Effects of Mobilization and Tactile Stimulation on Recovery

Transcription

1 ORIGINAL ARTICLE Effects of Mobilization and Tactile Stimulation on Recovery of the Hemiplegic Upper Limb: A Series of Replicated Single-System Studies Susan M. Hunter, PhD, Peter Crome, MD, PhD, Julius Sim, PhD, Valerie M. Pomeroy, PhD 1,2 ABSTRACT. Hunter SM, Crome P, Sim J, Pomeroy VM. with an initial severely affected UL. Of those beginning Effects of mobilization and tactile stimulation on recovery ofrehabilitation with marked impairment of function, only ap- 15% are reported to regain useful function. 6 the hemiplegic upper limb: a series of replicated single-systemproximately studies. Arch Phys Med Rehabil 2008;89: Poor recovery of the hemiplegic UL has been attributed to 7,8 inadequate and inappropriate therapy, with claims that the Objective: To explore the effects on motor function and 7 UL is neglected in rehabilitation. Despite apparent recovery of impairment of mobilization and tactile stimulation for the paretic arm and hand after stroke. 5 some motor activity, the majority of people with stroke continue to describe their paretic UL as useless and dysfunctional, Design: Replicated single-system series, ABA design. being unable to involve that limb automatically in everyday Setting: The stroke rehabilitation ward of a community activities of daily living. Consequently, its noninvolvement in hospital in the United Kingdom. regular activity is likely to result in further weakness, contracture, learned nonuse, 9 sensory loss, 10 and loss of cortical rep- Participants: Consecutive sample, men and women 6) (N with stroke (left or right), within 3 months of onset. resentation. 11 Furthermore, it has been suggested that UL rehabilitation is hindered by inaccurate beliefs that recovery occurs Intervention: Sixty minutes of daily mobilization and tactile stimulation to the paretic arm and hand for 6 weeks in addition uniquely proximally to distally, stability of the shoulder girdle to the usual rehabilitation program. 8,12 being required before the hand can become active. Main Outcome Measures: Focal disability (Action Research Arm Test [ARAT]) and motor impairment (Motricity 13 In a longitudinal study of UL recovery after stroke 47; (N follow-up at 6wk, 3mo, 6mo), Hunterreported that 93% of Index arm section). patients scoring 0 on the ARAT at 6 weeks continued to score Results: All participants showed visual change in 1 or more of 0 at 3 and 6 months. This was despite conventional routine trend, level, or slope between baseline and intervention phases for therapy provided on a stroke rehabilitation ward, suggesting both the ARAT and the Motricity Index. The visual analysis was that current routine packages of therapy are ineffective in confirmed through statistical testing c ( statistic and/or Mannrestoring functional activity in the majority of people with Whitney U test) for 5 of 6 participants (statistical analysis was severe UL paresis. precluded for 1 participant). No further improvements were made Yet, certain specific interventions may be effective in improving outcomes in the hemiplegic UL. In the presence of on intervention withdrawal. Conclusions: This study shows proof of concept for using voluntary motor activity, repetitive practice of task-related mobilization and tactile stimulation to improve motor recovery 14,15 activities is effective in improving function, an approach after severe paresis, justifying conducting dose-finding studies 6,16 most appropriate for less severe stroke. Imaging studies as a precursor to multicenter phase III clinical trials. have shown reorganization of sensory and motor cortices after Key Words: Cerebrovascular accident; Hemiplegia; Rehabilitation; Upper extremity. intensive practice and repetition of voluntary movement in the hemiparetic UL. 11,17 However, in the absence of voluntary 2008 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and 18 muscle activity, this approach, involving predominantly direct muscle activation, is somewhat less successful. Rehabilitation Specific sensory retraining programs have been shown to improve long-term UL sensory dysfunction after severe stroke. FTER STROKE, AROUND 60% to 70% of people experience initial loss of UL function, with approximately These programs include electrical and mechanical stimuli 1-3 A 19 ; texture discrimination and proprioception retraining 20 ; and localization, identification of object characteristics, and number half classified as having severe paresis (Scandinavian Stroke Scale) Furthermore, a substantial proportion (50% 70%) continue to experience long-term (2 4y) UL dysfunction, and and shape tactile discrimination. In each study, the recovery 1,4,5 of motor activity was reported informally by participants, but most reports of UL recovery suggest it is minimal in people the change in motor activity as a result of sensory retraining was not specifically measured. Tactile stimulation and sensory stimulation are components of routine therapy, and these ther- From the Research Institute for Life Course Studies, Keele University, Keele (Hunter, Crome, Sim); and Faculty of Health, University of East Anglia, Norwich (Pomeroy), UK. Supported by Research into Ageing (grant no. 211). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Reprint requests to Susan M. Hunter, PhD, School of Health and Rehabilitation, Keele University, Keele, Staffordshire, ST5 5BG, UK, s.m.hunter@shar. keele.ac.uk /08/ $34.00/0 doi: /j.apmr apeutic tools are used to facilitateor guide 8 movements of the UL. Five specific sensory facilitation techniques have been shown to improve the activation of wrist extensors after stroke 23 : (1) cuta- ARAT UL List of Abbreviations Action Research Arm Test upper limb 2003

2 2004 MOBILIZATION AND TACTILE STIMULATION IN A SINGLE CASE SERIES, Hunter neous and proprioceptive stimulation by tapping and rubbing skin over the target muscle, (2) weight bearing through the hemiplegic limb, (3) active elevation of the hemiplegic shoulder, (4) maximal isometric contraction of the nonhemiplegic hand and finger extensors against resistance, and (5) direct voluntary activation of the hemiplegic wrist and fingers. Although all 5 techniques increased activity in corticospinal pathways, the direct activation of muscle was most effective in achieving this. However, when voluntary activation was initially absent, cutaneous and proprioceptive stimulation appeared to provide a more effective excitatory or facilitatory influence. Although these exploratory studies examined small components or single interventions used in therapy, they do not reflect the combinations of techniques used in routine clinical practice for the hemiplegic UL. To identify the content of actual routine therapy for the hemiplegic UL, senior physical therapists with expertise in neurologic rehabilitation were asked to describe in detail what they do to treat the severely affected hemiplegic UL and hand. 24 As a result, 1 module of current routine therapy, known as mobilization and tactile stimulation, was identified, consisting of different techniques applied in combination by a skilled therapist, according to the patient s initial problems and individual response to handling. Techniques included soft-tissue mobilization including massage and active soft-tissue stretch, passive and accessory movements of joints, active movement and assisted or guided movement when active movement is not possible, and cutaneous sensory input through touch and pressure including weight bearing through the hand. 24 This module reflects current practice in treating biomechanic changes in the UL such as soft-tissue shortening and loss of range of movement, neurophysiologic problems such as hypersensitivity and sensory loss, and perceptual problems such as unilateral neglect and inattention. 24 However, the effects of this treatment module for the hemiplegic UL have not been explored or evaluated. The aim of this study was to explore the effects of mobilization and tactile stimulation for the hemiplegic hand on motor impairment and functional ability in the subacute phase of rehabilitation. METHODS We performed a series of 6 replicated single-system studies (by using an ABA design) with people with stroke, taking repeated measures from each participant before (initial A phase [baseline]), during (B phase [intervention]), and after (final A phase [withdrawal]) a program of 6 weeks of daily intervention using mobilization and tactile stimulation. Single-system methodology is appropriate to observe changes in a person s performance over time, providing clinically relevant information about individual subjects or, if replicated, groups of subjects with specified problems, and is of particular value for therapists in evaluating practice or exploring the effects of specific treatments for specific problems. 25 The initial baseline phase serves as a control, and data from this phase can be compared with data from the intervention phase to see if there is a difference in performance on chosen outcome measures. Three or 4 direct replications of a single-case experiment that exhibit a predictable pattern and produce the same result provide strong evidence of a causal relationship. 26 The accumulation of results across patients significantly increases the external validity of the findings. 