The Battery of Rehabilitation Assessments and INterventions (BRAIN): A Case of Successful Knowledge Translation
J. Moore, L. Pickering, G. Mathur, Rehabilitation Institute of Chicago, Chicago, Illinois. T.G. Hornby, University of Illinois at Chicago, Chicago, Illinois. K. Van Der Laan, Northwestern University, Chicago, Illinois.
Background and Purpose: Clinicians struggle with application of evidence-based practice (EBP) because of limited accessibility to research, lack of education on critical appraisal, and time constraints. Several facilitators to EBP have been identified, including providing clinicians with research that is written in an understandable manner and publications in summary forms that can be accessed quickly and understood easily. To overcome many of the common barriers to EBP, the Battery of Rehabilitation Assessment and INterventions (BRAIN) was developed for physical, occupational, and speech therapists throughout the Rehabilitation Institute of Chicago (RIC). The BRAIN provides RIC staff with summaries of evidence, including concise descriptions of how to appropriately use an assessment or intervention in the clinic. Assessment descriptions include psychometrics and clinical utility. Intervention summaries include the parameters for successful application (dosage, intensity, etc) and links to the appropriate outcome measures to monitor progress.
Case Description: Therapists volunteered to participate as a BRAIN specialist or champion. Specialists were responsible for the development of outcome measurement or intervention summaries. Within small groups, the specialists discussed the summaries and determined the most salient details for staff to use in clinical practice. One champion was designated for each site of care. These individuals were educated on the summaries and facilitated adoption of the practices at the site in which they worked. All of the BRAIN summaries were stored in a database on the RIC intranet and accessible at the point of care. A survey was completed before and 2 years after implementation of the BRAIN and its dissemination (champion) model. Survey results were compared to identify the impact of the project. A chart audit to compare utilization of EBP before and after the implementation of the BRAIN is now underway.
Outcomes: Since the development of the BRAIN, 58 to 90 therapists have volunteered annually to participate in the project as specialists, and 47 to 52 individuals have participated as champions. In total, 170 topics have been reviewed, and over 20 outcome measurements have been implemented throughout the system of care. Survey results indicate a 77% increase in utilization of outcome measures in clinical practice and a 52% increase in utilization of evidence-based interventions as a result of the BRAIN. Approximately 58% use the BRAIN to educate other clinicians and 54% use it to educate patients. Clinicians also indicated that the BRAIN has led to more discussions about EBP with colleagues (72%), and a better understanding of clinical translation of research (65%), of research (53%) and statistics (42%).
Discussion: As a result of the BRAIN, RIC clinicians are more routinely using EBP. The impact of this project has reached beyond our clinical practice, to staff and patient education, and a greater engagement in and understanding of research among the physical therapy staff.
Investigating the Immediate and Retention Effects of Contralesional Inhibitory rTMS and Functional Task Practice to Enhance Arm Recovery Poststroke
D.K. Rose, C. Patten, M. Pastula, C. Reese, Physical Therapy, University of Florida, Gainesville, Florida.
Purpose/Hypothesis: The interhemispheric competition model hypothesizes that the balance of cortical excitability between the 2 hemispheres observed in health is altered poststroke, producing a hyperexcitability of the contralesional primary motor cortex (M1), which in turn exerts an abnormally increased interhemispheric inhibition on the ipsilesional M1. Low-frequency repetitive transcranial magnetic stimulation (rTMS) can downregulate cortical excitability. The purposes of this study were to determine if low-frequency rTMS applied to the contralesional M1 would reduce hemispheric cortical hyperexcitability and augment upper extremity (UE) functional task practice (FTP) in those with chronic stroke following a 4-week intervention and at 1-month retention.
Number of Subjects: Twenty individuals (12 male; 9 LCVA; mean age: 63.2 ± 8.1 years; mean time poststroke: 47.3 ± 38.9 months).
Materials/Methods: Participants were randomly assigned to receive either real-experimental (EXP [n = 10]) or sham-control (CON [n = 10]) 1 Hz rTMS for 20 minutes (1200 pulses) to the contralesional hemisphere, followed by 1 hour of UE FTP 4×/week for 4 weeks. Contralesional short intracortical inhibition, reflecting inhibitory interneuronal modulation, was assessed using paired-pulse TMS, CL resting motor threshold with single-pulse TMS. Upper extremity motor ability was evaluated with the Wolf Motor Function Test (WMFT), the UE Fugl-Meyer Assessment, and Grip Strength pre- and postintervention and at a 1-month retention assessment. The WMFT performance time mean value was transformed to natural logarithm to normalize the skewed data distribution. Group differences were tested using Wilcoxon's rank sum test.
Results: Groups were equivalent at baseline on all parameters (Ps > 0.05). Postintervention, the magnitude of contralesional short intracortical inhibition increased, as hypothesized, with significant differences detected between EXP (26%) and CON groups (1%) (P < 0.05), but differences were not retained at 1 month. CL hemisphere resting motor threshold did not change following the intervention. Small, yet statistically significant treatment effects were observed in clinical measures at the posttest and retained at 1 month (P < 0.05), although no differences were detected between groups (change scores: Fugl-Meyer Assessment: EXP, 4.4 ± 1.5; CON, 3.6 ± 1.5; WMFT: EXP, 0.13 ± 0.13; CON, 0.27 ± 0.10; Grip Strength: EXP, 1.2 ± 1.1; CON 1.6 ± 0.7 [Ps > 0.05]).
Conclusions: Inhibitory rTMS induced significant effects on the contralesional circuitry in chronic poststroke participants, but these were not retained at 1 month. The potential effect of rTMS on paretic UE motor ability may have been limited in the presence of the small observed behavioral changes. A more robust rehabilitation intervention to induce greater behavioral gains may be prerequisite for an augmentative effect of rTMS to be produced.
Clinical Relevance: The role of rTMS as an adjuvant to behavioral intervention in this population, although promising, has not been equivocally determined. Repetitive transcranial magnetic stimulation should continue to be investigated as a potential rehabilitation modality.
An Intense Running Program to Retrain Gait and Mobility in a 23-Year-Old Woman Poststroke
M. Petrosky, P.M. Spigel, Brooks Rehabilitation Hospital, Jacksonville, Florida. E.J. Fox, Department of Physical Therapy, University of Florida, Gainesville, Florida.
Background and Purpose: Hemorrhagic stroke is most common in individuals 30 to 50 years of age, and returning to recreational activities is a common goal. Therapeutic interventions poststroke focus on intensity and task-specificity to promote neural plasticity and recovery. These strategies are particularly effective in younger adults and suggest running may be a salient and effective intervention for young adults poststroke. Running requires increased muscle activation and force production compared with walking. Although running may be an effective intervention, it is usually not included as part of rehabilitation poststroke due to many factors. The purpose of this case study was to examine the use of running as a treatment intervention to improve gait, balance, and strength in a 23-year-old female post hemorrhagic stroke.
Case Description: Four months after her stroke, the patient participated in 10 visits of outpatient physical therapy. Her goal was to complete a 5K run within 6 months. An interval running program was initiated using partial body weight support over a treadmill and wearing a posterior leaf spring AFO. The patient then progressed to over ground with the use of a Bioness 300 perioneal nerve stimulator and eventually used a custom dorsiflexion-assist AFO. Each session, the patient was challenged to run for longer periods of time. The patient used an application on her phone to track her running as part of a home exercise program. Her vital signs were monitored during each session and at home. Following the training program, walking function was assessed while the patient walked on a 12-ft instrumented walkway (GaitRITE). Strength, balance, and falls also were assessed.
Outcomes: The patient improved her gait speed from 1.0 m/s at fastest selected walking speed to 1.6 m/s. She did not report any falls during the intervention period and no adverse reactions during running. Her balance improved based on a Berg Balance score increase from 26 to 54 of 56. The patient's left lower extremity strength increased (dorsiflexion from 2+/5 to 4/5; plantar flexion from 3/5 to 4/5). The patient reported that she felt safer and did not use the AFO when ambulating at home or during short distances.
Discussion: An intense running program increased strength, balance, and walking function in a 23-year-old following stroke. The patient was able to continue with a home exercise program and community running program, and she achieved her goal of completing a 5K. Running affords young adults poststroke an intense and gait-specific activity that may improve strength, walking function, and cardiovascular health. Communication with the patient's physician, as well as continued monitoring of vital signs, is pivotal when using a running program.
