Cost-effectiveness analysis of overground robotic training versus conventional locomotor training in people with spinal cord injury.

BACKGROUND: Few, if any estimates of cost-effectiveness for locomotor training strategies following spinal cord injury (SCI) are available. The purpose of this study was to estimate the cost-effectiveness of locomotor training strategies following spinal cord injury (overground robotic locomotor training versus conventional locomotor training) by injury status (complete versus incomplete) using a practice-based cohort. METHODS: A probabilistic cost-effectiveness analysis was conducted using a prospective, practice-based cohort from four participating Spinal Cord Injury Model System sites. Conventional locomotor training strategies (conventional training) were compared to overground robotic locomotor training (overground robotic training). Conventional locomotor training included treadmill-based training with body weight support, overground training, and stationary robotic systems. The outcome measures included the calculation of quality adjusted life years (QALYs) using the EQ-5D and therapy costs. We estimate cost-effectiveness using the incremental cost utility ratio and present results on the cost-effectiveness plane and on cost-effectiveness acceptability curves. RESULTS: Participants in the prospective, practice-based cohort with complete EQ-5D data (n = 99) qualified for the analysis. Both conventional training and overground robotic training experienced an improvement in QALYs. Only people with incomplete SCI improved with conventional locomotor training, 0.045 (SD 0.28), and only people with complete SCI improved with overground robotic training, 0.097 (SD 0.20). Costs were lower for conventional training, $1758 (SD $1697) versus overground robotic training $3952 (SD $3989), and lower for those with incomplete versus complete injury. Conventional overground training was more effective and cost less than robotic therapy for people with incomplete SCI. Overground robotic training was more effective and cost more than conventional training for people with complete SCI. The incremental cost utility ratio for overground robotic training for people with complete spinal cord injury was $12,353/QALY. CONCLUSIONS: The most cost-effective locomotor training strategy for people with SCI differed based on injury completeness. Conventional training was more cost-effective than overground robotic training for people with incomplete SCI. Overground robotic training was more cost-effective than conventional training for people with complete SCI. The effect estimates may be subject to limitations associated with small sample sizes and practice-based evidence methodology. These estimates provide a baseline for future research.

Accuracy of self-reported severity and level of spinal cord injury.

STUDY DESIGN: Observational. OBJECTIVES: To assess accuracy of self-reported level of injury (LOI) and severity in individuals with chronic spinal cord injury (SCI) as compared with clinical examination. SETTING: An SCI Model System Hospital. METHODS: A 20-item survey evaluated demographics, physical abilities, and self-reported injury level and severity. A decision tree algorithm used responses to categorize participants into injury severity groups. Following the survey, participants underwent clinical examination to determine current injury level and severity. Participants were later asked three questions regarding S1 sparing. Chart abstraction was utilized to obtain initial injury level and severity. Injury level and severity from self-report, decision tree, clinical exam, and chart abstraction were compared. RESULTS: Twenty-eight individuals participated. Ninety-three percent correctly self-reported anatomical region of injury (ROI). Self-report of specific LOI matched current clinical LOI for 25% of participants, but matched initial LOI for 61%. Self-report of ASIA Impairment Scale (AIS) matched clinical AIS for 36%, but matched initial AIS for 46%. The injury severity decision tree was 75% accurate without, but 79% accurate with additional S1 questions. Self-report of deep anal pressure (DAP) was correct for 86% of participants, while self-report of voluntary anal contraction (VAC) was correct for 82%. CONCLUSION: Individuals with SCI are more accurate reporting ROI than specific LOI. Self-reported injury level and severity align more closely with initial clinical examination results than current exam results. Using aggregate data from multiple questions can categorize injury severity more reliably than self-report. Using this type of decision tree may improve injury severity classification in large survey studies.

Characterizing the Experience of Spasticity after Spinal Cord Injury: A National Survey Project of the Spinal Cord Injury Model Systems Centers.

