Citation
Functional and Descriptive Predictors of Outcomes following Constraint-Induced Movement Therapy for Individuals with Post-Stroke Hemiparesis

Material Information

Title:
Functional and Descriptive Predictors of Outcomes following Constraint-Induced Movement Therapy for Individuals with Post-Stroke Hemiparesis
Creator:
FRITZ, STACY L. ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Fingers ( jstor )
Hands ( jstor )
Modeling ( jstor )
Motor ability ( jstor )
Movement therapy ( jstor )
Paresis ( jstor )
Regression analysis ( jstor )
Strokes ( jstor )
Upper extremity ( jstor )
Wrist ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Stacy L. Fritz. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
8/31/2007
Resource Identifier:
73306807 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

FUNCTIONAL AND DESCRIPTIVE PRED ICTORS OF OUTCOMES FOLLOWING CONSTRAINT-INDUCED MOVEMENT TH ERAPY FOR INDIVIDUALS WITH POST-STROKE HEMIPARESIS By STACY L. FRITZ A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Stacy L. Fritz

PAGE 3

This document is dedicated to my family and friends.

PAGE 4

ACKNOWLEDGMENTS While I have reached the destination of my graduate studies, I want to thank all the people who made this journey so meaningful, memorable and eventful. First, I would like to acknowledge my family for their support. From the day I decided to go back to graduate school, my parents, my brothers and their families, and my grandmother have been nothing but supportive. They never questioned my ability or my desire to complete the program. With a family like mine, failure is truly impossible. I want to thank them for all their love and unending support. Many people enter a doctoral program because they have always wanted to receive a PhD. I entered because I wanted to teach. I wanted to share my knowledge and love of the profession of physical therapy with students, to watch them learn and grow and develop into excellent clinicians who can make sound clinical decisions. As I spent the last five years here, this commitment to mentoring has strengthened, as has my relationship with a host of mentors. Kathye Light is not only a wonderful mentor but also a good friend. I have prospered in my career because of what she has allowed me to do while at the University of Florida. As I tell many prospective students, her “long-comings” way outweigh her shortcomings. She was fundamental in helping me build a firm foundation on which to build a career, and I truly admire her for knowledge and scientific curiosity. Craig Velozo has spent endless hours enduring my half-minded ideas and poor grammar. His support through this process has truly been priceless. Having the chance to work with iv

PAGE 5

him has truly been one of the best opportunities of my graduate studies. Andrea Behrman was been supportive of me from the first day I came to the University. I appreciate her ideas, her support, and her guidance through this process. She has been very influential in my desire to be a strong teacher and researcher. Jim Cauraugh joined my committee late in the process, but has been a mentor since my first years at UF. I appreciate his support of my career, his sincerity, and his obvious love for stroke rehabilitation. Leslie Gonzalez-Rothi has been an amazing supporter of my career since the first years also. I appreciate her support of me for the pre-doctoral fellowship and my time within the VA. Without the support of my fellow graduate students and the faculty at the University of Florida, none of this would have been possible. I want to thank the Stroke LabMatt, Po, Tara, April, Gauri, Shalaka, Vicky, Cristinafor all contributing significantly to my success and to the completion of this dissertation. Po has taught me so much about life, cultural differences and similarities, and friendship over the last five years. Although Matt left me early, he did not escape far enough to avoid helping me in my dissertation year. Matt, Po and I started on the CIMT projects early on together, and there is no one I would have rather worked with so closely. These men are wonderful friends, clinicians and researchers. Tara joined us shortly thereafter and was a great addition to the team. Tara’s support of me and my project has been unyielding, and I truly appreciate this. Sheryl has been my biggest support from the start of my life in Gainesville. I know without her I would not be where I am today. I thank her for her support, love, and friendship throughout all my years here. I am such a better person for having known her v

PAGE 6

and so blessed to have spent so much time with her. Mary, Gwen, and Portia have helped me to grow as a person. Their friendship and guidance are so appreciated and loved. Phil, Skip, Woo, Weller, Brad, Kellie, Sarah, Carrie, AC, Karen, Amy and Kelly have all been there to support me through my decision to return to graduate school and to provide diversions when needed. Over the last year, Nikki has provided me with incredible personal and professional support. I want to thank Nik for her help “teaching” me statistics, her unbelievable patience, and most importantly for her love and support through this process. This process would have been so much harder without her to help me along. I also wish to acknowledge those financial contributions that helped me pursue my research and academic goals: the VA Division of Rehabilitation Research and Development, the VA Brain Rehabilitation Research Center, the Shands Hospital Board of Directors, Department of Physical Therapy at the University of Florida, and the Foundation for Physical Therapy. Last and most importantly, I thank God for his presence in my life, and for the answering of so many prayers. I continue to pray that He will allow comfort for all individuals who have suffered a stroke and be present in their lives. vi

PAGE 7

TABLE OF CONTENTS Page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...............................................................................................................x LIST OF FIGURES...........................................................................................................xi ABSTRACT.....................................................................................................................xiii CHAPTER 1 INTRODUCTION: CONSTRAINT-INDUCED MOVEMENT THERAPY.............1 Research Aims and Hypotheses....................................................................................2 Experiment I..........................................................................................................2 General aim 1: Determine functional predictors of upper-extremity Movement Capability for CIMT...............................................................2 General aim 2: Determine functional predictors of upper-extremity Amount of Use for CIMT..........................................................................2 Experiment II.........................................................................................................3 General aim 3: Determine descriptive predictors of upper-extremity Movement Capability for CIMT...............................................................3 General aim 4: Determine descriptive predictors of upper-extremity Amount of Use for CIMT..........................................................................3 Background and Significance.......................................................................................3 Animal Studies......................................................................................................6 Experimental testing of learned non-use........................................................8 Model and translation to human patients after stroke..................................10 Predictors of Success with CIMT........................................................................12 Functional potential predictors.....................................................................13 Descriptive potential predictors...................................................................17 Summary..............................................................................................................26 2 FUNCTIONAL PREDICTORS OF CIMT OUTCOMES.........................................27 Introduction.................................................................................................................27 Methods......................................................................................................................28 Participants..........................................................................................................28 Procedure.............................................................................................................30 vii

PAGE 8

Outcomes Measures.............................................................................................31 Selected Prospective Predictors:..........................................................................32 Minimal motor criteria level........................................................................33 Finger extension and grasp release...............................................................33 Grip strength.................................................................................................34 Upper-extremity Fugl-Meyer motor score...................................................34 Frenchay activities index..............................................................................35 Data Analysis.......................................................................................................36 Results.........................................................................................................................38 Power Analysis....................................................................................................38 Descriptive Characteristics..................................................................................38 Intention-to-Treat Analysis.................................................................................39 Multiple Regression Modeling............................................................................40 Aim 1: Wolf motor function test modeling.................................................40 Aim 2: Motor activity log amountmodeling.............................................40 Discussion...................................................................................................................42 Multiple Regression Modeling............................................................................43 Wolf Motor Function Test model........................................................................43 Motor Activity Log model...................................................................................44 Other outcomes....................................................................................................45 Drop-outs.............................................................................................................46 Statistical Analysis Discussion............................................................................47 Limitations...........................................................................................................48 Conclusions.................................................................................................................49 3 DESCRIPTIVE PREDICTORS OF CIMT OUTCOMES.........................................50 Introduction.................................................................................................................50 Methods......................................................................................................................51 Participants..........................................................................................................51 Procedure.............................................................................................................52 Outcomes Measures.............................................................................................53 Selected Prospective Predictors...........................................................................54 Side of stroke................................................................................................55 Time since stroke.........................................................................................58 Hand dominance...........................................................................................59 Age...............................................................................................................60 Gender..........................................................................................................61 Ambulatory status........................................................................................62 Data Analysis.......................................................................................................63 Results.........................................................................................................................64 Power Analysis....................................................................................................64 Intention-to-Treat Analysis.................................................................................65 Multiple Regression Modeling............................................................................66 Aim 1: Wolf Motor Function Test modeling..............................................66 Aim 2: Motor Activity Logamount modeling...........................................67 viii

PAGE 9

Discussion...................................................................................................................68 Multiple Regression Modeling............................................................................68 Other Potential Predictors....................................................................................70 Statistical Analysis Discussion............................................................................72 Limitations...........................................................................................................75 Conclusions.................................................................................................................75 4 GENERAL SUMMARY AND CONCLUSIONS.....................................................77 Overview.....................................................................................................................77 Experiment I Summary...............................................................................................78 Experiment II Summary.............................................................................................83 General Conclusions and Integration..........................................................................86 APPENDIX A PILOT DATA.............................................................................................................89 Pilot Study on CIMT: Functional Predictors of Outcomes for CIMT........................89 Methods for Pilot Study 1....................................................................................90 Data Analyses for Pilot Study 1..........................................................................90 Results for Pilot Study 1......................................................................................91 Discussion for Pilot Study 1................................................................................94 Tests of body function and structure...................................................................95 Tests of activity...................................................................................................97 Tests of participation...........................................................................................99 Pilot Study on CIMT: Descriptive Predictors of Outcomes for CIMT....................100 Methods for Pilot Study 2..................................................................................101 Data Analyses for Pilot Study 2........................................................................101 Results for Pilot Study 2....................................................................................101 Discussion for Pilot Study 2..............................................................................101 Conclusion.........................................................................................................105 B BEHAVIORAL TESTS FORMS, LOGS, AND CONTRACTS.............................107 BIOGRAPHICAL SKETCH...........................................................................................129 ix

PAGE 10

LIST OF TABLES Table Page 2-1. Descriptive statistics of the continuous independent variables.................................38 2-2. Descriptive statistics of the categorical independent variables.................................38 2-3. Chi-Square and Fisher’s exact test for the differences across categorical independent variables between those participants who completed the follow-up evaluation and those that did not complete the evaluation.......................................39 2-4. T-test for difference across continuous independent variables between those participants who completed the follow-up evaluation and those that did not complete the evaluation............................................................................................39 2-5. The adjusted R squared for the [ln] WMFT models, the B weights, the p-values for all the predictors, and the confidence intervals for the significant predictors are listed...................................................................................................................41 2-6. The adjusted R squared for the MAL amount models, the B weights, the p-values for all the predictors, and the confidence intervals for the significant predictors are listed...................................................................................................................42 3-1. Chi-Square and Fisher’s Exact Test for the differences across categorical independent variables between those participants who completed the follow-up post-test and those that did not complete the evaluation..........................................65 3-2. T-test for difference across continuous independent variables between those participants who completed the follow-up post-test and those that did not complete the evaluation............................................................................................66 3-3. The adjusted R squared for the [ln] WMFT models, the B weights, the p-values for all the predictors, and the confidence intervals for the significant predictors are listed...................................................................................................................67 3-4. The adjusted R squared for the MAL amount models, the B weights, the p-values for all the predictors, and the confidence intervals for the significant predictors are listed...................................................................................................................68 x

PAGE 11

LIST OF FIGURES Figure Page 1-1 Model for development of learned non-use 1 ..............................................................9 1-2 Model of mechanism for overcoming learned non-use 1 ...........................................9 2-1 Descriptive characteristics........................................................................................29 2-2 Participant response choices for MAL amount scale...............................................32 2-3 Timeline for testing of dependent variables.............................................................37 3-1 Descriptive characteristics........................................................................................52 3-2 Participant response choices for MAL amount scale...............................................55 3-3 Timeline for testing of dependent variable..............................................................64 4-1 A graphical example of WMFT post model for Experiment I.................................79 4-2 A graphical example of WMFT follow-up model for Experiment I........................80 4-3 A graphical example of the MALa post-test model for Experiment I......................81 4-4 A graphical example of the MALa follow-up test model for Experiment I.............82 4-5 A graphical example of the MALa post-test model for Experiment II....................84 4-6 A graphical example of the MALa follow-up model for Experiment II..................85 A-1 Demographics of Participant in Pilot Study 1 and 2................................................90 A-2 List of Outcomes and Abbreviations for Pilot Study 1 & 2.....................................91 A-3 List of Predictors and Abbreviations for Pilot Study 1 & 2.....................................92 A-4 Results Table for Tests of Body Function and Structure.........................................92 A-5 Results Table for Tests of Activity..........................................................................93 A-6 Results Table for Tests of Participation...................................................................93 xi

PAGE 12

A-7 Results Table for Descriptive Predictors................................................................102 B-1 Wolf Motor Function Test score sheet...................................................................107 B-2 Motor Activity Log score sheet..............................................................................108 B-3 Fugl-Meyer Upper Extremity Motor score sheet...................................................109 B-4 Frenchay Activities Index score sheet....................................................................110 B-5 Home Diary............................................................................................................111 B-6 Daily Activity Log – General Form.......................................................................112 B-7 Daily Activity Log – Timed Trial Form.................................................................113 B-8 Daily Activity Log – Behavioral Contract Form...................................................116 xii

PAGE 13

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy FUNCTIONAL AND DESCRIPTIVE PREDICTORS OF OUTCOMES FOLLOWING CONSTRAINT-INDUCED MOVEMENT THERAPY FOR INDIVIDUALS WITH POST-STROKE HEMIPARESIS By Stacy L. Fritz August 2004 Chair: Kathye E. Light Major Department: Rehabilitation Sciences Constraint Induced Movement Therapy (CIMT) is a rehabilitative strategy used primarily with the post-stroke population to increase the functional use of the neurologically-weaker upper-extremity through massed practice while restraining the lesser-involved upper-extremity. While solid research evidence supports CIMT, limited evidence exists regarding the specific characteristics of individuals who benefit most from this intervention. The goal of this study was to determine functional and descriptive predictors for CIMT outcomes. A convenience sample of 55 individuals post-stroke was recruited that met specific inclusion and exclusion criteria. These individuals participated in CIMT 6-hours per day, for two weeks with restraint of their more-affected upper-extremity. Pre-test, post-test and follow-up assessments were performed to assess the outcomes for the Wolf Motor Function Test (WMFT) and the amount section of the Motor Activity Log (MALa). The potential functional predictors for Experiment I were xiii

PAGE 14

minimal motor criteria, finger extension, grip strength, Fugl-Meyer upper-extremity motor score, and the Frenchay score. The potential descriptive predictors for Experiment II were side of stroke, time since stroke, hand dominance, age, gender, and ambulatory status. A step-wise regression analysis was used in which the group of potential predictors was entered in a linear regression model with simultaneous entry of the dependent variables’ pre-test score as the covariate. Two regressions models were determined for each dependent variable per experiment, one for the immediate post-test and one for the follow-up post-test. For Experiment I, finger extension was the only significant predictor of WMFT outcomes. Minimal motor criteria and upper-extremity Fugl-Meyer motor score were predictive of the MALa immediately following therapy, but only the Fugl-Meyer score was predictive of outcomes at the follow-up post-test. For Experiment II, ambulatory status and age were the only descriptive characteristics predictive of MALa outcomes for the immediate post-test and follow-up post-test, respectively. None of the potential descriptive predictors showed a predictive relationship with the WMFT in Experiment II. These experiments provide the most comprehensive investigation of predictors of CIMT outcomes to date. Further substantiation of these findings in more diverse samples is warranted in order to meet the urgent need of determining the appropriate candidates for CIMT. xiv

PAGE 15

CHAPTER 1 INTRODUCTION: CONSTRAINT-INDUCED MOVEMENT THERAPY The number of stroke survivors has almost doubled over the last 25 years, 1,2 and is predicted to double again in the next 50 years. 3 Currently, stroke is the leading cause of disability in the United States 3 and costs the American public more than 43 billion dollars per year. 4,5 More than half 4 of the 4.7 million stroke survivors 3 have residual motor disability, and of these, 30-66 percent have a nonfunctional paretic arm. 5 Undeniably, rehabilitative strategies aimed at reducing stroke related disabilities are important to this growing population. Currently, few traditional rehabilitation methods have been proven effective in the treatment of stroke survivors. 6 Constraint Induced Movement Therapy (CIMT) is a rehabilitative strategy used primarily with the post-stroke population. This therapy increases the functional use of the neurologically weaker upper-extremity through massed practice while restraining the lesser involved upper-extremity. 1 CIMT is reported to significantly improve functional use of the upper extremity in 20 to 25 percent of people with chronic stroke disability. 1 Limited evidence exists, however, regarding the specific characteristics of individuals who benefit most from this intervention. The goal of this research was to determine the predictive factors of functional improvements following traditional CIMT. The identification of clinical predictors for outcomes of CIMT is of value to both research and clinical settings. 7 This will help target the individuals who benefit most from this intervention. The goal of this study was to establish clinically relevant and usable regression models, not extensive or comprehensive models. Extensive 1

PAGE 16

2 multivariate models are often difficult to use and, therefore, have little value because they are less readily incorporated. 8 In addition, more comprehensive models often have little more predictive value then simple clinical predictors. 9 The goal of this research was to establish a predictive model that can identify individuals, based on their functional level and descriptive factors, who have better outcomes with CIMT. The following aims will be addressed. Research Aims and Hypotheses Experiment I General aim 1: Determine functional predictors of upper-extremity Movement Capability for CIMT Specific aim 1. To determine the relationship of five functional predictors (minimal motor criteria level, finger extension, grip strength, upper extremity Fugl-Meyer, and Frenchay Activities Scale) to the timed component of the Wolf Motor Function Test, a test of movement capability. Hypothesis 1. All five predictors will be included as significant predictors in the regression model for the test of movement capability at the immediate post-test and follow-up post-test. General aim 2: Determine functional predictors of upper-extremity Amount of Use for CIMT Specific aim 2. To determine the relationship of the same five predictors to the amount component of the Motor Activity Log, a test of perceived use. Hypothesis 2. All five predictors will be included as significant predictors in the regression model for the test of perceived use at the immediate post-test and follow-up post-test.

PAGE 17

3 Experiment II General aim 3: Determine descriptive predictors of upper-extremity Movement Capability for CIMT Specific aim 3. To determine the relationship of six descriptive predictors (side of stroke, time since stroke, dominance, age, gender, and ambulatory status) to the timed component of the Wolf Motor Function Test, a test of movement capability. Hypothesis 3. All six predictors will be included as significant predictors in the regression model for the test of movement capability at the immediate post-test and follow-up post-test. General aim 4: Determine descriptive predictors of upper-extremity Amount of Use for CIMT Specific aim 4. To determine the relationship of the same six descriptive predictors to the amount component of the Motor Activity Log, a test of perceived use. Hypothesis 4. All six descriptive predictors will be included as significant predictors in the regression model for the test of perceived use at the immediate post-test and follow-up post-tests Background and Significance Research evidence supports Constraint-Induced Movement Therapy (CIMT), but many questions persist about who can benefit from this therapy. 1,2,6,10-25 CIMT is mainly used with the post-stroke population to increase the functional use of the neurologically-weaker, upper-extremity (UE) through massed practice of hand and arm tasks while restraining the lesser involved UE. 6 The goal of CIMT is to overcome learned non-use and to improve functional use of the more-affected UE. 10 Originally tested in an animal model, the results of CIMT studies have demonstrated significant and lasting improvements of UE movement function. 1,2,6,10-25 Edward Taub, Steven Wolf and their

PAGE 18

4 colloborators have published extensively on the effectiveness of CIMT. 1,2,6,10,11,13-18,20,25 Together they have shown that CIMT significantly improves functional use of the UE of chronic stroke patients who meet what is termed “minimal motor criteria.” 1,14,26 Minimal motor criteria are defined as 20 degrees of wrist extension from a fully flexed position and 10 degrees of metacarpo-phalangeal (MCP) and proximal inter-phalangeal (IP) joint extension of the thumb and two fingers. 27 Approximately 20 to 25 percent of individuals with stroke meet minimal motor criteria. 1 Participants that initially have lower functional recovery who cannot meet these range requirements have had less improvements with traditional CIMT than patients with higher levels of motor ability. 6 While Taub’s definition of CIMT is broad, encompassing a family of rehabilitation techniques, there are two main components that define CIMT: 1) constraint of the less-involved UE 90 percent of waking hours forcing the use of the more-involved UE, and 2) massed practice of the more-involved UE. 2 CIMT has primarily been researched in individuals who have chronic disability from stroke. 1,2,6,11,12,15,16 Many rehabilitation specialists, however, accept the convention that motor function plateaus at approximately 6 to 12 months post-injury. In fact, this conviction has been consistent for many years. Ogden and Franz (pg. 32), in 1917, stated “it has long been believed that if improvement in motor ability does not occur in man within a period of two years following their cerebral accident, the paralysis is permanent.” 28 The clinical stroke guidelines for physical therapy practice state that functional recovery occurring in the first six weeks after stroke is predictive of functional outcomes. 29 This convention is currently being challenged by results from CIMT studies. CIMT focuses on the plasticity of the cortex and assumes that the nervous system always

PAGE 19

5 remains plastic; thus, functional improvements can occur at any time following the insult. 1 In fact, individuals as many as 18-years post-stroke have demonstrated improvements following CIMT. 2 By engaging the hemiparetic UE in massed practice of functional tasks, CIMT is believed to alter the representation of this UE within the primary motor cortex. 2 Studies in human and non-human primates have shown that the neural representation of hand muscles becomes enlarged when trained on a discrete motor skill. 30-32 Two transcranial magnetic stimulation studies support activity-dependent, neurological changes following CIMT in survivors of stroke. 33,34 Consequently, changes following CIMT have also been observed at the level of pathology/disease, as well as the level of body function/structures, activity, and participation. 35 CIMT results have recently been labeled the most promising evidence that motor recovery can occur in the post-stroke hand in patients who have some residual purposeful movement. 36 While the evidence for CIMT is promising, many questions persist concerning who will benefit from this form of therapy. Wolf et al. (2002, pg 331) stated that voluntary movements of wrist and finger extensors were “predictors of future acquisition of independent limb use” stating that the criteria established from electromyography studies would include about 20 to 25 percent of individuals post-stroke, which has more recently been defined as those individuals who meet minimal motor criteria. 18,26 The methodology used to determine this criteria, however, involved a small number of participants, and has not been replicated. 1,26 Furthermore, CIMT has been investigated primarily in right-hand dominant individuals, who have had only one stroke at least a year prior to the intervention. 17 This sample is not representative of all 730,000 people per year that have a stroke, 17 which begs the question, "Is CIMT a

PAGE 20

6 potentially effective rehabilitation technique for other stroke populations?” Recent pilot studies by Light (unpublished) have shown improvements in participants who do not meet the minimal traditional requirements. 37 Initial studies by Taub et al. (1998) demonstrated that individuals with less than minimal motor criteria improve; however, these participants must present with a minimum of 10 degrees of extension of the wrist, thumb, and two fingers. 6 According to Taub et al., this increases the applicability of CIMT to almost 50 percent of those who have had a stroke. 6 Selection criteria for participation should be carefully examined to determine who benefits most from this intervention and to determine what items are predictive of CIMT success. Stroke is the most common disabling condition with 30 to 66 percent of the individuals surviving losing functional ability in their more-affected arm and hand. 2,38 The need for innovative rehabilitation is clear. Currently, the individuals living with chronic stroke have few well-researched and effective therapies available to them. Those individuals who benefit the most from CIMT need to be determined. Animal Studies The origins of CIMT are based on Taub’s early work with deafferented primates. 39 This was then followed by translational work in collaboration with Wolf on studies with humans post-stroke. 11 Over 40 years ago Knapp et al. (1963) observed that monkeys did not use their UEs, in ”free situations,” following unilateral forelimb somatosensory deafferentation. 40 When the intact UE was restrained, however, the deafferented limb became the only means by which the monkey could function. The monkeys were, therefore, “forced” to use the deafferented limb and functional return was observed. Furthermore, if the constraint was used for 1-2 weeks, functional motor return could be permanent. 1,2

PAGE 21

7 These observations led to the formulation of a hypothesis to explain why restraint of an intact UE improves functional return of a weaker extremity. Taub et al. first hypothesized that the functional return occurred because of a theory he termed “learned non-use,” and believed that it correlated with functional return observed in humans following an insult to the central nervous system. 2 A substantial neurological injury results in a depressed condition of motorneurons. This causes a shock-like phenomenon that usually results in a greater initial deficit, when compared to an individual’s condition following spontaneous recovery. 1,6 The shock can either be at the level of the spinal cord, spinal shock, or at the brain, diaschisis or cortical shock. A similar condition results with experimental-induced deafferentation in monkeys. The experimental deafferentation causes neural shock rendering the monkey unable to perform motorically. 1 After a period of time, however, the spinal shock decreases. This results in less edema allowing greater potential for movement. The period for shock in monkeys lasts from 2 to 6 months, during which time they are able to progressively regain the ability to use the affected limb. 6 During this period of shock, the monkeys are unable to use the surgically deafferented limb. The animal attempts to use it immediately postoperatively and cannot. With continued attempts to use the arm, the monkey experiences pain, limited success and ineffective movement. This leads to decreased use of that extremity. The monkey learns that it can function quite well in its environment with its remaining three limbs, and is positively reinforced for this behavior. For example, while using the unaffected arm the monkey is able to retrieve food, groom, and be active. Conversely, attempts with the affected UE resulted in loss of food, possible falls, frustration, and general failure. 2,22