Auditory Impairments and Their Impact on Postural Control
N. Bugnariu, R.M. Patterson, Physical Therapy, University of North Texas Health Science Center, Fort Worth, Texas. L. Thibodeau, R. Roeser, P. Wilson, Callier Center for Communication Disorders, University of Texas at Dallas, Dallas, Texas.
Purpose/Hypothesis: We investigated the relationship between hearing loss and balance in adults, using advanced virtual reality technologies that provide realistic, ecologically valid yet controlled and safe testing conditions. A second aim was to evaluate the effects of 2 types of hearing aid (HA) technologies on measures of balance and gait. We used a regular HA that amplifies sound from all directions and frequencies and a frequency modulator (FM) system designed to be work in conjunction with the regular HA and to selectively amplify only 1 frequency of interest and not the ambient noise.
Number of Subjects: Twelve adults newly diagnosed with hearing loss, without vestibular or other neurologic impairment, and 12 age- and gender-matched healthy controls participated in this study.
Materials/Methods: Participants were tested for balance, gait, and functional activities, at the time of hearing loss diagnosis and enrollment in the study and after a 2-month accommodation and use of their HAs. Outcome measures included standing center-of-pressure sway, performance of dual task involving cognitive decisions, and self-selected gait speed on flat and uneven terrain in the virtual environment. Testing conditions were no HA, HA, Ha+FM; auditory task conditions were either listening only or repeating back sentences from standard audiology tests. Clinical tests of Dynamic Gait Index, Timed Up and Go, Activities-Specific Balance Confidence Scale, and Short Physical Performance Battery were also administered. Analysis of variance was conducted for each of the dependent variables with respect to group, condition of HA, and condition of auditory task.
Results: Center-of-pressure sway variability in M/L direction was increased (P < 0.05) in participants with hearing loss vs controls when subjects had to perform a dual standing/cognitive task. Without HA, self-selected gait speed was lower (P < 0.05) in individuals with hearing loss vs controls, as long as they attended to the auditory task of repeating back sentences. Use of HA+FM significantly improved (P < 0.01) performance on auditory repeating back sentences task but also increased self-selected speed. Clinical measures showed no difference between groups.
Conclusions: Hearing loss negatively impacts postural control particularly in dual task conditions when individuals attend to both auditory and postural tasks. Use of HAs, especially the FM system, significantly improves not only speech recognition but also measures of balance and gait, and ability to successfully perform dual tasks.
Clinical Relevance: Age-related hearing loss affects a large percentage of the older-adult population. Individuals with hearing loss may be at greater risk of falling than individuals without hearing loss. One possible explanation could be that reduced or conflicting auditory information increases the cognitive load and thus, attentional resources may be taxed particularly in older adults. Majority of current physical therapy clinical outcome measures are not designed to evaluate the impact of hearing loss on measures of postural control, therefore further studies are necessary.
Sleep Promotes Off-line Motor Skills Learning and Visuospatial Memory Consolidation in People With Multiple Sclerosis: A Pilot Study
C. Siengsukon, A. Al-Sharman, M. Aldughmi, Physical Therapy & Rehabilitation Science, University of Kansas Medical Center, Kansas City, Kansas. S. Lynch, Neurology, University of Kansas Medical Center, Kansas City, Kansas. J. Bruce, Psychology, University of Missouri-Kansas City, Kansas City, Missouri.
Purpose/Hypothesis: Evidence demonstrates that sleep promotes learning of new motor skills off-line (without further practice) in people with neurologic injury. However, the effect of sleep on motor skill learning and declarative memory consolidation has never been examined in people with multiple sclerosis (MS). The purpose of this study was to examine if sleep enhances motor skill learning and declarative memory consolidation in people with MS.
Number of Subjects: Seven individuals with primary-progressive or secondary-progressive MS (59 ± 5.57 years of age) participated in this cross-over design study.
Materials/Methods: Participants practiced a continuous tracking task and took 3 declarative memory tests (Rey Auditory Verbal Learning Test [RAVLT], Brief Visuospatial Memory Test [BVMT], and Symbol Digit Modalities Test, Oral form [SDMT]), and then underwent retention testing approximately 12 hours later following sleep (sleep condition) or following a period of being awake (no-sleep condition). The sleep and no-sleep conditions were counterbalanced in order and separated by 2 to 3 weeks. To examine off-line learning, paired t tests were used to assess change in performance from the last practice block to the retention test for the tracking task, RAVLT, and BVMT for the sleep and no-sleep conditions. A 1-way analysis of variance was used to assess group difference in the number of symbols recalled at retention on the SDMT.
Results: The sleep group demonstrated a significant improvement in performance, with less error on the tracking task (P = 0.033), whereas the no-sleep group failed to demonstrate an improvement in tracking performance (P = 0.513). The sleep group demonstrated a stabilization in performance on the BVMT (P = 1.00), while the no-sleep group demonstrated a significant worsening (P = 0.001). Although both groups demonstrated a worsening of performance on the RAVLT (P = 0.031 for sleep group, and P = 0.002 for no-sleep group), the sleep group retained more information than the no-sleep group. The sleep group recalled more symbols at retention on the SDMT than the no-sleep group, although this difference was not statistically significant (P = 0.348).
Conclusions: This pilot study is the first to demonstrate that sleep produces off-line improvements in performance of a motor learning task and sleep appears to stabilize or enhance consolidation of visuospatial types of declarative memories in people with MS. Additional studies are needed to verify these conclusions.
Clinical Relevance: Multiple sclerosis is a devastating disease with progressive worsening of physical and cognitive functions. Individuals with MS frequently experience sleep issues, physical impairments, and cognitive difficulties. Interventions designed to improve sleep quality and quantity may enhance cognition and improve motor learning in individuals with MS.
The Relationship Between the Energy Cost of Transport and Walking Activity in Individuals Poststroke
K.A. Danks, T. Wright, M. Roos, S. Binder-Macleod, D. Reisman, Physical Therapy, University of Delaware, Newark, Delaware. E. Matthews, W. Farquhar, Kinesiology and Applied Physiology, University of Delaware, Newark, Delaware.
Purpose/Hypothesis: Daily walking activity in individuals poststroke is very low, well below that of sedentary adults (<5000 steps/day). Stroke survivors also have elevated energy demands, specifically, the energy cost of transport (CT) or the oxygen consumption per unit distance walked (mL O2/kg/m). Cost of transport is known to have a negative relationship with gait speed; however, it has not been investigated if CT is related to decreased activity levels after stroke. Identifying the relationship between activity levels and CT may provide insight as to why stroke survivors are inactive. The purpose of this study was to examine the relationship between CT and daily walking activity after stroke. We hypothesized that individuals with the greatest CT would show the greatest impairment in daily walking activity. We also hypothesized that the relationship between CT and walking activity may be mediated by walking speed.
Number of Subjects: Fifty subjects with chronic (>6 months) stroke.
Materials/Methods: To obtain activity levels, subjects wore a step activity monitor during waking hours for at least 3 days. Oxygen consumption was measured as they walked at their self-selected speed on a treadmill. Oxygen consumption was normalized to body mass and speed, resulting in energy cost per kilogram of body mass, per meter walked (mL/kg/m) (CT). The descriptors of step activity were averaged across the days monitored, and were used to analyze the relationship to CT. Descriptors of step activity included the number of steps per day (SPD) and walking bouts per day (BPD), total time walking per day, and percent time walking. To investigate “real-world” community mobility, we analyzed how community distances and short bouts (<40 steps/bout), medium bouts (40-500 steps/bout), and long bouts (>500 steps/bout) were related to CT. All data was normally distributed; Pearson's r correlations were utilized to examine the relationship between CT and walking activity. Partial correlations between CT and step activity, controlling for walking speed, were also completed.
Results: There was a negative correlation between CT and SPD (r = −0.31, P < 0.05), BPD (r = −0.30, P < 0.05), and medium walking bouts (r = −0.34, P < 0.05). When controlling for speed, these relationships disappeared (r = −0.059, P = 0.67; r = −0.219, P = 0.14; r = −0.158, P = 0.28, respectively). There was a relationship between speed and SPD (r = 0.40, P < 0.01) and medium walking bouts (r = 0.329, P < 0.05). The correlation between speed and BPD was not significant (r = 0.21, P = 0.14).
Conclusions: There were significant but small correlations between CT and SPD, BPD, and the number of medium bouts of walking. When controlling for speed, these relationships disappeared, suggesting that walking speed is a mediator between CT and walking activity poststroke. However, the speed-activity relationship is also weak. These results suggest that neither walking speed nor CT play a substantial role in poststroke walking activity.
Clinical Relevance: Poststroke walking activity does not appear to be strongly related to the CT or walking speed.
