A Training Program to Improve Gait While Dual Tasking in Patients With Parkinson's Disease: A Pilot Study

A Training Program to Improve Gait While Dual Tasking in Patients With Parkinson's Disease: A Pilot Study
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  CLINICAL NOTE  A Training Program to Improve Gait While Dual Tasking inPatients With Parkinson’s Disease: A Pilot Study Galit Yogev-Seligmann, MScPT, Nir Giladi, MD, Marina Brozgol, PT, Jeffrey M. Hausdorff, PhD ABSTRACT. Yogev-Seligmann G, Giladi N, Brozgol M,Hausdorff JM. A training program to improve gait while dualtasking in patients with Parkinson’s disease: a pilot study. ArchPhys Med Rehabil 2011;xx:xxx.Impairments in the ability to perform another task whilewalking (ie, dual tasking [DT]) are associated with an in-creased risk of falling. Here we describe a program we devel-oped specifically to improve DT performance while walkingbased on motor learning principles and task-specific training.We examined feasibility, potential efficacy, retention, andtransfer to the performance of untrained tasks in a pilot studyamong 7 patients with Parkinson’s disease (PD). Seven patients(Hoehn and Yahr stage, 2.1  0.2) were evaluated before, after,and 1 month after 4 weeks of DT training. Gait speed and gaitvariability were measured during usual walking and during 4DT conditions. The 4-week program of one-on-one trainingincluded walking while performing several distinct cognitivetasks. Gait speed and gait variability during DT significantlyimproved. Improvements were also seen in the DT conditionsthat were not specifically trained and were retained 1 monthafter training. These initial findings support the feasibility of applying a task-specific DT gait training program for patientswith PD and suggest that it positively affects DT gait, even inuntrained tasks. The present results are also consistent with thepossibility that DT gait training enhances divided attentionabilities during walking. Key Words:  Attention; Case report; Gait; Parkinson dis-ease; Rehabilitation; Training; Transfer.©  2011 by the American Congress of Rehabilitation Medicine A LTERATIONS IN THE ABILITY to walk while simulta-neously dual tasking (DT) (ie, performing another task)have been related to gait and balance impairment. 1,2 The neg-ative effects of DT have been associated with an increased risk of f alling, underscoring the clinical importance of DT abili-ties. 3-6 DT while walking frequently occurs in many everydaysituations. Given its impact on gait, safe ambulation, and fallrisk, it is important to develop rehabilitation-like programs forreducing the negative impacts of DT. To date, only a fewstudies have attempted to specifically address this issue, andthe findings have not always been consistent. 7-10 Still, promis-ing results suggest that training improves DT gait in older adults with balance impairment 9 and patients with dementia. 10 Patients with Parkinson’s disease (PD) have both motor andcognitive impairments, making them especially vulnerable toDT. Motor learning is also impaired in PD, although it is stillpossible, 11,12 raising questions about the possibility of DT gaittraining in this patient population. An investigation 13 of theeffects of a single training session showed improvement in gaitspeed and step length during many of the DT conditions thatwere tested, but little effects on gait variability. The authorsconcluded, “The gait variable emphasized during DT trainingappears to be an important factor in enabling the transfer of training improvements across tasks.” (p.229) 13 A small pilotstudy 14 reported improvement in gait speed after DT training,while transfer to untrained tasks and the effects of training ongait variability were not examined. These studies suggest thatDT training is feasible, but many questions remain. Evidence-based training programs specifically designed to improve gaitthat could be readily applied in the clinical setting are notcurrently available, and it is not yet clear whether such pro-grams can be practically applied in patients with both motorand cognitive deficits such as patients with PD. In addition,there is currently a lack of a standardized protocol directlyimplementing motor learning principles.In the present study, we aimed to evaluate (1) the feasibilityof task-specific training of gait