PAGE 22

8 These adverse conditions are understood as punishment, and thus result in a suppression of further attempts. This cycle is known as learned non-use, because the monkey has learned not to use the affected UE. This response of non-use continues several months into the time when the UE is no longer in shock and may be potentially functional. The movement, therefore, is available just not accessed. (See Figure 1-1, adapted from Morris et al. 1997.) 2 The goal of CIMT is to overcome learned non-use and to improve functional use of the affected UE. Restraint of the unaffected arm, following the period of shock, necessitates use of the affected arm to perform activities. The animal either uses the arm, or it cannot carry out the majority of daily activities. This forced shift in motivation overcomes the early learned non-use. Taub et al. 2 found that the restraint must be used for one to two weeks to overcome the learned non-use phenomenon (See Figure 1-2, adapted from Morris et al. 1997). 1,2 Experimental testing of learned non-use Taub performed experiments to directly test the theory of non-use. 39 Monkeys’ affected UEs were restrained immediately following a deafferentation surgery, during the period of spinal shock, thereby prohibiting the monkeys from using their injured arm during the period of spontaneous recovery. This restraint prohibited the monkey from experiencing the negative reinforcement that corresponds with attempted use of the affected extremity during the shock period. The hypothesis was that this forced non-use of the affected extremity should prevent the monkey from developing learned non-use. Following the period of spinal shock, the restraint was removed and the animals then used the affected side with more success and less pain because they were no longer

PAGE 23

Unsuccessful motor attempts Punishment (pain, failure, incoordination ) Behavior suppression Masked ability Positive r e inf o r ce m e n t Less effective behavior strengthened Compensatory behavior patterns Learned nonusenormally permanent reversal possible Injury e.g. stroke, dorsal rhizotom y DEVELOPMENT OF LEARNED NONUSE Figure 1-1. Model for development of learned non-use (Figure adapted from Morris et al. 1997) 1 9 Masked Recovery of Limb use Increased Motivation Accesses Function Affected Limb Use Positive Reinforcement Further Practice More Reinforcement Limb Used in Life SituationPermanent Learned Nonuse OVERCOMING LEARNED NONUSE Figure 1-2. Model of mechanism for overcoming learned non-use (Figure adapted from Morris et al. 1997) 1

PAGE 24

10 limited by the effects of spinal shock. The animals were never limited by learned non-use. 2 An additional study also directly tested the theory of learned non-use. A small sample of monkeys received forearm deafferentation during the gestation period. 41 The spinal shock period, therefore, was experienced during gestation while prenatal monkeys were constrained in the uterine environment. These monkeys did not experience the negative reinforcement for attempted use of the affected UE. At birth, the animals exhibited spontaneous and purposeful use of the affected UE, therefore demonstrating that the learned non-use did not occur because monkeys were able to utilize the deafferented arm at birth. This constitutes another line of evidence for the learned non-use theory. 1,2 Model and translation to human patients after stroke The early studies involving deafferented monkeys led to a testable hypothesis in humans with stroke. The learned non-use principles could partially explain the typical motor recovery seen in individuals following stroke. Whereas the deafferented monkeys experience spinal shock, patients suffering from stroke or brain injury would have a period of cortical shock. 1 While a sensory deafferentation is not equivalent to that of a stroke, in which both motor and sensory pathways are disrupted, both result in a period of shock. The original attempts to utilize CIMT in humans to improve UE function following stroke and brain injury were performed with 25 subjects who were at least one year post-lesion. 11 The subjects chosen for the study had varying levels of residual motor control, but still demonstrated significant disability. The subjects wore a sling on the uninvolved arm for two weeks, forcing use of the involved UE. Participants were allowed to remove

PAGE 25

11 the restraint for half an hour for exercise, bathing, and sleep. No specific training was given for the therapy, just encouragement to use the affected UE. 1 For this initial study, the CIMT included preand post-tests using the Wolf Motor Function Test. Wolf’s results demonstrated 90 percent improvement of timed scores. There was no apparent difference between right and left-sided hemiparesis and no difference noted between patients with stroke and with traumatic brain injury. These data suggest that learned non-use can be reversed using CIMT in this sample of participants. 11 Taub et al. have also performed pilot work with CIMT. 16 Nine subjects ranging from 1-18 years post stroke were included using similar inclusion requirements as Wolf. The subjects were randomly assigned to one of two groups. The experimental group wore a restraint on the unaffected UE for all waking hours for 12 days, unless unsafe to do so. They signed a behavioral contract, kept an activity log, and had guided activities at the clinic for seven hours per day for eight days of the 12 day session. The attention-control group did not wear a restraint, were encouraged to use the involved UE in activities during clinic hours, and had two placebo physical therapy sessions and instruction in a home exercise program, consisting of passive range of motion exercises. 16 The experimental group demonstrated significant improvements in motor function tests and actual amount of use tests. The attention-control group did not show significant improvement. More importantly, the gains the experimental group demonstrated were completely maintained two years after the intervention. 1,16 Continued clinical trials are being performed to address other populations, optimal time periods and different modes of CIMT. 42,43

PAGE 26

12 In addition to the restraint of the UE, various training techniques were researched to understand which may complement the condition of restraint and improve functional use of the UE. Shaping, a training technique, was used as an adjunct to the restraint to complement the forced-use and in an attempt to regain improved motor skills. Shaping is a training procedure that uses successive steps to gain a desired motor outcome and allows for performance gains with small amounts of motor return. Shaping was found to advance motor recovery more than the restraint alone. 1,2 Rehabilitation researchers have yet to identify a truly effective treatment for upper-limb hemiparesis 6 and, therefore, they are continuously searching for improved approaches. Consequently, new treatment methods, such as CIMT, are often accepted before the relevance of the therapy to particular clients is clearly understood. Understanding the functional and descriptive predictors of CIMT can help researchers and clinicians target the appropriate populations for this intervention. The goal of this project was to determine the characteristics of those who benefit most from CIMT. Predictors of Success with CIMT The following possible predictors for outcomes following CIMT were investigated. For Experiment 1, five functional predictors are investigated: 1) minimal motor criteria level, 2) amount of finger extension, 3) amount of grip strength, 4) upper extremity Fugl-Meyer score and, 5) score of the Frenchay Activities Scale. For Experiment 2, six descriptive predictors are investigated: 1) side of stroke, 2) time since stroke, 3) hand dominance, 4) age, 5) gender, and 6) ambulatory status. These predictors were included in the regression model for several reasons. First, many of these predictors appear in stroke rehabilitation research as predictors of other outcomes, such as differing therapeutic interventions, return to function, and return to life roles. Second, these

PAGE 27

13 predictors are often discussed in CIMT literature as having a potential to impact outcomes. Third, some predictors demonstrated strong predictive value in the pilot studies (Appendix A). Each of these are further defined and explained in the following text. Functional potential predictors Minimal motor criteria level. CIMT appears to be effective in improving movement capabilities in a subset of approximately 20 to 25 percent stroke survivors who meet minimal motor criteria, demonstrating a relatively high level of hand and wrist control. 17 In a small sample Taub and Uswatte found that lower functional recovery of the hand wrist was related to poorer improvements compared with those with higher levels of motor ability. 44 Despite this finding, individuals with lower functional recovery of the hand and arm may still have significant success from participating in CIMT. Since current selection criterion for CIMT needed further attention, we have expanded selection criteria in the current CIMT research to include individuals that did not meet traditional minimal motor criteria level. Participants were divided into one of two functional categories, identified as high functioning and low functioning. High functioning participants met traditional minimal-motor criteria of 20 degrees of extension at the wrist from a relaxed flexed position, and 10 degrees of extension of two fingers and the thumb at the MCP and PIP joints. The low functioning participants had to demonstrate trace evidence of wrist extension from a relaxed flexed position a minimum of three times in a one-minute period. In addition, they needed to slightly extend (no degree requirements) two fingers at any joint from the same relaxed flexed position. Inclusion of this lower functioning group, those that do not meet minimal motor criteria, will aid in clarifying the necessity of this baseline criteria for participation in CIMT studies. Specifically, if

PAGE 28

14 minimal motor criteria is not a predictor, it may not be an appropriate determinant of CIMT participation. Finger extension and grasp release. Finger extension is essential to functional movement of the hand. Individuals post-stroke, however, often present with limited or reduced ability to perform both fine and gross movements of the hand. 45 Secondary to the intrinsic hand muscle hypertonia, which is commonly present post-stroke, controlled grasping is often very difficult. Individuals frequently can grasp objects via a massed grasp; however, this often requires recruitment of abnormal movement patterns. Furthermore, individuals post-stroke often have greater difficulty releasing an object than grasping. 45 Difficulty releasing objects may be due to a flexor synergy that effectively limits isolation of joint movements out of synergy. For the purpose of this study, finger extension and grasp release was defined as the ability to actively release a mass flexion grasp as defined by Fugl-Meyer assessment. It is graded as follows. A ” was given if the individual is unable to release the grasp, a ” if the participant could partially release the mass flexion grasp, and a ” if the participant could fully extend their fingers from the starting grasp position. 46 Finger extension/grasp release; therefore, is defined as release of a mass flexion grasp and hand opening. Finger extension plays a role in functional use of a hand; therefore, it was included as a predictor for CIMT. Grip strength. Grip strength was used as a potential predictor in the regression model because of its prevalence in the literature as a measure of outcome post-stroke 47-49 and as a predictor of outcomes. 50-56 For example, the absence of measurable grip strength at one-month post stroke is predictive of poor upper-extremity functional recovery. 5,57-59

PAGE 29

15 In contrast, early return of active grasping is a positive prognostic sign. 5,58,59 Although grip strength is strongly associated with chronological age, 60 independent of this relationship, it is a powerful predictor of disability, morbidity, and mortality. 53 Low grip strength is associated with disability, while there are strong relationships between mid-life measurements of grip-strength to predictions of long-term survival. 52,61 For this project, grip strength was assessed using standard grip strength dynamometry techniques 62 using a Jamar dynamometer at level three setting. Standard methods for handheld dynamometry and the reliability of the testing have been well established. 62 Upper-extremity Fugl-Meyer motor score. The Fugl-Meyer Assessment is a measurement of the percent recovery of a person following a stroke that provides objective and quantifiable assessment of motor function63 and was designed primarily for use in rehabilitation settings. The test is organized according to Brunnstrom levels of recovery and is divided into five sequential recovery stages.63 The Fugl-Meyer motor component is an assessment of ability to move in/out of synergy, reflexes, wrist stability, grasping ability, and coordination (Appendix B). The validity of the Fugl-Meyer is good and overall reliability is high (intraclass correlation coefficient of 0.96).46 This standardized assessment is also often used as a measure of function in stroke rehabilitation studies1,18,64-70 and has been shown to be predictive of dependency and functional level following stroke.5,7,71 For this project, the UE motor portion of this exam was used as a predictor of outcomes for CIMT. The total possible score for the UE motor section is 66. Using the

PAGE 30

16 Fugl-Meyer score allowed for the regression model to encompass a potential predictor that represents many aspects of UE recovery following stroke. Frenchay Activities Index. The Frenchay Activities Index is a survey that measures extended activities of daily living and was developed specifically for measuring disability and handicap in individuals with stroke. It includes 15 items that measure level of activity in areas such as domestic chores, leisure/work, and outdoor activities. During a short interview, the participant is asked to determine how often they perform a given activity using a scale of 1 to 4 (Appendix B). The item scores are totaled to determine the final score, which can range from 15 (inactive) to 60 (highly active). 72,73 The reliability is 0.78 to 0.87. 73 The Frenchay Activities Index was included as a potential predictor in the regression model for many reasons. First, it is frequently cited for determining outcomes post-stroke. 73-82 Second, the Frenchay was the only potential predictor used in this model that represents questions from the participation level of the International Classification of Functioning, Disability and Health model (ICF). 35 The Frenchay, however, still has overlap with the activities component of the ICF model, with questions like: “do you walk outside for more than 15 minutes.” In addition, this assessment is strongly affected by environmental and personal factors. 35,83 For example, if an individual can functionally walk, a personal factor would exist if the individual avoids walking because they do not want to be seen by the neighbors. An environmental factor that could be limiting would be if they live on a gravel road and can not walk safely on the terrain. In either case, the Frenchay represents questions including different areas of the ICF model that were not previously included as predictors.

PAGE 31

17 Third, the Frenchay was frequently correlated to outcomes of CIMT in the pilot study. Inclusion of the Frenchay will clarify how important ‘being active’ determines outcomes with CIMT. The relationship between the Frenchay score and CIMT outcomes could lend insight into involvement and independence. For example, an individual who scores high on the Frenchay is considered more independent because of their involvement in his/her own self-care and activities. In contrast, an individual with a lower Frenchay score may lack motivation or may not seek independence from a caregiver. Level of activity could be an important predictor of success with such a time-demanding, attention-driven therapy as CIMT. Descriptive potential predictors Side of stroke. Side of stroke is an obvious factor to consider when studying the effects of CIMT. A great deal of research has assessed right versus left-brain functions, rightsided versus left-sided strokes, and the varying effects that these have on patients’ presentation, functioning, and outcomes. Ornstein (1997) states that over the last 25 years over 45,000 articles and books have been written on the two hemispheres. 84 While some CIMT studies have included participants with both left and right-hemispheric damage, there was no main effect found for side of hemiparesis. 15 Studies, therefore, showed no differences in outcomes based on an individual’s side of hemiparesis. Generally speaking, the left hemisphere is concerned with analytical processing of individual components, sequencing of tasks, and language, while the right hemisphere is more focused on perception of whole and spatial tasks. 85 An individual who sustains a left-brain stroke may present with an inability to solve problems, is often more easily angered and frustrated, has impaired retention of information, and may present with language difficulties and/or apraxia. 85 If an individual has difficulty with language, they

PAGE 32

18 may have more difficulty understanding the directions for the therapy. They may be limited in keeping track of their home-activity log or they may misunderstand directions. This limitation could be augmented by the fact that individuals with left-brain damage also have impaired retention of information. With communication difficulty, there may be increased levels of frustration, especially given the social nature of CIMT that develops due to extensive time with trainers. Furthermore, frustration is often greater in individuals with left-brain damage, so a frustrating therapy, such as CIMT, may easily add to the already pre-existing tendencies towards increased levels of frustration. This frustration could lead to decreased levels of motivation or possibly difficulties with compliance. Individuals with left-brain stroke also have more difficulty with problem-solving. Again, this could lead to more difficulties with the therapy, as problem solving for new compensatory strategies are often essential to success with the therapy. An additional limitation that may affect individuals with a left-brain stroke is apraxia. Nothing is reported in the CIMT literature regarding outcomes for individuals with apraxia. Apraxia is defined as a disorder of skilled movement not caused by weakness, ataxia, akinesia, deafferentation, inattention to commands, or poor comprehension. 86 Apraxia is a disorder of a praxis system, and this praxis system involves memory representations that are formed based upon experiences with purposive actions. 87 People with apraxia predominantly present with right hemiparesis, affecting their dominant right hand. Although apraxia presents bilaterally, patients rarely complain about their disorder and are often unaware of it. This is because the person with apraxia often blames their clumsiness on use of their left or non-dominant hand. Clinically, those with ideomotor apraxia present with normal dexterity, in the non-hemiparetic hand. 86

PAGE 33

19 The errors made vary and spatial errors are most characteristic. Spatial errors include activating movements at incorrect joints. 87,88 hose with ideomotor apraxia may also have timing errors, such as a delay in initiation or inappropriate pauses, noticed especially when the plane or trajectory needs to be changed. Timing errors also present as failing to coordinate appropriate speed of movement with that required of the task. 89 These types of errors may impede acquisition 90 of new motor skills, and thus, result in increased difficulty with therapies such as CIMT. As presented, individuals with left-brain stroke may have unique challenges that could influence success with CIMT. Persons with right-brain stroke, however, also have distinctive problems that could limit success with CIMT. They present with left-side neglect, often have difficulty with spatial-perceptual tasks, are more impulsive, and frequently have greater balance problems. 85 Unilateral neglect is common following right-hemispheres stroke, with a reported incidence varying significantly from 10 to 82 percent having left-neglect. 91 Neglect is the inability to attend, or orient to, meaningful stimuli contralateral to the lesion. 89 Neglect may be either a problem recognizing sensory stimuli (referred to as sensory, attentional, inattention, or perceptual neglect), or it may be a problem with motor planning in response to stimuli (referred to as motor, intentional or output neglect). 91 Whether a CIMT participant presenting with either type of neglect would benefit from the therapy is unknown. Individuals with neglect may have significant difficulties attending to the CIMT tasks. Neglect would then result in interference with motor acquisitions, leading to poorer outcomes. Conversely, individuals may be able to

PAGE 34

20 overcome the inattention caused my neglect, due to the constant attention-driven methods of CIMT, and demonstrate improvement. Individuals with right-brain stroke also present with spatial-perceptual deficits which may result in difficulty performing some components of the training. This difficulty could lead to increased levels of frustrations with the therapy and affect overall success. In addition, individuals with right-brain stroke often have increased balance problems. The balance problems could lead to decreased time wearing the mitt and issues with compliance. Mitt time could be decreased in individuals with balance problems because of increased safety concerns and or use of an assistive device. The various presentations of pathology attributed to side of lesion could affect outcomes following CIMT, therefore, it is included as a potential predictor. Time since stroke. Recovery of function after stroke is multi-causal. Spontaneous recovery, related to time since stroke, contributes to improvement in motor function following stroke. 92 CIMT, given acutely during this period, could possibly enhance spontaneous recovery and boost functional recovery. Conversely, early attempts at forcing the use of the hemiparetic upper extremity could lead to worsened learned non-use. The learned non-use could be enhanced at such an early time when an individual may have limited function, because increased use of their weakened hand may lead to failure or pain. This would result in negative reinforcement and subsequent increased learned non-use. Some would argue against worsened learned non-use, however, by insisting that CIMT is gauged to the functional level of an individual. The positive reinforcement that is a key component to CIMT would, therefore, circumvent development of learned non-use. Furthermore, CIMT is based on the assumption that

PAGE 35

21 the nervous system always remains plastic, thus the time of administration of therapy would not be a factor. Improvements would, then, be possible at any time following the insult. 1 While people who are as much as 18-years post stroke have demonstrated functional improvements, 2 the literature results are still unclear if individuals in the chronic phase of stroke can improve to the same extent as those who have sustained a stroke more recently. Beginning CIMT in the acute, sub-acute or chronic phase after stroke may affect outcomes. Time since stroke is often used as a predictor for stroke outcomes. 5,57-59,93-95 Findings from longitudinal studies, with repeated measures across time, demonstrate that neurological recovery shows a nonlinear progression pattern as a function of time, but few individuals show additional improvement after three months post-stroke. 5,57-59,93-95 The length of time, in which there is a lack of improvement following a stroke, reflects the intrinsic cerebral damage and should be seen as an important predictor of poor outcome. 5,92 While the greater the time since stroke is associated with poorer functional recovery, the attention-driven methods of CIMT may counteract this temporal relationship, resulting in improvements at any time post stroke. Hand dominance. Hand dominance affects an individual’s functional ability after a stroke, yet current CIMT studies have not sufficiently addressed dominance and its role in functional recovery. 96 Hand dominance is a behavioral manifestation of hemispheric asymmetry. Typically, people who are considered left-hand dominant perform some tasks with their right hand, while right-hand dominant individuals rarely perform tasks primarily with their left hand. 96 The right-hand dominant population has stronger lateralization of the hemispheres while people who are left-hand dominant seem to

PAGE 36

22 present with more bilateral use of hemispheres with decreased specializations. A left-handed individual who sustains a stroke resulting in left hemiparesis, therefore, may recover function faster due to the lack of hemispheric lateralization. This decreased hemispheric lateralization is due to a long history of performing tasks with both hands which utilizes both sides of their brain. Conversely, if a right-handed individual, with right hemiparesis, loses the function of the left brain, they are less capable of compensating with the right brain due to the extensive pre-morbid lateralization. 96 CIMT’s effectiveness, as it relates to upper-extremity dominance, is unclear. The early CIMT studies excluded individuals with left-hand dominance or left hemiplegia, in order to test and analyze with less variance. 10,16 The generalizability, therefore, to non-right handers is limited. Another interesting, yet unexplored question, is whether CIMT is of greater benefit to individuals with dominant-side hemiparesis as opposed to non-dominant hemiparesis. Intuitively, one may consider that people with dominant-side hemiparesis would be more motivated to regain function of the affected extremity because this is the extremity of functional preference and practiced motor skill. Moreover, the dominant hemisphere has more intricate motor programs with better-developed coordination and skill, thus translating to improved neural representation. Does this pre-morbid increased representation of the dominant hand remain post-stroke, allowing for increased ease in recovery of function? In contrast, an individual with a non-dominant-side stroke could possibly return to a previous level of functioning at a quicker pace as compared to a dominant-side stroke. This possibility exists because of the decreased functional demands on the non-dominant extremity. The relationship of functional return following

PAGE 37

23 CIMT to pre-morbid handedness and motivation to regain use of a dominant extremity is unclear. Perhaps, a more interesting question is to investigate if dominance plays a role in predicting recovery at the follow-up assessment. The intense methods of CIMT may allow for no difference to be observed between individuals with dominant and non-dominant hemiparesis immediately following CIMT, however, this may change at follow-up. Obviously, dominance is an important issue that needs to be investigated as a predictor of function following CIMT. Age. The research literature to date presents conflicting results regarding the role of age in rehabilitation. On one hand, motor performance factors have been demonstrated to be greatly influenced by age. 97-100 Certainly age could be an important predictor of recovery potential following stroke. 98-100 According to Jongbloed (1986), age was identified as a significant prognostic factor in numerous studies. 101-104 Specifically, these studies indicate that age is negatively correlated with functional return. In studies reporting age as a factor, however, additional contributors of functional limitations such as co-morbidities, presence of a caregiver, and severity of stroke are frequently not accounted for. On the other hand, conflicting studies have shown that age does not have a negative impact on function over time. 101 In fact, much research has demonstrated the benefits of intensive stroke rehabilitation programs regardless of age. 4,100,105,106 Kugler et al. (2003) states that age should not be a limiting factor in the early rehabilitation of stroke patients. 98 The role age plays as a predictor of outcomes post-stroke is uncertain. The inclusion of age in the regression as a potential predictor is important to help clarify the role age plays in CIMT.

PAGE 38

24 Gender. The relationship between gender and stroke rehabilitation is not well explored. Scarce information exists about gender differences in the management of individuals with stroke. 107 Research on gender differences following stroke is needed in order to provide useful insight for long-term intervention and establishment of appropriate therapy. 107 The direct relationship between incidence of stroke and advancing age means that female patients, because they have a longer life expectancy than males, may bear the burden of much of the disease and subsequent disability. 107 Wyller et al. (1997) studied gender difference in functional outcomes following stroke. 108 Utilizing age-adjusted odds, these authors concluded that women seem to be functionally more impaired by stroke than men. 108 In addition, women are diagnosed with depression twice as frequently, which is strongly correlated with more impairment and loss of function. 109 Conversely, Twigg et al. (1998) found no significant correlation between gender and functional outcomes following stroke. 110 Time since stroke differences in the previously mentioned studies could explain some of the contradictory results. For example, the study that observed differences between genders 108 was performed in a chronic population, one-year, post-stroke, while the study that found no differences between genders utilized the acute and sub-acute population at a rehabilitation facility. 110 Gender differences may not emerge until further in the recovery process. Nonetheless, the question regarding gender influence on outcomes following stroke is important and remains unclear. Ambulatory status. Ambulatory status was included as a potential predictor of upper-extremity recovery for several reasons. This predictor is included in the regression model because cumulative deficits post-stroke can affect individuals functional

PAGE 39

25 outcomes. 111 Individuals who are non-ambulatory, therefore, may have poorer outcomes. Moreover, while the most rapid recovery for both upper and lower extremities occurs within the first 30 days after stroke, the severity of motor impairments and the subsequent patterns of recovery are similar for both extremities. 112 A relationship exists, therefore, between upper and lower extremity motor recovery. 112 In addition, the ability to walk is a strong predictor of functional outcomes 112 and has been included in regression models previously for stroke recovery. 113 Individuals who are able to ambulate may be able to use their hand in more functional tasks. If a participant is able to stand and walk, they have more synergies available to accomplish gross motor tasks. Furthermore, if someone is walking they may spontaneously find themselves involved in more activities due to increased access. Ambulatory status could lead to increased independence with tasks, and therefore, ambulators may be better candidates for CIMT. Including ambulatory status allows for a simple predictor that could potentially be very useful for determination of outcomes following CIMT. For the purposes of this study, ambulation was used as a dichotomous variable. If the participant presented to the clinic ambulating, with or without an assistive device, then they were considered functionally ambulatory. If they presented in a wheelchair the majority of the time, they were categorized as non-ambulatory. The identification of clinical predictors for outcomes of CIMT is essential and of value to both research and clinical settings. 7 The goal of this study was to establish clinically relevant and usable models, not extensive or comprehensive models. Extensive multivariate models are often difficult to use and, therefore, have little value because they are less readily utilized. 8 In addition, more comprehensive models often

PAGE 40

26 have little predictive value over simple clinical predictors. 9 Other predictive variables could have been included in this model, such as cognitive function or motivational variables, which may have increased the predictive capability of these models. Modern imaging techniques could have been included to identify type or location of lesion, since they are valuable in prediction of stroke outcomes, 114 these predictors, however, are unlikely to become easily accessible for clinical use. 9 For this reason, they were not included in the model. The importance of a predictor study for upper-extremity recovery is well expressed by Kwakkel (2003). 5 Knowledge on outcome of the upper limb is of paramount interest to clinicians to optimize their treatment goals and to inform patients properly. In those cases in which some return of dexterity is expected, training the paretic arm is justified. However, if the prognosis is poor, teaching the patient to deal with existing deficits may be more realistic, thus allowing for the use of compensating strategies. 5 Summary Limited evidence exists regarding the specific characteristics of individuals who benefit most from CIMT. This intervention is reported to significantly improve functional use of the UE in only 20 to 25 percent of people with chronic stroke disability. 1 There is an urgent need to determine appropriate selection criteria for CIMT participation. Rehabilitation professionals continuously search for improved approaches for treatment, and consequently, new treatment methods, such as CIMT, are often accepted before the relevance of the therapy to particular clients is clearly understood. Who benefits most from CIMT needs to be understood, in order to target the correct population. The following experiments address the goal of this project which was to determine the characteristics of individuals who benefit most from CIMT.