26 Sample The sample included consenting adult men and women inpatients admitted to a stroke rehabilitation ward with impaired UL function who had been living independently before admission and who could be recruited to the study within 3 months of the onset of a first stroke. We excluded people who were unable to understand or follow a simple 1-stage command (eg, place your [nonparetic] hand on your head). The timing of recruitment (3mo since stroke onset) was to ensure that participants were recruited at similar stages of chronicity with similar rehabilitation experiences and programs. Procedure We screened consecutive new admissions to the rehabilitation ward for inclusion and gave verbal and written information about the study to patients selected for possible participation. We recruited the first 6 eligible participants to the study and assigned a number to each participant to ensure anonymity. Throughout all 3 phases (A1, B, A2), all participants continued to receive routine rehabilitation and therapy from the clinical stroke team, the focus of which was reported as being to achieve general function and mobility. It is recommended that as many data points as possible should be collected in each phase at equal intervals, such as 7 days a week. 27 However, to mirror current clinical practice, we administered treatment on consecutive working days (Monday Friday); for practical reasons, this meant that data could only be collected with the same frequency. We endeavored to record observations at the same time each day to avoid instability caused by diurnal variation. 26,27 We planned the initial baseline (A1) phase to last for a minimum of 2 weeks to ensure a minimum of the 8 data points in the baseline phase recommended for statistical analysis 28 or when the baseline performance was considered to be unstable (judged by high level of variability on visual inspection of baseline data) until a stable pattern of performance emerged, up to a maximum of 4 weeks. On completion of the baseline (A1) phase, we began the intervention (B) phase, which we planned to last for 6 weeks. Each day, after the outcome measures were completed, participants received mobilization and tactile stimulation to the hemiplegic limb for a maximum of 1 hour. After discharge from the ward, we continued to apply the intervention and record outcome measures in the participant s home with prior agreement. During the withdrawal (A2) phase, which lasted for 2 further weeks, we withdrew mobilization and tactile stimulation, and participants were specifically asked not to self-administer the treatment during this phase. We continued to record performance daily. Intervention An experienced physical therapist with specialist skills in neurologic rehabilitation provided mobilization and tactile stimulation daily for a maximum of 60 minutes to the paretic arm and hand (upper limb) over 6 weeks. This was considered to be a maximum clinically feasible dose of an intervention based on discussion with senior clinicians. 22 On each occasion, this therapist recorded detailed content of the treatment session by using the treatment schedule. 24 Treatment was performed in a quiet area of the stroke rehabilitation ward or in the patient s own home after discharge from the ward. Although we did not record a clear description of the routine program of rehabilitation that was continued throughout the series because clear descriptions of routine therapy have not yet been identified, 29 therapists working routinely with the participants involved in the study reported informally that their treatment was focused on regaining general mobility and function.

3 MOBILIZATION AND TACTILE STIMULATION IN A SINGLE CASE SERIES, Hunter 2005 All participants, apart from participant number 6, were discharged home from the ward during the intervention phase. On discharge from the ward, only participants 1 and 3 actually received follow-up therapy, attending an outpatient unit during the withdrawal phase. Outcome Measures We used the ARAT 30 and Motricity Index (arm section) 31 to record focal disability and motor impairment, respectively. One author (S.M.H.) took these measurements. Data We recorded results for all 6 participants individually and presented data in time series plots according to each outcome, with the baseline, intervention, and withdrawal phases differentiated visibly. We analyzed data initially by using visual inspection, which is the most commonly used and accepted method of analyzing single-system studies, 26,32 to look for changes in trend, level, slope, or variability. 26,32 Trend refers to the direction in which performance on a series of scores is progressing 32 and may be accelerating, decelerating, quadratic, or curvilinear. 32,33 We also used statistical analysis to support the visual findings. We investigated the presence of serial dependency in the data by using the autocorrelation coefficient (r), 32 and, if no serial dependency was found (r 1.96/ n, where n is the number of observations 32 ) in either the baseline or the intervention phases, we analyzed differences between the 2 sets of data (A1, B) by using the nonparametric Mann-Whitney U test. In addition, for the data in these 2 phases, we used the c statistic. 28,32 This was for 2 reasons. First, the Mann-Whitney U test rests on the assumption of no serial dependency. However, the test used to assess serial dependency will have little power when the number of observations is small. Accordingly, it may not be wise to rely wholly on the results of the Mann-Whitney U test. Second, the Mann-Whitney U test only assesses a change in median performance (ie, a change in level) between phases, whereas the c statistic provides information on a change in trend. The c statistic is able to determine a statistically significant change in trend between serially dependent data in 2 phases but can be used with much smaller data sets than are required by most other time series analyses. 28 We set the cutoff for statistical significance as P less than or equal to.05 (2- tailed). In addition to considering a statistically significant change between phases, we took account of the applied functional or clinically significant importance of any change in performance. 26 We considered a change of score by 9 points or more on the ARAT to be functionally significant. 34 Reliability Because the treating therapist also administered the daily outcome measures immediately before treatment on each occasion, there was potential for investigator bias because the assessor was not blind to phase of treatment. Therefore, to eliminate this potential bias, we performed an independent reliability check of the test results. 26 A second therapist who was trained in the use of these outcome measures and who remained blinded to the results obtained by the main tester did this. This revealed perfect agreement on 91 of 91 pairs of scores (100%) for the ARAT and 89 of 91 pairs of scores (98%) for the Motricity Index. We did not perform statistical analysis because reliability was clearly high. Research Ethics We gained approval for this study from ethics committees at Keele University and Shropshire and Staffordshire Health Authority. RESULTS Sample All 6 participants recruited to the study completed all phases. The men-to-women ratio was 2:1, and the number of participants with left or right hemiplegia was equal. All participants were right-hand dominant. Ages ranged from 64 to 87 years (mean age SD, y). Table 1 summarizes participant characteristics with salient information about their clinical presentation. Participant No. Table 1: Participant Characteristics Age (y) Sex Stroke Type 35 CT Scan Time from Stroke Onset to Recruitment (d) 1 78 Female Right PACS Recent infarction right external capsule and lentiform nucleus 2 87 Female Left POCS Recent infarction left upper pons and left cerebral peduncle 3 64 Male Right TACS Large infarct right middle cerebral artery territory Male Right LACS Large (3mm) infarction ganglionic region, specifically head of caudate nucleus, lentiform nucleus and anterior limb of internal capsule as well as periventricular white matter 5 76 Male Left PACS Recent infarct left frontoparietal region, with further low attenuation slightly more posteriorly, in left parietal lobe, consistent with relatively recent infarct 6 72 Male Left PACS Infarction left frontal/parietal lobe and posterior aspect left lentiform nucleus, left middle cerebral artery territory Initial ARAT Score Initial MI Arm Score Abbreviations: CT, computed tomography; LACS, lacunar stroke; MI, Motricity Index; PACS, partial anterior circulation stroke; POCS, posterior circulation stroke; TACS, total anterior circulation stroke. 35

4 2006 MOBILIZATION AND TACTILE STIMULATION IN A SINGLE CASE SERIES, Hunter Fig 1. ARAT scores for participants (P1 P6). Action Research Arm Test Figure 1 shows time series plots of the ARAT total scores for all 6 participants across the 3 phases. Table 2 summarizes the results of visual and statistical analysis for all participants. Increases in ARAT scores from baseline to intervention phases ranged from 4 to 29 points, with a mean increase SD of points. All 6 plots show a visible difference in 1 or more of trend, level, and slope (gradient) between the baseline and intervention phases. These differences were confirmed by the c statistic for participant 2 (z , P.001), participant 5 (z , P.001), and participant 6 (z , P.001). Baseline data for participants 1, 3, and 4 were constant at 0 throughout the phase, and, therefore, the c statistic could not be calculated. Serial dependency (autocorrelation coefficient) was not detected in the data for participants 2, 5, or 6, and the Mann-Whitney U test further confirmed a statistically significant difference between phases (participant 2: U 1.5, P.001; participant 5: U 1.5, P.001; participant 6: U.50, P.001). The number of treatment sessions received before a change in performance was seen ranged from 1 (participant 6) to 14 (participant 5), with a mean number of 8 sessions. On withdrawal of the intervention, visual analysis showed that scores remained stable for all participants apart from participant 4, in whom scores decreased.

5 MOBILIZATION AND TACTILE STIMULATION IN A SINGLE CASE SERIES, Hunter 2007 Table 2: Summary of Analysis of ARAT Scores for Participants 1 Through 6 Indicating the Nature of Visible Change Participant Visible Change P* P Score Mean Baseline Mean Intervention Phase Score Impression 1 Trend NA NA 0 4 Difference between A1 and B phases 2 Slope Difference between A1 and B phases 3 Trend NA NA 0 10 Difference between A1 and B phases 4 Trend NA NA 0 5 Difference between A1 and B phases 5 Slope Difference between A1 and B phases 6 Trend Difference between A1 and B phases NOTE. P values from a *Mann-Whitney U test and/or c statistic and an overall impression of the results. Abbreviation: NA, not applicable (baseline data constant). Motricity Index Arm Section Figure 2 shows the Motricity Index scores for all participants. Table 3 summarizes the results of visual and statistical analysis. Visually, there was a clear change in 1 or more of trend, level, and slope between baseline and intervention phases for all participants, although this was less marked for participant 2 and 6. Increases in Motricity Index arm section scores ranged from 6 to 49 points, with a mean increase SD of points. Statistical analysis (c statistic) supported this difference between phases for participant 3 (z , P.001) and participant 5 (z , P.001). The c statistic could not be calculated for participant 1 (constant baseline), participant 2 (decelerating trend in baseline), participant 4, and participant 5 (both showing accelerating trends in baseline). Differences between phases were confirmed with the Mann-Whitney U test (serial dependency not found) for participant 2 (U 24, P.001), participant 3 (U 0, P.001), participant 4 (U 22.5, P.001), participant 5 (U 8, P.001), and participant 6 (U 9, P.001). No further improvements were made on withdrawal of the intervention, with scores remaining stable on visual analysis for all 6 participants. DISCUSSION This study has explored the effects of mobilization and tactile stimulation for the hemiplegic arm and hand on motor impairment and functional ability in the subacute phase of rehabilitation after stroke. We found improvements in UL activity and motor impairment in all 6 participants, with clear differences in total ARAT and Motricity Index (arm) scores between baseline and intervention phases after 6 weeks of daily mobilization and tactile stimulation applied to the paretic limb. Because these improvements have been replicated in all 6 participants, we attribute them to the intervention. Latency of improvement has varied, with some immediate changes and some delayed changes after the onset of the intervention phase. The majority of people with severe UL dysfunction poststroke are not expected to recover manipulation skills for useful functional activity, 1,4,5 such as those required for successful completion of the ARAT. 13 Indeed, it has been shown that total ARAT scores of 0 (severe dysfunction) at 6 weeks poststroke are unlikely to increase at 3 months or 6 months 13 despite the patient receiving a routine program of therapy. Three of our participants (1, 3, 4) started this study with baseline ARAT scores of 0. Consequently, we did not expect them to show substantial improvements in performance in the absence of specific intervention. However, in contrast to the findings suggested in previous studies, 13 all 3 increased their ARAT scores following the addition of mobilization and tactile stimulation intervention to routine therapy, from 0 to between 13 and 17 points. According to other work 34 in which a score of 9 on the ARAT is considered to reflect the difference in functional ability between being unable to grasp, grip, pinch, or move the limb and being able to perform parts of these tasks, this is considered to be of clinical significance. Once active movement can be initiated, repetition of the direct activation of muscle is most effective in gaining further recovery of activity. 23 In the absence of initial voluntary activity (eg, after severe stroke), cutaneous and proprioceptive stimulation (brushing the skin over the target muscle) has been shown to be an effective means of facilitating muscle activity. 23 The results of the present study further support these findings. We hypothesize that where motor impairment and dysfunction are severe, precluding patients from undertaking repetitive task retraining to improve activity, mobilization and tactile stimulation is an appropriate and effective therapy intervention that potentially will kick start the process of activating muscle after stroke by providing significant proprioceptive and somatosensory information to the brain, facilitating direct activation of the primary motor cortex and the corticospinal system to increase motor activity. In this study, 5 of the 6 participants showed carryover effects of the intervention without any further acceleration in the trend. Only 1 participant showed a decline in performance. This is strong evidence that the intervention has been effective in causing an improvement in performance in all 6 participants. However, this warrants further study. Further work might involve reintroducing the intervention after the withdrawal phase (ABAB design), which would provide valuable information about the effects of mobilization and tactile stimulation at a later stage of recovery. The results of this study show that there was a latency of improvement in performance for some participants, yet there was an immediate improvement in performance for others on the 2 outcome measures. In most, but not all cases, an improvement was seen on the Motricity Index before an improvement in the ARAT (participants 1, 2, 3, 4, 5). This is to be expected; the Motricity Index shows an improvement in motor impairment that is required before there is a similar improvement in function (ARAT). Consequently, the latency of improvement appeared to be solely associated with the ARAT scores. Again, this is to be expected given the higher threshold of motor improvement required in the severely paretic UL in order to execute a functional task. The magnitude of the latency was variable, and this is likely to be attributable to this sample being a heterogeneous group of patients with differing clinical presentations. Study Limitations The main purpose of this study was to explore the difference in performance between baseline and intervention phases. We

6 2008 MOBILIZATION AND TACTILE STIMULATION IN A SINGLE CASE SERIES, Hunter Fig 2. Motricity Index arm section scores for participants (P1 P6). adopted an ABA design to allow us to focus on any change in behavior after the initial onset of therapy. Although we examined performance between intervention and withdrawal phases, the issue of carryover effects of therapy needs to be explored in greater depth. Traditionally, if performance changes, such as decreases on withdrawal of an intervention, this is considered to add weight to the argument that there is a causal effect of the intervention. 26 However, although this may be an appropriate analysis for interventions that are not anticipated to effect a long-term change in behavior or performance, in the field of rehabilitation, in which it is hoped that the process of motor learning during rehabilitation will effect some long-term change (carryover), it is perhaps more appropriate to look for a decrease or a maintenance of performance on withdrawal and not a continued accelerating trend, which might indicate some process of improvement despite the intervention. All testing was undertaken and scored by the same tester who was also the only therapist providing the intervention. Although having the same therapist providing all the treatment reduced therapist bias, the potential for investigator and measurement bias remained. Blinded assessment would have increased the internal validity of the results if the assessor had been blind to the stage of the design (A or B phases). However, the blinded independent reliability check on 91 pairs of scores showed that investigator and measurement bias was not an issue. This was an initial exploratory study, and full blinding

7 MOBILIZATION AND TACTILE STIMULATION IN A SINGLE CASE SERIES, Hunter 2009 Table 3: Summary of Analysis of Motricity Index Arm Scores for Participants 1 Through 6 Indicating the Nature of Visible Change Participant Visible Change P* P Score Mean Baseline Mean Intervention Phase Score Impression 1 Trend NA NA 1 24 Difference between A1 and B phases 2 Trend; level.001 NA Difference between A1 and B phases 3 Trend; level; slope Difference between A1 and B phases 4 Trend.001 NA Difference between A1 and B phases 5 Level Difference between A1 and B phases 6 Trend; level; variability.001 NA Difference between A1 and B phases NOTE. P values from a *Mann-Whitney U test and/or c statistic and an overall impression of the results. Abbreviation: NA, not applicable (baseline constant or significant trend in baseline data). was considered to be neither feasible nor essential as long as there was some means of checking the reliability of the scores. We undertook testing and treatment on a daily basis (Monday Friday) to mirror the maximum clinically feasible dose of an intervention. Consequently, there were some gaps in the data series on weekends and when normal public holidays occurred. This is not ideal 27 ; however, for exploratory work based in clinical settings in which patients have additional health care needs to be met (real-world research 36 ), measurement on a strict daily basis is often not feasible. Furthermore, limited resources would have precluded recording performance 7 days a week. We did not consider the impact of missing data from weekends and public holidays to be large because we repeated the measures for a minimum total of 10 weeks for each participant. For 3 of the participants (1, 3, 4), it was not possible to perform a c statistic analysis because of constant baseline performance. However, in these circumstances, evidence of a trend on visual analysis is particularly persuasive, thereby diminishing the need for confirmatory statistical analysis. Replication of the results across all 6 participants increases the external validity of our findings. 26 However, the sample was a heterogeneous group, and further exploratory work should evaluate the effects of this intervention on specific problems in specific groups of patients, according to clinical presentation and at different stages of recovery, such as in chronic stroke. The dose chosen for this study was based on what was considered to be the maximum clinically feasible dose. 13 Further work is currently being undertaken to identify the optimum dose of mobilization and tactile stimulation through a dose-finding study. Opportunities for collaborative translational research need to be sought in order to discover more about the physiologic processes underpinning the effects of mobilization and tactile stimulation seen in this study. CONCLUSIONS In this exploratory study, we have shown proof of concept for using mobilization and tactile stimulation to improve motor recovery in people with severe paresis, and this justifies conducting dose-finding studies as a precursor to a multicenter phase III clinical trial. We have provided valuable insight into the effects of a module of contemporary routine therapy (mobilization and tactile stimulation) on UL motor impairment and function. The results have not only served to show that this treatment improved UL function, particularly motor activity as measured by the ARAT and the Motricity Index (arm section), but have also generated hypotheses and identified questions requiring further study. In summary, subsequent work should explore the effects of this treatment on specific problems in specific groups of people with stroke at different stages of recovery and study the dose effects of the module. An optimal dose of therapy using this module and the most appropriate target population needs to be identified before phase III controlled clinical trials 37 can be undertaken. Further exploratory trials should compare the effects of this treatment alone with the combination of this treatment and concomitant therapies. However, many of these additional therapies may first need to be described in phase I modeling studies. This single-system series provides important clinical information about the recovery of performance and changes in behavior after stroke. However, to fully evaluate the intervention, phase III clinical trials need to be developed. Clearly, this is dependent on the completion of the essential modeling work currently being undertaken as a follow-up to this study. References 1. Wade DT, Langton-Hewer R, Wood VA, Skilbeck CE, Ismail HM. The hemiplegic arm after stroke: measurement and recovery. J Neurol Neurosurg Psychiatry 1983;46: Nakayama H, Jorgensen HS, Raaschou HO, Olsen TS. Recovery of upper extremity function in stroke patients: the Copenhagen Stroke Study. Arch Phys Med Rehabil 1994;475: Jorgensen HS, Nakayama H, Raaschou HO, Pedersen PM, Houth J, Olsen TS. Functional and neurological outcome of stroke and the relation to stroke severity and type, stroke unit treatment, body temperature, age, and other risk factors: the Copenhagen Stroke Study. Top Stroke Rehabil 2000;6: Parker VM, Wade DT, Langton Hewer R. Loss of arm function after stroke: measurement, frequency, and recovery. Int Rehabil Med 1986;8: Broeks JG, Lankhorst GJ, Rumping K, Prevo AJ. The long-term outcome of arm function after stroke: results of a follow-up study. Disabil Rehabil 1999;21: Sunderland A, Tinson DJ, Bradley EL, Fletcher D, Langton- Hewer R, Wade DT. Enhanced physical therapy improves recovery of arm function after stroke. A randomised controlled trial. J Neurol Neurosurg Psychiatry 1992;55: Davies PM. Steps to follow. Berlin: Springer-Verlag; Carr J, Shepherd R. Neurological rehabilitation: optimizing motor performance. Oxford: Butterworth-Heinemann; Taub E, Crago JE, Burgio LD, et al. An operant approach to rehabilitation medicine: overcoming learned nonuse by shaping. J Exp Anal Behav 1994;61: Carey LM. Somatosensory loss after stroke. Crit Rev Phys Rehabil Med 1995;7: Liepert J, Graef S, Uhde I, Liedner O, Weiller C. Traininginduced changes in motor cortex representations in stroke patients. Acta Neurol Scand 2000;101: Tyson SF, Chillala J, Hanley M, Selley AB, Tallis RC. Distribution of weakness in the upper and lower limbs post-stroke. Disabil Rehabil 2006;28:715-9.

8 2010 MOBILIZATION AND TACTILE STIMULATION IN A SINGLE CASE SERIES, Hunter 13. Hunter SM. Definition and effects of physical therapy treatment for sensorimotor dysfunction in the hemiplegic upper limb after stroke [dissertation]. Keele: Keele Univ; Butefisch C, Hummelsheim H, Denzler P, Mauritz KH. Repetitive training of isolated movements improves the outcome of motor rehabilitation of the centrally paretic hand. J Neurol Sci 1995;130: Asanuma H, Pavlides C. Neurobiological basis of motor learning in mammals. Neuroreport 1997;8:i-vi. 16. Parry RH, Lincoln NB, Vass CD. Effect of severity of arm impairment on response to additional physiotherapy early after stroke. Clin Rehabil 1999;13: Johansen-Berg H, Dawes H, Guy C, Smith SM, Wade DT, Matthews PM. Correlation between motor improvements and altered fmri activity after rehabilitative therapy. Brain 2002; 125: Taub E, Morris DM. Constraint-induced movement therapy to enhance recovery after stroke. Curr Atheroscler Rep 2001;3: Dannenbaum RM, Dykes RW. Sensory loss in the hand after sensory stroke: therapeutic rationale. Arch Phys Med Rehabil 1988;69: Carey LM, Matyas TA, Oke LE. Sensory loss in stroke patients: effective training of tactile and proprioceptive discrimination. Arch Phys Med Rehabil 1993;74: Yekutiel M, Guttmann E. A controlled trial of the retraining of the sensory function of the hand in stroke patients. J Neurol Neurosurg Psychiatry 1993;56: Lennon S, Baxter D, Ashburn A. Physiotherapy based on the Bobath concept in stroke rehabilitation: a survey within the UK. Disabil Rehabil 2001;23: Hummelsheim H, Hauptmann B, Neumann S. Influence of physiotherapeutic facilitation techniques on motor evoked potentials in centrally paretic hand extensor muscles. Electroencephalogr Clin Neurophysiol 1995;97: Hunter SM, Donaldson C, Crome P, Sim J, Pomeroy V. Development of treatment schedules for research: a structured review to identify methodologies used and a worked example of mobilisation and tactile stimulation for stroke patients. Physiotherapy 2006;92: Riddoch J, Lennon S. Single subject experimental design: one way forward? Physiotherapy 1994;80: Barlow DH, Hersen M. Single case experimental designs. 2nd ed. Boston: Allyn & Bacon; Sim J, Wright C. Research in health care: concepts, designs and methods. Cheltenham: Stanley Thornes; Tryon WW. A simplified time-series analysis for evaluating treatment interventions. J App Behav Anal 1982;15: Pomeroy VM, Niven DS, Barrow S, Faragher EB, Tallis RC. Unpacking the black box of nursing and therapy practice for post-stroke shoulder pain: a precursor to evaluation. Clin Rehabil 2001;15: Lyle RC. A performance test for assessment of upper limb function in physical rehabilitation treatment and research. Int J Rehabil Res 1981;4: Demeurisse G, Demol O, Robaye E. Motor evaluation in vascular hemiplegia. Eur Neurol 1980;19: Ottenbacher KJ. Evaluating clinical change: strategies for occupational and physical therapists. Baltimore: Williams & Wilkins; Goodwin N, Sunderland S. Intensive time-series measurement of upper limb recovery in the subacute phase following stroke. Clin Rehabil 2003;17: Church C, Price C, Pandyan AD, Huntley S, Curless R, Rodgers H. Randomized controlled trial to evaluate the effect of surface neuromuscular electrical stimulation to the shoulder after acute stroke. Stroke 2006;37: Bamford J, Sandercock P. Classification and natural history of clinically identifiable subtypes of cerebral infarction. Lancet 1991; 337: Robson C. Real world research: a resource for social scientists and practitioner-researchers. 2nd ed. Oxford: Blackwell; Medical Research Council. A framework for development and evaluation of randomised controlled trials for complex interventions to improve health. London: Medical Research Council; 2000.