Comparison of Outcomes and Exercise Behaviors in Persons With Parkinson Disease Who Participate in Different Modes of Exercise: A Mixed-Method Study
S.A. Combs, D. Diehl, M.E. Bentz, K.T. Hojnacki, L.D. Sewell, S.L. Steider, Krannert School of Physical Therapy, University of Indianapolis, Indianapolis, Indiana.
Purpose/Hypothesis: The purposes of this study were to (1) investigate differences in physical function and participation between regular exercisers with Parkinson disease (PD) who either participate in boxing training or other modes of exercise, and to (2) explore perceptions related to initiation and adherence to exercise, as well as perceived health-related benefits between groups.
Number of Subjects: Eighty-three people with PD completed testing as part of a larger, longitudinal study. Of these, 63 were categorized as regular exercisers on the Stages of Readiness to Exercise Questionnaire and included in the analysis. Regular exercisers were dichotomized into boxers (n = 39) and nonboxers (n = 24) depending on self-reported type of exercise participation.
Materials/Methods: A mixed-method design was used to analyze a subset of quantitative and qualitative data. Demographics (age, gender), PD characteristics (time since diagnosis, Hoehn and Yahr stage), physical measures including grip strength, Functional Reach Test (FRT), and 6-Minute Walk Test, and participation on the Activities-Specific Balance Confidence Scale were collected. Participants were asked questions regarding exercise initiation, adherence, and perceived health-related changes due to exercise habits. Comparison of quantitative data between boxers and nonboxers were analyzed with independent t tests and effect sizes (Cohen's d). Qualitative responses were analyzed using content analysis.
Results: Demographics and PD characteristics were not significantly different between boxers and nonboxers (P < 0.05). The boxers demonstrated significantly greater distance reached on the FRT (P = 0.01). Moderate between-group effect sizes in favor of the boxers were found on all physical measures (dominant grip strength, d = 0.41; nondominant grip strength, d = 0.44; FRT, d = 0.65; 6-Minute Walk Test, d = 0.42) and for participation (Activities-Specific Balance Confidence, d = 0.54). Primary themes related to what initially attracted participants to their exercise program common to both groups were program structure and advantages of exercise. Themes unique to boxers were camaraderie and type of exercise, while enjoyment was distinctive to nonboxers. Primary themes related to exercise adherence discussed by both groups were social support/interaction, health benefits, and source of motivation. Improvement in fitness and health was a primary theme common to both groups regarding perceived health-related changes due to exercise. Increased confidence in activities and participation was an additional primary theme identified for the boxers.
Conclusions: Physical function and participation were enhanced in the boxers compared to nonboxers. Qualitative analysis revealed that higher exercise intensity, camaraderie within a disease-specific group atmosphere, and invested trainers contributed to positive perception of health benefits, exercise adherence, and self-efficacy in the boxers.
Clinical Relevance: These factors may be critical for clinicians developing exercise programs for people with PD that encourage long-term adherence.
Outcomes Following a Student-Led Group Exercise Class for Individuals With Parkinson Disease
M. Lomaglio, K. Mallini, Doctor of Physical Therapy, University of St Augustine for Health Sciences, St Augustine, Florida.
Purpose/Hypothesis: Access to inexpensive, supervised exercise classes is necessary to meet the needs of individuals living with Parkinson's disease (PD). A physical therapy school is an ideal environment to provide free, supervised exercise classes that target the impairments and activity limitations of individuals with PD. The primary purpose of this study was to examine the feasibility and effects of a student-led group exercise class on gait, balance, and quality of life in individuals with PD. The second purpose was to determine whether any improvements are maintained after the exercise is withdrawn.
Number of Subjects: Eleven community-dwelling volunteers with PD (Hoehn and Yahr stages I-III).
Materials/Methods: Participants exercised for up to 60 minutes, 2 times a week, for 8 weeks. All classes were led by 8 students under the direct supervision and assistance of a physical therapy faculty member. Sessions focused on repetitive functional activities and whole-body movements to improve flexibility, strength, gait, and balance. All sessions ended with progressive treadmill walking with a target of 2.0 miles per hour and 26 continuous minutes. Gait was measured with the Timed Up and Go test, at both comfortable and fast walking speeds (TUGc and TUGf, respectively), and the 2-Minute Walk Test. Balance and balance confidence were measured with the 5 Times Sit to Stand Test and the Activity-Specific Balance Confidence Scale. Quality of life was measured by the Nottingham Health Profile. All outcomes were assessed before, after, and at an 8-week follow-up. A repeated-measures analysis of variance was performed on each outcome measure and, when appropriate, followed by paired t tests to determine whether significant changes over the 3 time points occurred.
Results: There was a 90% attendance rate, and no injurious adverse events occurred. A comparison of the measures taken before and after the intervention revealed significant improvements on the TUGc, TUGf, 2-Minute Walk Test, 5 Times Sit to Stand test, and the Activity-Specific Balance Confidence Scale (P = 0.001-0.037). At the 8-week follow-up, there were no significant differences on any of the measures when compared with baseline values (P = 0.103-0.441), suggesting a loss of the gains over time. Exercise had no effect on the Nottingham Health Profile (P = 0.572).
Conclusions: Student-led group exercise is safe and feasible and can be provided without cost to the participant. The results suggest that improvements in gait and balance occur with exercise but there may be no effect on quality of life. In addition, the results suggest that the exercise needs to be ongoing, as the effects may be lost when the exercise is withdrawn.
Clinical Relevance: Student-led group exercise provides a win-win situation in that students are offered the opportunity to interact with individuals living with PD, and individuals with PD are offered free, supervised exercise that may improve their functional mobility and slow down the negative effects of disease progression.
Improvements in Locomotor Kinematics and Kinetics in Individuals Poststroke Following High-Intensity Stepping Training in Variable Contexts
T. Hornby, Physical Therapy, University of Illinois at Chicago, Chicago, Illinois. C. Kinnaird, C. Holleran, A. Leddy, Sensory Motor Performance Program, Rehabilitation Institute of Chicago, Chicago, Illinois.
Purpose/Hypothesis: Previous work suggests that high-intensity forward stepping training on a treadmill improves locomotor performance in individuals poststroke, although gains in walking function are not consistently observed. Despite recent animal data to suggest its utility, few studies have focused on variable stepping practice at high intensities in individuals poststroke, perhaps secondary to the substantial difficulty of performing these tasks and generation of kinematic errors. If individuals poststroke are able to correct such errors, however, stepping may be improved. Kinematic and kinetic locomotor strategies following high-intensity stepping training in variable contexts have not been documented.
Number of Subjects: Twenty-two individuals with subacute or chronic hemiparesis poststroke.
Materials/Methods: Participants completed an 8- to 10-week stepping training paradigm (<40 sessions) on a treadmill and overground at high aerobic intensities (70%-80% heart rate reserve) within variable contexts (altered directions, speeds, or perturbations). Lower limb kinematics and kinetics were captured using a motion capture system and a split-belt force treadmill during graded treadmill testing prior to and following training. Primary kinematic outcomes included spatiotemporal measures and sagittal plane lower extremity joint excursions during stepping at the highest possible speeds. Kinetic measures included sagittal moments and powers, including average positive and negative power in both limbs (eg, average positive power calculated by determining positive joint work performed by integrating joint power within normalized gait cycles and dividing by average step time). Joint powers were also analyzed within specific phases of the gait cycles consistent with power generation in unimpaired individuals.
Results: Peak treadmill speed improvements from 0.53 ± 0.40 m/s to 0.94 ± 0.48 m/s were accompanied by 20% to 40% improvements in stride length and cadence (all P < 0.01). Specific improvements included increased peak paretic hip and knee flexion and bilateral joint excursions (all P < 0.01), and consistent improvements in peak paretic extensor moments (27%-106% above baseline, P < 0.01) with large but variable changes in flexor moments. Averaged joint powers in the paretic limb improved significantly (P < 0.01), with similar improvements only in positive hip and negative knee nonparetic limb powers. Despite smaller improvements, regression analysis revealed that changes in peak treadmill speed were significantly correlated only to altered positive nonparetic hip and ankle powers.
Conclusions: These preliminary results suggest improvements in gait kinematics and kinetics with stepping training at high intensities in variable contexts, with potential changes related to nonparetic limb powers.
Clinical Relevance: This work may advance our understanding of allowing movement errors during stepping training and potential movement strategies underlying increased gait speed.