PAGE 41

CHAPTER 2 FUNCTIONAL PREDICTORS OF CIMT OUTCOMES Introduction Stroke is the most common disabling condition with 30 to 66 percent of individuals losing functional ability in their more-affected arm and hand. 2,38 The need for innovative rehabilitation is clear since few well-researched and effective therapies are available to individuals post-stroke. Constraint Induced Movement Therapy (CIMT), however, has shown great promise for individuals with hemiparesis. CIMT is a relatively new rehabilitative strategy used primarily with the post-stroke population. It encompasses a family of rehabilitation techniques having two main components 1) constraint of the less-involved upper extremity (UE), forcing the use of the more-involved UE, and 2) massed practice of the more-involved UE. 2 CIMT results have recently been labeled the most promising evidence that motor recovery can occur in the hands of individuals post-stroke that have some residual purposeful movement. 36 Significant scientific evidence supports the efficacy of CIMT in improving function after stroke, but the functional characteristics of those who will benefit most from this therapy is unknown. 1,2,6,10-25,115 To date, CIMT studies have demonstrated significant and lasting improvements in UE movement function, 1,2,6,10-25,115 especially for those 20 to 25 percent 1,18 of all individuals with post-stroke hemiparesis who meet what is termed ‘minimal motor criteria.’ 1,14,26 Minimal motor criteria is defined as 20 degrees of wrist extension from a fully flexed position and 10 degrees of metacarpo-phalangeal (MCP) and proximal inter-phalangeal (IP) joint extension of the thumb and two fingers. 27 The term ‘minimal motor 27

PAGE 42

28 criteria’ was coined following an experiment performed by Wolf (1983) in which voluntary movements of wrist and finger extensors were shown to be “predictors of future acquisition of independent limb use.” The methodology used to determine this criteria, however, involved a small number of participants, and has not been replicated. 1,26 Participants whose function falls below minimal motor criteria improve less with traditional CIMT than those with higher levels of motor ability. 44 These participants, however, must present with a minimum of 10 degrees of extension of the wrist, thumb, and two fingers. 44 According to Taub, this increases the applicability of CIMT to almost 50 percent of those who have had a stroke. 44 In light of the recent explosion of CIMT onto the rehabilitation scene, and the profound effect CIMT is beginning to have on rehabilitation strategies, scientists are charged with determining what functional characteristics are predictive of positive CIMT outcomes. The identification of clinical predictors for outcomes of CIMT can help to target the individuals who benefit the most from this intervention. 7 The aim of this study was to establish a predictive model for CIMT outcomes based on functional characteristics of individuals post-stroke. Methods Participants A convenience sample of 55 participants was recruited from two CIMT projects at the University of Florida and the Malcom Randall VA Medical Center, Brain Rehabilitation Research Center. Figure 2-1 presents the main participant characteristics. A post-hoc power analysis was conducted to determine the power for the given group sample size (see results section). Participants signed an informed consent prior to participation in the study.

PAGE 43

29 frequency%L brain stroke2851%R brain stroke2749%Dominant side involved2647%Non-dominant side involved2953%Female2240%Male3360%Ambulatory4480%Non-ambulatory1120%Ambulatory StatusGenderDominanceSide of Stroke NMinMaxMedianMean sdTime Since Stroke (in days) Ag e(in years)16060 (45 yrs)94062.0514.6229552378167355228366 Figure 2-1. Descriptive characteristics All participants met the following inclusion criteria: 1) diagnosis of at least one stroke and not more than three strokes on the same side of the brain, 2) stroke at least six months prior to study participation, 3) ability to follow simple instructions, 4) a score of 20 or higher on the Mini Mental State Exam, 116 5) the ability to sit independently without back or arm support for five minutes, 6) the ability to stand with support of a straight cane, quad cane or hemiwalker for two minutes, 7) the ability to actively participate for six hours of therapy without long rest or nap periods, and 8) passive range of motion of all UE motions of at least half the normal range. Participants were stratified according to their ability to meet minimal motor criteria to insure that the experiments contained participants of differing ability levels. Inclusion in the high functioning group (n= 32) required active movement of the wrist through at least 20 degrees of the flexion-extension range from a relaxed flexed position, and movement of at least three fingers 20 degrees at the metacarpal-phalangeal and inter-phalangeal joints (minimal motor criteria). Inclusion in the low functioning group (n=

PAGE 44

30 23) required evidence of trace wrist extension and extension of two fingers from a fully flexed position, but did not meet the criteria established in the high functioning group. Exclusion criteria for both groups included: any health problems that put the participant at significant risk of harm during the study; any other neurological conditions such as Multiple Sclerosis or Parkinson’s Disease; drugs for spasticity; and pain limiting participation in the study. Procedure Following a pre-test, participants received two-weeks, 10 consecutive weekdays, of intensive treatment, six hours per day. For this period, the unaffected hand was immobilized in a padded mitt for 90 percent of their waking hours. The mitt was used at all times except when performing a minimal amount of agreed upon activities (e.g., bathroom activities, naps, when the unaffected limb was used for an assistive device in walking, of other circumstances when safety was compromised). The trainer and participants signed a behavioral contract establishing agreed upon amounts of mitt use, task effort, activity logs and home diaries (Appendix B). The participants were strongly encouraged to continue to use their weaker hand during activities throughout the day and while at home. After the six hours of intensive therapy, the participant returned home and maintained a diary documenting activities and mitt time use. During the weekends, there were no assigned tasks, but the participants were instructed to continue to wear their mitt and maintain a home diary (Appendix B). During the ten consecutive weekdays, participants received supervised task practice using their affected hand and arm. CIMT activities were chosen from a task menu and an activity log was kept by the trainer to demonstrate what tasks had been attempted and how the tasks were progressed during training (Appendix B). CIMT consisted of a set of

PAGE 45

31 tasks to be performed with the affected upper extremity, such as picking up pencils, moving beans from one container to another, stacking blocks and using utensils. The treatment was focused on performance of frequent movement repetitions while performing functional activities. To remain challenging, as performance improved, tasks were increased in complexity and difficulty. For example, as the time to complete a task decreased, the task was raised to a higher surface, or farther away. The tasks were functional in nature, but were modified to allow some success. The ten-days of training were followed by an immediate post-test and a 4 to 6 month follow-up post-test. Outcomes Measures Two main outcome measures, commonly reported in CIMT studies, were used for this study: 1) the Wolf Motor Function Test (WMFT), a test of movement capability and 2) the amount component of the MAL (MALa), a test of perceived use. The WMFT, developed by Stephen Wolf and modified for use in CIMT trials, has been used successfully as CIMT outcome measures for several years. 12,117-126 This test evaluates movement capability through a series of 15 timed tasks and two strength tasks. Only the timed tasks were used in this study (Appendix B). The tasks, which are modeled from the Jebson Taylor Hand Test, 127 progress from joint specific to multi-joint movements. 13-17,115 The reliability of the WMFT has been reported with inter-rater reliability established at r=0.93, as measured by an intraclass-correlation coefficient. 128 The WMFT is administered first to the less involved UE and then to the more-involved UE. The WMFT outcome measure is reported as a mean of the affected task times minus the mean of the unaffected task times. The MAL is a commonly used CIMT outcome measure. 12,117-126 It is a 30-question structured interview in which the participants respond with a number corresponding to a

PAGE 46

32 given amount of use or perception of how well they have used their affected arm when away from the laboratory environment. 13-17,115 For example, the participant would respond to the question: “How much do you use your more affected arm to turn on a light switch?” by choosing the appropriate response from the MAL amount scale (Figure 2-2). The mean of the “amount” section of the MAL was used as an outcome measure. The “how well” scale was not used in this study (Appendix B). The interrater reliability for the MAL is 0.94. 13,15,16 0.0Did not use m y weaker arm for the activit y ( not used ) 0. 5 1.0Occasionall y tried to use m y weaker arm for that activit y ( ver y rarel y) 1. 5 2.0Sometimes used m y weaker arm for that activit y but did most of the activit y with m y stron g er arm ( rarel y) 2. 5 3.0Used m y weaker arm for that activit y about half as much as before the stroke ( half p re-stroke ) 3. 5 4.0Used m y weaker arm for that activit y almost as much as before the stroke ( 3/4 or 75% p re-stroke ) 4. 5 5.0Used m y weaker arm for that activit y as much as before the stroke ( same as p re-stroke ) Motor Activity Log Amount Scale Figure 2-2. Participant response choices for MAL amount scale Selected Prospective Predictors: Five prospective functional predictors were investigated: 1) minimal motor criteria level, 2) amount of finger extension/grasp release, 3) amount of grip strength, 4) upper-extremity Fugl-Meyer score and, 5) score of the Frenchay Activities Index. These predictors were included in the regression model for several reasons. First, many of these predictors appear in stroke rehabilitation research as predictors of other outcomes, such as differing therapeutic interventions, return to function, and return to life roles. Second, these predictors are often discussed in CIMT literature as having a potential to impact outcomes. Third, some predictors demonstrated strong predictive value in the pilot studies (Appendix A).

PAGE 47

33 Minimal motor criteria level CIMT appears to be effective in improving movement capabilities in a subset of approximately 20 to 25 percent of stroke survivors who meet minimal motor criteria, demonstrating a relatively high level of hand and wrist control. 17 In a small sample, Taub found that lower functional recovery of the hand wrist was related to poorer improvements compared with those with higher levels of motor ability. 44 Despite this finding, individuals with lower functional recovery of the hand and arm may still have significant success from participating in CIMT. Since current selection criterion for CIMT needed further attention, in this study we expanded selection criteria to include individuals that did not meet traditional minimal motor criteria level. Inclusion of this lower functioning group, those who did not meet minimal motor criteria, will aide in clarifying the necessity of this baseline criteria for positive outcomes with CIMT. Specifically, if minimal motor criteria is not a predictor, it may not be an appropriate determinant of CIMT participation. Finger extension and grasp release Finger extension and grasp release are essential to functional movement of the hand. Individuals post-stroke, however, often present with limited or reduced ability to perform both fine and gross movements of the hand. 45 Secondary to the intrinsic hand muscles hypertonia, which is commonly present post-stroke, controlled grasping is often very difficult. Individuals frequently can grasp objects via a massed grasp, however, this often requires recruitment of abnormal movement patterns. Furthermore, individuals post-stroke often have greater difficulty releasing an object. 45 Difficulty releasing objects may be due to a flexor synergy that effectively limits isolation of joint movements out of synergy.

PAGE 48

34 For the purpose of this study, finger extension/grasp release was defined as the ability to actively release a mass flexion grasp as defined by Fugl-Meyer assessment. It is graded as follows. A ’ was given if the individual is unable to release the grasp, a ’ if the participant could release the mass flexion grasp, and a ’ if the participant could fully extend their fingers from the starting grasp position. 46 Due to the role that finger extension plays in the functional use of a hand, it was included as a predictor for CIMT. Grip strength Grip strength was used as a predictor in the regression model due to its prevalence in the literature as a measure of outcomes post-stroke 47-49 and as a predictor of outcomes. 50-56 For example, the absence of measurable grip strength one-month post stroke is predictive of poor upper-extremity functional recovery. 5,57-59 In contrast, early return of active grasping is a positive prognostic sign. 5,58,59 Although grip strength is strongly associated with chronological age, 60 independent of this relationship, it is a powerful predictor of disability, morbidity, and mortality. 53 Low grip strength is associated with disability, while there are strong relationships between mid-life measurements of grip-strength to predictions of long-term survival. 52,61 For this project, grip strength was assessed using standard grip strength dynamometry techniques 62 with a Jamar dynamometer at level three setting. Standard methods for handheld dynamometry and the reliability of the testing have been well established. 62 Upper-extremity Fugl-Meyer motor score The Fugl-Meyer Assessment is a measurement of the percent recovery of a person following a stroke that provides objective and quantifiable assessment of motor function 63 and was designed primarily for use in rehabilitation settings. The test is

PAGE 49

35 organized according to Brunnstrom levels of recovery and is divided into five sequential recovery stages. 63 The Fugl-Meyer motor component is an assessment of ability to move in/out of synergy, reflexes, wrist stability, grasping ability, and coordination (Appendix B). The validity of the Fugl-Meyer is good and overall reliability is high (intraclass correlation coefficient of 0.96). 46 This standardized assessment is also often used as a measure of function in stroke rehabilitation studies 1,18,64-70 and has been shown to be predictive of dependency and functional level following stroke. 5,7,71 For this project, the UE motor portion of this exam was used as a predictor of outcomes for CIMT. The total possible score for the UE motor section is 66. Using the Fugl-Meyer score allowed for the regression model to encompass a predictor that represents many aspects of UE recovery following stroke. Frenchay activities index The Frenchay Activities Index is a survey that measures extended activities of daily living and was developed specifically for measuring disability and handicap in individuals with stroke. It includes 15 items that measure level of activity in areas such as domestic chores, leisure/work, and outdoor activities. During a short interview, the participant is asked to determine how often they perform a given activity using a scale of 1 to 4 (Appendix B). The item scores are totaled to determine the final score, which can range from 15 (inactive) to 60 (highly active). 72,73 The reliability is 0.78 to 0.87. 73 The Frenchay Activities Index was included as a predictor in the regression model for many reasons. First, it is frequently cited for determining outcomes post-stroke. 73-82 Second, the Frenchay was the only predictor used in this model that represents questions from the participation level of the International Classification of Functioning, Disability and Health model (ICF). 35 Third, the Frenchay was frequently correlated to outcomes of

PAGE 50

36 CIMT in the pilot study. Inclusion of the Frenchay will clarify how important “being active” is in determining outcomes with CIMT. The relationship between the Frenchay score and CIMT outcomes could lend insight into involvement and independence. For example, an individual who scores high on the Frenchay is considered more independent because of their involvement in his/her own self-care and activities. In contrast, an individual with higher motor capabilities, but a lower Frenchay score may lack motivation or may not seek independence from a caregiver. Level of activity could be an important predictor of success with such a time-demanding, attention-driven therapy as CIMT. Data Analysis Demographic and clinical presentation characteristics of the study sample are described using the mean, median and standard deviation for continuous variables, and using frequencies and percentages for categorical variables. For the intention-to-treat analysis demographic and clinical characteristics of the patients who were admitted to the program, but who did not participate in the 4-month follow-up evaluation, were compared with participants who completed the follow-up to determine differences that could result in a bias. The continuous variables from the intention-to-treat analysis were analyzed using t-tests and categorical variables using Chi square tests or Fisher’s exact tests. All data were analyzed using an intention-to-treat approach in which the pre-test scores were used as the follow-up post-test scores for those participants that did not return for the follow-up. The normality of the data for the WMFT and the MALa, the dependent variables, was visually verified with probability plots (P-P) and statistically verified with the Kolmogorov-Smirnov f test. The MALa met assumptions of normality; however, the

PAGE 51

37 WMFT required transformation using the natural log [ln] to meet assumptions of normality. The predictive relationships of various functional scores on the traditional measures of CIMT were investigated. The independent variables were used to develop a general linear model for the dependent variables individually: the WMFT (aim 1) and the MALa (aim 2). These analyses were performed for the immediate post-test and the follow-up post-test. The immediate post-test regression allowed for determination of predictors of immediate success. The follow-up post-test represented changes overall from the first test period to the last testing period allowing insight into predictors of long-term success (Figure 2-3). Follow-up post-test Time Frame 2 Time Frame 1 Immediate p ost-test Pre-test 14 days 4-6 months Figure 2-3. Timeline for testing of dependent variables Potential predictors were selected for the development of a general linear model for the prediction of outcomes with CIMT. The pretest scores for each of the dependent measures were used as covariates to statistically control for group differences that existed before treatment. A forward stepwise procedure was employed in which the variables were examined at each step for entry or removal from the model. The least significant variables were removed from the model, based on their level of association with the dependent variable at each step. Adjusted R-square values, p-values and 95 percent confidence intervals (CIs) were calculated. Thorough regression diagnostics were run

PAGE 52

38 and jackknife residual analyses were performed to verify the basic assumptions. Multicollinearity among predictor variables in the regression models was assessed using a variance inflation factor (VIF). Results Power Analysis A post-hoc power analysis was conducted to determine the power for the sample size (n=55). The effect size (f 2 ) for the WMFT immediate post-test and follow-up post-test and the MALa immediate post-test and follow-up post-test was determined from the data and they are 3.67, 2.82, 1.20, and 1.48, respectively. Using these effect sizes, the sample size of 55 participants at an =0.05, for five predictors, met an average power level = 1.0. This strong power can be attributed primarily to the presence of the covariate in the model. Descriptive Characteristics Tables 2-1 and 2-2 provide the descriptive statistics of the independent variables. Table 2-1. Descriptive statistics of the continuous independent variables Independent VariableNMinMaxMedianMean sdGrip Stren g th55029.76.68.757.39Fugl-Meyer UE Moto r 5516623335.2711.4Frencha y 5518533737.057.89 Table 2-2. Descriptive statistics of the categorical independent variables Independent Variablefrequency%High Meets minimal criteria3258%Low Does not meet criteria2342%0Cannot release a mass flexion grasp1425.5%1Can release paritally2749.0%2Can fully extend fingers1425.5%GroupFugl-Meyer Finger ExtensionMinimal Motor Criteria Level

PAGE 53

39 Intention-to-Treat Analysis An intention-to-treat analysis was used because of the significant differences apparent between the group that completed the follow-up post-test and the group that did not. Nine participants did not return for the follow-up post-test evaluation and, therefore, their pre-test score was used as the follow-up post-test measure. 129-132 Three of the independent variables, (1) grip strength, (2) Fugl-Meyer upper-extremity motor, and the (3) Frenchay Activities Index (Table 2-3 and 2-4) demonstrated significant differences between the individuals who returned for the follow-up post-test and those who did not return. Individuals with lower ability levels showed significantly higher drop-out rates. None of the descriptive characteristics from Figure 2-1 demonstrated significant differences between the group that completed the follow-up post-test and the group that did not. Table 2-3. Chi-Square and Fisher’s exact test for the differences across categorical independent variables between those participants who completed the follow-up evaluation and those that did not complete the evaluation Drop-outsCompletedChi-squareFisher's Exact TestHigh Meets minimal criteria329Low Does not meet criteria6170Cannot release a mass flexion grasp4101Can release paritally5222Can fully extend fingers014Mimal Motor Criteria Level0.0980.143Independent Categorical VariablesFugl-Meyer Finger Extension0.113na Table 2-4. T-test for difference across continuous independent variables between those participants who completed the follow-up evaluation and those that did not complete the evaluation Drop-outsCompletedGrip Strength3.08 (4.82)9.86 (7.33).003*Fugl-Meyer UE Moto r 29.22(6.04)36.46 (11.86)0.013*Frenchay29.78 (6.18)38.48 (7.43).003**significant differencesIndependent Continous Variablesmean (sd)t-test (sig)

PAGE 54

40 Multiple Regression Modeling Aim 1: Wolf motor function test modeling Minimal motor criteria (0=low, 1=high), finger extension/grasp release, grip strength, Fugl-Meyer UE motor score, and the Frenchay score were entered into a linear multiple regression model with stepwise entry using the [ln]WMFT immediate post-test as the dependent variable and entry of the WMFT pre-score as the covariate. Not all predictor variables made significant contributions to the WMFT immediate post-score. The only significant predictor was finger extension/grasp release, all other variables were removed during the regression analysis. Including the covariate, the model accounted for 0.786 of the variance in [ln]WMFT at immediate post-test. The p-values for all the independent variables are listed in Figure 2-5. The final regression equation is as follows. [ln]WMFTpost’= (1.608) +(.03575)WMFTpre – (.410)finger extension The same independent variables were entered into a second linear multiple regression model with stepwise entry using the [ln]WMFT follow-up post-test score as the dependent variable. Similarly, the only significant predictor variable was finger extension/grasp release. This model accounted for 0.738 of the variance in WMFT at follow-up post-test with the pre-test as a covariate. This model is very similar to the WMFT immediate post-test model. The p-values for all the independent variables are listed in Figure 2-5. [ln]WMFTfu’= (1.804) + (.03285)WMFTpre – (.402)finger extension. Aim 2: Motor activity log amountmodeling The same independent variables were entered into a third linear multiple regression model with stepwise entry using the MAL amount immediate post-test as the

PAGE 55

41 Table 2-5. The adjusted R squared for the [ln] WMFT models, the B weights, the p-values for all the predictors, and the confidence intervals for the significant predictors are listed. [ln] WMFT MODELS Adjusted R squareF ( si g) INDEPENDENT VARIABLES B p-value (CI) B p-value (CI)Covariate ( WMFT p re-score ) 0.0360.000 ( .028,.043 ) 0.0330.000 ( .025,.041 ) Minimal motor criteria level0.0410.6930.0070.949Fin g er extension-0.4100.027 ( -.771, -.048 ) -0.4020.04 ( -.785, -.020 ) Grip stren g th-0.0480.570-0.0630.504UE Fu g l-Me y er motor score-0.1260.214-0.1850.098Frenchay0.0640.376-0.0320.687Constant1.6080.000 (.956, 2.261)1.8040.000 (1.113,2.494)Follow-up post-test Immediate post-test0.786100.099 (.000)0.73877.037 (.000) dependent variable. The Fugl-Meyer UE motor score and minimal motor criteria level were significant predictor variables for the MALa. All other variables were removed during the regression analysis. This model accounted for 0.546 of the variance in MALa at the immediate post-test. The p-values for all the independent variables are listed in Figure 2-6 and the regression equation is as follows. MALapost’= (-.0301)+ (.204)MALapre + (.05634)F-M UE motor + (1.116)Min Motor The same independent variables were entered into a fourth linear multiple regression model with stepwise entry using the MALa follow-up post-test score as the dependent variable. At the follow-up post-test, the baseline UE Fugl-Meyer motor was the only significant predictor variable. This model accounted for 0.596 of the variance in MALa at follow-up post-test. The p-values for all the independent variables are listed in Figure 2-6 and the regression equation is as follows. MALafu’= (.645) + (.409)MALapre + (.0643)F-M UE motor The largest value of the variance inflation factors was <1.7, indicating that multicollinearity among the predictors did not unduly influence the regression

PAGE 56

42 Table 2-6. The adjusted R squared for the MAL amount models, the B weights, the p-values for all the predictors, and the confidence intervals for the significant predictors are listed. MAL amount MODELS Adjusted R squareF ( si g) INDEPENDENT VARIABLES B p-value (CI) B p-value (CI)Covariate ( MALa p re-score ) 0.2040.478 ( -.369, .777 ) 0.4090.049 ( .003,.816 ) Minimal motor criteria level1.1160.004 ( .362, 1.870 ) 0.0670.558Fin g er extension0.0220.8620.0680.526Gri p stren g th0.2020.0890.0800.475UE Fu g l-Me y er motor score0.0560.004 ( .019,.094 ) 0.0640.000 ( .037,.091 ) Frencha y 0.1820.0790.0450.651Constant-0.030.954 (-1.071, 1.011)-0.645.104 (-1.427, .136)Follow-up post-test Immediate post-test0.54622.638 (.000)0.59640.832 (.000) estimates. 133 The adequacy of the final regression model was examined. Upon visual examination of the histogram, the residuals appeared to be normally distributed. Presence of outliers was assessed using Jackknife residuals. The sample contained two outliers for the WMFT model. The influence and accuracy of these data points were assessed and they remained in the model. 134,135 These post-hoc regression diagnostics results suggested that the regression analysis was appropriate. Discussion While CIMT has been proven to be a beneficial therapy for individuals post-stroke, little is know about the functional characteristics that are predictive of CIMT outcomes. 1,2,6,10-25,115 The identification of these clinical predictors for outcomes of CIMT is needed to target the appropriate populations. The goal of this study was to establish usable models, not extensive or comprehensive models. While detailed models serve a function, they are less readily utilized and have little more predictive capability than simple models. 8,9 Other predictive variables could have been included in this model, potentially making it more inclusive. For example, cognitive function or motivational variables may have increased the predictive capability of these models. In addition,

PAGE 57

43 modern imaging techniques could have been included to identify type or location of lesion, since they are valuable in prediction of stroke outcomes. 114 These additional predictors would add to the complexity of the model, making it more difficult to use and clinically less practical, while potentially adding little strength to the model. Multiple Regression Modeling The goal of this study was to determine the significant functional predictors of outcomes for CIMT. The outcomes were divided into two essential time frames necessary to establish the functional predictors of those who improved most during the treatment phase and who maintained or improved functional gains four to six-months after the CIMT intervention. 100 Previous CIMT studies suggest that functional gains are persistent, 1,16 signifying individuals maintain over time the improvements gained while in the training program. Identifying the predictors of individuals who are able to maintain functional gains at the long-term is of primary importance. 100 Wolf Motor Function Test model Finger extension/grasp release is the only predictor of positive outcomes for movement capability of the more affected UE, as measured by the WMFT, immediately following therapy and at 4-6 months after therapy. In fact, the regression models were very similar for the WMFT at both the immediate post-test and follow-up post-test. When interpreting the model, recall that the WMFT score is a time score for performance of tasks. The higher the WMFT score the longer it will take to complete the activities, thus, the lower the ability level. For example, the coefficient of [-.410 finger extension] results in a lower predicted WMFT time score (better score) for individuals with higher finger extension scores. The predictive capability of finger extension/grasp release is logical due to its role in the functional use of a hand. This measure, while related to

PAGE 58

44 minimal motor criteria, does not include the requirements of wrist extension. In this sample, minimal motor criteria did not predict outcomes for the WMFT. This finding is important since the primary screening criteria used for the majority of CIMT studies is an individual’s ability to meet minimal motor criteria. Establishing minimal motor criteria has been an advantageous and an appropriate screening technique for traditional CIMT until this time. The component of wrist extension, however, may be too restrictive. An individual’s ability to extend his/her fingers may be a better screening criterion. Many of the tasks practiced in CIMT involve grasp and release, and while wrist extension is helpful in this capacity, it appears not to be essential for positive outcomes with CIMT. Although wrist extension is used in normal grasp, perhaps, presence of finger extension is enough to justify participation in CIMT. Minimal motor criteria should be investigated further to determine its appropriateness for screening of potential CIMT participants. Motor Activity Log model The results suggest, however, that minimal motor criteria is predictive of the MALa immediately following therapy, but that it is not predictive of outcomes for long-term success with CIMT. While the predictors for the immediate post-test are of interest, the most important predictors are those of the follow-up post-test. Specifically, the predictors of how individuals perform four to six months after the treatment. While individuals who meet minimal motor criteria demonstrate improved amount of use of the affected UE immediately following therapy, there are no significant differences between groups at the long-term follow-up. The Fugl-Meyer UE motor score, however, is predictive of the MALa immediately following therapy and at the long-term follow-up. This indicates that individuals with higher function, according to the Fugl-Meyer, result in prediction of higher scores on the

PAGE 59

45 MALa following CIMT. One component of the Fugl-Meyer evaluates an individual’s ability to move in and out of synergy, quite possibly an individuals ability to function out of synergy is an important predictor of success with CIMT. This would allow increased ease in practicing of tasks that involve reaching and multi-joint coordination, thus translating to greater success following CIMT. The Fugl-Meyer UE motor test could be further broken down to evaluate if just one component is predictive of improvements. The significance of the Fugl-Meyer UE motor test was expected due to the predictive value it has shown to have in other stroke rehabilitation studies. 5,7,71 Other outcomes The Frenchay Activities Index score and grip strength were included in the regression analysis as potential predictors, however, neither were significant to the prediction of success in CIMT. Knowing what variables are not predictive is also valuable and necessary when screening a potential client for CIMT. This offers insight into what predictors should not be considered when determining if a participant can improve with this therapy. For example, denying participation to individuals who demonstrate a low grip strength would be inappropriate, as it is not a predictor of success. If clinicians only want to offer traditional CIMT to individuals who will make the most long-term gains in UE movement capability and amount of use of the affected UE, the results suggest enrolling individuals who are able to actively extend their fingers from a massed grasp and who have greater Fugl-Meyer UE motor scores. While improvements occur following CIMT with even a very low UE Fugl-Meyer score, the higher the baseline score the greater the improvement following therapy. To determine the predicted immediate post-test scores for a given individual, the regression equations should be utilized.

