PD regardless of the employment of the clock-turn strategy. In our experience, participants could learn the skill after 4 to 8 sessions of supervised practice. Trials turning both to the left and to the right should be practiced to encourage maximum functionality. Although our participants learned the skill without visual enhancement to minimize study confounding, a cross mark on the floor specifying the directions of 12, 3, 6, and 9 o'clock may be a helpful visual cue.7
The clock-turn strategy reduced turning time and FOG during turning, probably by lowering step time variability and asymmetry. These beneficial effects could reduce the risk of imbalance and fall during turning. Dual task, however, compromises the effects of the clock-turn strategy, suggesting a competition for attentional resources.
The authors appreciate the assistance from orthotist/prosthetist Hung-Bin Chen, as well as physical therapists Hsin-Mao Yu and Li-Han Chang.
1. Crenna P, Carpinella I, Rabuffetti M, et al. The association between impaired turning
and normal straight walking in Parkinson's disease
. Gait Posture. 2007;26:172–178.
2. Bloem BR, Hausdorff JM, Visser JE, Giladi N. Falls and freezing of gait
in Parkinson's disease
: a review of two interconnected, episodic phenomena. Mov Disord. 2004;19:871–884.
3. Grimbergen YA, Munneke M, Bloem BR. Falls in Parkinson's disease
. Curr Opin Neurol. 2004;17:405–415.
4. Yogev G, Giladi N, Peretz C, Springer S, Simon ES, Hausdorff JM. Dual tasking, gait rhythmicity, and Parkinson's disease
: which aspects of gait are attention demanding? Eur J Neurosci. 2005;22:1248–1256.
5. Yogev G, Plotnik M, Peretz C, Giladi N, Hausdorff J. Gait asymmetry in patients with Parkinson's disease
and elderly fallers: when does the bilateral coordination of gait require attention? Exp Brain Res. 2007;177:336–346.
6. Plotnik M, Hausdorff JM. The role of gait rhythmicity and bilateral coordination of stepping in the pathophysiology of freezing of gait
in Parkinson's disease
. Mov Disord. 2008;23:S44–S50.
7. Morris ME. Movement disorders in people with Parkinson's disease
: a model for physical therapy. Phys Ther. 2000;80:578–597.
8. Keus SH, Bloem BR, Hendriks EJ, Bredero-Cohen AB, Munneke M. Evidence-based analysis of physical therapy in Parkinson's disease
with recommendations for practice and research. Mov Disord. 2007;22:451–460.
9. Tariq SH, Tumosa N, Chibnall JT, Perry MH, Morley JE. Comparison of the Saint Louis University Mental Status Examination and the Mini-Mental State Examination for detecting dementia and mild neurocognitive disorder—a pilot study. Am J Geriatr Psychiatry. 2006;14:900–910.
10. Giladi N, Tal J, Azulay T, et al. Validation of the freezing of gait
questionnaire in patients with Parkinson's disease
. Mov Disord. 2009;24:655–661.
11. Spildooren J, Vercruysse S, Desloovere K, Vandenberghe W, Kerckhofs E, Nieuwboer A. Freezing of gait
in Parkinson's disease
: the impact of dual-tasking and turning
. Mov Disord. 2010;25:2563–2570.
12. Zhao YJ, Wee HL, Chan YH, et al. Progression of Parkinson's disease
as evaluated by Hoehn and Yahr stage transition times. Mov Disord. 2010;25:710–716.
13. Steffen T, Seney M. Test-retest reliability and minimal detectable change on balance and ambulation tests, the 36-Item Short-Form Health Survey, and the Unified Parkinson's Disease
Rating Scale in people with Parkinsonism. Phys Ther. 2008;88:733–746.
14. Han B, Enas NH, McEntegart D. Randomization by minimization for unbalanced treatment allocation. Stat Med. 2009;28:3329–3346.
15. Dorociak RD. A comparison of normal kinematics for Vicon clinical manager and plug in gait. Gait Posture. 2002;16:S102–S103.
16. Podsiadlo D, Richardson S. The Timed “Up & Go”: a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 1991;39:142–148.
17. Yardley L, Gardner M, Bronstein A, Davies R, Buckwell D, Luxon L. Interference between postural control and mental task performance in patients with vestibular disorder and healthy controls. J Neurol Neurosurg Psychiatry. 2001;71:48–52.
18. Huang SL, Hsieh CL, Wu RM, Tai CH, Lin CH, Lu WS. Minimal detectable change of the Timed “Up & Go” test and the dynamic gait index in people with Parkinson's disease
. Phys Ther. 2011;91:114–121.
19. Conover WJ, Iman RL. Rank transformations as a bridge between parametric and nonparametric statistics. Am Stat. 1981;35:124–129.
20. Perneger TV. What's wrong with Bonferroni adjustments. BMJ. 1998;316:1236–1238.
21. Nieuwboer A, Rochester L, Müncks L, Swinnen SP. Motor learning in Parkinson's disease
: limitations and potential for rehabilitation. Parkinsonism Relat Disord. 2009;15:S53–S58.
22. Abbruzzese G, Trompetto C, Marinelli L. The rationale for motor learning in Parkinson's disease
. Eur J Phys Rehabil Med. 2009;45:209–214.
23. Canning CG. The effect of directing attention during walking under dual-task conditions in Parkinson's disease
. Parkinsonism Relat Disord. 2005;11:95–99.
24. Baker K, Rochester L, Nieuwboer A. The immediate effect of attentional, auditory, and a combined cue strategy on gait during single and dual tasks in Parkinson's disease
. Arch Phys Med Rehabil. 2007;88:1593–1600.
25. Smith BA, Jacobs JV, Horak FB. Effects of amplitude cueing on postural responses and preparatory cortical activity of people with Parkinson's disease
. J Neurol Phys Ther. 2014;38(4):207–215.
26. Nutt JG, Bloem BR, Giladi N, Hallett M, Horak FB, Nieuwboer A. Freezing of gait
: moving forward on a mysterious clinical phenomenon. Lancet Neurol. 2011;10:734–744.
27. Nieuwboer A, Dom R, De Weerdt W, Desloovere K, Fieuws S, Broens-Kaucsik E. Abnormalities of the spatiotemporal characteristics of gait at the onset of freezing in Parkinson's disease
. Mov Disord. 2001;16:1066–1075.
28. Chee R, Murphy A, Danoudis M, Georgiou-Karistianis N, Iansek R. Gait freezing in Parkinson's disease
and the stride length sequence effect interaction. Brain. 2009;132:2151–2160.
29. Lohnes CA, Earhart GM. The impact of attentional, auditory, and combined cues on walking during single and cognitive dual tasks in Parkinson's disease
. Gait Posture. 2011;33:478–483.
30. Baltadjieva R, Giladi N, Gruendlinger L, Peretz C, Hausdorff JM. Marked alterations in the gait timing and rhythmicity of patients with de novo Parkinson's disease
. Eur J Neurosci. 2006;24:1815–1820.
31. Lewek MD, Poole R, Johnson J, Halawa O, Huang X. Arm swing magnitude and asymmetry during gait in the early stages of Parkinson's disease
. Gait Posture. 2010;31:256–260.
32. Fritz NE, Cheek FM, Nichols-Larsen DS. Motor-cognitive dual-task training in persons with neurologic disorders: a systematic review. J Neurol Phys Ther. 2015;39(3):142–153.