and DT in patients with PD,applying basic elements of motor learning principles; and (2)the suitability of the selected outcome measures for furtherresearch. This program differs from previous training pro-grams 7,9 in that it is focused only on DT gait training, applyingthe task-specific training principle—that is, to improve a skill,one should train that specific skill. We also explored whethersuch training shows evidence of motor learning as reflected inretention and transfer to the performance of untrained tasks,and examined both gait speed and variability. METHODSParticipants Seven patients with idiopathic PD 15 were recruited fromlocal outpatient clinics. Patients were included if their disease From the Movement Disorders Unit, Department of Neurology, Tel-Aviv SouraskyMedical Center, Tel-Aviv (Yogev-Seligmann, Giladi, Brozgol, Hausdorff); the Dr.Miriam and Sheldon G. Adelson Graduate School of Medicine (Yogev-Seligmann),and Departments of Neurology (Giladi) and Physical Therapy (Hausdorff), SacklerFaculty of Medicine, Tel-Aviv University, Tel-Aviv, Israel; and Department of Medicine, Harvard Medical School, Boston, MA (Hausdorff).No commercial party having a direct financial interest in the results of the researchsupporting this article has or will confer a benefit on the authors or on any organi-zation with which the authors are associated.Correspondence to Jeffrey M. Hausdorff, PhD, Laboratory for Gait Analysis &Neurodynamics, Movement Disorders Unit, Tel Aviv Sourasky Medical Center, 6Weizman St, Tel Aviv, 64239 Israel, e-mail:  jhausdor@tasmc.health.gov.il.  Reprintsare not available from the authors.0003-9993/11/xxx-00426$36.00/0doi:10.1016/j.apmr.2011.06.005 List of Abbreviations DT dual taskingFAB Frontal Assessment BatteryKP knowledge of performanceKR knowledge of resultsMoCA Montreal Cognitive AssessmentPD Parkinson’s diseaseTMT Trail-Making TestUPDRS Unified Parkinson’s Disease Rating ScaleVF verbal fluency 1 Arch Phys Med Rehabil Vol xx, Month 2011  stage was 2 to 3 on the Hoehn and Yahr scale, if they weretaking antiparkinsonian medications, and if they were between50 and 90 years of age, able to ambulate independently, and didnot have dementia (using  Diagnostic and Statistical Manual of  Mental Disorders, Fourth Edition  criteria). Subjects were ex-cluded if they had clinically significant musculoskeletal, car-diovascular, or respiratory disease, other neurologic disease,major depression, or uncorrected visual disturbances. Overview of Assessment Protocol After providing written informed consent as approved by thelocal human studies committee, subjects were evaluated withrespect to cognitive function, balance, and gait. At the end of the initial evaluation meeting, subjects were scheduled for12 training sessions (3 times a week for 4 weeks), for the post-training testing (2–3d after the last training session), and fortesting of retention effects (30d after the end of the training). Cognitive Assessment The Montreal Cognitive Assessment (MoCA) 16 provided ageneral measure of cognitive function. The Frontal Assessment Battery (FAB) 17 and the Trail-Making Test (TMT) 18 evaluatedexecutive function, a cognitive domain related to DT abili-ties. 1,19 Walking Protocol During the pretesting and posttesting, gait was assessedunder 6 walking conditions:1.  Usual walking with no DT:  This condition was alwaysfirst and allowed subjects to become familiar with thewalkway. In the following 3 walking conditions (walk-ing conditions 2–4), subjects were asked to perform 3different cognitive tasks that were later included in thetraining program.2.  Verbal fluency (VF) task:  Subjects were asked to recallas many words as possible beginning with a predefinedletter, during 1 minute, while walking.3.  Serial 3 subtractions task:  Subjects walked while per-forming serial 3 subtractions out loud, starting from a3-digit number (eg, 487).4.  Information processing task:  This task was adopted fromthe Mindstreams cognitive assessment battery. 20 Partic-ipants are presented with simple arithmetic problems andare instructed to respond as quickly as possible and statewhether the outcome is larger or smaller than 4.5.  