PAGE 60

46 An important clarification is that these reported results apply to only traditional CIMT, when delivered 6 hours per day for 10 consecutive weekdays as delineated in the procedure. They may not be appropriate predictors of success for other methods of CIMT delivery, such as modified CIMT, 67,69 home-based CIMT, and CIMT delivered in a less-intense manner or for a different duration. Additional considerations for future studies include investigating the predictors for other types of CIMT, when administered in differing intensities or duration. Another research avenue would be investigating the predictors of individuals who do not improve with therapy. More specifically, defining predictors that can identify both responders and non-responders would be advantageous to proper application of CIMT. Another direction for future research is establishing if the changes in outcomes represent a clinically meaningful difference. In other words, what amount of improvement is needed on the outcome measures for the change to be clinically meaningful. The next step, then, would be determining the predictors for individuals who make a clinically important change with this therapy. Drop-outs An additional important finding is the significant difference in the individuals who did not complete the follow-up post-test and those that did complete the follow-up post-test. Individuals who dropped-out scored significantly lower for grip strength, Fugl-Meyer UE motor, and the Frenchay Activities Index, indicating lower ability levels. Minimal motor criteria and finger extension/grasp release approached significance substantiating that lower-functioning individuals have a higher tendency to drop-out from the intensive, time-dependent, and often frustrating therapy. This demonstrates that lower-functioning participants are more likely not to return for long-term follow-up. It

PAGE 61

47 does not suggest, however, that lower-functioning individuals cannot improve following CIMT Statistical Analysis Discussion The predictive relationship of various functional predictors on the traditional measures of CIMT was investigated using linear regression. Including the pre-test scores as a covariate in the models may be questioned because it increases complexity of the model use. Specifically, a baseline test is required in order to utilize the regression equations for prediction of outcomes for CIMT. The use of the covariate, however, was essential to form accurate regression models. A pre-test for each measure was used as a covariate to statistically control for differences that existed before treatment. Another statistical option considered was the use of change score or the percent change as the dependent measure across time frames. The use of change scores, however, is poor statistical practice for these dependent measures. 136,137 The measures are not of equal intervals and have ceiling and floor effects, which would lead to inaccuracy in model development and interpretation. According to Allison et. al (1995), the use of ratios as response variables in regression should be avoided, if possible in favor of including the baseline measure as a covariate in the regression. 138 Controversy exists surrounding the use of step-wise regression. 139 While stepwise methods are frequently used in behavioral research, Thompson (1995) argues that severe limitations exist with this method. 139 For example, he claims that incorrect degrees of freedom are often used by the statistical software. This misuse, however, is accentuated when a large number of predictors are ‘thrown’ into the model. In this study restricting the stepwise equation to five predictors decreases the chance of a type I error. Another limitation is that stepwise regression tends to capitalize on sampling error. In other

PAGE 62

48 words, if your sample is not representative, your results will not be replicable. 139 Every effort was made for the sample to contain a diverse population of individuals with stroke across descriptive and functional categories. Many recently published articles include step-wise regression for predictive modeling. 5,7,140,141 According to Jonsson (2000) researchers need to be aware of the possible dangers of stepwise model building, use it as appropriately as possible, and realize that this kind of approach is an “essential component in our efforts to make sense out of the data we are to analyze.” 142 Stepwise regression was the appropriate methodology to use in this study. This allows for the most efficient and usable models. If simultaneous entry was used instead of stepwise, it would result in including non-significant predictors in the regression equations. This would cause greater clinical demand to measure all the predictors in order to employ the model. Path analysis or structural equation modeling are techniques that could have also been used in analyzing this data. These methods, however, are more confirmatory than exploratory. In path analysis, a directional pathway is constructed from pre-existing knowledge and/or data and then the strength of the path is confirmed. Since this was the first study of its type for CIMT, the objective was to explore the factors that should be included in a predictive model. Path analysis could be a future method to investigate and confirm the strength and directionality of this study. Limitations This study was performed in individuals who met strict inclusion and exclusion criteria. While, this sample was more diverse than most CIMT studies, it is not representative of the entire stroke population. Furthermore, individuals were not randomly selected, but were respondents to inquiries for participants. Second, sample size decreased at the follow-up post-test evaluation, leading to a bias for withdrawals.

PAGE 63

49 While this bias was dealt with during analysis, using an intention-to treat approach, its presence as a limitation is still significant. Third, due to the relatively small sample size, we were unable to evaluate the stability of the model using a portion of the sample as a reserve. These results should be replicated in another sample. Fourth, the outcome measures were chosen due to their abundant use in the CIMT literature, however, predictors based on other, more readily used stroke outcome measures should be considered. Finally, a predictor model for longer-term follow-up should also be carried out. This is essential to know predictors of long-term success. Conclusions CIMT is reported to significantly improve functional use of the upper-extremity in 20 to 25 percent of people with chronic stroke disability. 1 Limited evidence exists, however, regarding the specific functional characteristics of individuals who benefit most from this intervention. Significant predictive ability was discovered with finger extension/grasp release, minimal motor criteria level, and Fugl-Meyer UE Motor scores. These items, when used in the regression equations along with the covariate, can predict an individual’s score on the dependent measures. Selection criteria for participation need to be carefully examined to ensure proper inclusion in CIMT studies. Further substantiation of these findings in larger and more diverse samples is warranted in order to meet the urgent need of determining the appropriate candidates for CIMT.

PAGE 64

CHAPTER 3 DESCRIPTIVE PREDICTORS OF CIMT OUTCOMES Introduction Limited evidence exists regarding characteristics of individuals who are most successful with Constraint-Induced Movement Therapy (CIMT). CIMT is a rehabilitative strategy used primarily with the post-stroke population to increase the functional use of the neurologically, weaker upper-extremity through massed practice while restraining the lesser-involved upper extremity. 1 Although stroke is the most common disabling condition in America, with 30 to 66 percent of the individuals losing functional ability in their more-affected arm and hand, there are few well-researched and effective therapies available to individuals post-stroke. 2,38 The need for innovative rehabilitation is clear. Solid research evidence supports CIMT, but many questions persist about who can benefit from this therapy. 1,2,6,10-19,21-24,34,115,143 Originally tested in an animal model, the results of CIMT studies have demonstrated significant and lasting improvements of upper-extremity movement function. 1,2,6,10-19,21-24,34,115,143 Edward Taub and Steven Wolf have published extensively on the effectiveness of CIMT. 1,2,6,10,13-18,34,44,115,143 Together they have shown that CIMT significantly improves functional use of the upper extremity of some individuals with chronic stroke. 1,14,26 CIMT results have recently been labeled the most promising evidence that motor recovery can occur in the hand of a post-stroke individual who has some residual purposeful movement. 36 While the evidence for CIMT is promising, many questions persist concerning who will benefit from this form of therapy. Selection 50

PAGE 65

51 criteria for participation should be carefully examined to determine who benefits most from this intervention what descriptive characteristics are predictive of positive CIMT outcomes. The aim of this study was to establish a simple predictive model for CIMT outcomes based on descriptive characteristics of individuals post-stroke. The identification of clinical predictors for outcomes of CIMT is essential and of value to both researchers and clinicians. 7 The goal of this study was to establish a clinically usable model, not a comprehensive predictor model. Extensive multivariate models are often difficult to use and, therefore, have little value because they are less readily incorporated than simple models. 8,9 Methods Participants A convenience sample of 55 participants was recruited from two CIMT projects at the University of Florida and the Malcom Randall VA Medical Center, Brain Rehabilitation Research Center. Figure 3-1 presents the main participant characteristics. A post-hoc power analysis was conducted to determine the power for the given group sample size (see results section). Participants signed an informed consent prior to participation in the study. All participants met the following inclusion criteria: 1) diagnosis of at least one stroke and not more than three strokes on the same side of the brain, 2) stroke at least six months prior to study participation, 3) ability to follow simple instructions, 4) a score of 20 or higher on the Mini Mental State Exam, 116 5) the ability to sit independently without back or arm support for five minutes, 6) the ability to stand with support of a straight cane, quad cane or hemiwalker for two minutes, 7) the ability to

PAGE 66

52 frequency%L brain stroke2851%R brain stroke2749%Dominant side involved2647%Non-dominant side involved2953%Female2240%Male3360%Ambulatory4480%Non-ambulatory1120%Ambulatory StatusGenderDominanceSide of Stroke NMinMaxMedianMean sdTime Since Stroke (in days) Ag e(in years)16060 (45 yrs)94062.0514.6229552378167355228366 Figure 3-1. Descriptive characteristics actively participate for six hours of therapy without long rest or nap periods, and 8) passive range of motion of all upper extremity motions of at least half the normal range. Exclusion criteria included: any health problems that put the participant at significant risk of harm during the study; any other neurological conditions such as Multiple Sclerosis or Parkinson’s Disease; drugs for spasticity; and pain limiting participation in the study. Procedure Following a pre-test, participants received two-weeks, 10 consecutive weekdays, of intensive treatment, six hours per day. For this period, the unaffected hand was immobilized in a padded mitt for 90 percent of their waking hours. The mitt was used at all times except when performing a minimal amount of agreed upon activities (e.g., bathroom activities, naps, when the unaffected limb is used for an assistive device in walking, of other circumstances when safety was compromised). The trainer and participants signed a behavioral contract establishing agreed upon amounts of mitt use, task effort, activity logs and home diaries (Appendix B). The participants were strongly

PAGE 67

53 encouraged to continue to use their weaker hand during activities throughout the day and while at home. After the six hours of intensive therapy, the participant returned home and maintained a diary documenting activities and mitt time use. During the weekends, there were no assigned tasks, but the participants were instructed to continue to wear their mitt and maintain a home diary (Appendix B). During the ten consecutive weekdays, participants received supervised task practice using their affected hand and arm. CIMT activities were chosen from a task menu and an activity log was kept by the trainer to demonstrate what tasks had been attempted and how the tasks were progressed during training (Appendix B). CIMT consisted of a set of tasks to be performed with the affected upper extremity, such as picking up pencils, moving beans from one container to another, stacking blocks and using utensils. The treatment was focused on performance of frequent movement repetitions while performing functional activities. To remain challenging, as performance improved, tasks were increased in complexity and difficulty. For example, as the time to complete a task decreased, the task was raised to a higher surface, or farther away. The tasks were functional in nature, but were modified to allow some success. The ten-days of training were followed by an immediate post-test and a 4-6 month follow-up post-test. Outcomes Measures Two main outcome measures, commonly reported in CIMT studies, were used for this study: 1) the Wolf Motor Function Test (WMFT), a test of movement capability and 2) the amount component of the MAL (MALa), a test of perceived use. The WMFT, developed by Stephen Wolf and modified for use in CIMT trials, has been used successfully as CIMT outcome measures for several years. 12,117-126 This test evaluates movement capability through a series of 15 timed tasks and two strength tasks.

PAGE 68

54 Only the timed tasks were used in this study (Appendix B). The tasks, which are modeled from the Jebson Taylor Hand Test, 127 progress from joint specific to multi-joint movements. 13-17,115 The reliability of the WMFT has been reported with inter-rater reliability established at r=0.93, as measured by an intraclass-correlation coefficient. 128 The WMFT is administered first to the less-involved UE and then to the more-involved UE. The WMFT outcome measure is reported as a mean of the affected task times minus the mean of the unaffected task times. The MAL is a commonly used CIMT outcome measure. 12,117-126 It is a 30-question structured interview in which the participants respond with a number corresponding to a given amount of use or perception of how well they have used their affected arm when away from the laboratory environment. 13-17,115 For example, the participant would respond to the question: “How much do you use your more affected arm to turn on a light switch?” by choosing the appropriate response from the MAL amount scale (Figure 3-2). The mean of the “amount” section of the MAL was used as an outcome measure. The “how well” scale was not used in this study (Appendix B). The interrater reliability for the MAL is 0.94. 13,15,16 Selected Prospective Predictors Six prospective descriptive predictors are investigated: 1) side of stroke, 2) time since stroke, 3) hand dominance, 4) age, 5) gender, and 6) ambulatory status. These predictors were included in the regression model for several reasons. First, many of these predictors appear in stroke rehabilitation research as predictors of other outcomes, such as differing therapeutic interventions, return to function, and return to life roles. Second, these predictors are often discussed in CIMT literature as having a potential to impact

PAGE 69

55 0.0Did not use m y weaker arm for the activit y ( not used ) 0. 5 1.0Occasionall y tried to use m y weaker arm for that activit y ( ver y rarel y) 1. 5 2.0Sometimes used m y weaker arm for that activit y but did most of the activit y with m y stron g er arm ( rarel y) 2. 5 3.0Used m y weaker arm for that activit y about half as much as before the stroke ( half p re-stroke ) 3. 5 4.0Used m y weaker arm for that activit y almost as much as before the stroke ( 3/4 or 75% p re-stroke ) 4. 5 5.0Used m y weaker arm for that activit y as much as before the stroke ( same as p re-stroke ) Motor Activity Log Amount Scale Figure 3-2. Participant response choices for MAL amount scale outcomes. Third, some predictors demonstrated strong predictive value in the pilot studies (Appendix A). Side of stroke Side of stroke is an obvious factor to consider when studying the effects of CIMT. A great deal of research has assessed right versus left-brain functions, right-sided versus left-sided strokes, and the varying effects that these have on patients’ presentation, functioning, and outcomes. Ornstein (1997) states that over the last 25 years greater than 45,000 articles and books have been written on the two hemispheres. 84 While some CIMT studies have included participants with both left and right-hemispheric damage, there was no main effect found for side of hemiparesis. 15 Specifically, studies showed no differences in outcomes based on an individual’s side of hemiparesis. Generally speaking, the left hemisphere is concerned with analytical processing of individual components, sequencing of tasks, and language, while the right hemisphere is more focused on perception of whole and spatial tasks. 85 An individual who sustains a left-brain stroke may present with an inability to solve problems, is often more easily angered and frustrated, has impaired retention of information, and may present with language difficulties and/or apraxia. 85 If an individual has difficulty with language, they may have more difficulty understanding the directions for the therapy. They may be

PAGE 70

56 limited in keeping track of their home-activity log or they may misunderstand directions. In addition, this limitation could be augmented by the fact that individuals with left-brain damage also have impaired retention of information. With communication difficulty, there may be increased levels of frustration, especially given the social nature of CIMT that develops due to extensive time with trainers. Furthermore, frustration is often greater in individuals with left-brain damage, so a frustrating therapy, such as CIMT, may easily add to the already pre-existing tendencies towards increased levels of frustration. This frustration could lead to decreased levels of motivation or possibly difficulties with compliance. Individuals with left-brain stroke also have more difficulty with problem-solving. Again, this could lead to more difficulties with the therapy, as problem solving for new compensatory strategies are often essential to success with the therapy. An additional limitation that may affect individuals with a left-brain stroke is apraxia. Nothing is reported in the CIMT literature regarding outcomes for individuals with apraxia. Apraxia is defined as a disorder of skilled movement not caused by weakness, ataxia, akinesia, deafferentation, inattention to commands, or poor comprehension. 86 Apraxia is a disorder of a praxis system, and this praxis system involves memory representations that are formed based upon experiences with purposive actions. 87 People with apraxia predominantly present with right hemiparesis, affecting their dominant right hand. Those with apraxia may have spatial or timing errors, such as a delay in initiation or inappropriate pauses, noticed especially when the plane or trajectory needs to be changed. Timing errors also present as failing to coordinate appropriate speed of movement with that required of the task. 89 These types of errors may impede acquisition 90 of new motor skills, and thus, result in increased difficulty with

PAGE 71

57 therapies such as CIMT. As presented, individuals with left-brain stroke may have unique challenges that could influence success with CIMT. Persons with right-brain stroke, however, also have distinctive problems that could limit success with CIMT. They present with left-side neglect, often have difficulty with spatial-perceptual tasks, are more impulsive, and frequently have greater balance problems. 85 Unilateral neglect is common following right-hemispheres stroke, with a reported incidence varying significantly from 10 to 82 percent having left-neglect. 91 Neglect is the inability to attend, or orient to, meaningful stimuli contralateral to the lesion. 89 Neglect may be either a problem recognizing sensory stimuli or it may be a problem with motor planning in response to stimuli. 91 Whether a CIMT participant presenting with either type of neglect would benefit from the therapy is unknown. Individuals with neglect may have significant difficulties attending to the CIMT tasks. Neglect would then result in interference with motor acquisitions, leading to poorer outcomes. Conversely, individuals may be able to overcome the inattention caused my neglect, due to the constant attention-driven methods of CIMT, and demonstrate improvement. Individuals with right-brain stroke also present with spatial-perceptual deficits that may result in difficulty performing some components of the training. This difficulty could lead to increased levels of frustrations with the therapy and affect overall success. In addition, individuals with right-brain stroke often have increased balance problems. The balance problems could lead to decreased time wearing the mitt and issues with compliance. Mitt time could be decreased in individuals with balance problems because of increased safety concerns and or use of an assistive device. The various presentations

PAGE 72

58 of pathology attributed to side of lesion could affect outcomes following CIMT, therefore, it is included as a potential predictor. Time since stroke Recovery of function after stroke is multi-causal. Spontaneous recovery, related to time since stroke, contributes to improvement in motor function following stroke. 92 CIMT, given acutely during this period, could possibly enhance spontaneous recovery and boost functional recovery. Conversely, early attempts at forcing the use of the hemiparetic upper extremity could lead to worsened learned non-use. The learned non-use could be enhanced at such an early time when an individual may have limited function, because increased use of their weakened hand may lead to failure or pain. This would result in negative reinforcement and subsequent increased learned non-use. Some would argue against worsened learned non-use, however, by insisting that CIMT is gauged to the functional level of an individual. The positive reinforcement that is a key component to CIMT would, therefore, circumvent development of learned non-use. Furthermore, CIMT is based on the assumption that the nervous system always remains plastic, thus the time of administration of therapy would not be a factor. Improvements would, then, be possible at any time following the insult. 1 While people who are as much as 18-years post stroke have demonstrated functional improvements, 2 the literature results are still unclear if individuals in the chronic phase of stroke can improve to the same extent as those who have sustained a stroke more recently. Beginning CIMT in the acute, sub-acute or chronic phase after stroke may affect outcomes. Time since stroke is often used as a predictor for stroke outcomes. 5,57-59,93-95 Findings from longitudinal studies, with repeated measures across time, demonstrate that neurological recovery shows a nonlinear progression pattern as a function of time, but

PAGE 73

59 few individuals show additional improvement after three months post-stroke. 5,57-59,93-95 The length of time, in which there is a lack of improvement following a stroke, reflects the intrinsic cerebral damage and should be seen as an important predictor of poor outcome. 5,92 While the greater the time since stroke is associated with poorer functional recovery, the attention-driven methods of CIMT may counteract this temporal relationship, resulting in improvements at any time post stroke. Hand dominance Hand dominance affects an individual’s functional ability after a stroke, yet current CIMT studies have not sufficiently addressed dominance and its role in functional recovery. 96 Hand dominance is a behavioral manifestation of hemispheric asymmetry. CIMT’s effectiveness, as it relates to upper-extremity dominance, is unclear. The early CIMT studies excluded individuals with left-hand dominance or left hemiplegia, in order to test and analyze with less variance. 10,16 The generalizability, therefore, to non-right handers is limited. Another interesting, yet unexplored question, is whether CIMT is of greater benefit to individuals with dominant-side hemiparesis as opposed to non-dominant hemiparesis. Intuitively, one may consider that people with dominant-side hemiparesis would be more motivated to regain function of the affected extremity because this is the extremity of functional preference and practiced motor skill. Moreover, the dominant hemisphere has more intricate motor programs with better-developed coordination and skill, thus translating to improved neural representation. Does this pre-morbid increased representation of the dominant hand remain post-stroke, allowing for increased ease in recovery of function? In contrast, an individual with a non-dominant-side stroke could possibly return to a previous level of functioning at a quicker pace as compared to a

PAGE 74

60 dominant-side stroke. This possibility exists because of the decreased functional demands on the non-dominant extremity. The relationship of functional return following CIMT to pre-morbid handedness and motivation to regain use of a dominant extremity is unclear. Perhaps, a more interesting question is to investigate if dominance plays a role in predicting recovery at the follow-up assessment. The intense methods of CIMT may allow for no difference to be observed between individuals with dominant and non-dominant hemiparesis immediately following CIMT, however, this may change at follow-up. Obviously, dominance is an important issue that needs to be investigated as a predictor of function following CIMT. Age The research literature to date presents conflicting results regarding the role of age in rehabilitation. On one hand, motor performance factors have been demonstrated to be greatly influenced by age. 97-100 Certainly age could be an important predictor of recovery potential following stroke. 98-100 According to Jongbloed (1986), age was identified as a significant prognostic factor in numerous studies. 101-104 Specifically, these studies indicate that age is negatively correlated with functional return. In studies reporting age as a factor, however, additional contributors of functional limitations such as co-morbidities, presence of a caregiver, and severity of stroke are frequently not accounted for. On the other hand, conflicting studies have shown that age does not have a negative impact on function over time. 101 In fact, much research has demonstrated the benefits of intensive stroke rehabilitation programs regardless of age. 4,100,105,106 Kugler et al. (2003) states that age should not be a limiting factor in the early rehabilitation of stroke patients. 98 The role age plays as a predictor of outcomes post-stroke is uncertain. The

PAGE 75

61 inclusion of age in the regression as a potential predictor is important to help clarify the role age plays in CIMT. Gender The relationship between gender and stroke rehabilitation is not well explored. Scarce information exists about gender differences in the management of individuals with stroke. 107 Research on gender differences following stroke is needed in order to provide useful insight for long-term intervention and establishment of appropriate therapy. 107 The direct relationship between incidence of stroke and advancing age means that female patients, because they have a longer life expectancy than males, may bear the burden of much of the disease and subsequent disability. 107 Wyller et al. (1997) studied gender difference in functional outcomes following stroke. 108 Utilizing age-adjusted odds, these authors concluded that women seem to be functionally more impaired by stroke than men. 108 In addition, women are diagnosed with depression twice as frequently, which is strongly correlated with more impairment and loss of function. 109 Conversely, Twigg et al. (1998) found no significant correlation between gender and functional outcomes following stroke. 110 Time since stroke differences in the previously mentioned studies could explain some of the contradictory results. For example, the study that observed differences between genders 108 was performed in a chronic population, one-year, post-stroke, while the study that found no differences between genders utilized the acute and sub-acute population at a rehabilitation facility. 110 Gender differences may not emerge until further in the recovery process. Nonetheless, the question regarding gender influence on outcomes following stroke is important and remains unclear.