The Association of Rehabilitation Service Use on Hospital Readmission for Patients With Acute Stroke
A.W. Andrews, Physical Therapy Education, Elon University, Elon, North Carolina. D. Li, J.K. Freburger, Sheps Center for Health Services Research, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina.
Purpose/Hypothesis: To determine if the intensity of speech, physical, and occupational therapy services for patients with acute stroke was associated with 30-day and 90-day hospital readmissions.
Number of Subjects: Patients aged 45 years and older residing in Arkansas or Florida who were hospitalized with an incident stroke in 2010 or 2011 (N = 64 065) and who survived their acute hospitalization.
Materials/Methods: This was a retrospective, cohort study using data on acute care hospital admissions from the State Inpatient Databases for Arkansas and Florida. The independent variable, intensity of rehabilitation services, was determined by examining charges for speech, physical, and occupational therapy during the acute admission. These charges were summed for each patient and categorized as No Therapy or as Low Intensity, Medium Low Intensity, Medium High Intensity, or High Intensity, based on the quartile distribution of the charges. The dependent variables were the presence of a readmission within 30 or 90 days of hospital discharge for the incident stroke. Multilevel Cox regression analyses, adjusting for demographic variables, stroke severity, comorbidities, hospital quality indicators, and other hospital effects, were used to examine the association between rehabilitation use and readmission.
Results: The risk of 30-day and 90-day hospital readmission decreased as intensity of rehabilitation services received during the acute care hospitalization increased. Relative to the Low Intensity group, the odds of 30-day readmission ranged from 0.91 (0.85-0.98) for the Medium Low intensity group to 0.87 (0.80, 0.94) for the High Intensity group. Findings for 90-day readmission were similar, with odds of readmission ranging from 0.95 (0.89-1.00) for the Medium Low intensity group to 0.90 (0.85, 0.96) for the High Intensity group, relative to the Low Intensity group. Individuals who received no therapy had also an increased risk of 30-day and 90-day readmission, odds ratios of 1.31 (1.22, 1.41) and 1.40 (1.33, 1.48) respectively, relative to the Low Intensity group.
Conclusions: Increased intensity of rehabilitation service use in the acute setting was associated with a lower risk of 30-day and 90-day hospital readmission for patients with an incident stroke.
Clinical Relevance: Interventions, education, and recommendations made by physical therapists and other rehabilitation professionals may positively influence the probability of readmission in patients who are admitted to an acute-care hospital with an incident stroke.
Body-Weight–Supported Treadmill Training Is no Better Than Overground Gait Training During Intensive Mobility Training for Individuals With Chronic Stroke, a Randomized Control Trial
S. Fritz, A. Middleton, D.M. Peters, E.L. Blanck, Exercise Science–Physical Therapy Program, University of South Carolina, Columbia, South Carolina. A.R. Merlo, Physical Therapy, Northern Arizona University, Phoenix, Arizona.
Purpose/Hypothesis: To determine if an intensive intervention (intensive mobility training) including body-weight–supported treadmill training (BWSTT) provided superior gait, balance, and mobility outcomes as compared with an intervention of equal intensity, using overground gait training in place of BWSTT.
Number of Subjects: Forty-three individuals with chronic stroke (average age, 61.5 ± 13.5 years; average time since stroke, 3.3 ± 3.8 years).
Materials/Methods: Participants were randomized to either a treatment group incorporating BWSTT (n = 23) or a control group incorporating overground gait training (n = 20). All participants received 3 hours of intervention for 10 consecutive weekdays, totaling 30 hours of therapy. Each 3-hour session consisted of 1 hour of gait training (experimental group = BWSTT; control group = overground gait training), 1 hour of balance activities, and 1 hour of strength, range of motion, and coordination. All treatment activities were specific to individual deficits and progressed as appropriate. Assessments were conducted by an evaluator (blinded to group) prior to intervention, immediately following the intervention, and at a 3-month follow-up. Outcome assessments included step length differential, self-selected and fast walking speed, 6-minute walk, Berg Balance Scale, Dynamic Gait Index (DGI), Activities-Specific Balance Confidence (ABC) Scale, single-limb stance, Timed Up and Go (TUG), Fugl-Meyer (FM), and perceived recovery (PR). A repeated-measures analysis of variance for each outcome measure followed with a post hoc Tukey-Kramer analysis was used to locate differences between and within treatment groups. Effect size was calculated for each outcome.
Results: No significant differences (α = 0.05) were found between groups on any of the outcome measures assessed either immediately following training or at follow-up. For this reason, groups were combined for all remaining analyses. Pre- to posttest analyses found significant differences (α = 0.05) for fast walking speed, Berg Balance Scale, DGI, ABC, TUG, Fugl-Meyer, and PR. Four of these measures (DGI, ABC, TUG, and PR) remained significant at follow-up. Calculated effect sizes ranged from small to moderate, moving in the direction of improvement.
Conclusions: Individuals with chronic stroke did demonstrate improvements in gait, balance, and mobility following intensive therapy; however, only limited improvements were maintained at 3 months.
Clinical Relevance: This study supports previous research demonstrating that clinicians do not need expensive equipment, such as body-weight–supported treadmill systems, to make functional gains in individuals with chronic stroke. Body-weight–supported treadmill training was not more beneficial in this study than overground gait training. Predictors of who benefits most from certain therapies and optimal dosage for maximal benefit still need to be determined.
The Relationships Between Gait Speed, Efficiency, and Symmetry With Daily Step Counts in Individuals With Chronic Stroke
M.D. Lewek, Division of Physical Therapy, University of North Carolina, Chapel Hill, North Carolina.
Purpose/Hypothesis: Individuals poststroke walk less in the home and community on a daily basis than age-matched peers. Limited physical activity, in the form of daily step counts, has been attributed to reduced gait speed. Slow gait speed, however, has been linked with both gait asymmetry and greater energy cost of walking. It is important to determine how these factors relate to daily stepping to understand which factors to address in rehabilitation to improve daily stepping. The purpose of this study was to determine the association between gait speed, efficiency, and symmetry with daily step counts in individuals with chronic stroke. It was hypothesized that each of these variables would be related to daily step counts.
Number of Subjects: Twenty-one individuals (15 male/6 female; 62 ± 11 years) with chronic (>6 months) hemiparesis following stroke.
Materials/Methods: Subjects walked across a GAITRite mat to determine comfortable gait speed (CGS) and step length (SLA) and stance time asymmetry (STA) ratios. A 6-Minute Walk Test was performed with a portable metabolic cart to estimate cost of transport (COT). Subjects wore a Stepwatch monitor for 7 days to measure average daily step count. A stepwise regression analysis was performed to determine the primary factors (ie, CGS, SLA, STA, and COT) associated with daily step counts. Correlational analyses were performed to determine the presence of significant linear relationships between variables (α = 0.05).
Results: Subjects had a CGS of 0.45 ± 0.23 m/s (range: 0.10-0.94 m/s) and substantial spatiotemporal asymmetry ratios (SLA: 1.29 ± 0.21; STA: 1.18 ± 0.11). During the 6-Minute Walk Test, subjects completed 181 ± 98 m (range: 30-341 m) and had a COT of 7.6 ± 6.0 J/kg/m. Subjects walked an average of 1662 ± 1172 steps/day. The stepwise regression model found only SLA to predict average daily step count (r 2 = 0.44; P = 0.003). The collinearity between variables, however, was likely responsible for excluding other variables. In particular, CGS was significantly related to COT (r = −0.77; P < 0.001), SLA (r = −0.72; P = 0.001), and step count (r = 0.68; P = 0.003). Cost of transport was also related to SLA (r = 0.78; P < 0.001) and step count (r = −0.55; P = 0.012). Stance time asymmetry was not related to any variable.
Conclusions: Daily step counts were best predicted by SLA. This suggests that the inability to take equal step lengths may be an underlying factor in reduced home/community stepping following stroke. Previous literature has related SLA to propulsive asymmetry, which may require energetically inefficient movement compensations that increase COT and reduce CGS. The interrelationships between these variables, but not STA, suggests that spatial symmetries are more important for enhancing functional mobility than temporal symmetries.
Clinical Relevance: Rehabilitation to emphasize reduced SLA asymmetries may be important for improving gait speed, efficiency, and daily stepping.
Sensory and Cognitive Deficits Poststroke May Be Interrelated and Impact Motor Recovery
D.S. Nichols-Larsen, A. Borstad, P. Kaur, School of Health and Rehabilitaton Sciences, The Ohio State University, Columbus, Ohio. P. Schmalbrock, Radiology, The Ohio State University, Columbus, Ohio.