Performance of an additional DT not included in train-ing session:  This walk evaluated the performance of anadditional DT that was not included in the trainingsession and was only assessed in the preevaluation andpostevaluation sessions to assess transfer of learning to anontrained DT. This task consisted of open-ended, com-plex questions that simulated everyday conversation andrequired a level of reasoning. For example: “List 3reasons against or in favor of being vegetarian.” “Name3 traditional customs of a (certain) holiday.” “Explainhow to get by car from the hospital to your neighbor-hood.”6.  Usual walking with no DT:  This was always the lastwalking condition and could be used to monitor foreffects of fatigue or training.During these 6 walks (and during all training), subjects wereinstructed to walk at their usual walking, preferred pace onlevel ground in a well-lit, obstacle-free, 30 m-long corridor for1 minute. No explicit instruction for prioritization of either task was given. The order of the DT conditions was randomlyassigned. The Timed Up and Go test 21 and the motor part of theUnified Parkinson’s Disease Rating Scale (UPDRS) 22 evalu-ated mobility, disease severity, and parkinsonian symptoms. Quantification of Gait An ambulatory recorder and footswitches were used to quan-tify the temporal parameters of the gait cycle. The systemconsisted of a pair of insoles a and a recording unit. b Thelightweight recording unit (11  7  3cm, 230g) was carried onthe waist. Measurements were transferred to a computer forfurther analysis where stride time variability and swing timevariability were determined. 3,23 Briefly, variability was quan-tified using the coefficient of variation; for example, stride timevariability    100    (SD/average stride time) after automati-cally processing the time series. 3,23 The results were similar forstride time variability and swing time variability; therefore, forbrevity, we report only the former. Gait speed was determinedby measuring the average time the subject walked the middle10m of the corridor. To assess whether the training programimproved (ie, reduced) the interference effect, the DT interfer-ence was calculated as follows: ([single task parameter – DTparameter]/single task)  100%. Statistical Analyses To estimate and explore the pre-post training effects and anyretention effects, we used the Wilcoxon test for pairwise,nonparametric comparisons. A  P  value of .05 or less wasconsidered statistically significant. Effect sizes (Cohen’s  d  24 )were calculated for the changes in gait speed and variability inthe posttraining evaluation. According to Cohen’s conven-tion, 24 effect sizes greater than 0.8 and 0.5 are considered largeand medium, respectively. Intervention Description  Rationale and principles behind the training pro- gram.  The possibility of DT training (where both tasks arenot walking) has been examined in several neuropsychologicstudies. 25,26 Two main strategies emerge: the part-task strategyinvolves training each task separately, while the whole-task strategy involves simultaneous training of both components of the DT. Although the findings are somewhat controversial,there appears to be added benefits to training of both tasks atthe same time. Furthermore, DT training as a whole task wasapparently critical for acquiring attentional control and task coordination strategies. 26 Consistent with this, a functionalmagnetic resonance imaging study reported reduced activationin brain areas that were initially involved with DT processingafter whole-task training, interpreted as an “increase in neuralefficiency.” 27 In addition to the concept of practicing DT as a “wholetask,” several additional principles of motor learning wereintegrated into the construction of the present program: ●  Task-specific training:  Practice should be specific to thedesired function (ie, one has to practice DT walking toimprove DT walking). Both animal and clinical studieshave demonstrated that task-specific training yields betteroutcomes and induces and maintains larger corticalchanges, compared with traditional therapies that targetmore general abilities. 28 ●  Feedback may enhance learning.  Since practicing DTinvolves implicit processes, we chose to provide feedback of knowledge of results of the cognitive tasks that werepracticed. ●  Intensity:  Erickson et al 27 demonstrated that 5 hours of DTtraining (not while walking) resulted in improvement, 2  DUAL TASK TRAINING IN PARKINSON’S DISEASE, Yogev-SeligmannArch Phys Med Rehabil Vol xx, Month 2011  compared with control subjects who did not receive anypractice. However, the intensity needed to practice DTwhile walking is unknown. Studies of treadmill training inPD showed improvement in usual-walking gait (ie, notduring DT) after at least 4 weeks with practice 3 times aweek. 