PAGE 76

62 Ambulatory status Ambulatory status was included as a potential predictor of upper-extremity recovery for several reasons. This predictor is included in the regression model because cumulative deficits post-stroke can affect individuals functional outcomes. 111 Individuals who are non-ambulatory, therefore, may have poorer outcomes. Moreover, while the most rapid recovery for both upper and lower extremities occurs within the first 30 days after stroke, the severity of motor impairments and the subsequent patterns of recovery are similar for both extremities. 112 A relationship exists, therefore, between upper and lower extremity motor recovery. 112 In addition, the ability to walk is a strong predictor of functional outcomes 112 and has been included in regression models previously for stroke recovery. 113 Individuals who are able to ambulate may be able to use their hand in more functional tasks. If a participant is able to stand and walk, they have more synergies available to accomplish gross motor tasks. Furthermore, if someone is walking they may spontaneously find themselves involved in more activities due to increased access. Ambulatory status could lead to increased independence with tasks, and therefore, ambulators may be better candidates for CIMT. Including ambulatory status allows for a simple predictor that could potentially be very useful for determination of outcomes following CIMT. For the purposes of this study, ambulation was used as a dichotomous variable. If the participant presented to the clinic ambulating, with or without an assistive device, then they were considered functionally ambulatory. If they presented in a wheelchair the majority of the time, they were categorized as non-ambulatory.

PAGE 77

63 Data Analysis Demographic and clinical presentation characteristics of the study sample are described using the mean, median and standard deviation for continuous variables, and using frequencies and percentages for categorical variables. For the intention-to-treat analysis demographic and clinical characteristics of the patients who were admitted to the program, but who did not participate in the 4-month follow-up evaluation, were compared with participants who completed the follow-up to determine differences that could result in a bias. The continuous variables from the intention-to-treat analysis were analyzed using t-tests and categorical variables using Chi square tests or Fisher’s exact tests. All data were analyzed using an intention to treat approach in which the pre-test scores were used as the follow-up post-test scores for those participants that did not return for the follow-up. The normality of the data for the WMFT and the MALa as dependent variables was visually verified with probability plots (P-P) and statistically verified with the Kolmogorov-Smirnov f test. The MALa met assumptions of normality, however, the WMFT required transformation using the natural log [ln] to meet assumptions of normality. The predictive relationships of various descriptive variables on the traditional measures of CIMT were investigated. The independent variables were used to develop a general linear model for the dependent variables individually: the WMFT (aim 1) and the MALa (aim 2). These analyses were performed for the immediate post-test and the follow-up post-test. The immediate post-test regression allowed for determination of predictors of immediate success. The follow-up post-test represented changes overall

PAGE 78

64 from the first test period to the last testing period allowing insight into predictors of long-term success (Figure 3-3). Follow-up post-test Time Frame 2 Time Frame 1 Immediate p ost-test Pre-test 14 days 4-6 months Figure 3-3. Timeline for testing of dependent variable Potential predictors were selected for the development of a general linear model for the prediction of outcomes with CIMT. The pretest scores for each of the dependent measures were used as covariates to statistically control for group differences that existed before treatment. A forward stepwise procedure was employed in which the variables were examined at each step for entry or removal from the model. The least significant variables were removed from the model, based on their level of association with the dependent variable at each step. Adjusted R-square values, p-values and 95 percent confidence intervals (CIs) were calculated. Thorough regression diagnostics were run and jackknife residual analyses were performed to verify the basic assumptions. Multicollinearity among predictor variables in the regression models was assessed using a variance inflation factor (VIF). Results Power Analysis A post-hoc power analysis was conducted to determine the power for the sample size (n=55). The effect size (f 2 ) for the WMFT immediate post-test and follow-up post-test and the MALa immediate post-test and follow-up post-test was determined from the data and they are 3.30, 2.58, 0.70, and 0.89, respectively. Using these effect sizes, the

PAGE 79

65 sample size of 55 participants at an =0.05, for six predictors, met an average power level = 1.0. This strong power can be attributed primarily to the presence of the covariate in the model. Descriptive Characteristics The descriptive statistics of the independent variables are presented in Figure 3-1. Intention-to-Treat Analysis An intention-to-treat analysis was selected for this study because of the differences in functional level between those that completed the study and those that did not. Nine participants (16%) did not return for the follow-up post-test evaluation. There was no significant difference for the descriptive variables for the individuals that completed the study and those that dropped-out (Table 3-1 and 3-2). Individuals with lower ability levels, however, showed significantly higher drop-out rates. For example, individuals that dropped out had a significantly lower UE Fugl-Meyer motor score and grip strength than those that completed the follow-up post-test evaluation. For this reason, an intention-to-treat analysis was used due to the significant differences in functional level found between the group that completed the follow-up post-test and the group that did not. 129-132 Table 3-1. Chi-Square and Fisher’s exact test for the differences across categorical independent variables between those participants who completed the followup post-test and those that did not complete the evaluation Drop-outsCompletedChi-squareFisher's Exact TestL226R720dom422non524F319M627yes638no 38Independent Categorical VariablesGender0.060.078Side of strokeDominanceAmbulatory10.8530.2740.3620.6550.727

PAGE 80

66 Table 3-2. T-test for difference across continuous independent variables between those participants who completed the follow-up post-test and those that did not complete the evaluation Multiple Regression Modeling Aim 1: Wolf Motor Function Test modeling Side of stroke (right= 0, left=1), time since stroke, hand dominance (nondominant= 0, dominance=1), age, gender (male=0, female=1), and ambulatory status (no=0, yes=1) were entered into a linear multiple regression model with stepwise entry using the [ln]WMFT post-score as the dependent variable and entry of the WMFT pre-score as the covariate. None of the potential predictor variables made significant contributions to the WMFT post-score. The model accounted for 0.769 of the variance in [ln]WMFT at immediate post-test with the covariate alone. The B and p-values for all the independent variables are listed in Table 3-3. The final regression equation is as follows. [ln]WMFTpost’= .967 +(.04111)(WMFTpre) The same independent variables were entered into a second linear multiple regression model with stepwise entry using the [ln]WMFT follow-up post-test score as the dependent variable. Similarly, no significant independent variables emerged. This model accounted for 0.721 of the variance in WMFT at follow-up post-test with the covariate alone. This model is very similar to the WMFT immediate post-test model. The B and p-values for all the independent variables are listed in Table 3-3. The final regression equation is as follows. [ln]WMFTfu’= 1.174 +(.03811)(WMFTpre) Drop-outsCompletedTime since stroke1518 (1143)1703 (2559)0.73 A ge64.89 (17.861.5 (14.1)0.6mean (sd)t-test ( si g) Independent Continous Variables

PAGE 81

67 Table 3-3. The adjusted R squared for the [ln] WMFT models, the B weights, the p-values for all the predictors, and the confidence intervals for the significant im 2: Motor Activity Logamount modeling d into a third linear multiple regression modes. iple regress the est variables are listed in Table 3-4 and the regression equation is as follows. predictors are listed. WMFT MODELS Adjusted R squareF ( si g) INDEPENDENT VARIABLES B p-value (CI) B p-value (CI)Covariate ( WMFT p re-score ) 0.0410.000 ( .035,.047 ) 0.0380.000 ( .032,.045 ) Side of stroke (right=0, left=1)0.0730.2770.1400.055 T ime since stroke0.0140.8320.0440.552Hand dominance (non=0, dom=1)0.0220.7400.1150.110A g e-0.1140.089-0.0890.230Gender (male=0, female=1)0.0400.5490.0710.329Ambulator y status (no=0, yes=1)0.0410.5540.0090.907Constant0.9670.000 (.063,1.305)1.1740.000 (.819,1.529)0.7690.721180.791 (.000)140.459 (.000) [ln] Follow-up post-test Immediate post-test A The same independent variables were entere l with stepwise entry using the MAL amount immediate post-test as the dependent variable. Ambulatory status was the only significant predictor for the MALa. All other variables were removed during the regression analysis. This model accounted for 0.410 of the variance in MALa at the immediate post-test. The B and p-values for all the independent variables are listed in Table 3-4 and the regression equation is as followMALapost’= .994 +(1.148)(MALpre) + (1.033)(Ambulatory Status) The same independent variables were entered into a fourth linear mult sion model with stepwise entry using the MALa follow-up post-test score adependent variable. At the follow-up post-test, age was the only significant predictor variable. This model accounted for 0.472 of the variance in MALa at follow-up post-tand the regression equation is as follows. The B and p-values for all the independent

PAGE 82

68 MALafu’= 2.308 +(1.047)(MALpre) (0.0196)(Age) Table 3-4. The adjusted R squared for the MAL amount models, the B we ights, the p-values for all the predictors, and the confidence intervals for the significant nfluence the regression estimual . ere The goal of this study was to detgnificant descriptive predictors of gly, only two descriptive characteristics demonstrated a predicghts, the p-values for all the predictors, and the confidence intervals for the significant nfluence the regression estimual . ere The goal of this study was to detgnificant descriptive predictors of gly, only two descriptive characteristics demonstrated a predic predictors are listed. predictors are listed. MALafu’= 2.308 +(1.047)(MALpre) (0.0196)(Age) The largest value of the variance inflation factors was <1.07, indicating that MALafu’= 2.308 +(1.047)(MALpre) (0.0196)(Age) The largest value of the variance inflation factors was <1.07, indicating that INDEPENDENT VARIABLES B p-value (CI) B p-value (CI)Covariate ( MALa p re-score ) 1.148.000 ( .706,1.589 ) 1.0470.000 ( .717,1.378 ) Side of stroke (right=0, left=1)-0.0870.4130.0180.858Time since stroke0.1530.1440.0860.400Hand dominance (non=0, dom=1)-0.0490.6460.1520.127A g e-0.1370.239-0.020.026 ( -.037,-.002 ) Gender (male=0, female=1)0.0150.8860.0470.638Ambulator y status (no=0, yes=1)1.033.015 ( .205,1.861 ) 0.1040.350Constant0.994.012 (.225,1.764)2.3080.000 (1.148,3.467) MAL amount MODELSted R squareF Adjus ( si g) Follow-up post-test0.4100.472 Immediate post-test 19.749 (.000)25.093 (.000) multicollinearity among the predictors did not unduly imulticollinearity among the predictors did not unduly i ates.133 The adequacy of the final regression model was examined. Upon visexamination of the histogram, the residuals appeared to be normally distributedPresence of outliers was assessed using Jackknife residuals. The sample contained twooutliers for the WMFT model. The influence and accuracy of these data points wassessed and they remained in the model.134,135 These post-hoc regression diagnostics results suggested that the regression analysis was appropriate. Discussion Multiple Regression Modeling ates.133 The adequacy of the final regression model was examined. Upon visexamination of the histogram, the residuals appeared to be normally distributedPresence of outliers was assessed using Jackknife residuals. The sample contained twooutliers for the WMFT model. The influence and accuracy of these data points wassessed and they remained in the model.134,135 These post-hoc regression diagnostics results suggested that the regression analysis was appropriate. Discussion Multiple Regression Modeling ermine the siermine the si outcomes for CIMT. Interestinoutcomes for CIMT. Interestin tive relationship, ambulatory status and age. Both of these predictors showed a tive relationship, ambulatory status and age. Both of these predictors showed a

PAGE 83

69 relationship with the MALa outcome. None of the potential predictors showed a predictive relationship with the WMFT. Ambulatory status was a significan t predictor for the amount section of the MAL immeter aker descriptive characteristic that has a predictive relationship with amou the WMFT, however, is not affected by increasing age. diately following therapy. This means, if two individuals have the same MALa score at baseline, the individual who is a functional ambulator is predicted to have greaimprovements in amount of hand and arm use following the two-week intervention. Ambulation may allow individuals greater access and more opportunity to use the wehand, which would translate to improved scores on the amount section of the MAL. This said, ambulation is not a significant predictor for the MALa at follow-up post-test. While a functional ambulator may have better outcomes initially, these differences are not sustained long-term. Age is the second nt of use of the more affected UE following CIMT. Age is only a predictor for the amount section of the MAL at follow-up post-test. Assuming the same baseline MAL score, as age increases, the predicted MAL at follow-up post-test decreases. There is aninverse relationship, therefore, between MAL and age. Although there is a general beliefthat younger persons have greater potential for recovery,144,101-104 much research has demonstrated the benefits of intensive stroke rehabilitation programs regardless of age.4,100,105,106 For this study, age is only a predictor of long term amount of use of more affected UE, not movement capability of the affected UE. As an individual ages,therefore, the amount of use of their more affected arm and hand does not increase to thesame degree as for someone who is younger. Movement capability, as measured by the

PAGE 84

70 Other Potential Predictors Side of stroke, time since stroke, hand dominanc e, and gender were not predictive ediately after, or four months following, CIMT. Knowing what er. theref future, f al n this sample was approximately 7 months post-stroke. A sampntly debated and questioned topics. Intuitively, one may assume that indivion hich ell in of the outcome measures imm variables are not predictive is valuable and necessary when screening a potential client for CIMT. Expressly, individuals should not be discriminated against for participation in CIMT based upon side of stroke, time since stroke, dominance, or gendSide of lesion was not a significant predictor of outcomes despite the vast differences in presentation. Individuals with both right and left-sided hemiparesis, ore, should continue to be considered as potential CIMT candidates. In theside of stroke could be further delineated to see if apraxia or neglect are predictors ooutcome with this therapy. Time since stroke was not a CIMT outcome predictor in this study. The individuwith the most acute stroke i le that includes individuals with acute stroke, however, may have differing results. Incorporating a more diverse sample could lend further insight into this potential predictor. This result should be encouraging for those individuals with chronic disability for stroke. The role hand dominance plays in CIMT and stroke rehabilitation is one of the most freque duals with dominant-side hemiparesis would be more motivated to regain functiof the affected extremity. Our findings support those of Miltner et al. (1999) in wdominance does not seem to be a factor in outcomes for CIMT.15 Interestingly, dominance also did not play a role in long term-follow up. Assuming equal baseline scores, therefore, individuals with non-dominant hemiparesis would function as w

PAGE 85

71 the long term as those with dominant-side hemiparesis in both movement capability anamount of use. This implies that functional return following CIMT is not related to dominance and may not be related to motivation to regain use of a dominant extremity. Gender did not emerge as a significant predictor of outcomes for CIMT. To dathere is little information that exists regarding gender differences post-stroke.107 The d te resultedictors ediate post-test or at the follow-up post-test. Specifically, we know that nte hose for the follow-up post-tays as delineated in the s of this study indicate that gender is not an influential predictor for CIMT outcomes. Of interest is that none of the potential predictors emerged as significant prfor the WMFT at imm one of the descriptive variables were predictors of movement capability of the affected UE, as measured by the WMFT. This is a significant finding within itself for the development of more inclusive criteria for study participation. While the finding that ambulatory status is a significant predictor for the immediapost-test for the MALa, the predictors that are of most interest are t est. More specifically, the predictors of how individuals perform 4-6 months after the treatment. The most ideal candidates, the ones that will improve the most in the long-term, are those that are younger. If clinicians, therefore, only want to offer traditional CIMT to individuals who will make the most gains in amount of use of the affected UE in the long-term, the results of this study suggests offering it to individuals who are younger. Important to clarify, however, is that older individuals make meaningful gains with therapy, just not to the same extent as younger individuals. An important clarification is that these reported results apply to only traditional CIMT, when delivered 6 hours per day for 10 consecutive weekd

PAGE 86

72 proce dure. They may not be appropriate predictors of success for other methods of CIMT delivery, such as modified CIMT,67,69 home-based CIMT, and CIMT delivered in a less-intense manner or for a different duration. For example, age may not be a significant predictor if the therapy was less intense. An older individual, for instance, may be able to more actively participate in CIMT if it was only three hours per dainstead of six hours. The qualification is that this study represents strictly traditional CIMT, as reported in the research literature. Additional considerations for future studies include investigating the predictors other types of CIMT, when administered in d y for iffering intensities or duration. Another researnnext step, arious descriptive predictors on the traditional ted using linear regression. Including the pre-test scores as a ce s ch avenue would be investigating the predictors of individuals who do not improve with therapy. Specifically, defining predictors that can identify both responders and noresponders would be advantageous to proper application of CIMT. Another direction for future research is establishing if the changes in outcomes represent a clinically meaningful difference. In other words, what amount of improvement is needed on the outcome measures for the change to be perceived as clinically meaningful. Thethen, would be determining the predictors for individuals who make a clinically important change with this therapy. Statistical Analysis Discussion The predictive relationship of v measures of CIMT was investiga ovariate in the models may be questioned because it increases complexity of thmodel use. Specifically, a baseline test is required in order to utilize the regression equations for prediction of outcomes for CIMT. The use of the covariate, however, wa

PAGE 87

73 essential to form accurate regression models. A pre-test for each measure was used as a covariate to statistically control for differences that existed before treatment. Another statistical option considered was the use of change score or the p ercent changsures 1995), While stepwise metho ery it as epwise e as the dependent measure across time frames. The use of change scores, however, is poor statistical practice for these dependent measures.136,137 The meaare not of equal intervals and have ceiling and floor effects, which would lead to inaccuracy in model development and interpretation. According to Allison et. al (the use of ratios as response variables in regression should be avoided, if possible in favorof including the baseline measure as a covariate in the regression.138 Controversy exists surrounding the use of step-wise regression.139 ds are frequently used in behavioral research, Thompson (1995) argues that severe limitations exist with this method.139 For example, he claims that incorrect degrees of freedom are often used by the statistical software. This misuse, however, is accentuatedwhen a large number of predictors are ‘thrown’ into the model. In this study restricting the stepwise equation to 6 predictors decreases the chance of a type I error. Another limitation is that stepwise regression tends to capitalize on sampling error. In other words, if your sample is not representative, your results will not be replicable.139 Eveffort was made for the sample to contain a diverse population of individuals w ith stroke across descriptive and functional categories. Many recently published articles include step-wise regression for predictive modeling.5,7,140,141 According to Jonsson (2000) researchers need to be aware of the possible dangers of stepwise model building, use appropriately as possible, and realize that this kind of approach is an “essential component in our efforts to make sense out of the data we are to analyze.”142 St

PAGE 88

74 regression was the appropriate methodology to use in this study. This allo ws for the moefficient and usable models. If simultaneous entry were used instead of stepwise, it would result in including non-significant predictors in the regression equations. Thiswould cause greater clinical demand to measure all the predictors in order to employ tmodel. Pat st he h analysis or structural equation modeling are techniques that could have also been the d eficial therapy for individuals post-strokeIMT ls ls. In used in analyzing this data. These methods, however, are more confirmatory than exploratory. In path analysis, a directional pathway is constructed from pre-existing knowledge and/or data and then the strength of the path is confirmed. Since this wasfirst study of its type for CIMT, the objective was to explore the factors that should be included in a predictive model. Path analysis could be a future method to investigate anconfirm the strength and directionality of this study. While CIMT has been demonstrated to be a ben , the descriptive characteristics that are predictive of CIMT outcomes were unclear.1,2,6,10-25,115 The identification of these clinical predictors for outcomes of Cis needed to target the appropriate populations. The goal of this study was to establish usable models, not extensive or comprehensive models. While detailed models serve a function, they are less readily utilized and have little predictive capability over simple models.8,9 Often models with only a few predictors have almost equal strength to modewith mu lt iple predictors.8,9 Other predictive variables could have been included in this model, potentially making it more inclusive. For example, cognitive function or motivational variables may have increased the predictive capability of these modeaddition, modern imaging techniques could have been included to identify type or

PAGE 89

75 location of lesion, since they are valuable in prediction of stroke outcomes.114 Theadditional predictors would add to the complexity of the model, making it m se lt dy was performed in individuals who met strict inclusion and exclusion criter ple , arried CIMT is reported to significantltional use of the upper-extremity in 20 to t ors, and g o re difficuto use and clinically less practical, while potentially adding little strength to the model. Limitations This stu ia. While, this sample was more diverse than most CIMT studies, it is not representative of the entire stroke population. Furthermore, individuals were notrandomly selected, but were respondents to inquiries for participants. Second, samsize decreased at the follow-up post-test evaluation, leading to a bias for withdrawals. While this bias was dealt with during analysis, using an intention-to treat approach, its presence as a limitation is still significant. Third, due to the relatively small sample sizewe were unable to evaluate the stability of the model using a portion of the sample as a reserve. These results should be replicated in another sample. Fourth, the outcome measures were chosen due to their abundant use in the CIMT literature, however, predictors based on other, more readily used stroke outcome measures should be considered. Finally, a predictor model for longer-term follow-up should also be cout. This is essential to know predictors of long-term success. Conclusions y improve func 25 percent of people with chronic stroke disability.1 Limited evidence exists, however, regarding the specific descriptive characteristics of individuals who benefimost from this intervention. Significant predictive ability was discovered with ambulatory status and age. Ambulators had greater outcomes than non-ambulatan inverse relationship was demonstrated between age and amount of use. When enterin

PAGE 90

76 ambulatory status and age in the regression equations, along with the covariate score, an individual’s outcome can be predicted. Further substantiation of these findings in larger and more diverse samples is warranted in order to meet the urgent need of determining the appropriate candidates for CIMT.

PAGE 91

CHAPTER 4 GENERAL SUMMARY AND CONCLUSIONS Specific and detailed discussion sections have been provided previously concerning the two studies of functional and descriptive predictors for CIMT outcomes. The purpose of this final chapter is to summarize and integrate the key findings for each experiment. Overview While CIMT has been proven to be a beneficial therapy for individuals post-stroke, the functional and descriptive characteristics that are predictive of CIMT outcomes have been unclear. 1,2,6,10-25,115 The identification of these clinical predictors is needed to target the appropriate populations. The goal of these experiments was to establish meaningful and clinically usable models, not extensive or comprehensive models. Although extensive and detailed models serve a function, they are less readily utilized and have little predictive capability over simple models. 8,9 Often regression models with only a few predictors have almost equal strength to models with multiple predictors. 8,9 The two research papers discriminate between the influences of various functional and descriptive predictive factors on the outcomes of CIMT divided into two essential time frames. These timeframes were necessary to establish the potential predictors of those who improved most during the treatment phase and those individuals who maintained or improved functional gains four to six-months after the CIMT intervention. 100 Previous CIMT studies suggest that functional gains are persistent, 1,16 signifying individuals maintain improvements that are made during the training program. 77

PAGE 92

78 Identifying the predictors of individuals who are able to maintain functional gains at the long-term function is of primary importance. 100 Experiment I Summary The goal of this study was to determine the significant functional predictors of outcomes for CIMT. The potential functional predictors were: 1) minimal motor criteria level, 2) amount of finger extension/grasp release, 3) amount of grip strength, 4) upper-extremity Fugl-Meyer score and, 5) score of the Frenchay Activities Index. These predictors were included in the regression model for several reasons: some are shown to be predictors in other areas of stroke rehabilitation, many appear in CIMT research literature, and some demonstrated predictive ability in pilot work. Finger extension/grasp release is the only predictor of positive outcomes for movement capability of the more affected UE, as measured by the WMFT, immediately following therapy and at 4-6 months after therapy. In fact, the regression models were very similar for the WMFT at both the immediate post-test and follow-up post-test. The predictive capability of finger extension/grasp release is logical due to its role in the functional use of a hand. This measure, while related to minimal motor criteria, does not include the requirements of wrist extension. In this sample, minimal motor criteria did not predict outcomes for the WMFT. This finding is important since the primary screening criteria used for the majority of CIMT studies is an individual’s ability to meet minimal motor criteria. Establishing the minimal motor criteria has been an advantageous and appropriate screening technique for traditional CIMT until this time. The component of wrist extension, however, may be too restrictive. An individual’s ability to extend his/her fingers or release a mass flexion grasp may be a better screening criterion. Quite possibly, wrist extension is not as important as was previously believed.