Purpose/Hypothesis: Sensory and mild cognitive deficits often go undiagnosed poststroke, since appropriate tests are rarely done in the clinic setting. However, when these tests are conducted, almost 90% of stroke survivors demonstrate either sensory or mild cognitive deficits; studies have not looked at both in the same cohort, and no study has looked at the relationship between these deficits.
Purpose: To examine the relationship between measures of cognition, specifically working memory, and sensorimotor behavior of the hand. Hypothesis: (1) Poststroke subjects will perform worse on measures of sensory discrimination and cognition than age-matched peers; (2) Deficits in sensory and cognitive function will relate to poorer motor function; (3) Functional magnetic resonance imaging analyses will differentiate activation patterns between controls and poststroke participants.
Number of Subjects: Fourteen controls and 12 chronic poststroke subjects were tested in a single behavioral and imaging session. Inclusion criteria were (1) single hemiparetic documented stroke; (2) aged 21 to 85 years; (3) hand movement sufficient to grasp and release objects; and (4) ability to understand test instructions. Exclusion criteria were (1) prior diagnosis of cognitive (eg, dementia), sensory (eg, neuropathy), or neurologic (eg, Parkinsons) disorders; (2) metal device or other incompatibility with magnetic resonance imaging testing; (3) Mini-Mental State Examination <24.
Materials/Methods: Testing: Sensory—Hand Active Sensation Test, Brief Kinesthesia Test, Form Perception Test; Cognition—Paced Auditory Serial Addition Test, Choice Reaction Test; Motor—Box and Blocks Test (BBT), 9 Hole Peg; imaging—functional magnetic resonance imaging sensory (brush discrimination) and cognitive task (n-back).
Results: Significant group differences were found for the paretic hand versus the nondominant hand of the controls for all measures. Hand Active Sensation Test strongly correlated with Paced Auditory Serial Addition Test (r = 0.722, P < 0.0001). Paced Auditory Serial Addition Test strongly correlated with BBT (r = 0.685, P < 0.001); Hand Active Sensation Test moderately correlated with BBT (r = 0.431, P < 0.05). Networks supporting sensory discrimination and working memory overlap with the parietal lobe and dorsal frontal cortex, playing a key role in both; poststroke subjects demonstrated disruption of both networks.
Conclusions: Sensory and working memory deficits are common after stroke and may be related to disruption of common neural networks. These hidden deficits may impact motor recovery.
Clinical Relevance: Therapists need to include measures of sensory discrimination and cognition within their standard assessment procedures for stroke survivors.
Is Multitasking Impaired in Older Adults With Type 2 Diabetes Mellitus?
J. Rucker, A.J. Britton, N. Utech, P. Kluding, Physical Therapy and Rehabilitation Science, University of Kansas Medical Center, Kansas City, Kansas. J. McDowd, Psychology, University of Missouri–Kansas City, Kansas City, Missouri.
Purpose/Hypothesis: Deficits in the ability to multitask contribute to gait abnormalities and falls in many at-risk populations. However, it is unclear whether older adults with type 2 diabetes mellitus (DM) also demonstrate impairments in multitasking. This study examined whether multitasking was impaired in older adults with DM when compared with those without DM.
Number of Subjects: Forty individuals with type 2 DM (65% female; age, 72.9 ± 8.3 years) and 40 individuals without DM (65% female; age, 72.9 ± 7.7 years). Groups were matched for age, sex, education, and presence/absence of hypertension.
Materials/Methods: Multitasking was assessed via the Walking and Remembering Test, in which subjects ambulated along a 6.1 m, 19-cm wide path while attempting to remember a random number sequence of individually determined length. Four trials of the digit recall and walking tasks were completed both individually (eg, single-task conditions) and simultaneously (eg, multitask conditions), and the average number of digits recalled, walking speed, and the number of steps off path were recorded for each condition. Two-tailed paired t tests assessed between-group differences in digit recall, walking speed, and the number of steps off path, as well as the percentage change from single- to multitask conditions for each variable. The significance level was set at 0.05.
Results: Under single-task conditions, subjects with DM exhibited similar digit recall (6.7 ± 1.2 vs 6.5 ± 1.4 digits; P = 0.55), but ambulated more slowly (1.1 ± 0.3 vs 1.4 ± 0.4 m/s; P < 0.001) and took more steps off path (0.6 ± 0.7 vs 0.2 ± 0.3 steps off; P = 0.004) than control subjects. Likewise, under multitasking conditions, subjects with DM demonstrated similar digit recall (4.6 ± 1.3 vs 4.9 ± 1.9 digits; P = 0.48), but ambulated more slowly (1.1 ± 0.8 vs 1.4 ± 0.3 m/s; P < 0.001) and took more steps off path (1.2 ± 1.2 vs 0.4 ± 0.5 steps off; P < 0.001). Both groups exhibited similar changes in digit recall (29.5 ± 19.4% vs 24.2 ± 14.3%; P = 0.20) and walking speed (−0.1 ± 8.0% vs 2.3 ± 5.9%; P = 0.13) from single- to multitasking conditions; however, subjects with DM demonstrated a greater change in steps off path (107.5 ± 138.3% vs 45.4 ± 116.3%; P = 0.03).
Conclusions: This study indicates that older adults with type 2 DM exhibit decreased walking speed and stability while multitasking. Additionally, it suggests that those with DM may not allocate attention appropriately when multitasking—in this case, preserving cognitive function and walking speed at the expense of gait stability.
Clinical Relevance: Older adults with DM are known to exhibit gait abnormalities and a greater risk of falls than their peers without DM. Our study suggests that neuropsychological factors such as multitasking may negatively affect gait in this population. Clinicians should consider employing measures of multitasking when assessing gait and fall risk in older adults with DM.
Can Cool Ambient Temperature Affect Cognitive Performance in Persons With Tetraplegia?
J.P. Handrakis, S. Liu, M. Krajewski, Physical Therapy, New York Institute of Technology, Old Westbury, New York. D. Rosado Rivera, A.M. Spungen, W.A. Bauman, National Center of Excellence for the Medical Consequences of Spinal Cord Injury, Bronx, New York. C. Bang, James J Peters VA Medical Center, Bronx, New York.
Purpose: To determine the effects of cool ambient temperature (18°C) exposure on body core temperature, cognitive performance, and distal skin temperatures in persons with a cervical spinal cord injury (tetraplegia).
Primary Hypothesis: Body core temperature and cognitive performance will decline in persons with tetraplegia after cool temperature exposure but will be maintained in controls. Secondary Hypothesis: The decrease in distal skin temperatures after cool temperature exposure will be greater in controls than in persons with tetraplegia.
Number of Subjects: Seven male individuals with tetraplegia (C3-C7; AIS A-C; DOI 16.4 ± 6.5 years; mean age, 43 ± 6.4 years; body mass index, 23 ± 1.6 kg/m2) and 7 age- and gender-matched controls (mean age, 41 ± 4.2 years; body mass index, 27 ± 0.9 kg/m2) signed a consent form approved by the James J Peters Bronx VA Medical Center institutional review board.
Materials/Methods: Persons with tetraplegia and able-bodied controls were exposed to temperatures of 27°C baseline for 30 minutes, followed by up to 120 minutes of 18°C exposure (Cool Challenge) while wearing only shorts in the seated position. Rectal temperature (T core) and distal skin temperatures (T sk) were measured by thermocouples and continuously collected throughout. The protocol was terminated if T core ≤ 35°C (hypothermia) or the subject expressed discomfort. A neuropsychological battery, including Delayed Recall (working memory) and Stroop Color and Word tests (executive function), was administered at the end of baseline and after Cool Challenge.
Results: After Cool Challenge, T core decreased −1.2 ± 0.12°C (P < 0.0001) in the group with tetraplegia after an average of 109 ± 16 minutes compared with no change (0.07 ± 0.08°C) in able-bodied controls after 120 minutes. The average of distal skin temperatures declined in both groups, but the decline in controls was significantly greater than in those with tetraplegia (−31.6 ± 7.9% vs −8.6 ± 5.8%, respectively; P < 0.0001). In the group with tetraplegia, Delayed Recall and Stroop Interference scores declined −55 ± 47.4% (P < 0.05) and −3.9 ± 3.8% (P < 0.05), respectively, both of which were different (P < 0.05) than the consistent cognitive performance maintained by controls.
Conclusions: Even limited exposure to cool temperatures can overwhelm the impaired thermoregulatory mechanisms of persons with tetraplegia, as evidenced by the attenuated decline in T sk, which most likely contributed to accelerated heat loss and the inability to maintain T core. The decline in T core was associated with deterioration of cognitive performance in the areas of working memory and executive function.