29 ●  Variability of practice:  Training a variety of tasks en-hances learning and improves transfer. 30 Practice undervariable conditions is also relevant for transfer to dailylife. As described later, this principle was applied by usingdifferent cognitive tasks in the training program and add-ing obstacles into the walking path. ●  Increasing the level of difficulty:  To promote progressionand enhance motivation, our program was made up of elements that allowed for increasing the level of difficultyduring the 4 weeks of training. This included addingobstacles, reducing the time between rest breaks, andswitching from serial 3 subtractions to serial 7s.  Implementation of the training program.  The training tar-geted both the walking and the cognitive tasks and aimed to beintensive and progressive for both components. Subjectstrained 3 times a week for 4 weeks. In each training session,subjects completed 5 blocks of 5 minutes of walking (ie, a totalof 25min of walking in each training session). In each 5-minuteblock, subjects performed 3 different kinds of cognitive tasks:VF task, serial subtractions, and information processing. Theorder of the cognitive tasks in each 5-minute block was ran-dom. All tasks were recorded in advance on an MP3 device,and subjects were instructed through earphones. This promotesstandardization and allows the therapist to guard the subjectand provide appropriate feedback without having to implementthe DT training.  Prioritization and feedback.  Variable prioritization (ie, al-ternate focus on the cognitive and the motor tasks) may en-hance learning (eg, better retention). Therefore, we used vari-able prioritization instructions. To enhance learning andmotivation, knowledge of results (KR) feedback was given forthe VF and serial subtraction tasks, and knowledge of perfor-mance (KP) feedback was supplied for gait performance. KRfeedback consisted of information regarding performance (ie,how many words were generated; how many calculations andmistakes for the serial 3 task) and encouragement of betterperformance (ie, more words, more calculations, fewer mis-takes). KP feedback consisted of information regarding steplength or arm swing. Feedback was given to only 1 task in eachblock, the task that was prioritized (see table 1 for a summaryof the training). RESULTS Mean age, disease duration, and years of education were63.8  8.4 years (all men), 5.4  4.8 years, and 14.1  5.0 years,respectively. One patient had motor response fluctuations; theothers did not. Every effort was made to carry out the evalu-ation and training sessions in the “on” state. At baseline,UPDRS motor part scores were 20.1  3.2, consistent withthose of patients with moderate PD. The VF score while sitting,assessed at the preevaluation session, was 10.42  3.25. Thescores of the MoCA (26.6  1.2), TMT part A (54.1  19.0s),TMT part B (126.3  38.8s), and FAB (16.8  1.0) were similarto those of older adults. 31 The mean score on the Timed Up andGo test was 9.9  1.0 seconds. While this is below the 13.5-second threshold that indicates a high risk of f alls, it also isconsistent with parkinsonian motor impairment. 32 In the VFtest while sitting, subjects generated 10.4  3.3 words, similarto the performance of healthy older adults. 6 Gait speed in the posttraining evaluation significantly in-creased ( P  .02) in all 3 trained DT conditions, compared with pretraining values (table 2, fig 1). For example, in the serial 3 subtraction walking task, gait speed increased in all 7 subjects(increases ranged from .16 to .36 m/s; mean, .25  .10m/s). Gaitspeed also increased significantly in the task that was notspecifically trained (open-ended, complex questions), com-pared with pretraining values. In contrast, gait speed during Table 1: Description of a Single Training Session Training Block 1 Block 2 Block 3 Block 4 Block 5 Duration of walk(total of 25min ineach session)5min 5min 5min 5min 5minInstructions for taskprioritizationNone “Concentrate mainly onthe gait task.”