PAGE 93

79 Many of the tasks practiced in CIMT involve grasp and release, and while wrist extension is helpful in this capacity, it appears not to be essential for positive outcomes with CIMT. Although wrist extension is used in normal grasp, perhaps, presence of finger extension is enough to justify participation in CIMT. Minimal motor criteria should be investigated further to determine its appropriateness for screening of potential CIMT participants. The WMFT immediate post and follow-up post-test regression equations are graphically depicted in Figures 5 and 6 respectively. These figures are meant as examples to demonstrate and visually explain the interpretation of the regression equation. For ease, a WMFT pre-score of 20 seconds was chosen for this demonstration. If the pre-test is Figure 4-1. A graphical example of WMFT post model for Experiment I. If the WMFT constant the change at immediate post-test is due strictly to finger extension capability. pre-score is set at 20 seconds, the estimated WMFT post-test is shown for the The riteria is predictive of the MALa immediately following therapy. The prediction does not hold up at the long-term follow-up. While the predictors for the immediate post-test are of interest, the most important WMFT Immediate Post-test'0510152025WMFT Pre WMFT Imm Post'time in seconds 0cannot extend 1partially extend 2fully extend [ln]WMFTpost’= ( 1.608 ) + ( .03575 ) WMFTpre – ( .410 ) fin g er extension Finger Extension different levels of baseline finger extension. results suggest, however, that minimal motor c

PAGE 94

80 Figure 4-2. A graphical example of WMFT follow-up model for Experiment I. If the WMFT pre-score is set at 20 seconds, the estimated WMFT follow-up is shown for the different levels of baseline finger extension. predictors are those of the follow-up post-test. More specifically, the predictors of how 051015WMFT Pre WMFT Follow'time in secon 1partially extend 2fully extend [ln]WMFTfu’= (1.804) + (.03285)WMFTpre – (.402)finger extension individuals perform 4-6 months after the treatment. While individuals who meet minimria demonstrate improved amount of use of the affected UE al motor criteimmediately to the Fust E following therapy, the prediction does not remain at the long-term follow-up. The Fugl-Meyer UE motor score, however, is predictive of outcomes at both time frames for the MALa. This indicates that individuals with higher function, accordingthe Fugl-Meyer, result in prediction of higher scores on the MALa. One component of gl-Meyer evaluates an individual’s ability to move in and out of synergy. Quite possibly an individuals ability to function out of synergy is an important predictor of success with CIMT. This would allow increased ease in practicing tasks that involve reaching and multi-joint coordination, thus translating to greater success following CIMT. The Fugl-Meyer UE motor test could be further broken down to evaluate if juone component is predictive of improvements. The significance of the Fugl-Meyer U WMFT Follow-up Post-test' Finger Extension 2025ds 0cannot extend

PAGE 95

81 motor test was expected because of its predictive value that has been demonstrated iother stroke rehabilitation studies. 5,7,71 The MALa immediate post and follow-up post-test regression equations are graphically depicted in Figures 7 and 8, respectively. Thesefigures are meant as examples to demonstrate and visually explain the interpretation ofthe regression equation. For ease, a MALa pre-score of 1.0 was chosen for this demonstration. If the pre-test is constant, the change at immediate post-test is due strictly to the predictive variables. d iother stroke rehabilitation studies. 5,7,71 The MALa immediate post and follow-up post-test regression equations are graphically depicted in Figures 7 and 8, respectively. Thesefigures are meant as examples to demonstrate and visually explain the interpretation ofthe regression equation. For ease, a MALa pre-score of 1.0 was chosen for this demonstration. If the pre-test is constant, the change at immediate post-test is due strictly to the predictive variables. MALa Immediate Post-test' MAL Imm Po n gression analysis as potential predictors, however, neither proved to significantly predict out also valuable and necessary when screening a potential client for CIMT. This offers insight into n gression analysis as potential predictors, however, neither proved to significantly predict out also valuable and necessary when screening a potential client for CIMT. This offers insight into Figure 4-3. A graphical example of the MALa post-test model for Experiment I. The estimations are made with the MALa pre-test score set to 1.0. The Frenchay Activities Index score and grip strength were included in the Figure 4-3. A graphical example of the MALa post-test model for Experiment I. The estimations are made with the MALa pre-test score set to 1.0. The Frenchay Activities Index score and grip strength were included in the 01234253035404550556065253035404550556065Does NOT meet Minimal Motor CriteriaMeets Minimal Motor CriteriaUE Fugl-Meyer Motor Scoreamount of use MALapost’= (-.0301)+ (.204)MALapre + (.05634)F-M UE motor + (1.116)Min Motor 56 st' rere comes for CIMT. Knowing what variables are not predictive iscomes for CIMT. Knowing what variables are not predictive is

PAGE 96

82 Figure 4-4. A graphical example of the MALa follow-up test model for Experiment I. The estimations are made with the MALa pre-test score set to 1.0. Therefore the results are affected only by the baseline Fugl-Meyer UE motor scores. what predictors should not be considered when determining if a participant can improve with this therapy. For example, denying participation to individuals who demonstrate low grip strength would be inappropriate, as it is not a predictor of success. If clinicians only 0.511.52MAL PreMAL Follow'amou 45 55 65MALafu’= (.645) + (.409)MALapre + (.0643)F-M UE motor want to off er traditional CIMT to individuals who will make the most long-term gains in UE movement capability and amount of use of the affected UE, the results of ExperimentI suggest enrolling individuals who are able to actively extend their fingers from a massed grasp and who have higher Fugl-Meyer UE motor scores. While improvements occur following CIMT with even a very low UE Fugl-Meyer score, the higher the baseline score the greater the improvement following therapy. To determine the predicted immediate post-test scores for a given individual, the regression equationsshould be utilized. MALa Follow-up Post-test'2.53.5nt ofe 344.5 us 25 Fugl-Meyer UE Motor Score 35

PAGE 97

83 Experiment II Summary The goal of th is study was to determine the significant descriptive predictors of outcomes for CIMT. The potewere: 1) side of stroke, 2) time since and me MALa scorer n of ctor for the amount section of the MAL at followntial descriptive predictors stroke, 3) hand dominance, 4) age, 5) gender, and 6) ambulatory status. These predictors were included in the regression model for several reasons: some are shown to be predictors in other areas of stroke rehabilitation, many appear in CIMT research literature, and some demonstrated predictive ability in pilot work. Interestingly, only twodescriptive characteristics demonstrated a predictive relationship, ambulatory status age. Both of these predictors showed a relationship with the MALa outcome. None of the potential predictors showed a predictive relationship with the WMFT. Ambulatory status was a significant predictor for the amount section of the MAL immediately following therapy. This means, if two individuals have the sa at baseline, the individual who is a functional ambulator is predicted to have greateimprovements in amount of hand and arm use following the two-week intervention (Table 4-5). Ambulation may allow individuals greater access and more opportunity to use the weaker hand, which would translate to improved scores on the amount sectiothe MAL. This said, ambulation is not a significant predictor for the MALa at follow-uppost-test. While a functional ambulator may have better outcomes initially, interestingly, these differences are not sustained long-term. Age is the second descriptive characteristic that has a predictive relationship with outcomes following CIMT. Age is only a predi up post-test in Experiment II. Assuming the same baseline MAL score, as age

PAGE 98

84 MALa Immediate Post-test'0.511.522.533.5MAL PreMAL Imm Post'amount of use Ambulator Non-AmbulatorMALpost’= .994 + (1.148)(MALpre) + (1.033)(Ambulatory Status) Figure 4-5. A graphical example of the MALa post-test model for Experiment II. The estimations are made with the MALa pre-test score set to 1.0. Therefore the results are affected only by ambulation status. increases, the predicted MAL at follow-up post-test decreases. There is an inverse relationship, therefore, between MAL and age (Table 4-6). Although there is a general belief that younger persons have greater potential for recovery, 144,101-104 much research has demonstrated the benefits of intensive stroke rehabilitation programs regardless of age. 4,100,105,106 For this study, age is only a predictor of long term amount of use of the more affected UE, not movement capability of the affected UE. As an individual ages, the amount of use of their more affected arm and hand does not increase to the same degree as for someone who is younger. Movement capability, as measured by the WMFT, however, is not affected by increasing age. While the finding that ambulatory status is a significant predictor for the immediate post-test for the MALa, the predictors that are of most interest are those for the follow-up post-test. More specifically, the predictors of how individuals perform 4-6

PAGE 99

85 MALa Follow-up Post-test'0.511.522.53MAL PreMAL Follow'amount score 40 50 60 70 80MALfu’= 2.308 +(1.047)(MALpre) (0.0196)(Age) Figure 4-6. A graphical example of the MALa follow-up model for Experiment II. The estimations are made with the MALa pre-test score set to 1.0. Therefore the results are affected only by age. months after the treatment. The most ideal candidates, the ones that will improve the most in the long-term, are those that are younger. If clinicians, therefore, only want to offer traditional CIMT to individuals who will make the most gains in amount of use of the affected UE in the long-term, the results of this study suggests offering it to individuals who are younger. Important to clarify, however, is that older individuals make meaningful gains with therapy, just not to the same extent as younger individuals. Of interest is that none of the potential predictors emerged as significant predictors for the WMFT at immediate post-test or at the follow-up post-test. Specifically, we know that none of the descriptive variables were predictors of movement capability of the affected UE, as measured by the WMFT. This is a significant finding within itself for the development of more inclusive criteria for study participation.

PAGE 100

86 General Conclusions and Integration Limited evidence exists regarding the specific functional and descriptive characteristics of individuals who benefit most from CIMT. Significant predictive ability was discovered with finger extension/grasp release, minimal motor criteria level, Fugl-Meyer UE motor scores, ambulatory status, and age. These items, when used in their appropriate regression equations along with the pre-test scores, can predict an individual’s score on the WMFT or the MALa. This study can be used to help determine the appropriate individuals for participation in traditional CIMT. An important clarification is that these reported results apply to only traditional CIMT, when delivered 6 hours per day for 10 consecutive weekdays as delineated in the procedure. They may not be appropriate predictors of success for other methods of CIMT delivery, such as modified CIMT, 67,69 home-based CIMT, and CIMT delivered in a less-intense manner or for a different duration. For example, age may not be a significant predictor if the therapy was less intense. An older individual, for instance, may be able to more actively participate in CIMT if it was only three hours per day instead of six hours. The qualification is that this study represents strictly traditional CIMT, as reported in the research literature. The WMFT and MALa should be accepted as appropriate identifiers of success for CIMT. They are extensively cited in the CIMT literature as outcomes measures. 12,117-126 Furthermore, accepting the limitations that these regression models include only a limited group of potential predictors, we can identify the predictors that are influential for CIMT outcomes. Specifically, we know that the only predictor of movement capability of the affected UE, as measured by the WMFT, is finger extension/grasp release. Establishing the minimal motor criteria has been an advantageous and an appropriate screening

PAGE 101

87 technique for traditional CIMT until this time. Based on our research using minimal motor criteria for inclusion may be too restrictive. If clinicians only want to offer traditional CIMT to individuals who will make the most gains in UE functional capability, the results of this study suggest offering it to individuals who are able to actively extend their fingers from a massed grasp. This, however, is only a predictor of functional capability of the more affected arm and hand, not amount of use. The individuals who will demonstrate greater improvements in amount of use of the more affected UE are those that are young ambulators with higher Fugl-Meyer UE motor scores. While the predictors for the immediate post-test are of interest, the most important predictors are those of the follow-up post-test. Specifically, the predictors of how individuals perform four to six months after the treatment. The most ideal candidates, therefore, the ones that will improve both UE amount of use and movement capability in the long-term are: 1) those that are younger, 2) have higher Fugl-Meyer motor scores, and 3) are able to actively extend their fingers. A regression model that includes all these predictors together would have to be performed to determine the most significant determinants of movement capability and amount of use of the affected UE. Additional considerations for future studies include investigating the predictors for other types of CIMT, when administered in differing intensities or duration. Another research avenue would be investigating the predictors of individuals who do not improve with therapy. Specifically, defining predictors that can identify both responders and non-responders would be advantageous to proper application of CIMT. Another direction for future research is establishing if the changes in outcomes represent a clinically

PAGE 102

88 meaningful difference. In other words, what amount of improvement is needed on the outcome measures for the change to have a difference clinically. The next step, then, would be determining the predictors for individuals who make a clinically important change with this therapy. These experiments provide the most comprehensive investigation of predictors of CIMT outcomes to date. Taken together, these studies provide a foundation for the formation of new inclusion criteria for CIMT studies. Utilizing these regression formulas can help to predict a potential participant’s ability to improve with CIMT.

PAGE 103

APPENDIX A PILOT DATA A pilot study was conducted specifically for the purposes of this investigation. This pilot work was performed in the Stroke Rehabilitation Laboratory at the University of Florida. In this section evidence is presented to establish the author’s competence demonstrating past work in the area of CIMT research with direct relevance to the study. The relevant preliminary studies were performed prior to this proposal. Pilot Study on CIMT: Functional Predictors of Outcomes for CIMT The purpose of pilot study #1 was to conduct the proposed project with a smaller group of participants to determine functional predictors of outcomes in CIMT and feasibility of performing this experiment. There were 110 predictors and five outcomes for each of the three time frames (time frames are outlined in proposed project). The aim and hypotheses are to follow. Aim: To determine the tests of body function, activity, and participation which predict outcomes following CIMT. Hypothesis 1. The test of body function and structure which are most predictive of functional improvements will include: shoulder flexor strength, pursuit rotor task-time on target, active extension of fingers from fully flexed position (item on Fugl-Meyer), and movements out of synergy (section on Fugl-Meyer). Hypothesis 2. The tests of activity that are most predictive are the Box and Block Test and the WMFT item 8, “picking up a can.” 89

PAGE 104

90 Hypothesis 3. The tests of participation that will be predictive of outcomes for CIMT are the physical items for upper extremity on the Stroke Impact Scale. Methods for Pilot Study 1 Eleven participants meeting the inclusion/exclusion criteria listed in the proposed project were involved in this study (please refer to Figure A-1 for details). Each participant received CIMT for two weeks as described in the proposed project. Descriptive Characteristics Gender6 females5 malesSide of Lesion8 left3 rightDominance8 Dominant hand affected3 Non-dominant affected6 high 5 low (met minimal motor criteria)(did not meet minimal motor criteria)Level of Function Time Since StrokeAgeMean2.5 years67.7Stand Dev1.8 years12.63Min11 months39Max6 years83 Figure A-1. Demographics of Participant in Pilot Study 1 and 2 Data Analyses for Pilot Study 1 The predictive relationships of various functional assessments (pre-test) on the traditional measures of CIMT (WMFT, AAUT, and MAL) were analyzed in the following manner. Categorical predictors were analyzed using a nonparametric test on equal medians (Wilcoxon). Continuous predictors were analyzed using univariate regression. Continuous predictors are operationally defined as any predictor with more than ten ratings, because they function well in a regression model. These analyses were performed for the changes in outcomes from preto posttesting, from pretesting to follow-up testing and post-testing to follow-up testing. Due to the limited sample size

PAGE 105

91 this pilot data was used to establish trends toward significant results. The p-value of 0.10 was not corrected for multiple comparisons because the goal of this pilot project was to establish trends but not to prove significance. The author acknowledges that this increases the chance of a type I error (finding significance when it is not really present). Results for Pilot Study 1 The results are presented in table format in sequence with the proposed hypotheses. The independent variables (predictors) are presented in the left-most column followed by the dependent variables (outcomes of CIMT) in each of the three time frames. The following abbreviations will aide in interpretation of the three tables of results (Figure A-2 and A-3). #AbbrevAmount of Use1 A OUQuality of Motion2QOMMotor Activity Logamount3MAL A Motor Activity Log how well4MALHWWolf Motor Function Test5WOLFActual Amount of Use Test (AAUT)Motor Activity Log (MAL)OUTCOMES Figure A-2. List of Outcomes and Abbreviations for Pilot Study 1 & 2 The first results table (Table A-4) is the body function and structure tests that are significantly (p< 0.10) predictive of the outcomes identified in the table. Note that positively associated outcomes are in yellow, and negatively associated outcomes are in red. The p-value is listed in parentheses immediately following the outcome that is significant. Non-significant results are denoted with ‘ns.’ The tables are designed to be

PAGE 106

92 read from left to right. For example: “The better the pre-test score on shoulder abduction strength, the greater the change in MALHW in time frame 1.” #Abbrev1FM25WFAS2fmrom26-40W1 W153fmpain41BBT4fmsen5fmflsyn42-71 A M1AM30Fugl-Meyer Extenstion Synerg y 6fmexsyn72-101HW1-HW307fmcom8fmoos9fmwri102SISM10fmhand103%11fmcoor104SISMem12fmfingx105SISem13PR106SISsp107SISsoc14pinch108SISUE15sflex109CS16sext110Fre17sabd18elbflex111gender 19elbext112timestroke20wristflex113age21wristext114sidestroke115domin22Wweight116level23Wgrip117neglect24TMS118-119apraxiaStrength TestingWolf Motor Function TestPREDICTORSWolf Motor Function TestShoulder Extensor StrengthShoulder Abduction StrengthFugl-Meyer Finger Extension Fugl-Meyer CoordinatiionFugl-Meyer Wrist Strength/ROMFugl-Meyer Out of SynergyFugl-Meyer Combinign SynergiesTests of Body Function and StructureTests of ActivityFugl-Meyer Pain scoreFugl-Meyer SensationFugl-Meyer Range of MotionFugl-Meyer Wolf Functional Ability Scale Wolf item 115Box and Block TestFugl-Meyer Flexor SynergyTranscranial Magnetic StimulationElbow Flexor StrengthWrist Flexion StrengthWrist Extension StrengthSide of DominanceLevel of FunctionNeglectApraxiaMotor Activity Log Amount Items 1 -30Motor Activity LogTests of ParticipationStroke Impact Scale SocialStroke Impact ScaleStroke Impact Scale Motor Motor Activity Log How Well Items 1 -30AgeStroke Impact Scale Percent RecoveryStroke Impact Scale MemoryStroke Impact Scale EmotionStroke Impact Scale SpeechFrenchayDescriptive PredictorsGenderTime since StrokeShoulder Flexor StrengthStroke Impact Scale Upper Extremity MotorCaregiver Strain Wolf grip strengthFugl-Meyer Hand ScoreElbow Extension StrengthWolf weight to boxSide of StrokePursuit Rotar (PR)Pinch Strengh Figure A-3. List of Predictors and Abbreviations for Pilot Study 1 & 2 Shoulder abduction strengththe the in MALHW (.034)the the in AOU (.084)the the in MALHW (.074)MALA (.048)MALHW (.068)WOLF (.082)Shoulder flexion strengththe the in AOU (.012)Elbow flexion strengththe the in WOLF (.082)Elbow extension strengththe the in AOU (.009)Wrist flexion strengththe the in MALA (.079)AOU (.010)QOM (.09)FM ROMthe the in MALHW (.10)FM coordinationthe the in WOLF (.073)FM out of synergythe the in WOLF (.032)FM flexion synergythe the in WOLF (.079)Pursuit Rotarthe the in WOLF (.072)the the in MALHW (.045)nsnsTests of Body Function and StructureIndependent Variable: The better the pre-test score on:Time frame 1Time frame 2Time frame 3pre to postpre to followpost to followthe the in FM-sensationnsnsAOU (.066)the the in Shoulder extension strengthnsnsnsnsnsnsnsthe the in nsnsnsnsnsnsnsnsns Figure A-4. Results Table for Tests of Body Function and Structure

PAGE 107

93 Figure A-5 is the tests of activity that are significantly (p< 0.10) predictive of the CIMT outcomes identified in the table. Figure A-5. Results Table for Tests of Activity ignificantly (p< 0.10) predictive of the CIMT outcomes identified in the table. Wolf item 2 (forearm to box)the the in MALHW (.045)the the in MALHW (.065)Wolf item 8 (lift can)the the in Wolf (.035)Wolf Item 9 (lift pencil)the the in Wolf (.068)Wolf item 11 (stack checkers) the the in MALA (.020)the the in MALHW(.058)Wolf item 12 (flip card)the the in Wolf (.035)Wolf item 13 (turn key in lock)the the in Wolf (.083)Wolf item 14 (fold towel)the the in Wolf (.033)Wolf Functional Ability Scorethe the in MALA (.067)the the inWolf (.063)the the in MALHW (.054)Box and Block Testthe the in MALA (.036)the the in MALHW (.052)MAL (How Well) item 2 ( openin g a drawer ) the the in MALHW (.077)MAL (How Well) item 3 ( pickin g up an item of clothin g) the the in MALHW (.052)MAL (Amount) item 8 ( openin g a door ) the the in MALHW (.090)QOM (.018) AOU (.060)MAL (How Well) item 21 ( shavin g /makeup ) the the in MALHW (.077)QOM (.070)AOU (.090)MAL (How Well) item 24 ( stabilizin g while standin g) the the in QOM (.090)MAL (How Well) item 25 (carrying objects)the the in AOU (.070)nsnsthe the in MAL (How Well) item 10 (washing hands)nsnsMAL (Amount) item 24 (stabilizing while standing)nsnsnsnsnsnsnsnsnsTests of Activitynsnsnsnsnsthe the in pre to postTime frame 1Independent Variable: The better the pre-test score on:nsnsnsnsnsnsTime frame 2pre to follownsnsnsnsnsTime frame 3post to follow Figure A-6. Results Table for Tests of Participation SIS motorthe the in WOLF (.076)AOU ( .025 ) QOM (.028)MALA (.076)SISSocial scorethe the in MALA (.098)SISMemorythe the in QOM (.045)AOU ( .015 ) QOM (.027)MALA ( .040 ) MALHW (.046)nsthe the in AOU (.078)nsnsnsnsThe better the pre-test score on:pre to postpre to followpost to followFrenchaynsnsthe the in SIS speechthe the in nsthe the in QOM (.018) Tests of Participation Independent Variable: Time frame 1Time frame 2Time frame 3 Figure A-6 is the tests of participation that are s

PAGE 108

94 Discu ssion for Pilot Study 1 This study discriminated between the influences of various predictive factors on the hich showed a predictive relationship. The outcomes were dividegram. o postalue was chosected nt re outcomes of CIMT, many of w d into three main time frames as discussed in the methods of the proposed project. These time frames are essential to establish who may improve most during the treatment phase and who maintains/improves functional gains four-months after the CIMT intervention.100 Previous CIMT studies suggest that functional gains are persistent,145,146 suggesting individuals maintain the improvements gained while in the training proAccording to the preliminary data, however, individuals who improve at the follow -up may be both functionally and descriptively different than individuals who show improvement immediately following CIMT. For example, the greater an individual’s shoulder abduction strength was on the pre-test, the better the change from preton the MAL, but, the poorer they perform from post-test to follow-up. The time framesin which people improve on outcomes are integral to the proposed study. Although there appear to be a number of significant results, despite the small sample size, the reader is reminded of the inflated alpha (<. 10). This p-v n to simply demonstrate trends in the data. In addition, the p-value is not correfor the multiple comparisons, meaning that there is an increased chance the significaresults are by chance. Furthermore, for this pilot data, none of the descriptive factors were controlled. For example, perhaps the differences noted in sensation could also be due to the participant’s age. Possibly, all the participants with high sensation scores weolder than those with lower sensation scores, and therefore, the differences found may have been explained by age rather than sensation. This comparison will be controlled in the proposed project.

PAGE 109

95 Tests of body function and structure The tests of body f unction and structure that are predictive of outcomes following ugl-Meyer, and the pursuit rotar. Specific strengs is nt se four id not have ation (including proprioception), passive range of motion, coordrds, the CIMT are strength, sub-sections of the F th findings warrant further discussion. First, five of the eight strength measures were predictive of outcomes. The only strength measures not predictive of outcomewrist extension, grip, and pinch strength. Interestingly, wrist extension is one componethat defines ‘minimal motor criteria.’26 Wolf states that in order to be functional with CIMT a certain amount of range is required in the wrist.26 This, however, does not appear to relate to wrist strength. Second, a ll the strength measures, which are positively correlated to outcomes, are predictive only in preto posttimeframe. Therefore, individuals that present with more strength initially do better immediately followingtherapy. On the other hand, at the follow-up, all strength measures are negatively correlated with outcomes. Therefore, at the four-month follow-up, individuals with greater strength during the pre-test do not improve as much on the follow-up as thowith less strength at the pre-test. Possibly the individuals with less strength need themonths to start building their strength, which could then translate into larger improvements at the follow-up. Another possibility is that the individuals with higher pre-test strength reached a ceiling on the outcome measures, therefore, they das much room to improve Many subsections of the Fugl-Meyer are predictive of outcomes following CIMT. These subsections are sens ination, movements out of synergy, and the movements of flexion synergy. Sensation (time-frame one), coordination (time-frame one), and out of synergy movements (time frame two) are negatively correlated with outcomes. In other wo

PAGE 110

96 better pre-score on these items, the less they improve on certain outcomes. Of nsensation is negatively correlated for time frame one and positively correlated at time frame three. This means that the higher the sensation pre-score, the poorer the outcomes at the post-test, but the better the outcome from the post-test to the follow-up. These results are intriguing considering the original experiments with forced-use were performed on deafferented monkeys.2 There is limited evidence that participants with sensory disorders can make notable improvements in motor and functional measuThe coordination score is also of interest. The poorer someone’s coordination thmore they improve on the Wolf from the preto post-test. One possible explanation for ote is that res.12 e this iser g that individual who have high pursuit rotar scores on the pretest demo. r the floor effect on the Wolf. If participants cannot perform a task of the Wolf, their score is counted as 120 seconds. So if they are unable to pick up a pencil at the pre-test (120) but are able to do so in 20 seconds on post-, they improve 100 seconds. Conversely, if someone is able to pick up a pencil in 10 seconds at pre-test, and improve to three seconds at post-test, they only improved by seven seconds. Therefore, lowfunctioning participants (possibly those with low coordination scores) are less likely to meet the ceiling. The pursuit rotar test is positively correlated with Wolf scores from the pre-test tofollow-up, meanin nstrate more improvement on the Wolf than do individuals who have low scores at the pre-test. The pursuit rotar is conventionally used as a test of closed-loop processingIn this situation it may be more of a motor test because it performed with the affected extremity, and individuals with stroke have difficulty performing the movement. A betteway to judge closed-loop processing would be to use the unaffected extremity score.