OBJECTIVE: To characterize the qualities that individuals with spinal cord injury (SCI) associate with their experience of spasticity and to describe the relationship between spasticity and perceived quality of life and the perceived value of spasticity management approaches. DESIGN: Online cross-sectional survey. SETTING: Multicenter collaboration among 6 Spinal Cord Injury Model Systems hospitals in the United States. PARTICIPANTS: Individuals with SCI (N=1076). INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: Qualities of Spasticity Questionnaire, modified Spinal Cord Injury-Spasticity Evaluation Tool (mSCI-SET), and the modified Patient-Reported Impact of Spasticity Measure (mPRISM). RESULTS: Respondents indicated that spasms most often occurred in response to movement-related triggering events. However, spontaneous spasms (ie, no triggering event) were also reported to be among the most common types. Frequency of spasms appears to decline with age. The highest frequency of spasms was reported by 56% of respondents aged <25 years and by only 28% of those >55 years. Stiffness associated with spasticity was reported to be more common than spasms (legs, 65% vs 54%; trunk, 33% vs 18%; arms, 26% vs 15%). Respondents reported negative effects of spasticity more commonly than positive effects. Based on their association with negative scores on the mSCI-SET and the mPRISM, the 5 most problematic experiences reported were stiffness all day, interference with sleep, painful spasms, perceived link between spasticity and pain, and intensification of pain before a spasm. Respondents indicated spasticity was improved more by stretching (48%) and exercise (45%) than by antispasmodics (38%). CONCLUSIONS: The experience of spasticity after SCI is complex and multidimensional, with consequences that affect mobility, sleep, comfort, and quality of life. Stiffness, rather than spasms, appears to be the most problematic characteristic of spasticity. Physical therapeutic interventions to treat spasticity warrant in-depth investigation.

Predicting Duration of Outpatient Physical Therapy Episodes for Individuals with Spinal Cord Injury Based on Locomotor Training Strategy.

OBJECTIVE: To characterize individuals with spinal cord injuries (SCI) who use outpatient physical therapy or community wellness services for locomotor training and predict the duration of services, controlling for demographic, injury, quality of life, and service and financial characteristics. We explore how the duration of services is related to locomotor strategy. DESIGN: Observational study of participants at 4 SCI Model Systems centers with survival. Weibull regression model to predict the duration of services. SETTING: Rehabilitation and community wellness facilities at 4 SCI Model Systems centers. PARTICIPANTS: Eligibility criteria were SCI or dysfunction resulting in motor impairment and the use of physical therapy or community wellness programs for locomotor/gait training. We excluded those who did not complete training or who experienced a disruption in training greater than 45 days. Our sample included 62 participants in conventional therapy and 37 participants in robotic exoskeleton training. INTERVENTIONS: Outpatient physical therapy or community wellness services for locomotor/gait training. MAIN OUTCOME MEASURES: SCI characteristics (level and completeness of injury) and the duration of services from medical records. Self-reported perceptions of SCI consequences using the SCI-Functional Index for basic mobility and SCI-Quality of Life measurement system for bowel difficulties, bladder difficulties, and pain interference. RESULTS: After controlling for predictors, the duration of services for the conventional therapy group was an average of 63% longer than for the robotic exoskeleton group, however each visit was 50% shorter in total time. Men had an 11% longer duration of services than women had. Participants with complete injuries had a duration of services that was approximately 1.72 times longer than participants with incomplete injuries. Perceived improvement was larger in the conventional group. CONCLUSIONS: Locomotor/gait training strategies are distinctive for individuals with SCI using a robotic exoskeleton in a community wellness facility as episodes are shorter but individual sessions are longer. Participants' preferences and the ability to pay for ongoing services may be critical factors associated with the duration of outpatient services.

Modified PRISM and SCI-SET Spasticity Measures for Persons With Traumatic Spinal Cord Injury: Results of a Rasch Analyses.