“Concentrate mainly onthe cognitive task.”“Concentrate mainly onthe gait task.”“Concentrate mainly onthe cognitive task.”Kind of feedback andinstructionsNone Knowledge of performance:“Walk with largersteps.”Knowledge of results:“In block 1, you said Xwords. Try to saymore words than inblock 1.”Knowledge of performance:“Sway your arms asmuch as you can.”Knowledge of results:In block 1, youcalculated Xcalculations with Y errors. Calculatemore accurately andfaster than you did inblock 1.”Cognitive tasks thatwere practicedVerbal fluency, serial 3 subtractions, information processing task (applied in a random order ), recorded inadvance on an MP3 device.Instructions forwalkingThroughout the training, subjects were always instructed to walk at their self-selected, comfortable pace. Heartrate was measured to ensure that aerobic effort did not exceed the target heart rate for aerobic training (ie,[220 – age]  0.8). In general, subjects walked far below the level needed for aerobic training.Progression duringtraining1. Shortening of break times from 5min (1st week) to 2min (2nd week), to 1min (3rd week), to no breaks (4thweek).2. Adding obstacles in the 2nd week: 5 boxes were placed with a separation of 60cm at the first third of thewalkway. Subjects were asked to walk in slalom between the obstacles without touching them.3. Changing serial 3 to serial 7 subtractions in the 3rd week. 3 DUAL TASK TRAINING IN PARKINSON’S DISEASE, Yogev-SeligmannArch Phys Med Rehabil Vol xx, Month 2011  usual walking only tended to increase ( P  .09) at the end of thetraining program (see table 2). After training, gait variability significantly decreased( P  .02; ie, improved) in all 3 trained tasks and in the task thatwas not trained (see table 2). For example, in the serial 3 subtraction walking condition, stride time variability decreasedin 6 of 7 patients (   ranged from .64%–2.60%; mean,.83%  .87%). Gait variability did not change during usualwalking ( P  .39). The improvements in gait speed and vari-ability were generally retained 1 month later (table 2). There was no significant change in the performance of theVF or serial 3 subtraction tasks at the end of the training( P  .17) (see table 2). DT interference of gait speed signifi- cantly decreased (ie, improved) for all trained tasks except forthe open-ended, complex questions task (table 3). The effects were less pronounced for stride time variability, perhaps re-lated to the large within-subject variance.From a qualitative perspective, all participants fully com-plied with the program. Most subjects participated in eachtraining session with enthusiasm, and only 1 patient com-plained of boredom toward the end of the program. None of theparticipants said that the tasks were too difficult, and all sub- jects always tried to respond to the questions (ie, none of themgave up out of frustration). We did notice some individualdifferences in the way the DT affected gait and the response totraining. For example, some participants were more affected bythe VF task, whereas others were more affected by the math-ematical task. DISCUSSION The present study demonstrates that a task-specific trainingprogram designed specifically to reduce the negative effects of DT while walking is feasible among patients with mild tomoderate PD. This study also moves beyond previous work bydemonstrating positive effects of the training on both gait speedand gait variability, f eatures that have been related to diseaseseverity and fall risk, 1-3,5,23,33 and on transfer to a nontrainedDT, along with retention 1 month later.You et al 8 observed no changes in DT gait after intensivetraining, while we observed relatively large improvements. Theobserved training effect on gait speed was beyond the mini-mum clinically significant change (ie, .05m/s for gait speed)and often was “substantial” (  0.1m/s) 34 ; many of the effectssizes were large. In a sense, these findings are consistent withthe previous report of DT gait speed improvements in olderadults. 