Clinical Relevance: The findings of this pilot study call attention to the cognitive implications of thermoregulatory fragility in persons with tetraplegia and the need to address this impairment by efficacious medical interventions (targeting impaired thermoregulatory mechanisms), specific guidelines for safe temperature exposure for patient education, and novel bioengineering solutions (insulated outdoor transportation areas, wheelchair-mounted core temperature monitoring devices, etc).
The Content of “Rehabilitation Education for Caregivers and Patients” (RECAP) in Stroke Physical Therapy Practice
M. Danzl, Physical Therapy, Bellarmine University, Louisville, Kentucky. A. Harrison, P. Kitzman, P. McKeon, Rehabilitation Sciences, University of Kentucky, Lexington, Kentucky. G. Rowles, Graduate Center for Gerontology, University of Kentucky, Lexington, Kentucky. E.G. Hunter, Cardinal Hill Rehabilitation Hospital, Lexington, Kentucky.
Purpose/Hypothesis: Patient and caregiver education is suggested as a critical component of stroke rehabilitation and physical therapy (PT) practice. Research indicates that the informational needs of stroke survivors and caregivers are largely unmet. Previous studies provide insight into the frequency and type of educational statements made by physical therapists; however, they are limited to the outpatient setting and general patient population. As a first step toward developing optimal educational interventions in stroke PT practice, examination of the content of education provided by physical therapists to stroke survivors and their caregivers is needed. The purpose of this study was to describe the content of “Rehabilitation Education for Caregivers and Patients” (RECAP) in post–acute care stroke rehabilitation by physical therapists.
Number of Subjects: A purposeful, criterion-sampling paradigm was used, and 19 physical therapists from 2 health care organizations were recruited.
Materials/Methods: Qualitative research methods with a grounded theory approach were used in a study examining the phenomenon of RECAP by physical therapists in stroke rehabilitation. A component of this study involved an in-depth examination of RECAP content. As a preinterview activity, each participant listed all topics educated about to stroke survivors and/or caregivers. The master list of topics was analyzed for patterns, and domains of content emerged. Participants discussed the domains during interviews and focus groups. Verification strategies for credibility included member checking, peer debriefing, and reflexivity.
Results: The master list included 126 topics of stroke-related RECAP content. The analysis of the master list yielded 10 domains of content: Stroke Knowledge, Functional Mobility, Equipment and Devices, Safety and Precautions, Promoting Optimal Recovery, Psychological and Emotional Issues, Community Reintegration, Advocacy, Institutional Support and Resources, and Healthcare Continuum and Team. Participants shared perceptions regarding the importance, comfort level, prioritization, frequency, and ideal timing for each domain. Participants described providing education within the context of a multidisciplinary team.
Conclusions: The findings represent an extensive advancement in the description of stroke-related RECAP in PT practice. Ten domains of RECAP content, provided by physical therapists to stroke survivors and their caregivers, were identified. The findings provide a springboard for assessing the content of educational interventions in stroke rehabilitation. Further, the research design provides a model with which to assess RECAP in other patient populations.
Clinical Relevance: The stroke-related content domains of RECAP that emerged provide practicing clinicians with a framework to provide comprehensive education to stroke survivors and caregivers. The domains, as well as their descriptions and examples, can be used to inform entry-level curricula by providing a guide to a formerly abstract area of practice.
The Effects of Age and Parkinson's Disease on Temporal and Spatial Learning During a Posturally Demanding Implicit Motor Sequence Task
H. Hayes, N. Hunsaker, K.B. Foreman, L.E. Dibble, Physical Therapy, University of Utah, Salt Lake City, Utah. L. Boyd, Brain Behavior Lab, University of British Columbia, Vancouver, British Columbia, Canada.
Purpose/Hypothesis: Age and disease may adversely affect the performance of implicit motor learning tasks. In order to evaluate the effects of age and disease on the temporal and spatial error components of a standing implicit motor sequence task, we examined individuals with Parkinson's disease compared with healthy controls. We hypothesized that there would be differences in temporal and spatial error between groups.
Number of Subjects: Healthy young (HY) (N = 10); healthy elders (HE) (N = 10); and Parkinson's disease (PD) (N = 18).
Materials/Methods: HY, HE, and individuals with PD practiced tracking a series of random and repeating waves by shifting their center of pressure (COP) anteriorly and posteriorly to match a target wave. Both their COP and the target wave were projected on a screen in front of them. Two days of practice (each day with 6 blocks of 10 trials) were performed. Two dependent variables were calculated: temporal tracking accuracy (TTA) and spatial tracking accuracy (STA), as components of overall root mean square error. Temporal tracking accuracy was assessed using a time series analysis, by serially correlating the data points from the participant's tracking pattern with the target pattern until a maximum correlation coefficient was achieved. Spatial tracking accuracy was measured by assessing the remaining lag error that persists after the correction for TTA based on the root mean square error in centimeters that remains. Sequence-specific learning of TTA and STA was the difference in performance of the random and repeated waves across the 12 blocks. A 3 (group) × 2 (day) analysis of variance to assess learning was performed for both sequence-specific TTA and STA to examine interaction and main effects, using Tukey's HSD for post hoc analysis.
Results: H&Y, M = 1.81 (0.62). Temporal tracking accuracy results revealed significant day (P < 0.01), group (P < 0.001) effects but no significant interaction (P = 0.44) with the following between group differences; HY to HE (P < 0.001), HY to PD (P = 0.05), and HE to PD (P = 0.02). Overall, the PD group improved more than the HE group on the temporal accuracy (ES = 0.20). Spatial tracking accuracy results revealed significant day (P < 0.001), group (P < 0.001), and interaction (P = 0.038) effects, with the following between-group differences in STA; HY to HE (P < 0.001), HY to PD (P < 0.001), but not HE to PD (P = 0.95); furthermore, the interaction indicated that the HE improved more than the PD on spatial accuracy (ES = 0.91).
Conclusions: These results indicate that while all groups benefit from practice, both age and PD degrade spatial and temporal accuracy on this task. The magnitude of changes in temporal and spatial components of sequence-specific learning in persons with PD suggests that basal ganglia damage may have a differential effect on practice-mediated improvements in the timing and amplitude of movements.
Clinical Relevance: Recognizing difficulties in subcomponents of sequence-specific learning in individuals with PD and HE during an implicit motor task may aid clinicians in developing more appropriate motor learning strategies in the clinic.
Locomotor Requirements for Bipedal Locomotion: A Delphi Survey
L.D. Hedman, Physical Therapy, Northwestern University, Chicago, Illinois.
Purpose/Hypothesis: A variety of constructs have been proposed for physical therapy diagnostic classifications. Bipedal locomotor control requirements may be useful as classifications for walking dysfunction, because they go beyond traditional observational gait analysis to address all issues contributing to locomotor movement problems. The objective of this study was to determine if locomotor experts could achieve consensus about requirements for bipedal locomotion.
Number of Subjects: Expert panel: round 1: n = 115; round 2: n = 78; and round 3: n = 58.
Materials/Methods: Locomotor experts from physical therapy and other related professions were recruited using purposive and snowball methods to participate in an electronic mail Delphi survey. Experts recommended additions, deletions, rewording, and merges for 15 proposed locomotor requirements in round 1. In rounds 2 and 3, panelists commented on and rated the validity, mutual exclusiveness, and understandability of each requirement using Likert ratings. Consensus was defined a priori as follows: (1) 75% or more panelists agree or strongly agree that a requirement is valid, mutually exclusive, and understandable in round 3; (2) no difference between rounds 2 and 3 ratings with kappa coefficients >0.60; and (3) a reduction in panelists who commented and convergence of comments between rounds 1 and 3. Content analysis and nonparametric statistics were used (P < 0.05).
Results: The expert panel reached full consensus on 5 locomotor requirements: initiation, termination, anticipatory dynamic balance, multitask capacity, and walking confidence and partial consensus for 6 other requirements in their Likert ratings. There were no significant differences in Likert ratings between rounds 2 and 3 and a decrease in the percentage of panelists who commented between rounds 1 and 3. There was a correspondence in the issues that emerged from the qualitative comments and the Likert ratings of the requirements.
Conclusions: This is the first systematic attempt to achieve consensus among a cross-disciplinary group of experts in locomotion about fundamental requirements of bipedal locomotion as a precursor to developing a diagnostic classification framework. The results of this study provide initial face and content validation for the 5 requirements for which the experts reached consensus. Further work is needed to establish the validity of all of the locomotor requirements. Consensus on the underlying constructs and language should be prioritized. The next step will be to identify and test clinical measures and interventions for each requirement.
Clinical Relevance: A clinical framework based on locomotor requirements representing the spectrum of walking dysfunction could help physical therapists identify issues that may not be easily derived from traditional gait analysis. With validation, these requirements, along with associated clinical measures and interventions, can provide the framework for a clinically feasible and systematic diagnostic tool for physical therapists to categorize locomotor problems and standardize intervention for walking dysfunction.
Fall Risk Assessment at Admission and Discharge in Clients With Acquired Brain Injury in Post–Acute Inpatient Rehabilitation: A Theoretical and Clinical Model
L.N. Kerr, E.G. Hunter, Outpaitent Services, Cardinal Hill Rehabilitation Hospital, Lexington, Kentucky. C. Killian, P. Rundquist, Physical Therapy, University of Indianapolis, Indianapolis, Indiana.
Purpose/Hypothesis: The purpose of this study was to identify fall risk for clients with ABI in a post–acute inpatient rehabilitation center at admission and discharge. The information gained may enhance health professionals' understanding of fall risk assessment, ultimately assisting in the prevention of falls in this vulnerable population. These clinical measures are necessary for safe client progression toward functional independence.
Number of Subjects: Thirty-six adults who were admitted to the acute rehabilitation hospital brain injury unit with a primary diagnosis of ABI. The sample consisted of 19 (52.8%) males and 17 (47.2%) females, with a median age of 40.5 years.
Materials/Methods: A prospective exploratory cohort study in which participants were assessed for fall risk at admission and discharge. Fall risk inventory included 11 clinical cognitive, fall history, and motor measures routinely completed in this setting.
Results: The Wilcoxon signed rank test determined significant reduction in fall risk for all tests from admission to discharge (α = 0.005). The McNemar binominal exact test determined significant reduction in fall risk categories for all tests from admission to discharge. The significant fall risk category across the 11 measures at admission was 95.8%, and 55.6% at discharge. A principal component analysis at admission and again at discharge yielded a 3-component model supporting the importance of motor, cognitive, and an inventory including fall history in determination of fall risk. The 3-component model at admission accounted for 79% of the variance, and at discharge, it accounted for 74% of the variance. This result reduced the complexity of the 11 clinical tests to 3 main concepts.
Conclusions: An important finding of this study was the crucial importance of assessing motor, cognitive, and fall history to adequately evaluate fall risks for people with ABI in a post–acute inpatient rehabilitation setting. Individually each component does contribute in the assessment of fall risk, but clinicians must understand that the 3 components being assessed together gives a more inclusive comprehensive assessment for each patient. This study then identified the best clinical instruments, for each of the 3 components, to detect fall risk in clients at admission and discharge. Using this information, 2 models were developed, 1 clinical and 1 theoretical. One last issue was uncovered during this study: during the rehabilitation process, clients had significant reduction in fall risk from admission to discharge. However, it is unsettling that 55.6% of the clients remained at significant fall risk at discharge. With this in mind, it is incumbent that clinicians and discharge planners consider this risk of falls in the continuity of appropriate care.
Clinical Relevance: The clinical and theoretical models that were developed out of the findings in this project can be useful tools for providing best practice care for people with ABI to enhance their health and quality of life.
Development of an Electrophysiologic and Behavioral Methodology to Assess Consciousness After Severe Brain Injury
K.V. Day, J. Whyte, Moss Rehabilitation Research Institute, Elkins Park, Pennsylvania.
Purpose: A primary rehabilitation goal for patients with disorders of consciousness (DoC) is to assess level of awareness accurately; physical therapists play a key role in this interdisciplinary assessment. Command following serves as powerful evidence of consciousness, demonstrating a contingent relationship between environmental events and behavior unaccounted for by reflex mechanisms. Recent studies reported persons with DoC capable of following commands via voluntary modulation of brain activity on electroencephalogram (EEG) in the absence of overt movement. These findings were interpreted to suggest that covert command following may be a more sensitive indicator of consciousness than overt movement. However, systematic studies of the stability and sensitivity of these different forms of evidence have not been published. While our ultimate aim is to address the psychometrics of overt and covert command following during DoC recovery, initial steps are required to ensure accurate data collection and to refine analyses that characterize response accuracy. Thus, the purpose of this report was to discuss the development of an EEG and behavioral methodology to aid in assessment of awareness.
Description: We developed a mobile laboratory containing a state-of-the-art EEG system, as well as accelerometers, for quantitative measurement of subtle limb motion. This setup allows for efficient testing in patient rooms on our brain injury unit. As part of the refinement process to optimize our data acquisition protocol, we are testing 5 healthy controls and 5 participants with severe brain injuries who are conscious and can follow commands (eg, “move your hands” vs “hold still”). Subsequently, data are applied to a machine classifier to determine accuracy of responses. This strategy ensures our equipment accurately measures any movements and brain activity when commands are followed appropriately. Data will be presented, demonstrating optimization of EEG and accelerometry signals, as well as classification analyses.
Summary of Use: Evidence suggests that as many as 43% of persons with DoC are misdiagnosed as being in vegetative state based on clinical evaluation, when they indeed retain some degree of awareness. In an attempt to reduce this error using advanced technologies, development of a methodology for assessment of consciousness requires an initial effort to guarantee collection of the most accurate data possible. The testing paradigm developed here will be applied next in our longitudinal study for persons with DoC.
Importance to Members: Diagnostic errors may result from a clinical team briefly observing a patient's fluctuating level of awareness or using standardized tests that fail to circumvent sensorimotor confounds, thus limiting demonstration of consciousness. With literature reporting a much better prognosis for persons in minimally conscious state compared with vegetative state early after injury, such a misdiagnosis could have a profound impact on continuation of rehabilitation, subsequent recovery potential, quality of life, and cost of care.
Influence of Structural vs Functional Corticospinal Tract Integrity on Gait Performance Poststroke
V.L. Little, Physical Therapy, University of Florida, Gainesville, Florida. T.E. McGuirk, C. Patten, Brain Rehabilitation Research Center, Malcom Randall VA Medical Center, Gainesville, Florida. N. Lodha, Applied Physiology & Kinesiology, University of Florida, Gainesville, Florida.
Purpose/Hypothesis: Impaired paretic limb advancement (PLA) is among the most prominent manifestations of walking dysfunction poststroke. Consistent with previous findings, our recent investigation of gait neuromechanics points to plantar flexion (PF) dysfunction as a causal mechanism of impaired PLA. However, neither the underlying mechanisms of PF dysfunction nor its capacity for recovery following stroke are well understood. Damage to corticospinal projections may contribute to PF dysfunction following stroke. To link corticospinal integrity with functional PF performance, we investigated (a) the structural and functional corticospinal tract integrity serving paretic plantar flexors following stroke and (b) paretic plantar flexor muscle activity patterns during overground gait.
Number of Subjects: We studied 8 participants (age: 55.3 ± 5.1 years; 8 male) with chronic (91.7 ± 46.6 months) poststroke hemiparesis and 4 controls (age: 64 ± 15.1 years; 3 male).
Materials/Methods: We used single-pulse transcranial magnetic stimulation to evoke responses in paretic leg gastrocnemius (MG), soleus (SO), and tibialis anterior (TA) during isometric and dynamic PF at 3 positions: dorsiflexion, neutral, and plantar flexion. Dynamic contractions were preloaded (20% MVC) to control background activation. Motor-evoked potential area (MEParea) was normalized to maximal M wave. The greatest difference in normalized MEParea during dynamic, relative to isometric, PF quantified modulation. We also studied the same individuals using motion analysis during overground walking at self-selected speed to compare gait mechanics with MEP data. We calculated the proportion of integrated EMG (iEMG) activity present in a given phase of the gait cycle to evaluate the temporal pattern of activation for the MG, SO, and TA muscles.
Results: Normalized MEParea was comparable between groups during isometric PF, revealing spared connectivity to MG (P = 0.26) and SO (P = 0.30). However, modulation of cortical drive to the plantar flexors is markedly reduced during dynamic PF (MG: P < 0.01; SO: P = 0.01). Similarly, the proportions of MG and SO EMG activity during mid- and terminal stance are reduced following stroke (all Ps < 0.01). Interestingly, connectivity to (P = 0.63) and functional integrity of (P = 0.64) the TA appear to be preserved following stroke. Of note, individuals poststroke reveal a greater proportion of TA activity at the stance-to-swing transition (P < 0.05) than controls.
Conclusions: Profound impairment in modulating PF activity suggests inability to mediate cortical activity during movement, consistent with reduced plantar flexor EMG during gait. Finally, robust TA responses complement our prior work indicating dorsiflexor dysfunction does not underlie impaired PLA poststroke.
Clinical Relevance: Demonstration of spared connectivity to the plantar flexors reveals presence of a neurologic substrate for locomotor plasticity. Interventions that increase corticospinal tract activity serving the paretic plantar flexors may produce better gait outcomes.
Action Selection for Paretic Hand Movements Leads to Increased Activation in the Contralesional Hemisphere After Stroke
J.C. Stewart, P. Dewanjee, U. Shariff, S.C. Cramer, Departments of Neurology and Anatomy & Neurobiology, University of California Irvine, Irvine, California.
Purpose/Hypothesis: Action selection (AS) is a critical feature of voluntary movement. After stroke, damage to the motor system often leads to compensatory brain activation for simple movement tasks. It is not known, however, how the motor system responds after stroke when additional AS requirements are placed on movement. The purpose of this study was to determine the neural correlates of motor AS in individuals poststroke.
Number of Subjects: Ten individuals with right-sided hemiparesis (mean ± SD: age, 68.0 ± 8.2 years; upper extremity Fugl-Meyer motor score, 50.5 ± 9.7; time poststroke, 38.5 ± 25.5 months) and 16 age-matched controls (age, 65.0 ± 9.0 years).
Materials/Methods: Participants performed a right or left joystick movement with the right hand under 2 conditions. In the AS condition, participants moved right or left based on an abstract visual rule. In the execution only (EO) condition, participants moved in the same direction on every cue. After a practice period (3 blocks of 36 trials), the 2 conditions (AS, EO) were performed during functional magnetic resonance imaging in a 3T scanner.
Results: Response accuracy in the AS condition was >90% during functional magnetic resonance imaging. In both groups, reaction time (RT) in the AS condition was longer than in the EO condition (P < 0.001). Reaction time was longer in the stroke group but RT cost (AS RT–EO RT) did not differ between groups and did not correlate with degree of motor impairment. Strokes were both subcortical and cortical, but primary motor (M1) and PMd cortices were spared in all participants. Task performance across conditions and groups activated a motor network that included left M1, supplementary motor area, cingulate motor, and bilateral PMd, parietal cortex, and cerebellum. The 2 groups differed, however, in their neural response to the change in task condition. When contrasting the AS condition to the EO condition, the control group showed increased activation in the left inferior parietal lobule (IPL), while the stroke group increased activation in right hemisphere sites (PMd, anterior cingulate, IPL, visual cortex). Region-of-interest analyses confirmed these results; the stroke group but not the control group showed increased activation for the AS condition in right M1, PMd, and supplementary motor area.
Conclusions: Despite a similar RT cost in the stroke and control groups, the neural resources used to perform the AS task differed between groups. In the control group, the AS condition resulted in increased activation of the IPL, a region shown to play a role in visuomotor actions. In the stroke group, the AS condition resulted in increased activation of right, contralesional motor regions (M1, PMd) and areas that support visuomotor task performance (IPL, visual cortex). Engagement of the contralesional side after stroke may reflect compensatory activation to maintain task performance or a different phase of learning.
Clinical Relevance: Increased neural resource utilization in the contralesional hemisphere during AS after stroke may impact motor learning under complex conditions that require AS.
Dual-Task Walking Variability Relates to Stroop and Dual-Task Questionnaire Performance in Individuals With Multiple Sclerosis
N. Fritz, D. Kegelmeyer, D.S. Nichols-Larsen, A. Kloos, The Ohio State University, Columbus, Ohio.
Purpose/Hypothesis: Individuals with multiple sclerosis (MS) demonstrate cognitive dysfunction and gait impairments, which may increase fall risk. Although gait impairments in forward walking are well documented in MS, backward walking has not been examined. Cognitive deficits are associated with decreased performance on dual-task walking (eg, walking while talking) in older adults, but little is known about this relationship in MS. Therefore, we compared spatiotemporal gait measures during backward and dual-task walking between individuals with MS and healthy adults and examined the relationships between gait variability measures (ie, coefficients of variation [CVs]), neuropsychological measures, and Dual-Task Questionnaire (DTQ) scores in individuals with MS. We hypothesized that individuals with MS would demonstrate greater gait parameter variability during backward and dual-task walking than healthy adults and that greater variability would be related to poorer performance on neuropsychological tests and the DTQ.
Number of Subjects: Eleven individuals with relapsing remitting MS (mean age ± SD = 44.0 ± 9.4 [range 31-56] years, 11 females, EDSS mean ± SD = 2.8 ± 1.0 [range 1.5-4]) and 11 age- and gender-matched healthy adults (mean age ± SD = 43.7 ± 9.3 [range 31-55] years, 11 females) participated in the study.
Materials/Methods: Spatiotemporal gait measures and their CVs during forward, backward, and dual-task walking (walking with subtraction) were obtained using a GAITRITE system. Participants also completed neuropsychological tests of processing speed (Stroop and Symbol Digit Modality Test) and the DTQ, a subjective rating of dual-task ability.
Results: There were no significant differences between individuals with MS and healthy adults on any measures of backward or dual-task walking. Increased double support time CV in forward (ρ = −0.803, P = 0.005; ρ = −0.764, P = 0.006) and backward (ρ = −0.789, P = 0.005; ρ = −0.758, P = 0.011) dual-task walking conditions was correlated with poorer performance on the Stroop and Symbol Digit Modality Tests, respectively. Individuals who self-reported more difficulty with dual tasks (higher DTQ scores) demonstrated poorer performance on the Stroop (ρ = −0.803, P = 0.003) and increased double support time CV in backward (ρ = 0.843, P < .001) and backward dual-task walking (ρ = 0.754, P < 0.012). Poor Stroop performance was also associated with high backward dual-task gait variability; indeed 79% of Stroop test variance was explained by dual-task variability measures.
Conclusions: This study highlights the important relationship between mobility and cognition in individuals with MS. Poor performance on the Stroop and Symbol Digit Modality Tests and high DTQ scores were related to increased gait parameter variability in both backward and dual-task walking.
Clinical Relevance: Strong correlations between gait variability, neuropsychological test performance, and the DTQ suggest that increased gait variability may be related to cognitive and perceived dual-task deficits in MS.
Brief Kinesthesia Test Scores Are Poorer After Stroke
A. Borstad, Physical Medicine and Rehabilitation, The Ohio State University, Columbus, Ohio. D.S. Nichols-Larsen, School of Health and Rehabilitation Sciences, The Ohio State University, Columbus, Ohio.
Purpose/Hypothesis: Clinicians lack tools to quantify sensorimotor impairment following upper motor neuron damage. The Brief Kinesthesia Test (BKT), considered for inclusion in the National Institutes of Health Toolbox, was recently validated in healthy individuals across the life span. The objectives of this descriptive study were (1) to report BKT scores from a heterogeneous group of persons with chronic stroke and a control group matched for age, gender, and handedness; and (2) to examine the relationship between BKT and other valid sensorimotor measures.
Number of Subjects: Twenty-four.
Materials/Methods: A convenience sample of community-dwelling persons with chronic stroke (n = 12) were recruited for a one-time evaluation of sensorimotor function of both upper extremities. Control participants (n = 12) recruited by word of mouth and using Research Match, were matched within 3 years of age and for gender and handedness to chronic stroke participants. Criteria were a single clinical stroke diagnosed greater than 3 months ago, ≥10° active finger extension in the more involved hand. Participants were excluded if they had cognitive or language impairment that precluded following test directions. Outcome measures included the BKT, Wolf Motor Function Test, the Motor Activity Log (MAL), and the Box and Blocks Test.
Results: Poststroke participants performed significantly poorer than control participants on the BKT (t = 4.24; P < 0.001). Average deviation (mean, SD) was 14.2 (3.5) cm and 9.1 (2.0) cm, respectively. Impaired upper extremity BKT was strongly correlated with the MAL-how much (r = 0.84, P = 0.001), the MAL-how well (r = 0.76, P = 0.007), Wolf Motor Function Test (r = 0.80, P = 0.005), and the Box and Blocks Test (r = 0.77, P = 0.006).
Conclusions: The BKT was sensitive to kinesthetic differences between persons with stroke and matched controls. A strong relationship exists between valid upper extremity motor function measures and the BKT.
Clinical Relevance: The BKT is brief and freely available. Based on these preliminary data, the BKT may be a useful tool to screen for kinesthetic impairment in persons with stroke.