OBJECTIVE: To evaluate the psychometric properties of the Spinal Cord Injury Spasticity Evaluation Tool (SCI-SET) and Patient-Reported Impact of Spasticity Measure (PRISM) using Rasch analysis to optimize their validity and efficiency. DESIGN: Rasch analysis of the SCI-SET and PRISM represents a secondary analysis of data collected as part of a collaborative research project of the SCI Model Systems Centers. The overall survey was organized into 4 sections: (1) participant demographics and injury characteristics, (2) participant experiences of spasticity, (3) SCI-SET, and (4) PRISM. Participants were recruited from the community via multiple avenues. Data were collected and managed via an online survey tool using a secure web-based data management application. SETTING: Participating Spinal Cord Injury Model Systems Centers. PARTICIPANTS: Most participants (N=1239) had lived with their injury for more than 2 years and used a wheelchair as their primary mode of mobility. The majority of the sample (58%) sustained cervical injuries. INTERVENTIONS: None. MAIN OUTCOME MEASURES: SCI-SET and PRISM. RESULTS: The SCI-SET demonstrated strong measurement properties with acceptably high reliability and point-measure correlations and no evidence of multidimensionality. However, respondents underused some rating scale categories. Analyses of the PRISM demonstrated 3 distinct subscales relating to the physical, psychological, and social influences of spasticity; respondents underused some rating scale categories. Combining underused rating scale categories for both spasticity instruments resulted in increased reliability and reduced respondent burden compared with the original versions. Both the Modified SCI-SET (person separation reliability=0.93) and Modified PRISM (person separation reliability=0.85, 0.89, 0.83 for physical, psychological, and social subscores, respectively) display strong measurement properties. CONCLUSIONS: Measurement properties of the SCI-SET and PRISM improved from use of Rasch model methods. The SCI-SET required minor revisions, whereas the PRISM required definition of subscores. Both modified spasticity measures demonstrated adequate psychometric properties, and correlations among the modified measures were high, providing evidence of convergent validity. We recommend use of the Modified SCI-SET and Modified PRISM measures in future studies.

Experience of Robotic Exoskeleton Use at Four Spinal Cord Injury Model Systems Centers.

BACKGROUND AND PURPOSE: Refinement of robotic exoskeletons for overground walking is progressing rapidly. We describe clinicians' experiences, evaluations, and training strategies using robotic exoskeletons in spinal cord injury rehabilitation and wellness settings and describe clinicians' perceptions of exoskeleton benefits and risks and developments that would enhance utility. METHODS: We convened focus groups at 4 spinal cord injury model system centers. A court reporter took verbatim notes and provided a transcript. Research staff used a thematic coding approach to summarize discussions. RESULTS: Thirty clinicians participated in focus groups. They reported using exoskeletons primarily in outpatient and wellness settings; 1 center used exoskeletons during inpatient rehabilitation. A typical episode of outpatient exoskeleton therapy comprises 20 to 30 sessions and at least 2 staff members are involved in each session. Treatment focuses on standing, stepping, and gait training; therapists measure progress with standardized assessments. Beyond improved gait, participants attributed physiological, psychological, and social benefits to exoskeleton use. Potential risks included falls, skin irritation, and disappointed expectations. Participants identified enhancements that would be of value including greater durability and adjustability, lighter weight, 1-hand controls, ability to navigate stairs and uneven surfaces, and ability to balance without upper extremity support. DISCUSSION AND CONCLUSIONS: Each spinal cord injury model system center had shared and distinct practices in terms of how it integrates robotic exoskeletons into physical therapy services. There is currently little evidence to guide integration of exoskeletons into rehabilitation therapy services and a pressing need to generate evidence to guide practice and to inform patients' expectations as more devices enter the market.Video Abstract available for more insights from the authors (see Video, Supplemental Digital Content 1, available at: http://links.lww.com/JNPT/A231).