9 Interestingly, the magnitude of improvement seen inthe present study is about 3 times larger than that reported bySilsupadol et al. 9 Given the small numbers of subjects in boththe present and previous study, these differences must beinterpreted cautiously.The DT training program described here can readily beadapted into rehabilitation-like clinical settings for a variety of populations including older adults and other patients who haveDT effects on gait and fall risk. It requires minimal technicalsupport and short and simple training of the therapist-instruc- Table 2: Effects of Training on Gait Speed, Stride Time Variability, and Cognitive Tasks Variable Test Condition Pretraining Posttraining Retention Effect Size ‡ Gait speed (m/s) Usual walking 1.22  0.17 1.34  0.20 (.09)* 1.37  0.09 (.61) † 0.64VF 0.97  0.16 1.18  0.15 (.02)* 1.24  0.06 (.23) † 1.35Serial 3 subtractions 0.97  0.19 1.22  0.14 (.01)* 1.26  0.04 (.49) † 1.49Information processing 0.97  0.12 1.25  0.11 (.02)* 1.29  0.05 (.31) † 2.43Open-ended, complex questions 0.99  0.17 1.17  0.14 (.01)* 1.22  0.07 (.39) † 1.15Stride timevariability (%)Usual walking 2.00  0.52 1.87  0.59 (.39)* 1.72  0.49 (.08) † 0.23VF 2.67  0.75 1.98  0.40 (.01)* 2.15  0.94 (.60) † 1.14Serial 3 subtractions 3.14  0.78 2.31  0.74 (.02)* 1.90  0.41 (.24) † 1.09Information processing 2.76  0.72 1.81  0.30 (.02)* 1.73  0.22 (.50) † 1.70Open-ended, complex questions 2.97  0.50 2.20  0.63 (.02)* 1.81  0.45 (.07) † 1.35Cognitive tasks VF (no. of words generated in 1minwhile walking)15.29  6.42 17.14  9.13 (.66) NA 0.23Serial 3 subtractions (no. of calculationsgenerated while walking)24.86  9.92 24.83  7.82 (.17) NA   0.01NOTE. Values are mean  SD or as otherwise indicated.Abbreviation: NA, not applicable.*Numbers in parentheses are  P   values for comparisons between pretraining and posttraining speeds. † Numbers in parentheses are  P   values for comparisons between posttraining and retention speeds. ‡ Effect sizes refer to the change in gait measures at the posttraining evaluation. Rejection of the null hypothesis here suggests that the trainingeffects persisted 1 month later. 0.600.700.800.901. 321    G  a   i   t  s  p  e  e   d   (  m   /  s  e  c   )   Pre trainingPost trainingRetention Fig 1. Example of the effects of DT training on DT gait speed amongtheindividualsubjects.Gaitspeedmeasuredduringthepretraining,posttraining, and retention evaluations are presented for each par-ticipant (ie, each line represents a different subject) for the serial 3subtraction DT condition. Note that for all subjects, DT gait speedwas higher at posttraining and 1 month later, compared with base-line, pretraining values. 4  DUAL TASK TRAINING IN PARKINSON’S DISEASE, Yogev-SeligmannArch Phys Med Rehabil Vol xx, Month 2011  tor. The use of the MP3 device ensures uniformity of thecognitive training protocol among subjects. While training, allsubjects spent exactly the same time in each cognitive task. Italso allows a single therapist to carry out training while guard-ing against falls, and it enables the therapist to watch thesubject and provide feedback (this would have been impossibleto do if the therapist also had to generate the cognitive tasks inreal-time). The training can be applied one-on-one or adaptedto group settings. The training program described here couldalso be viewed as a general framework that could be modifiedto best fit the needs of the individual patient and clinician. Forexample, different tasks could be chosen for the training pro-gram, provided that there is adherence to the principles out-lined in designing the program (eg, variability of practice andchoosing tasks that challenge different cognitive domains). Study Limitations One of the limitations of the present study is that cognitiveperformance in sitting was not fully evaluated before trainingor at the end of the training program. Some of the cognitivetasks such as the open-ended, complex questions were difficultto quantify. These issues probably should be addressed infuture research.The present pilot study supports feasibility and suggests thattraining is efficacious and the outcome measures respond to thetraining. Future studies should evaluate the current programusing a controlled design. We estimate that a randomizedcontrolled trial will need 23 subjects in the training arm and 23in the control arm to show a statistically significant and clini-cally meaningful change (   .05;  90% power), both for gaitspeed and variability, posttraining and 1 month later in theuntrained DT (eg, assuming that the intervention group im-proves by .23  .17 m/s and the control by .05  .17 m/s). Futurestudies should examine whether DT