PAGE 111

97 This will be added to the proposed project. In addition, the Fitts tapping task will be added to the proposed project to assess the predictive value of an open-loop task. Tests of activity The tests of activity that are predictive of outcomes following CIMT are six individual Wolf it ems, the Wolf Functional Ability Score, the Box and Block Test, and eight item , orly outcomes across all three time frames, however, it is associated with three differ different MAL items. There are two specific results to discuss from the Wolfpredictor results. First, anytime an item is positively correlated with outcomes in time frame 1 (preto post-test), that same predictor is negatively correlated with that outcome in time frame 3 (postto follow). In individuals that score high on any of these Wolf items at pre-test demonstrate better change in the short term, but do not change as significantly as individuals with poor Wolf scores after leaving the therapy. Secondlyevery individual Wolf item that is predictive in the second time frame (from pretofollow-up) is only for the outcome Wolf (overall score). This suggests that people who score high on theses items improve more overall on the Wolf than those that score poon this items. Further analysis may warrant shortening the scale. If investigators know items 8,9,12,13,14 were predictive of CIMT success, they may be used for a screening tool. The predictor Wolf Functional Ability Score (WFAS) is the only score that is predictive of ent outcomes at each time frame. In time frame one and two, the WFAS is positively correlated with outcomes, but for time frame 3, it is negatively correlated withoutcomes. These findings further suggest that the greater improvement observed immediately, the less observed at the four-month follow-up.

PAGE 112

98 The last predictor to discuss in the ‘Tests of Activity’ section is the MAL. TMAL was used item-by-item to determine if any individual ite he ms were predictive of overated ore e o the ll outcomes. Half of the eight items that were predictive were negatively correlawith outcomes (all in time frame 2). Once again, the better an individual’s pre-test scon these four items, the less they improved from preto follow-up testing, suggest the presence of a ceiling effect. The only item that is predictive of outcomes in time frame one is item 10, washing hands (how well). This task is a true bilateral task, one of the few on the MAL. The outcomes that were positively correlated with this bilateral task included the QOM and AOU on the AAUT, a test that involves many bilateral tasks. Thlast point to discuss in this section is item 24, using the affected hand to stabilize while standing, essentially using the hand for support. This item was a significant predictor in both the ‘how well’ and the ‘amount’ components of the test. Worthy to note is that thisitem has been removed from the MAL, on more recent versions, and replaced with a more ‘appropriate item.’ This study warrants its inclusion. Of importance also is that forall the activity predictors, the only outcome that is significant in time frame two is theMAL-how well outcome. Furthermore, the “how well” outcome is always negatively correlated with the activity predictor. Therefore the better someone performs initially, the less they improve on the ‘how-well’ section of the MAL. This may suggest that individuals with less function make good candidates for CIMT. For the purposes of theproposed project, only the “how well” section of the MAL will be used. According tpilot results, the “how well” section had more predictive value. This will decrease the number of predictors, therefore, strengthening the power of the study.

PAGE 113

99 Tests of participation The tests of participation that are predictive of outcomes for CIM T are four sub-pact Scale (SIS) and the Frenchay Activities Index test. The SIS motor to he y the individual is with his/he is sections of the Stroke Im section is positively correlated with greater improvements on the Wolf from pre-post-test. This means the better an individual rates him/herself on the overall motor score (both upper and lower extremity scores) the more they improve on the Wolf. One of the most interesting outcomes is the SIS speech subsection predictor score. The better an individual’s speech, the better they improve from preto post(on three of the five outcome measures). This trend reverses at post-testing to follow-up, the poorer the speech initially, the more they improve after leaving therapy. This may be related toissues of hand-dominance and side of stroke, which are not controlled for in the preliminary data. The SIS scores also demonstrated that the more social someone is, tmore they improve on the amount score of the MAL from preto follow; possiblbecause they are putting themselves in more social situations and they are attempting to use the arm more often. Finally on the SIS, the better their memory on the pre-test, less they improve after leaving therapy. This is an interesting finding, which is difficult to explain. Findings in the proposed study may prove interesting. The only item that was predictive of four of the five tested outcomes in one time frame was the Frenchay Activities Index. The more independent an r instrumental activities of daily living (IADLs), as measured by the Frenchay Activities Index, the more they improved after leaving CIMT (postto follow-up). TheFrenchay Activities Index seems to be the strongest predictor of outcomes found in thpilot study. Worth mentioning is that the Frenchay Activities Index, however, was negatively correlated for one outcome (AOU) during time frame one. Less independent

PAGE 114

100 individuals improved more from preto post-testing. Caregiver support may explainthese differences. That said, the Caregiver Strain Index was not predictive of any outcomes for CIMT. Therefore, the Caregiver Strain Index will not be used as a predictor in the proposed project. Further trends include the results that the Wolf test is not a significant outcoany of the predictors in time frame 3. This could m me for ean that there is not a large enough changre to conduct the proposed project with a smaller group od feasib w f functional improvements. e in this outcome from postto follow-up testing across any of the predictors. This may be due to the small sample size used in the pilot study. In conclusion, although thewere significant associations between the predictors and the outcomes, this study should be performed with a larger sample size and with proper control of the variables. There is an urgent need to determine effective stroke rehabilitation that is appropriate to all stroke survivors’ individual level of function. Pilot Study on CIMT: Descriptive Predictors of Outcomes for CIMT The purpose of pilot study #2 was f participants to determine descriptive predictors of outcomes for CIMT an ility of performing this experiment. There were 9 predictors and 5 outcomes for each of the three time frames (outlined in the methods of the project). The followingwere the hypotheses as to which of the 9 predictors are positively associated with outcomes following CIMT. These were educated hypotheses developed from literaturereview and observation of CIMT participants. The aim and hypothesis are to folloAim. To determine descriptive factors that are predictive of functional outcomes for CIMT. Hypothesis 1. Dominance and level of function are the only descriptive factors predictive o

PAGE 115

101 Meth ods for Pilot Study 2 Eleven participants meeting the in clusion/exclusion criteria listed in the proposed study (please refer to Figure A-1 in pilot study one for detaile influence of the participant characteristics on the traditional outcomes T, and MAL) was analyzed in the following manner. Categere ng t column f CIMT (dependent variable) in each of the three time frames. ReferDiscunfluences of various descriptive predictors on the outcomes of CIMT. Many factors showed a predictive relationship with the outcomes for CIMT. As with pilot study 1, the results suggest that even with small project were involved in this s). Each participant received CIMT for 2 weeks as described in the proposed project. Data Analyses for Pilot Study 2 Th measures of CIMT (WMFT, AAU orical predictors were analyzed using an ANOVA. Continuous predictors wanalyzed using univariate regression. These analyses were performed for the changes inoutcomes from preto post-testing, from pre-testing to follow-up testing and post-testito follow-up testing. Due to the limited sample size this pilot data was used to determineif trends exist, therefore the p-value of 0.10 was not corrected for multiple comparisons. The author acknowledges that this increases the chance of a type I error. Results for Pilot Study 2 The predictors (independent variables) are presented in the left-mos followed by the outcomes o to experiment one for description of the time frames. Figure A-7 is the descriptive characteristics that are significantly (p< 0.10) predictive of the CIMT outcome identified in the table. ssion for Pilot Study 2 This study attempted to discriminate between the i

PAGE 116

102 Figure A-7. Results Table for Descriptive Predictors sample size significant results remain. Limitations exist with this assumption, please refer to experiment 1 for the limitations that also apply to pilot study 2. As in pilot 1, the outcomes for pilot study 2 were also divided into three main time frames as ts Table for Descriptive Predictors sample size significant results remain. Limitations exist with this assumption, please refer to experiment 1 for the limitations that also apply to pilot study 2. As in pilot 1, the outcomes for pilot study 2 were also divided into three main time frames as Females had in MALA (.055)than malesParticipants with affected dominant hands had in QOM (.083)than those with affected non-dominant handsAOU (.052) QOM (.035)than non-apraxicsnsQOM (.068)MALA (.083)The less items correct on the FAST-R (apraxia)nsnsthe the in The > the neglect scorethe the in AOU (.074)AOU (.041)QOM (.076)Participants with apraxia had in nsthan non-apraxics in the the in nsnsnsnsQOM (.047) The > the agethe the in AOU (.015)nsnsAOU (.097)post to followDescriptive PredictorsIndependent Variable: pre to followpre to post AOU (.069)QOM (.083)QOM (.088)nsthe the in the the in The > the time since strokerticipants with R brain stroke hadnsnsQOM (.029) in than L brain strokeTime frame 3Time frame 2Time frame 1 Pa described previously. These separate time frames aredescribed previously. These separate time frames are of extreme importance because rehabs l he primary screening criteria utilized for the majority of CIMT studies. If the of extreme importance because rehabs l he primary screening criteria utilized for the majority of CIMT studies. If the ilitation specialists must not only know who may improve the most during the treatment, but to know the predictors of who will maintain these functional gains after rehabilitation.100 Contrary to the hypothesis, eight of nine descriptive predictors were significant (<. 10). Level of function (high or low) was the only non-significant predictor. Thifinding is important because level of function (whether someone is able to meet minimamotor criteria ) is tilitation specialists must not only know who may improve the most during the treatment, but to know the predictors of who will maintain these functional gains after rehabilitation.100 Contrary to the hypothesis, eight of nine descriptive predictors were significant (<. 10). Level of function (high or low) was the only non-significant predictor. Thifinding is important because level of function (whether someone is able to meet minimamotor criteria ) is t se results hold true for proposed project, with a more substantial sample size, the findings could change criteria for who may be eligible for CIMT. se results hold true for proposed project, with a more substantial sample size, the findings could change criteria for who may be eligible for CIMT.

PAGE 117

103 The participants’ time since stroke was predictive of the AAUT (AOU and QOM) outcomes immediately following CIMT (time frame one). Interestingly, the greater the time since stroke the greater the improvement on the AAUT outcomes following CIMTIn other words, this data supports that those who have had a stroke . more recently will impros than re-ing the first testing period, therefore, younger participants do not have as mucha en ve less on amount and quality of spontaneous use. In the CIMT literature, no differences have been reported for time since stroke.2 Conversely, the correlation reversedirection at time frame three (change from post-testing to the follow-up). The greater the time since stroke the less improvement in the QOM portion of the AAUT test. The results suggest that chronic stroke participants can improve during the therapy more sub-acute; however, sub-acute participants seem to improve more by the four-month assessment, possibly suggesting increased ease at accessing plasticity in the sub-acute population. The age of the participant is predictive of outcomes during time frame two. This result suggests that the older a participant, the better they improve in AOU from the ptest to the follow-up test. Possibly the younger participants make more attempt to use their hand dur room to improve when tested at the follow-up. In addition, older individuals have more gradual progression in learning than do younger adults. They require more time to process the information; therefore, improvements in the older population may not be seimmediately following therapy, but at the follow-up.147,148 Furthermore, older adults may be experiencing more fatigue following the two weeks of therapy, resulting in poorer performances immediately after the therapy, but increased improvements at the follow-up.

PAGE 118

104 Another descriptive predictor that was found to be significant is side of stroke. Participants with right brain stroke had greater improvement on the AAUT outcome measures following the post-test (time frame one). Possibly this result is due to the factthat p eople with left brain stroke have more difficulty solving problems, are more easily angerthe o owing CIMT. Participants’ with affected dominls t demonstrated to be signif ed and frustrated, and have impaired retention of information.85 In the CIMT literature, no differences have been reported for side of stroke.149 This may be due to small sample sizes used in previous studies. In addition to age and side of stroke, the results from gender12 and dominance149 are not consistent with previous CIMT findings. In the third time fr a me (post-test tfollow-up) females improved more than males on the amount section of the MAL. Dominance is also predictive of outcomes foll ant hands had greater improvement in QOM from the post-test to the follow-u p test. This could be related to motivation, where individual with dominant side hemiparesis are more motivated to recover use of their dominant hand than individuawith non-dominant hemiparesis. Especially since the improvement is noted from the posto follow-up test. In other words individuals with dominant side hemiparesis may havecontinued to use their hemiparetic hand more after the therapy. Apraxia was used as a predictor in two different methods. A categorical analysis was performed to determine if the presence of apraxia would be a predictor. Secondly, the continuous score of number of items correct on the FAST-R was also used as a predictor. Both the categorical and continuous scores of apraxia icant predictors of outcomes. Individuals with apraxia had less improvements immediately following therapy. This trend reversed at the third time frame in which

PAGE 119

105 individuals with apraxia improved more than those without apraxia. Individuals witapraxia often need a lot or practice to show improvements,150 maybe two-weeks of practice is not enough, while the 4-months between the post-test and the follow-up alfor more practice and subsequent improvement in outcomes. If this holds true for a larsample size that has controlled variables the result could be very beneficial. Finally, the results demonstrated that the greater the h lows ger the latter is proved roke on. Conc riteria for participation needs to be carefully examined to determs is T. ne glect score, the greaterimprovement in outcomes. Individuals may be able to overcome the inattention due to neglect, secondary to the constant attention driven methods of CIMT. In a small trial, van der Lee found greater improvements in those with hemi-neglect.12 If the n accurate, neglect may be an important predictor of outcomes for CIMT. In conclusion, although there were many significant associations between the predictors and the outcomes, this study must be performed with a larger sample size anwith proper control of the variables. There is an urgent need to determine effective strehabilitation that is appropriate to all stroke survivors’ individual level of functi lusion CIMT is reported to significantly improves functional use of the upper extremity in20-25% of people with chronic stroke disability.145 Limited evidence exists, however, regarding the specific characteristics of individuals who benefit most from this intervention. Selection c ine who does benefit. Rehabilitation professionals continuously search for improved approaches for treatment, consequently, new treatment methods, such as CIMT, are often accepted before the relevance of the therapy to particular clientclearly understood. This needs to be understood, in order to target the correct population. The goal of this proposal is to determine who benefits most from CIM

PAGE 120

106 Although there were significant associations between the predictors and the outcomthe pilot study, this study needs to be performed with a larger sample size and with proper control of the variables. There is an urgent need to determine who are apcandidates for CIMT. es in propriate

PAGE 121

APPENDIX B BEHAVIORAL TESTS FORMS, LOGS, AND CONTRACTS WMFT UnaffectedCommentsAffected (R / L)CommentsFunctional Ability 1Forearm to table (side)0 1 2 3 4 52Forearm to box (side)0 1 2 3 4 53Extend elbow (side)0 1 2 3 4 54Extend elbow ( I lb.)0 1 2 3 4 55Hand to table (front)0 1 2 3 4 56Hand to box (front)0 1 2 3 4 57Weight to box (lbs.)0 1 2 3 4 58Reach and Retrieve0 1 2 3 4 59Lift can0 1 2 3 4 510Lift pencil0 1 2 3 4 511Lift paper clip0 1 2 3 4 512Stack checkers0 1 2 3 4 513Flip card0 1 2 3 4 514Grip Strength (kgs)0 1 2 3 4 515Turn key in lock0 1 2 3 4 516Fold towel0 1 2 3 4 517Lift basket0 1 2 3 4 5WOLF MOTOR FUNCTION TEST Figure B-1. Wolf Motor Function Test score sheet 107

PAGE 122

108 AmountHow WellComments1Turn light on2Open drawer3Pick up clothing4Pick up phone5Wipe counter6Open car door7Open refrigerator8Turn door knob9Operate TV remote10Wash hands11Dry hands12Put socks on13Take socks off14Put shoes on15Take shoes off16Get out of a chair17Pull the chair out18Pull the chair in19Pick up a glass20Brush teeth21Apply makeup, lotion, shaving cream22Use key to open door23Write on paper24Use hand to stabilize while standing25Carry object26Use fork27Use comb/ brush28Pick up cup with a handle29Button shirt30Eat finger foodsAverageAmount:How well:MOTOR ACTIVITY LOG 0.0Did not use m y weaker arm for the activit y ( not used ) 0. 5 1.0Occasionall y tried to use m y weaker arm for that activit y ( ver y rarel y) 1. 5 2.0Sometimes used m y weaker arm for that activit y but did most of the activit y with m y stron g er arm ( rarel y) 2. 5 3.0Used m y weaker arm for that activit y about half as much as before the stroke ( half p re-stroke ) 3. 5 4.0Used m y weaker arm for that activit y almost as much as before the stroke ( 3/4 or 75% p re-stroke ) 4. 5 5.0Used m y weaker arm for that activit y as much as before the stroke ( same as p re-stroke ) Motor Activity Log Amount Scale Figure B-2. Motor Activity Log score sheet

PAGE 123

109 biceps reflexbiceps, finger, and tricepstriceps reflextotaltotal0shoulder retractionshoulder elevationwrist stability, Elbow 90shoulder abductionwrist, flex/ext, elbow 90shoulder ERwrist stability elbow 0elbow flexionwrist flex/ext, elbow 0forearm supinationcircumductiontotaltotalshoulder adduction/IRelbow extensionfinger mass flexionforearm pronationfinger mass extensiontotalMCP ext, PIP and DIP flexedAdduct thumbopposition of thumb and index padshand to L-spinegrasp cylindershoulder flexion 0-90grasp sphericalelbow 90, pronation/supinationtotaltotalshoulder-abduction 0-90tremorflexion 90-180dysmetriaelbow 0, pron/supspeedtotaltotalTOTAL/661. REFLEX FUGL-MEYER UPPER-EXTREMITY MOTOR 5. OUT OF SYNERGY4. COMBINING SYNERGIES3. EXTENSOR SYNERGY2. FLEXOR SYNERGY9. COORDINATION8. HAND7. WRIST STABILITY6. NORMAL REFLEX ACTIVITY Figure B-3. Fugl-Meyer Upper Extremity Motor score sheet

PAGE 124

110 ActivityScoreCode1= Never2= Under once weekly3= 1-2 times/ week4 = Most days3Washing clothes4Light housework5Heavy housework1= Never6Local Shopping2= 1-2 times in three months7Social outings3= 3-12 times in three months8Walking outside > 15 mins4 = At least weekly9Actively pursuing hobby10Driving car/ travel on bus1= Never2= 1-2 times in six months3= 3-12 times in six months4 = At least twice weekly1= Never2= Light3= Moderate4= All necessary1= None2= One in 6 months3= Less than one a fortnight4= Over one in a fortnight (2 weeks)1= None2= Up to 10 hours/week3= 10-30 hours/ week4= Over 30 hours/ weekFRENCHAY ACTIVITIES INDEX12Preparing main mealsWashing upTravel outings/ car rides1112Gardening15Gainful workHousehold and/or car maintenance1314Reading books Figure B-4. Frenchay Activities Index score sheet

PAGE 125

111 SUBJECT ID:DATETIME ONTIME OFFACTIVITIES DONE DURING THIS PERIODHome Diary CIMT for Stroke RehabilitationFlorida Biomedical Grant Figure B-5. Home Diary

PAGE 126

112 SUBJECT ID:_____________________________DATETIME ACTIVITIES DONE DURING THIS PERIODDaily Activity Log CIMT for Stroke Rehabilitation Figure B-6. Daily Activity Log – General Form

PAGE 127

113 DAY # ______________ SUBJECT ID:___________Time of DayActivity DescriptionTrial 12345678910Time to Complete/ Number Completed CIMT for Stroke RehabilitationFlorida Biomedical Grant Figure B-7. Daily Activity Log – Timed Trial Form

PAGE 128

114 CIMT for Stroke Rehabilitation Florida Biomedical Grant BEHAVIORAL CONTRACT FORM I, ________________________________________, agree to wear the mitt on my good arm (i.e. the arm that was not affected by the stroke) as much as possible when I am away from the stroke rehabilitation lab during the 2 week training session. The purpose of the mitt is to remind me to use my weaker arm. I agree not to remove the mitt at any time or for any task for which I have agreed to wear it. An exception will be that I will NOT try to use my weaker arm alone if my safety could in any way be affected. Safety is the first consideration. I understand that this contract is to encourage me to participate in the therapy, and that if I am unable to follow this contract it can be adjusted at any time to fit my needs. AFFECTED ARM I agree to try and use ONLY my affected arm in all activities in which it is safe and possible to do so in the home and outside the home, including social situations. I will attempt to use my affected arm alone in all these activities, even if I had previously been using only my good arm for some of these tasks. The only activities for which I will NOT use my affected arm alone are: 1. Those which my safety could be affected 2. When a task is two-handed by nature 3. When I am using water These specific activities will be discussed with the trainer, but, in general, it is important to remember that safety and caution always need to be considered first before trying to perform a task with the affected arm alone. MITT ON GOOD ARM I agree to wear the mitt on my good or unaffected arm when I am way form the project as much as possible during the 2 weeks of training. I will wear the mitt for at least 90% of the time I am awake. The purpose of the mitt on my

PAGE 129

115 good arm is to prevent me from using the stronger arm, even if I have a strong urge or unconscious tendency to do so. ACTIVITIES IN WHICH I WILL USE MY AFFECTED ARM ONLY: I have agreed with my trainer that I will make a strong effort to use my affected arm as much as possible during the activities listed below. I also agree to wear my mitt on my good arm for these activities. The approximate times when I think these activities are most likely to occur are listed. I will start wearing my mitt when I wake up at ______a.m. A.M. ACTIVITES USING AFFECTED ARM ONLY TIMES P.M. ACTIVITES USING AFFECTED ARM ONLY TIMES

PAGE 130

116 ACTIVITIES IN WHICH I WILL USE BOTH ARMS My trainer and I have also agreed that I will use both arms in the following tasks, either because: 1. Safety would be a problem if I just used my affected arm 2. These tasks are, by their nature, two-handed ACTIVITES USING BOTH ARMS TIMES OTHER ACTIVITIES FOR WHICH I CAN REMOVE MY MITT After discussion with my trainer, I understand that I can remove my mitt for the activities listed below. These activities also include sleep, naps, and any task involving water. The approximate times at which I will carry out these activities are also listed. TIME MITT OFF GOOD ARM ONLY ACTIVITIES TIME MITT ON I agree to abide by the above terms to the best of my ability in all situations while I am away from the stroke rehabilitation lab. _____________________ __________ ________________________ Signature of Patient Date Signature of Trainer Figure B-8. Daily Activity Log – Behavioral Contract Form

PAGE 131

117 LIST OF REFERENCES 1. Morris D, Crago J, DeLuca S, Pidikiti R, Taub E. Constraint-Induced (CI) Movement Therapy for motor recovery after stroke. Neuror ehabilitation 1997;9:2943. 2. Taub E, Uswatte G, Pidikiti R. Constr aint-Induced Movement Therapy: a new family of techniques with broad applicati on to physical rehabilitation--a clinical review. J Rehabil Res Dev 1999;36(3):237-51. 3. American Stroke Association. ASA website. www.strokeassociation.org , Accessed 9-15-2003. 4. Stineman M, Maislin G, Fiedler R, Gra nger C. A prediction model for functional recovery in stroke. Stroke 1997;28(3):550-6. 5. Kwakkel G, Kollen B, Van Der Grond J, Prevo A. Probability of Regaining Dexterity in the Flaccid Upper Limb. Impact of Severity of Paresis and Time Since Onset in Acute Stroke. Stroke 2003. 6. Taub E, Crago J, Uswatte G. Constrai nt-induced movement therapy: a new approach to treatment in physical re habilitation. Rehabilitation Psychology 1998;43(2):152-170. 7. Hsieh CL, Sheu CF, Hsueh IP, Wang CH. Trunk control as an early predictor of comprehensive activities of daily living function in stroke patients. Stroke 2002;33(11):2626-30. 8. Gladman JR, Harwood DM, Barer DH. Pred icting the outcome of acute stroke: prospective evaluation of five multivaria te models and comparison with simple methods. J Neurol Neurosurg Psychiatry 1992;55(5):347-51. 9. Stone SP, Allder SJ, Gladman JR. Predicti ng outcome in acute stroke. Br Med Bull 2000;56(2):486-94. 10. Taub E, Crago JE, Burgio LD, Groomes TE, Cook EW, DeLuca SC, Miller NE. An operant approach to rehabilitation medicine: overcoming learned nonuse by shaping. J Exp Anal Behav 1994;61(2):281-93.

PAGE 132

118 11. Wolf S, Lecraw D, Barton L. Forced use in hemiplegic upper extremities to reserve the effect of learned nonuse among chronic stroke and head-injured patients. Experimental Neurology 1989;104:125-132. 12. van der Lee JH, Wagenaar RC, Lankhorst GJ, Vogelaar TW, Deville WL, Bouter LM. Forced use of the upper extremity in chronic stroke patients: results from a single-blind randomized clinical trial. Stroke 1999;30(11):2369-75. 13. Kunkel A, Kopp B, Muller G, Villringer K, Villringer A, Taub E, Flor H. Constraint-induced movement therapy for motor recovery in chronic stroke patients. Arch Phys Med Rehabil 1999;80(6):624-8. 14. Blanton S, Wolf S. An application of upper-extremity constraint-induced movement therapy in a patient with subacute stroke. Phys Ther 1999;79(9):847-53. 15. Miltner W, Bauder H, Sommer M, Dettmers C, Taub E. Effects of constraint-induced movement therapy on patients with chronic motor deficits after stroke: a replication. Stroke 1999;30(3):586-92. 16. Taub E, Miller N, Novack T, Cook E, Fleming W, Nepomuceno C, Connell J, Crago J. Technique to improve chronic motor deficit after stroke. Arch Phys Med Rehabil 1993;74(4):347-54. 17. Taub E, Wolf S. Constraint induced movement techniques to facilitate upper extremity use in stroke patients. Topics in Rehabil 1997;3(4):38-61. 18. Wolf SL, Blanton S, Baer H, Breshears J, Butler AJ. Repetitive task practice: a critical review of constraint-induced movement therapy in stroke. Neurolog 2002;8(6):325-38. 19. Dromerick A, Edwards D, Hahn M. Does the application of constraint-induced movement therapy during acute rehabilitation reduce arm impairment after ischemic stroke? Stroke 2000;31(12):2984-8. 20. Liepert J, Miltner W, Bauder H, Sommer M, Dettmers C, Taub E, Weiller C. Motor cortex plasticity during constraint-induced movement therapy in stroke patients. Neuroscience Letters 1998;250(1):5-8. 21. Liepert J, Uhde I, Graf S, Leidner O, Weiller C. Motor cortex plasticity during forced-use therapy in stroke patients: a preliminary study. J Neurol 2001;248(4):315-21. 22. Russo S. Hemiplegic upper extremity rehabilitation: A review of the forced-use paradigm. Neurology Report 1995;19:17-21. 23. Sabari JS, Kane L, Flanagan SR, Steinberg A. Constraint-induced motor relearning after stroke: a naturalistic case report. Arch Phys Med Rehabil 2001;82(4):524-8.

PAGE 133

119 24. Schaechter JD, Kraft E, Hilliard TS, Dijkhuizen RM, Benner T, Finklestein SP, Rosen BR, Cramer SC. Motor recovery and cortical reorganization after constraint-induced movement therapy in stroke patients: a preliminary study. Neurorehabil Neural Repair 2002;16(4):326-38. 25. Taub E, Morris D. Constraint-induced movement therapy to enhance recovery after stroke. Curr Atheroscler Rep 2001;3(4):279-86. 26. Wolf SL, Binder-MacLeod SA. Electromyographic biofeedback applications to the hemiplegic patient. Changes in upper extremity neuromuscular and functional status. Phys Ther 1983;63(9):1393-403. 27. Wolf S. Extremity constraint induced therapy evaluation (EXCITE). National Institutes of Health. 28. Ogden R, Franz S. On cerebral motor control: The recovery from experimentally produced hemiplegia. Psychobiology 1917;1:33-47. 29. Gresham G, Duncan PW, Stason W. Post Stroke Rehabilitation. Clinical Practice Guideline #6. Rockville, MD: US Department of Health and Human Services, Public Health Service, Agency for health care policy and research, 1995. 30. Friel KM, Heddings AA, Nudo RJ. Effects of postlesion experience on behavioral recovery and neurophysiologic reorganization after cortical injury in primates. Neurorehabilitation and Neural Repair 2000;14(3):187-198. 31. Friel KM, Nudo RJ. Recovery of motor function after focal cortical injury in primates: compensatory movement patterns used during rehabilitative training. Somatosens Mot Res 1998;15(3):173-89. 32. Nudo RJ, Wise BM, Sifuentes F, Milliken GW. Neural substrates for the effects of rehabilitative training on motor recovery after ischemic infarct. Science 1996;272:1791-1794. 33. Liepert J, Bauder H, Miltner WHR, Taub E, Weiller C. Treatment-induced cortical reorganization after stroke in humans. Stroke 2000;31(6):1210-1216. 34. Liepert J, Miltner WHR, Bauder H, Sommer M, Dettmers C, Taub E, Weiller C. Motor cortex plasticity during constraint-induced movement therapy in stroke patients. Neuroscience Letters 1998;250(1):5-8. 35. www3.who.int/icf. International Classification of Functioning (ICF), Disability and Health. World Health Organization. Vol. 2002, 2001. 36. Duncan P. Synthesis of intervention trials to improve motor recovery following stroke. Topics in Stroke Rehabilitation 1997;3:1-20.

PAGE 134

120 37. Light K, Principal investigator. Florida Biomedical Research Program, Grant No. BM042, Constraint Induced Movement Therapy for Stroke Rehabilitation. 38. Kwakkel G, Kollen B, Wagenaar R. Therapy impact on functional recovery in stroke rehabilitation: a critical review of the literature. Physiotherapy 1999;13:457-470. 39. Taub E. Movement in nonhuman primates deprived of somatosensory feedback. Exercise and Sport Science Reviews, Santa Barbara: Journal Publishing Affiliates 1977:335-374. 40. Knapp H, Taub E, Berman A. Movements in monkeys with deafferented forearms. Experimental Neurology 1963;7:305-315. 41. Taub E, Perrella PN, Miller EA, Barro G. Diminution of early environmental control through perinatal and prenatal somatosensory deafferentation. Biol Psychiatry 1975;10(6):609-26. 42. Light K, Principal investigator. Florida Biomedical Research Program, Grant No. BM042, Constraint Induced Movement Therapy for Stroke Rehabilitation. 2001. 43. Wolf S. Extremity constraint induced therapy evaluation (EXCITE). National Institutes of Health, 2000. 44. Taub E CJ, Uswatte G. Constraint-induced movement therapy: a new approach to treatment in physical rehabilitation. Rehabilitation Psychology 1998;43(2):152-170. 45. Duncan P, Badke M. Stroke rehabilitation: The recovery of motor control. Chicago: Year Book Medical Publishers, 1988. 46. Sanford J, Moreland J, Swanson LR, Stratford PW, Gowland C. Reliability of the Fugl-Meyer assessment for testing motor performance in patients following stroke. Physical Therapy 1993;73(7):447-454. 47. Duncan P, Studenski S, Richards L, Gollub S, Lai SM, Reker D, Perera S, Yates J, Koch V, Rigler S, Johnson D. Randomized clinical trial of therapeutic exercise in subacute stroke. Stroke 2003;34(9):2173-80. 48. Richards LG, Olson B, Palmiter-Thomas P. How forearm position affects grip strength. Am J Occup Ther 1996;50(2):133-8. 49. Stevens JA, Stoykov ME. Using motor imagery in the rehabilitation of hemiparesis. Arch Phys Med Rehabil 2003;84(7):1090-2. 50. Prigatano GP, Wong JL. Speed of finger tapping and goal attainment after unilateral cerebral vascular accident. Arch Phys Med Rehabil 1997;78(8):847-52.

PAGE 135

121 51. Rantanen T, Guralnik JM, Foley D, Masaki K, Leveille S, Curb JD, White L. Midlife hand grip strength as a predictor of old age disability. Jama 1999;281(6):558-60. 52. Rantanen T, Harris T, Leveille SG, Visser M, Foley D, Masaki K, Guralnik JM. Muscle strength and body mass index as long-term predictors of mortality in initially healthy men. J Gerontol A Biol Sci Med Sci 2000;55(3):M168-73. 53. Syddall H, Cooper C, Martin F, Briggs R, Aihie Sayer A. Is grip strength a useful single marker of frailty? Age Ageing 2003;32(6):650-6. 54. Rantanen T, Volpato S, Ferrucci L, Heikkinen E, Fried LP, Guralnik JM. Handgrip strength and cause-specific and total mortality in older disabled women: exploring the mechanism. J Am Geriatr Soc 2003;51(5):636-41. 55. Giampaoli S, Ferrucci L, Cecchi F, Lo Noce C, Poce A, Dima F, Santaquilani A, Vescio MF, Menotti A. Hand-grip strength predicts incident disability in non-disabled older men. Age Ageing 1999;28(3):283-8. 56. Rantanen T. Muscle strength, disability and mortality. Scand J Med Sci Sports 2003;13(1):3-8. 57. Heller A, Wade DT, Wood VA, Sunderland A, Hewer RL, Ward E. Arm function after stroke: measurement and recovery over the first three months. J Neurol Neurosurg Psychiatry 1987;50(6):714-9. 58. 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(6):521-4. 59. Sunderland A, Tinson D, Bradley L, Hewer RL. Arm function after stroke. An evaluation of grip strength as a measure of recovery and a prognostic indicator. J Neurol Neurosurg Psychiatry 1989;52(11):1267-72. 60. Bassey EJ, Harries UJ. Normal values for handgrip strength in 920 men and women aged over 65 years, and longitudinal changes over 4 years in 620 survivors. Clin Sci (Lond) 1993;84(3):331-7. 61. Laukkanen P, Heikkinen E, Kauppinen M. Muscle strength and mobility as predictors of survival in 75-84-year-old people. Age Ageing 1995;24(6):468-73. 62. Bohannon R. Make tests and proper muscle strength measurement. Stroke 2000;81(10):1442-1443. 63. O'Sullivan S, Schmitz T. Physical Rehabilitation, Assessment and Treatment 4th Ed. Philadelphia: FA Davis Company, 2001.

PAGE 136

122 64. 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(8):357-64. 65. Bonifer N, Anderson KM. Application of constraint-induced movement therapy for an individual with severe chronic upper-extremity hemiplegia. Phys Ther 2003;83(4):384-98. 66. van der Lee JH, Beckerman H, Lankhorst GJ, Bouter LM. Constraint-induced movement therapy. Arch Phys Med Rehabil 1999;80(12):1606-7. 67. Page S, Sisto S, Levine P, Johnston M, Hughes M. Modified constraint induced therapy: a randomized feasibility and efficacy study. J Rehabil Res Dev 2001;38(5):583-90. 68. Page SJ, Elovic E, Levine P, Sisto SA. Modified constraint-induced therapy and botulinum toxin A: a promising combination. Am J Phys Med Rehabil 2003;82(1):76-80. 69. Page SJ, Sisto SA, Levine P. Modified constraint-induced therapy in chronic stroke. Am J Phys Med Rehabil 2002;81(11):870-5. 70. Page SJ, Sisto S, Johnston MV, Levine P, Hughes M. Modified constraint-induced therapy in subacute stroke: a case report. Arch Phys Med Rehabil 2002;83(2):286-90. 71. Chae J, Johnston M, Kim H, Zorowitz R. Admission motor impairment as a predictor of physical disability after stroke rehabilitation. Am J Phys Med Rehabil 1995;74(3):218-23. 72. Holbrook M, Skillbeck CE. An activities index for use with stroke patients. Age and Ageing 1983;12(2):166-170. 73. Schuling J, de Haan R, Limburg M, Groenier KH. The Frenchay Activities Index. Assessment of functional status in stroke patients. Stroke 1993;24(8):1173-7. 74. Carod-Artal FJ, Gonzalez-Gutierrez JL, Herrero JA, Horan T, De Seijas EV. Functional recovery and instrumental activities of daily living: follow-up 1-year after treatment in a stroke unit. Brain Inj 2002;16(3):207-16. 75. Indredavik B, Bakke F, Slordahl SA, Rokseth R, Haheim LL. Stroke unit treatment improves long-term quality of life: a randomized controlled trial. Stroke 1998;29(5):895-9. 76. Patel MD, Coshall C, Rudd AG, Wolfe CD. Cognitive impairment after stroke: clinical determinants and its associations with long-term stroke outcomes. J Am Geriatr Soc 2002;50(4):700-6.

PAGE 137

123 77. Buck D, Jacoby A, Massey A, Ford G. Evaluation of measures used to assess quality of life after stroke. Stroke 2000;31(8):2004-10. 78. Dorman PJ, Waddell F, Slattery J, Dennis M, Sandercock P. Is the EuroQol a valid measure of health-related quality of life after stroke? Stroke 1997;28(10):1876-82. 79. Pedersen PM, Jorgensen HS, Nakayama H, Raaschou HO, Olsen TS. Orientation in the acute and chronic stroke patient: impact on ADL and social activities. The Copenhagen Stroke Study. Arch Phys Med Rehabil 1996;77(4):336-9. 80. Rodgers H, Mackintosh J, Price C, Wood R, McNamee P, Fearon T, Marritt A, Curless R. Does an early increased-intensity interdisciplinary upper limb therapy programme following acute stroke improve outcome? Clin Rehabil 2003;17(6):579-89. 81. Mant J, Carter J, Wade DT, Winner S. Family support for stroke: a randomised controlled trial. Lancet 2000;356(9232):808-13. 82. Piercy M, Carter J, Mant J, Wade DT. Inter-rater reliability of the Frenchay activities index in patients with stroke and their careers. Clin Rehabil 2000;14(4):433-40. 83. Stucki G, Cieza A, Ewert T, Kostanjsek N, Chatterji S, Bedirhan Ustun T. Application of the International Classification of Functioning, Disability and Health (ICF) in clinical practice. Disability and Rehabilitation 2002;24(5). 84. Ornstein R. The Right Mind, Making Sense of the Hemispheres. New York: Harcourt Brace & Company, 1997. 85. Umphred D. Neurological Rehabilitation. St Louis: Mosby, 1995. 86. Heilman KM, Rothi LJG. Apraxia. Cinical Neuropsychology. New York: Oxford University Press, 1993. 87. Rothi L, Heilman K. Apraxia : the neuropsychology of action. Hove: Psychology Press, 1997. 88. Poizner H, Mack L, Verfaellie M, Rothi L, Heilman K. Three-dimensional computergraphic analysis of apraxia. Neural representations of learned movement. Brain Cognit 1990;113(Pt 1):85-101. 89. Heilman KM, Watson RT, Valenstein E. Neglect and related disorders. Clinical Neuropsychology. New York: Ofxord University Press, 1993;279-386. 90. Heilman KM, Schwartz HD, Geschwind N. Defective motor learning in ideomotor apraxia. Neurology 1975;25(11):1018-20.

PAGE 138

124 91. Plummer P, Morris ME, Dunai J. Assessment of unilateral neglect. Phys Ther 2003;83(8):732-40. 92. Twitchell TE. The restoration of motor function following hemiplegia in man. Brain 1951;74(4):443-80. 93. Nakayama H. Recovery of upper extremity function in stroke patients: the Copenhagen Stroke Study. Arch Phys Med Rehabil 1995;76:27-32. 94. Loewen SC, Anderson BA. Predictors of stroke outcome using objective measurement scales. Stroke 1990;21(1):78-81. 95. Duncan PW, Goldstein LB, Matchar D, Divine GW, Feussner J. Measurement of motor recovery after stroke. Outcome assessment and sample size requirements. Stroke 1992;23(8):1084-9. 96. Hellige J. Hemispheric Asymmetry: What's Right and What's Left? Cambridge, Massachusetts: Harvard University Press, 1993. 97. Ward NS, Frackowiak RS. Age-related changes in the neural correlates of motor performance. Brain 2003;126(Pt 4):873-88. 98. Kugler C, Altenhoner T, Lochner P, Ferbert A. Does age influence early recovery from ischemic stroke? A study from the Hessian Stroke Data Bank. J Neurol 2003;250(6):676-81. 99. Hankey GJ. Long-term outcome after ischaemic stroke/transient ischaemic attack. Cerebrovasc Dis 2003;16 Suppl 1:14-9. 100. Bagg S, Pombo AP, Hopman W. Effect of age on functional outcomes after stroke rehabilitation. Stroke 2002;33(1):179-85. 101. Lehmann JF, DeLateur BJ, Fowler RS, Warren CG, Arnhold R, Schertzer G, Hurka R, Whitmore JJ, Masock AJ, Chambers KH. Stroke rehabilitation: Outcome and prediction. Arch Phys Med Rehabil 1975;56(9):383-9. 102. Jongbloed L. Prediction of function after stroke: a critical review. Stroke 1986;17(4):765-76. 103. Kalra L. The influence of stroke unit rehabilitation on functional recovery from stroke. Stroke 1994;25(4):821-5. 104. Nakayama H, Jorgensen HS, Raaschou HO, Olsen TS. The influence of age on stroke outcome. The Copenhagen Stroke Study. Stroke 1994;25(4):808-13. 105. Kalra L. Does age affect benefits of stroke unit rehabilitation? Stroke 1994;25(2):346-51.

PAGE 139

125 106. Garraway W, Whisnant J, Drury I. The changing pattern of survival following stroke. Stroke 1983;14:699. 107. Di Carlo A, Lamassa M, Baldereschi M, Pracucci G, Basile AM, Wolfe CD, Giroud M, Rudd A, Ghetti A, Inzitari D. Sex differences in the clinical presentation, resource use, and 3-month outcome of acute stroke in Europe: data from a multicenter multinational hospital-based registry. Stroke 2003;34(5):1114-9. 108. Wyller TB, Sodring KM, Sveen U, Ljunggren AE, Bautz-Holter E. Are there gender differences in functional outcome after stroke? Clin Rehabil 1997;11(2):171-9. 109. Paradiso S, Robinson RG. Gender differences in poststroke depression. J Neuropsychiatry Clin Neurosci 1998;10(1):41-7. 110. Twigg A, Cifu D, Keyser-Marcus L, Swartz Z. The association between gender, race and marital status on functional outcome at rehabilitation discharge after thromboembolic stroke: a prospective analysis. Neurorehabilitation 1988;11(3). 111. Patel AT, Duncan PW, Lai SM, Studenski S. The relation between impairments and functional outcomes poststroke. Arch Phys Med Rehabil 2000;81(10):1357-63. 112. Duncan PW, Goldstein LB, Horner RD, Landsman PB, Samsa GP, Matchar DB. Similar motor recovery of upper and lower extremities after stroke. Stroke 1994;25(6):1181-8. 113. Counsell C, Dennis M, McDowall M, Warlow C. Predicting outcome after acute and subacute stroke: development and validation of new prognostic models. Stroke 2002;33(4):1041-7. 114. Baird AE, Dambrosia J, Janket S, Eichbaum Q, Chaves C, Silver B, Barber PA, Parsons M, Darby D, Davis S, Caplan LR, Edelman RE, Warach S. A three-item scale for the early prediction of stroke recovery. Lancet 2001;357(9274):2095-9. 115. Wolf SL, Barton LA. Forced use in hemiplegic upper extremities to reserve the effect of learned nonuse among chronic stroke and head-injured patients. Experimental Neurology 1989;104:125-132. 116. Gimenez-Roldan S, Novillo M, Navarro E, Dobato J, Gimenez-Zuccarelli M. Mini-mental state examination: proposal of protocol to be used. Rev Neurol 1997;25(140):576-83. 117. Blanton S, Wolf S. An application of upper-extremity constraint-induced movement therapy in a patient with subacute stroke. Phys Ther. Vol. 79, 1999;847-53. 118. van der Lee JH, Beckerman H, Lankhorst GJ, Bouter LM. Constraint-induced movement therapy. Phys Ther. Vol. 80, 2000;711-3.

PAGE 140

126 119. Kunkel A, Kopp B, Muller G, Villringer K, Villringer A, Taub E, Flor H. Constraint-induced movement therapy for motor recovery in chronic stroke patients. Arch Phys Med Rehabil. Vol. 80, 1999;624-8. 120. Levy CE, Nichols DS, Schmalbrock PM, Keller P, Chakeres DW. Functional MRI evidence of cortical reorganization in upper-limb stroke hemiplegia treated with constraint-induced movement therapy. Am J Phys Med Rehabil. Vol. 80, 2001;4-12. 121. Miltner W, Bauder H, Sommer M, Dettmers C, Taub E. Effects of constraint-induced movement therapy on patients with chronic motor deficits after stroke: a replication. Stroke. Vol. 30, 1999;586-92. 122. Morris D, Crago J, DeLuca S, Pidikiti R, Taub E. Constraint-Induced (CI) Movement Therapy for motor recovery after stroke. Neurorehabilitation. Vol. 9, 1997;29-43. 123. Boyd RN, Morris ME, Graham HK. Management of upper limb dysfunction in children with cerebral palsy: a systematic review. Eur J Neurol 2001;8 Suppl 5:150-66. 124. Sterr A, Elbert T, Berthold I, Kolbel S, Rockstroh B, Taub E. Longer versus shorter daily constraint-induced movement therapy of chronic hemiparesis: an exploratory study. Arch Phys Med Rehabil. Vol. 83, 2002;1374-7. 125. Taub E, Crago J, Uswatte G. Constraint-induced movement therapy: a new approach to treatment in physical rehabilitation. Rehabilitation Psychology. Vol. 43, 1998;152-170. 126. Wolf SL, Blanton S, Baer H, Breshears J, Butler AJ. Repetitive task practice: a critical review of constraint-induced movement therapy in stroke. Neurolog. Vol. 8, 2002;325-38. 127. Jebson R. An objective and standardized test of hand function. Arch Phys Med Rehabil. Vol. 50:311-319, 1969;311-319. 128. Morris DM, Uswatte G, Crago JE, Cook EW, 3rd, Taub E. The reliability of the wolf motor function test for assessing upper extremity function after stroke. Arch Phys Med Rehabil. Vol. 82, 2001;750-5. 129. Hollis S, Campbell F. What is meant by intention to treat analysis? Survey of published randomised controlled trials. Bmj 1999;319(7211):670-4. 130. Gibaldi M, Sullivan S. Intention-to-treat analysis in randomized trials: who gets counted? J Clin Pharmacol 1997;37(8):667-72.

PAGE 141

127 131. Hanger HC, Whitewood P, Brown G, Ball MC, Harper J, Cox R, Sainsbury R. A randomized controlled trial of strapping to prevent post-stroke shoulder pain. Clin Rehabil 2000;14(4):370-80. 132. Evans A, Harraf F, Donaldson N, Kalra L. Randomized controlled study of stroke unit care versus stroke team care in different stroke subtypes. Stroke 2002;33(2):449-55. 133. Neter J, Kutner M, Nachtsheim C, Wasserman W. Applied Linear Statistical Models. 4th ed. Chicago, 1996. 134. Buzas J. Fast estimators of the jackknife. American Statistician 1997;51(3):235-240. 135. Pigeot I. The jackknife and bootstrap in biomedical research Common principles and possible pitfalls. Drug Informational Journal 2001;35(4):1431-1443. 136. Hurst H, Bolton J. Assessing the clinical significance of change scores recorded on subjective outcome measures. J Manipulative Physiol Ther 2004;27(1):26-35. 137. Stucki G, Daltroy L, Katz JN, Johannesson M, Liang MH. Interpretation of change scores in ordinal clinical scales and health status measures: the whole may not equal the sum of the parts. J Clin Epidemiol 1996;49(7):711-7. 138. Allison DB, Paultre F, Goran MI, Poehlman ET, Heymsfield SB. Statistical considerations regarding the use of ratios to adjust data. Int J Obes Relat Metab Disord 1995;19(9):644-52. 139. Thompson B. Stepwise regression and stepwise discriminant analysis need not apply here: a guidelines editorial. Educational and Psychological Measurement 1995;55(4):525-534. 140. Desrosiers J, Noreau L, Rochette A, Bravo G, Boutin C. Predictors of handicap situations following post-stroke rehabilitation. Disabil Rehabil 2002;24(15):774-85. 141. Mackenzie AE, Chang AM. Predictors of quality of life following stroke. Disabil Rehabil 2002;24(5):259-65. 142. Jonsson N. Online stepwise regression discussions. www.cognigencorp.com/nonmem/nm/99nov212000.html , accessed 5-30-04, 2000. 143. Taub E, Morris DM. Constraint-induced movement therapy to enhance recovery after stroke. Curr Atheroscler Rep 2001;3(4):279-86. 144. Craik R. Recovery Processes: Maximizing Function. In: MJ L, ed. Contemporary Management of Motor Control Problems: Proceedings of the II STEP Conference. VA: Bookcrafters, 1991.

PAGE 142

128 145. Morris D, Crago, J., DeLuca, S., Pidikiti, R., & Taub E. Constraint-Induced (CI) Movement Therapy for motor recovery after stroke. Neurorehabilitation 1997;9:29-43. 146. Taub E, Miller NE, Novack TA, Cook EW, 3rd, Fleming WC, Nepomuceno CS, Connell JS, Crago JE. Technique to improve chronic motor deficit after stroke. Arch Phys Med Rehabil 1993;74(4):347-54. 147. Light KE, Reilly MA, Behrman AL, Spirduso WW. Greater benefits of practice on reaction time in older versus younger participants. Journal of Aging and Physical Activity 1996;4(1):27-41. 148. Light KE, Spirduso WW. Effects of adult aging on the movement complexity factor of response programming. J Gerontol 1990;45(3):P107-9. 149. Miltner WH, Bauder H, Sommer M, Dettmers C, Taub E. Effects of constraint-induced movement therapy on patients with chronic motor deficits after stroke: a replication. Stroke 1999;30(3):586-92. 150. Goldenberg G, Daumuller M, Hagmann S. Assessment and therapy of complex activities of daily living in apraxia. Neuropsychological Rehabilitation 2001;11(2):147-169.

PAGE 143

BIOGRAPHICAL SKETCH After receiving my Bachelor of Health Science and Master of Science in physical therapy from the University of Kentucky in 1997, I began a rewarding career as a physical therapist. At the outset, I worked as a traveling therapist in a variety of settings including acute care, rehabilitation, outpatient and home care. Through these diverse experiences I became acutely aware of the health care spectrum and of the importance in determining which patients would benefit most from physical therapy. Furthermore, I questioned the theory and research underlying many of the interventions we utilize. It was these questions that led me to the University of Florida in 1999, where I pursued a doctoral degree in rehabilitation science. Within the Rehabilitation Science Program, my major area of concentration was movement dysfunction in neurological disorders and my minor concentration was gerontology. In my five years at the University of Florida, I was involved in numerous neurologically based research projects, particularly in the area of stroke rehabilitation. I worked extensively with Dr. Kathye E. Light on a series of Constraint-Induced Movement Therapy (CIMT) studies designed to improve hand and arm function in stroke survivors. As a pre-doctoral fellow at the VA Brain Rehabilitation Research Center in Gainesville, Florida, I investigated predictors of outcomes for CIMT. Research is an integral part of my profession, and my short term goals are to continue advancing my skills of becoming a better researcher and to seek more opportunities of presenting and 129

PAGE 144

130 publishing clear, scientifically-based, clinically relevant research. The area of research I plan to pursue is therapeutic interventions for neurologically impaired patients. I believe rehabilitation will continue to improve and advance when supported by scientifically based research. As an educator, I plan to instill this philosophy in my students; as a researcher, I plan to pursue this philosophy to progress rehabilitation science. My experience thus far as a teaching assistant, researcher, and clinician, along with my future goals is evidence of my commitment and intent to contribute to the science and advancement or rehabilitation literature and knowledge. In August 2004, I will join the physical therapy faculty at the University of South Carolina in Columbia.