training has additive valueover single-task training in PD, and explore the effects of thisprogram on freezing of gait. We suggest testing transfer effectsto tasks in which cognitive performance can be directly quan-tified, and evaluating self-efficacy, satisfaction from the pro-gram, and qualitative information regarding the interest anddifficulty of the program. From the perspective of a patient-targeted program, it might be helpful to create a “pool” of secondary tasks, ordered in increasing level of difficulty, thatcould be adjusted to each patient’s abilities and could enableprogression during the training program. Motor tasks such ascarrying objects and dialing a number on a cellular phone couldalso be included. CONCLUSIONS The present feasibility study provides the basis for additionalresearch on DT training and sets the stage for clinical imple-mentation of a program that may help to reduce the negativeimpacts of DT on gait and fall risk. References 1. Woollacott M, Shumway-Cook A. Attention and the control of posture and gait: a review of an emerging area of research. GaitPosture 2002;16:1-14.2. Yogev-Seligmann G, Hausdorff JM, Giladi N. The role of exec-utive function and attention in gait. Mov Disord 2008;23:329-42.3. Herman T, Mirelman A, Giladi N, Schweiger A, Hausdorff JM.Executive control deficits as a prodrome to falls in healthy olderadults: a prospective study linking thinking, walking, and falling.J Gerontol A Biol Sci Med Sci 2010;65:1086-92.4. 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 Neurosci2005;22:1248-56.5. Springer S, Giladi N, Peretz C, Yogev G, Simon ES, Hausdorff JM. Dual-tasking effects on gait variability: the role of aging,falls, and executive function. Mov Disord 2006;21:950-7.6. Yogev-Seligmann G, Rotem-Galili Y, Mirelman A, Dickstein R,Giladi N, Hausdorff JM. How does explicit prioritization alterwalking during dual-task performance? Effects of age and sex ongait speed and variability. Phys Ther 2010;90:177-86.7. Yang YR, Wang RY, Chen YC, Kao MJ. Dual-task exerciseimproves walking ability in chronic stroke: a randomized con-trolled trial. Arch Phys Med Rehabil 2007;88:1236-40.8. You JH, Shetty A, Jones T, Shields K, Belay Y, Brown D. Effectsof dual-task cognitive-gait intervention on memory and gait dy-namics in older adults with a history of falls: a preliminaryinvestigation. NeuroRehabilitation 2009;24:193-8.9. Silsupadol P, Shumway-Cook A, Lugade V, et al. Effects of single-task versus dual-task training on balance performance inolder adults: a double-blind, randomized controlled trial. ArchPhys Med Rehabil 2009;90:381-7.10. Schwenk M, Zieschang T, Oster P, Hauer K. Dual-task perfor-mances can be improved in patients with dementia: a randomizedcontrolled trial. Neurology 2010;74:1961-8.11. Onla-or S, Winstein CJ. Determining the optimal challenge pointfor motor skill learning in adults with moderately severe Parkin-son’s disease. Neurorehabil Neural Repair 2008;22:385-95.12. Rochester L, Baker K, Hetherington V, et al. Evidence for motorlearning in Parkinson’s disease: acquisition, automaticity and re-tention of cued gait performance after training with externalrhythmical cues. Brain Res 2010;1319:103-11.13. Brauer SG, Morris ME. Can people with Parkinson’s diseaseimprove dual tasking when walking? Gait Posture 2010;31:229-33. Table 3: Effects of Training on DT Interference* Variable Test Condition Pretraining Posttraining † Effect Size Gait speed VF 20.31  6.06 11.38  6.87 (.02) 1.37Serial 3 subtractions 20.87  9.14 8.01  6.03 (.01) 1.66Information processing 17.40  8.47 5.47  7.34 (.02) 1.50Open-ended, complex questions 19.20  6.53 11.78  7.46 (.12) 1.05Stride time variability VF –35.50  28.77 –17.29  42.91 (.31) 0.49Serial 3 subtractions –48.40  35.19 –12.92  34.50 (.09) 1.01Information processing –32.59  24.74 –8.82  36.58 (.91) 0.76Open-ended, complex questions –43.47  32.03 –28.95  45.83 (.23) 0.36NOTE. Values are mean  SD or as otherwise indicated.*Interference is defined as the % change with respect to value measured in the single-task, usual-walking condition. † Numbers in parentheses are  P   values for comparisons between pretraining and posttraining interference. 5 DUAL TASK TRAINING IN PARKINSON’S DISEASE, Yogev-SeligmannArch Phys Med Rehabil Vol xx, Month 2011
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks