Abstract
Introduction
1Department of Health Sciences and Research, Medical University of South Carolina; NIH/NICHD/NCMRR Medical Rehabilitation Research Resource Network National Coordinating Center, Charleston, South Carolina
The Medical Rehabilitation Research Resource (MR3) Network is supported by the National Institutes of Health through National Center for Medical Rehabilitation Research (NCMRR) as part of the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD). The MR3 Network comprises six national rehabilitation research resource centers that provide infrastructure and access to expertise, technologies, and resources to foster clinical and translational research in medical rehabilitation. The centers offer expertise across the translational spectrum from basic science to implementation science, including: regenerative rehabilitation, neuromodulation, pediatric rehabilitation, technology for real-world assessment, translational rehabilitation, and learning health systems research.
One primary role the MR3 National Coordinating Center and the six centers is to organize annual scientific retreats addressing critical areas in medical rehabilitation research. The Second Annual MR3 Network Scientific Retreat was entitled “Rehabilitation Clinical Trials: Innovations, Designs, and Tribulations,” and was presented over two days on September 29-30, 2022, in a virtual format. The planning committee issued a national call for abstracts from which the following themes emerged, and the retreat agenda was organized: novel trial design, barriers and difficulty of success, outcomes, student blitz, ethics and diversity, telerehabilitation, and adaptive trial design. The retreat was extraordinarily well attended, and we think provided insights into emerging best practices, important scientific questions needing further investigation, avenues of collaboration among colleagues, and a reminder of the vast array of research being conducted under the medical “rehabilitation” umbrella.
Day 1 Abstracts
Novel Trial Design
Distinctive Challenges in Design and Data Analysis of Pediatric Rehabilitation Trials
1Virginia Tech. 2The University of Virginia. 3The Ohio State University
Rehabilitation science builds on both theory and empirical evidence from case histories, pilot studies, and basic research. There are many important issues for determining success of a pediatric clinical trial including how to determine what is clinically and personally ‘meaningful change’. The leadership team of investigators at C-PROGRESS have worked collaboratively on 5 multi-site randomized clinical trials, including an ongoing Phase 3 RCT. Using an interactive symposium format, we will outline key issues in studying treatment efficacy and effectiveness in pediatric populations. These include: issues of disaggregating chronological age and maturation from treat-induced changes; lack of continuity/isomorphism in outcome measures across ages; well-documented multi-domain changes beyond targeted primary outcomes; high levels of intra- and inter-subject variability (impacting precision of measurement and power in clinical trials); determining ’clinical significance’ or meaningful clinical changes; role of parent or child subjective reports; and complexity of healthcare issues and combined treatments; understanding multiple endpoints; determining composite(profile) outcomes in treatment-induced changes; and cross-study comparisons. A new issue in current pediatric trials concerns both equipoise and the changing landscape of ‘usual and customary’ care for particular clinical populations. We will focus on proposed solutions and invite active exchange of information with conference participants. Based on participant interest and contributions, we will produce a set of references and offer individualized follow-up for those currently engaged in pediatric clinical trials through C-PROGRESS supports.
We Found a Way: Conducting and Modifying an In-person Randomized Controlled Trial for Vestibular Disorders During the COVID-19 Pandemic in New York City
1New York University. 2New York Eye and Ear Infirmary of Mount Sinai. 3Icahn School of Medicine at Mount Sinai
We created a clinical virtual reality application for a Head Mounted Display to provide contextual sensory integration (C.S.I.) for patients with vestibular dysfunction in a functional and non-threatening context. We began a pilot randomized controlled trial in Fall 2019 to compare C.S.I. training to traditional vestibular rehabilitation. The study was designed as a collaboration between a university research lab (for assessments) and hospital clinic (for both interventions). The protocol included 2 pre assessments, 8 intervention sessions over 8 weeks and 1 post assessment. We recruited 9 patients pre pandemic. Two completed (1 traditional, 1 app), 1 dropped out (traditional) and 6 had to stop because the study shut down in March 2020. In-person research was paused for 6 months at our hospital and most of our clinical team was deployed to treat covid patients. In September 2020 the hospital allowed for in-person research to resume while the university research lab was still closed. We decided to adapt the original protocol to the new circumstances. The trial was completed in April 2022. We were able to recruit 20 additional patients. There were no differences between patients pre and post covid, no difference between patients who dropped out or completed and no difference between group in attrition %. In addition to the trial results, in this presentation we will discuss all trial’s modifications including changes in baseline and post assessments from lab to clinic, changes in blinding of assessors, allowing breaks in the protocol when patients needed to quarantine etc.
Application of Hybrid Effectiveness-Implementation Designs to Bridge the Research-Practice Divide for mRehab Interventions
1University of Illinois at Chicago. 2Northwestern Illinois University. 3Northwestern University
Mobile rehabilitation (mRehab) interventions hold great potential to close the access gap in much needed rehabilitation medicine services, yet organizational, clinical, and consumer barriers to implementation persist. Given the pace of technological innovation, there is a critical need to bridge the divide between mRehab interventions created by researchers and those in which patients and professionals are using in practice. The traditional research pipeline that encourages a staged approach to moving an intervention from efficacy trials to the real world can take a long time. To address this issue, hybrid effectiveness-implementation designs were developed to promote the intentional examination of effectiveness and implementation outcomes within a single study. This presentation will explore the range of hybrid designs available to rehabilitation researchers with illustrative mRehabilitation exemplars. There are three types of hybrid designs, all of which vary in their primary focus and relative emphasis on effectiveness versus implementation outcomes. A type 1 hybrid focuses primarily on the effectiveness outcomes of an intervention while exploring its “implementability.” A type 2 hybrid design has a dual focus on effectiveness and implementation outcomes by allowing for the simultaneous testing or piloting of implementation strategies during an effectiveness trial. Finally, a type 3 hybrid design focuses primarily on implementation outcomes while also collecting effectiveness outcomes related to the intervention’s uptake or fidelity. Hybrid effectiveness-implementation designs may help realize the democratizing potential of mRehab by getting evidence-based interventions into the hands of the people who can most benefit from them when they need them most.
Barriers and Difficulty of Success
Standardization to Achieve Optimal Data Quality in Multi-Site Rehabilitation Clinical Trials
1Georgia State University. 2Medical University of South Carolina. 3Duke University. 4UMCMS-Baystate. 5University of Cincinnati. 6National Institute of Neurological Disorders and Stroke. 7University of South Carolina. 8Emory University
While the standardization of a clinical trial protocol, investigational treatment, and data collection procedures are important in all clinical trials, unique challenges emerge in the implementation of large-scale multi-site rehabilitation clinical trials. TRANScranial direct current stimulation for Post-stroke mOtor Recovery (TRANSPORT2) is an ongoing phase II multi-site clinical trial (NCT03826030) with the primary goal of determining whether there is an initial overall treatment effect in three dosing groups: sham tDCS + modified Constraint-Induced Movement Therapy (mCIMT), 2 mA tDCS + mCIMT, and 4 mA tDCS + mCIMT. The TRANSPORT2 investigators have implemented a number of protocol standardization and quality of control strategies with the goal of minimizing errors in protocol execution, therapy administration, and outcome collection and interpretation at the 12 clinical sites participating in the trial. These strategies include:
-A research training study in which each site enrolls training participates in order to standardize mCIMT, outcome assessments, and tDCS prior to enrolling subjects into the TRANSPORT2 trial
-Ongoing demonstration of fidelity of treatment and evaluation to the central cores after every fourth subject is enrolled at a particular site
-Centrally collected and scored primary outcome measure (Fugl Meyer - Upper Extremity)
-Centrally collected and analyzed exploratory outcome measures (TMS and MRI)
These standardization strategies, and lessons learned through implementation, will be discussed.
Multi-Site Clinical Trials: Treatment Fidelity and Beyond
1Louisiana State University Health Sciences Center. 2Seattle Children’s Research Institute
Multi-site clinical trials are becoming more and more prevalent in rehabilitation research as a way to improve diversity, increase recruitment numbers, and improve generalizability, among others. In this talk, the authors will draw on their experiences from conducting multi-site rehabilitation clinical trials and discuss the intersections between fidelity, rigor, and reproducibility. Treatment fidelity and assessment or outcomes fidelity are both important components of conducting a successful clinical trial. When interpreting the results of a clinical trial, it is important to examine the treatment fidelity to uncover the extent to which the participants received the intended dosage for the intervention. The 5 domains of fidelity (study design, training, delivery, receipt, and enactment) will be discussed with an emphasis on training, delivery, and receipt. Further, the importance of documenting the intervention-specific fidelity measures beyond a single metric, such as attendance, will be emphasized. An example from a resistance training intervention will be provided in order to illustrate the adequate documentation and reporting of the delivery of all key dosing parameters (frequency, duration, volume, intensity, and velocity of movement) and the receipt based on progressions in training load/intensity. A successful secondary analysis can then determine which parameters were the key ingredients associated with improved outcomes. Treatment fidelity will be expanded to include individualized dosing and progression to account for subject heterogeneity. Examples of successful decision-tree algorithms for progression based on individualized subject responses from two on-going multi-site clinical trials will be shared with the audience.
Participant Retention, Compensation, and Compliance during Knee Joint Immobilization Research: An Analysis of Two Clinical Trials
1University of Central Florida. 2Oklahoma State University. 3Pfeiffer University 4Syracuse University
Investigators studying muscle disuse often rely on joint immobilization models because they provide excellent internal validity. However, asking healthy adults to place a limb in a brace or cast and refrain from weightbearing to induce weakness presents unique challenges. Our laboratory has completed two knee immobilization clinical trials to study neuromuscular plasticity. Methods: In Study 1 (2017-2018), we recruited females for a two-week immobilization study. Laboratory testing was conducted 48 and 72 hours, as well as one and two weeks, following immobilization. Compensation for completing Study 1 was $350 (Visa/Mastercard). In Study 2 (2021-2022), we recruited males and females for a one-week study that featured neural and muscular interventions, as well as a post-immobilization rehabilitation program. Compensation for completing Study 2 was ≤ $220 (Amazon) and a t-shirt. In both studies, compliance was monitored via accelerometers. A researcher was responsible for checking in with participants via daily phone calls and/or text messages. Results: Twenty-six females enrolled in Study 1, but 11 withdrew (57.7% completion). Reasons for withdrawing included study duration, medication changes, difficulty accomplishing daily activities, and brace discomfort. Fifty participants enrolled in Study 2 (28 males, 22 females), but five males and three females withdrew (84% completion). Reasons for withdrawing from Study 2 only included discomfort with laboratory tests (electrical stimulation and transcranial magnetic stimulation) and scheduling conflicts. Conclusion: Study duration has a substantial impact on completion rates of knee immobilization studies. We recommend ≤ one-week studies, fostering a supportive environment for participants, and providing ~$200/week compensation and t-shirts.
Outcomes
A Story of Those Who Get Left Behind
1Cleveland Clinic Foundation. 2Kessler Institute of Rehabilitation. 3Louis Stokes Cleveland VA Medical Center. 4University of Texas. 5MetroHealth Center for Rehabilitation Research
There is a tremendous burden of severe UE paresis after stroke and spinal cord injury (SCI). Persons with severe paresis are typically excluded from RCTs because interventions cannot be made feasible for them, and outcome assessments are not designed to accommodate significant impairment. Unfortunately, this perpetuates the (vicious) cycle of disability- because persons with severe paresis are excluded, what works and does not work for them cannot be determined. Here we offer a window into challenges and successes of studying patients with severe paresis from ongoing single- and multi-site RCTs in stroke and cervical SCI (NIHR01HD098073, SCIRPW81XWH1810530) to develop a roadmap for future. Topics will include: 1) scientific challenges in identifying treatments specific to mechanisms of and feasible to administer among persons with severe paresis; 2) delivery of rehabilitation tailored in content and length to meet unique challenges posed by severe paresis; 3) outcomes assessments guided by their ability to identify impairments/limitations in the presence of severe weakness, feasibility to administer without exaggerating fatigue/burden, and sensitivity to capture small, albeit meaningful, effects; 4) patient expectations, where in the absence of distal improvements statistically significant gains seem meaningless and clinically-meaningful gains calibrated to severity are unknown; 5) co-morbidities and worsening effects of chief diagnoses; 6) familial/personal burden worsened dramatically for those with severe paresis and 7) logistics of travel and caregiver availability that ensure gainful participation. A dedicated and long-standing commitment to evaluate treatments in this disadvantaged population is necessary to overcome one of the most complex challenges facing rehabilitation research.
Beyond the Clinic: Measuring Walking Performance and Mobility in Daily Life
1Seattle Children’s Research Institute. 2University of Washington. 3Urban Farm Lab. 4Louisiana State University
Walking activity is known to be predictive of participation in mobility, education, and social relations for children with cerebral palsy (CP) per cohort and population-based studies. Traditionally, clinic or laboratory-based outcome measures of walking activity have been implemented to measure the effectiveness of rehabilitation interventions to optimize walking and community mobility. However, these outcomes do not provide knowledge of translation or carryover to walking in natural settings. Synchronizing accelerometry data with global positioning system (GPS) data allows the combined measurement of temporally explicit stride counts and rates with the precise location of the walking activity. This novel approach represents a unique departure from the status quo by using wearable sensors to directly measure walking and mobility-based participation in the real-world (e.g., home, school, park). Community walking activity levels and intensity of children with CP relative to a typically developing cohort will be presented. We will describe a process for synchronizing accelerometry-based walking and GPS-based location information by matched time stamps. This temporally and spatially precise walking information will be reported as an outcome of interventions to enhance walking (i.e., orthotics, home-based treadmill training) in children with CP through exemplar group and individual research data (i.e., daily Google Map). The novel combination of sensor and monitoring technology has potential to inform the documentation of rehabilitation strategy effectiveness in the context of daily life.
OpenCap: 3D Human Movement Dynamics from Smartphone Video
(*Authors contributed equally)
1Stanford University
Measures of human movement dynamics, such as loading of bones and muscles, are sensitive biomarkers of neurological and musculoskeletal disease status and progression. However, movement dynamics are rarely quantified in large-scale clinical trials due to the prohibitive cost, time, and expertise required to make these measurements in a motion capture laboratory. Instead, functional endpoints for clinical trials typically comprise general movement evaluations that are evaluated visually or with a stopwatch. The lack of fast, precise, and repeatable functional biomarkers limits the statistical power of outcomes. To address these limitations, we developed OpenCap, an open-source platform for evaluating movement dynamics in minutes using two smartphones. OpenCap’s web application enables users to collect synchronous videos and visualize movement data that is automatically processed in the cloud, thereby eliminating the need for specialized hardware, software, and expertise. Through a validation study comparing OpenCap to gold-standard laboratory techniques, we show that OpenCap estimates muscle activations, joint loads, and joint moments with sufficient accuracy to inform research and clinical decisions for applications including osteoarthritis, age-related muscle weakness, and post-surgical functional recovery. Additionally, we demonstrate OpenCap’s practical utility through a field study in which a clinician estimated movement dynamics of 100 individuals in real-world settings for about 300 times less money and in about 25 times less time than was previously possible. By reducing the barriers to evaluating human movement dynamics, OpenCap facilitates the development of more sensitive functional endpoints for rehabilitation clinical trials and the incorporation of these measures into clinical practice.
Student Session: Idea Blitz
Utilization of EPIC Slicer Dicer in Achilles Tendinopathy - A Recruitment Case Study
1New York University
The use of electronic health record (EHR) tools such as Slicer Dicer has substantial potential to improve clinical trials planning. Even though musculoskeletal diseases account for more than 50% of disabling health conditions, there is limited literature exploring the use of these tools in musculoskeletal rehabilitation. The purpose of this study is to examine potential utility of using Slicer Dicer for cohort exploration in Achilles Tendinopathy (AT). Methods: Using EPIC, we searched over 350 sites affiliated with NYU Langone Health using ICD-10-based diagnoses. Results were sliced by year, provider, age, race, comorbid conditions. Results: Over a 12 month period (2021-present), 6310 patients were diagnosed with AT. Number of patients dropped to 5347 in 2020 due to the COVID-19 pandemic, but prior to that the number of patients had risen from 662 in 2010 to 5458 in 2019. 78.5% were within 18-70 years, 18.28% were >70 years. First encounter ranked by provider department indicated that 53.4% presented to internal medicine and 15.2% presented to orthopedic surgery. 59.6% were white, 11.3% were African American and 0.03% were Asian. 30% of patients had primary hypertension, 25% had hyperlipidemia, 14% had osteoarthritis, 10.2% had Type 2 Diabetes Mellitus. Conclusion: At our academic medical center, the number of AT diagnoses has steadily increased, keeping with the growth in ambulatory care sites. Slicer Dicer is a powerful tool for cohort exploration with significant potential to reduce recruitment bias by allowing for recruitment across multiple geographic locations and providers.
Changes in the Fronto-central Brain activity for Balance Control in Chronic Stroke Survivors
1University of Houston. 2Yale University
Nearly 50% of stroke survivors experience a fall within 6-12 months after discharge from hospital, which leads to significant complications and financial burden on the society. Balance control is a key indicator of mobility and independence in ADLs, and its impairment is an important factor contributing to falls in stroke patients. Therefore, it is important to understand the mechanisms underlying impaired balance control following stroke. Remarkably, the contribution of cortical reorganization following stroke to impaired balance control remains unknown. In this study, we investigated the changes in activation over balance related brain areas in individuals with chronic stroke. We have recruited 7 stroke patients with mild-to-moderate severity and 4 age/gender matched healthy adults. Clinical assessment was performed using Berg Balance Scale and Time Up and Go tests. Participants performed a standing balance task in presence of balance perturbations with simultaneous neuroimaging using electroencephalography (EEG). On the clinical tests and the laboratory-based balance task, stroke patients showed poor performance when compared with healthy controls. Our preliminary EEG findings suggest differences in Delta and Theta band EEG power over the frontocentral region between patients and healthy controls during the balance task. These findings suggest changes in the activation within frontocentral brain areas during a balance task following stroke. Ongoing work is investigating the changes in functional coupling between the frontocentral activity and lower limb muscle activity during the balance task in stroke patients. These results will aid in the identification of targets for neurorehabilitation of balance control in stroke patients.
Regenerative Rehabilitation for Enhancing Muscle Recovery Following Volumetric Muscle Loss
1Saint Louis University
Volumetric muscle loss (VML) injury causes irreversible deficits in muscle mass and function and often results in permanent disability. The current standard of care is physical therapy but it is limited in mitigating functional deficits. We have previously optimized a rehabilitation technique using electrically stimulated eccentric contraction training (EST) that improved muscle mass, strength, and size in VML injured rats. A biosponge scaffold composed of extracellular matrix proteins has previously enhanced muscle function post-VML. This study aimed to determine whether combined application of a regenerative therapy (i.e., biosponge) with a novel rehabilitation technique (i.e., EST) could enhance muscle mass and strength recovery in a rat model of VML. A VML defect was created by removing ~20% muscle mass from the tibialis anterior muscle in adult male Lewis rats. Experimental groups included VML injured rats treated with biosponge+EST or biosponge alone (n=6). EST was implemented 2 weeks post-injury and was continued for 4 weeks. A repeat-bout effect was observed with a linear increase in eccentric torque over 4 weeks. The combined application of biosponge+EST improved muscle mass by ~10% (NS) and peak isometric torque by ~52% (p=0.0411) compared to biosponge treatment alone at 6 weeks post-injury. Qualitative and quantitative analysis of muscle cross-sections also suggests improvements in muscle structure; specifically in mean cross-sectional area of fast twitch (Type 2B) myofibers. Thus far, these findings show the potential for a combined regenerative and rehabilitative therapy to improve muscle recovery following VML. Ongoing work will determine the extent of immunomodulation, angiogenesis, regeneration and fibrosis through additional histology and gene expression analysis.
Day 2 Abstracts
Novel Trial Design
Assessing Organizational and Provider Context and Readiness to Inform National, Multi-site Pragmatic Trials: Methods and Lessons Learned
1University of Pittsburgh. 2University of Colorado. 3UPMC. 4Encompass Health
Implementation of standardized, evidence-based interventions in rehabilitation settings requires a careful examination of organizational and provider context and readiness. Guided by the Consolidated Framework for Implementation Research, we conducted a two-phase cross-sectional descriptive study, in partnership with a large for-profit national rehabilitation corporation. In phase I, we characterized variations among 139 inpatient rehabilitation facilities with respect to geographic region, size, organizational structure, staffing, operations, and clientele using a combination of surveys and electronic health record information. Our team of scientists, corporate leaders, rehabilitation providers, and patient advocates then used nominal group technique to select 30 facilities with sufficient variation in region, size, and clientele to participate in phase II, oversampling facilities serving high proportions of patients from historically marginalized race and ethnicity groups. In Phase II, we recruited the director of therapy operations and six rehabilitation providers (i.e., occupational, physical, and speech therapy) from each of the 30 facilities to examine readiness for intervention implementation using convergent mixed methods. Participants completed validated surveys assessing attitudes and behaviors related to evidence-based practice as well as organizational culture, implementation climate, and readiness for change. Participants then engaged in focus groups to discuss barriers and facilitators to implementation of an efficacious intervention that shows promise for reducing disability among people with cognitive impairments in this setting. Using the findings and the Expert Recommendations for Implementing Change Strategy Matching Tool we identified a core set of implementation strategies to be examined in a future hybrid I effectiveness-implementation pragmatic trial.
Moving Infant Rehabilitation Forward Through Rigorous Research Design: Moving from Single Site Single Variable Trials to Multi-arm, Multi-phase Intervention
1University of Southern California. 2University of Virginia. 3Virginia Commonwealth University
Advances in life saving procedures for infant born preterm, international guidelines on early detection of cerebral palsy and US federal education policy all support the need to start intervention to prevent or ameliorate disability early. However, intervention is often delayed months or years due to complicated policy, lack of access, and limited information on why early matters. Combined with a lack of efficacy data from large scale efficacy trials pediatric and infant rehabilitation is being left behind. We are not generating research fast enough to serve our youngest clients in need of habilitation services. Clinical trials that have been completed are often years away from effectiveness or implementation research due to the need for further testing of dose, timing, or other parameters. This presentation will highlight the use of a 3 arm clinical trial to address the question of efficacy and timing of intervention for infants born very preterm during the transition from NICU to Home. The speaker will pose important questions on how the field of rehabilitation can move forward. A comparison of the pros and cons of multiple serial clinical trials prior to implementation vs. multi-arm, multi-phase interventions trial will support the need for alternative designed. While testing multiple factors in a single study designs may require larger samples and will cost more per study, they are needed to move the field toward implementation of effective rehabilitation or habilitation.
Responder Analysis vs Mean Change Analysis in Stroke Rehabilitation Clinical Trials
1Medical University of South Carolina. 2Duke University. 3UMCMS-Baystate
Planning stroke recovery clinical trials requires nontrivial consideration of the most appropriate design to determine whether or not a clinically meaningful treatment effect exists. The stroke recovery community often focuses on mean changes on a continuous outcome, for example the Fugl Meyer - Upper Extremity (FM-UE), to measure treatment effect. Responder analysis, in which a continuous or ordinal outcome is dichotomized to define “responders” and “non-responders”, has emerged as an alternative to assess clinical relevance. In responder analysis, the definition of responder can be consistent for all subjects or baseline severity can be incorporated. By using a so called “sliding dichotomy”, the definition of responder varies according to baseline severity which may be desirable if criterion for success differs by prognosis. Responder analysis may be favored by clinicians as it can be more straightforward for individual patients to interpret and understand the risk and benefit of treatment. While this approach has gained popularity amongst clinicians, it has been criticized by statisticians largely due to cost in efficiency. The benefits and pitfalls of responder analysis will be discussed.
Ethics and Diversity
Recruiting Under-Represented Minorities (U-RM) for Rehabilitation Research
1The Ohio State University. 2Nationwide Children’s Hospital. 3Virginia Tech. 4Fralin Biomedical Research Institute
Ethical Considerations in Neurorehabilitation Clinical Trials: Emerging Principles & Priorities
1Massachusetts General Hospital
As the rehabilitation clinical trial landscape for patients with brain injuries and other neurological disorders expands, consideration of associated ethical dimensions is of mounting importance. Here I assess central ethical considerations in rehabilitation clinical trials involving participants with neurological disorders, including (1) autonomy, respect for persons and informed consent of individuals with neurological disorders; (2) balancing unknown benefits and risks in the especially sensitive context of novel neurointerventions to promote rehabilitation (3) disclosure to surrogates and clinical teams of investigational results pertaining to recovery and neuroprognosis; (4) justice and equity considerations, including fair access to rehabilitation clinical trial enrollment across patient communities and backgrounds; (5) post-trial responsibilities and data handling. Guiding principles and research opportunities for rehabilitation clinicians, researchers, ethicists, and other scholars engaged in rehabilitation clinical trials are described to advance ethical study design and deployment in this uniquely complex yet vital area of study.
Stakeholder Engagement and Integration in Clinical Trials
1The Ohio State University. 2I-ACQUIRE Parent Council. 3Fralin Biomedical Research Institute, Virginia Tech. 4Nationwide Children’s Hospital
Clinical trials in rehabilitation present unique challenges and opportunities with interventions that are often complex and time-intensive. Researchers must collaborate with prospective participants, families, and caregivers to recruit, accrue, and retain study participants. When a study team includes both researchers and participant stakeholders as integral members, vital and prolonged partnerships form that ensure participant-centered research processes and outcomes and enrich the research experience for all parties. The C-PROGRESS leadership team has engaged with families in multiple pediatric clinical trials, culminating in the integration of a Parent Advisory Council in an ongoing multisite, phase 3 randomized controlled trial. In an interactive, discussion-based session, our team of parents and scientists will describe the distinct advantages and common challenges when integrating a parent advisory council into trial design and implementation. Topics include: timeline (e.g. at which research stage to initiate stakeholders-scientist partnership), roles and responsibilities (e.g. consent & enrollment recruitment training), communication (e.g. participation in executive steering committee investigator liaisons), scientific integrity (e.g. bias, equipoise), parent-reported and family-centric outcomes (e.g. family feedback and family input into relevant and meaningful outcomes), and family-centered procedures (e.g. respectful communication sensitivity to family needs and experiences). In addition, we will share recommendations based on experience with integrating stakeholder councils into grant submissions, structuring the council, establishing and funding council payments, designing systems of communication and feedback, and navigating regulatory and compliance issues. Following the session, individuals interested in further information can access C-PROGRESS investigators for consultation about integrating stakeholder advisory groups into their clinical trials.
Telerehabilitation
Overcoming Barriers During COVID-19: A Completely Virtual Tele-Exercise Intervention Study for Adults with Chronic Neurological Impairments
1Burke Neurological Institute. 2State University of New York Upstate Medical University. 3Columbia University Irving Medical Center. 4Weill Cornell Medicine
Exercise is a critical component of a healthy lifestyle, yet many with chronic neurological impairments (CNI) do not regularly exercise. Reasons include inaccessibility of gyms and equipment, transportation, and costs. COVID-19 magnified these barriers. Study goal: to feasibly provide seated, home-based exercise to people with CNI using a completely virtual study design. Participants were screened and consented over Zoom. Intervention: 12-week, 3x/week seated exercise program. One group (synchronous, n=33) attended classes at a specified time, on Zoom with instructor and classmates. Classes were recorded. The other group (asynchronous, n=30) was given recorded videos to complete on their own time. Participants were mailed a blood pressure (BP) monitor and Polar OH1 heart rate (HR) recording device that synched to a smartphone app. The study team downloaded HR data from the Polar website. Before and after each class, participants completed surveys on REDCap that queried pain level, BP, exertion, motivation, and satisfaction. Before the first session, midpoint, after the last session, and one month later, participants completed a Reasons for Exercise Inventory, Physical Activity Log, Perceived Wellness Survey, Physical Activity Enjoyment Scale (PACES), and SF-36v2, on REDCap virtually. Participants reported satisfaction with the study. Percent of participants satisfied or extremely satisfied with: use of OH1 (90%), use of Zoom (84%), ease of surveys online (97%), indicating strong feasibility of the protocol. Both groups improved in PACES (p=0.02), which is a measure of physical activity enjoyment. These findings demonstrate that virtual study designs can be feasible and enjoyable for people with CNI.
Novel Clinical Trial of Portable Near-Infrared Spectroscopy-based (fNIRS) Brain Computer Interface to Augment Upper Extremity Motor Recovery after Stroke
1Harvard Medical School. 2Massachusettes General Hospital. 3Department of Veteran Affairs Medical Center. 4Northeastern University. 5Axem Neurotechnology Inc. 6Laboratory for Brain Recovery and Function
Post-stroke rehabilitation has been shown to be beneficial in helping to restore stroke survivors’ movement abilities and improve their level of function. Rehabilitation practice is increasingly in need of more quantifiable and real-time measures of therapeutic interventions to optimize treatment approaches and facilitate motor recovery processes. However, most technologies capable of producing neural biomarkers are currently limited to the laboratory due to size, cost and complexity of operation (e.g., functional magnetic resonance imaging). There is indeed increasing evidence that high dose, individualized, engaging therapy maximizes motor recovery. However, stroke survivors often do not receive adequate rehabilitation services due to challenges including transportation and financial burden-issues exacerbated in rural areas where access to specialized outpatient rehabilitation may be inadequate or non-existent. Solutions to post-stroke rehabilitation that incorporate portable neurophysiological measurement technologies and telehealth strategies to maximize practice are critically needed to facilitate best possible motor recovery. Here we describe a novel pilot clinical trial of telerehabilitation paired with a portable near-infrared spectroscopy (fNIRS) biofeedback system for at-home upper extremity motor rehabilitation following stroke. Our rehabilitation intervention provides a combination of asynchronous app-based exercise sessions guided by fNIRS-generated neural biomarker data and synchronous therapist-led telerehabilitation sessions. We discuss the rationale and design of our clinical trial, how fNIRS signals are integrated for participants and therapists, and selection of our study outcome measures. We describe experiences with the first cohort of participants in this ongoing study, and highlight early lessons learned in designing and deploying this novel clinical trial.
Innovative Telerehabilitation Model for Slowing Disease Progression in Amyotrophic Lateral Sclerosis (ALS) Using Non-invasive Brain Stimulation
1University of Illinois at Chicago
Innovative telerehabilitation model for slowing disease progression in amyotrophic lateral sclerosis (ALS) using non-invasive brain stimulation ALS is the third most common adult-onset neurodegenerative disease with no known cause or cure. Current treatment is based primarily on symptom management and palliative care, making rehabilitation an integral part of clinical management. Although symptom management has improved in the recent years, there is lack of evidence-based clinically meaningful therapies. Recent advances in ALS research have reinforced the pathology of ALS as primarily a disease of the cerebral cortex. We proposed a novel approach that targets hypoexcitable cortical circuits using remotely supervised non-invasive facilitatory transcranial direct current stimulation (tDCS). Telerehabilitation is a relatively new approach to broaden access to rehabilitation services in a cost-effective manner, particularly for those with limited mobility and transportation barriers. In this symposium, we will present our NIH-funded clinical trial protocol to determine safety and feasibility of long-term treatment (6-months, 3 times/week) with remotely supervised tDCS in persons with ALS. We will outline procedures for training staff and patients in tDCS administration and remote supervision, actively monitoring patient safety, and share barriers and opportunities to participation in telerehabilitation research for persons living with neurodegenerative conditions. Pilot data from 5 participants enrolled in the study will be presented. The current study is the first to examine remotely supervised tDCS in persons with ALS. Successful implementation of a remotely supervised protocol will broaden access to research participants and healthcare by providing therapy in the comfort of one’s own home and enable the intensive dosages of treatment that is needed to slow disease progression in persons with ALS.
Adaptive Trials Design
Improving functional outcomes in skilled nursing facilities nationwide: a hybrid I effectiveness-implementation pragmatic clinical trial approach
1VA Eastern Colorado Health Care System. 2University of Colorado Anschutz Medical Campus. 3Rocky Mountain Regional VA Medical Center. 4Aegis Therapies
Each year, approximately 1.35 million patients receive skilled nursing facility (SNF) rehabilitation to address hospitalization-associated functional deficits. In SNFs, “usual care” rehabilitation consists of low-intensity interventions, which are often physiologically inadequate to induce improvements in muscle strength and function. In published work, a high-intensity resistance training program (i-STRONGER) resulted in better physical function, increased community discharge rates, and reduced lengths-of-stay among patients receiving SNF care. To further evaluate the role of i-STRONGER in optimizing SNF rehabilitation, we will use a pragmatic clinical trial design to compare patient outcomes between 16 SNFs utilizing i-STRONGER principles and 16 Usual Care sites. We will evaluate i-STRONGER effectiveness and mechanisms underlying successful i-STRONGER implementation using the RE-AIM (Reach, Effectiveness, Adoption, Implementation, and Maintenance) framework. Implementation strategies (e.g., champion selection) have been selected from the Expert Recommendations for Implementation Change (ERIC) to improve program success. During this study, i-STRONGER-site clinicians will undergo a 3-month, self-paced, online i-STRONGER training program post-training, physical performance data (i.e., gait speed, Short Physical Performance Battery scores) will be collected across all sites for 12 months. Changes in physical performance between patient admission and discharge will be compared across sites to determine Effectiveness. Clinician surveys and focus groups will inform factors influencing Reach (proportion of patients treated with i-STRONGER), Adoption (proportion of clinicians utilizing i-STRONGER), Implementation (i-STRONGER fidelity), and Maintenance (i-STRONGER sustainment). Study findings have the potential to shift SNF rehabilitation care paradigms, optimize patient outcomes and independence, and critically inform future work aimed at large-scale i-STRONGER implementation in rehabilitation settings.
The complexities, challenges, and promise of rehabilitation clinical trials
1Washington University School of Medicine
While rehabilitation clinical trials have improved in quality over the past decades, there are key opportunities for improvement. First, we need to conceptualize our rehabilitation interventions with respect to their active ingredients, inactive ingredients, and mode of delivery. Improved conceptualization of interventions allows more rigorous trial design to evaluate causal relationships between ingredients and desired outcomes. Second, we need to be creative in our trial design process. Trial design starts with a critical scientific question and known or hypothesized active and inactive ingredients, and then carefully considers the target population, selection of one or more comparator interventions, measurement of outcomes, threats to internal validity, and state-of-the art analytic options. Creative trial designs, especially in early phase trials, will maximize the information gained and improve later phase trials. Third, we may use surrogate outcomes in early phase trials, but definitive, later phase trials must measure outcomes that are of interest to the populations we serve. Nearly 90% of self-identified patient goals are for better performance of activities in daily life, not to get better on the in-clinic tests and measures that are most often primary trial endpoints. And fourth, we need to anticipate that any given experimental rehabilitation intervention has a low probability of eventual success, as is true with other biomedical interventions. Anticipating stumbles along the clinical trial pipeline will: 1) allow rehabilitation research groups to nimbly pivot to new ideas when hypotheses are not supported, and 2) ultimately help the rehabilitation field advance knowledge more quickly and efficiently.
Adaptive Clinical Trials in Spinal Cord Injury
1The Miami Project to Cure Paralysis
Spinal cord injury (SCI) is a heterogeneous condition due to the level and severity of the injury, individual genetic factors, variability in treatment and acute care complications, rehabilitation intensity, and other environmental factors. This complicates the design of clinical trials and has led to a lack of clarity regarding efficacy due to the effect of large standard deviations on statistical comparisons between groups. Other challenging issues include relatively small effect sizes, lack of validated biomarkers, and long enrollment durations. These factors create significant financial risk for clinical trial sponsors. The heterogeneity is particularly notable when using the internationally standardized assessment of sensory and motor scores wherein floor and ceiling effects can render some participants non-contributory to testing the experimental hypothesis. These issues likely underlie some apparently failed clinical trials.
A number of solutions have been proposed, several of which are based on stratification methods to generate more homogeneous cohorts for comparison. A machine learning technique known as recursive partitioning separates binary pairs recursively based on dichotomous separation of independent variables into decision tree nodes whose known responses are similar until no further separation is meaningful. Other more established techniques include adaptations to the trial design, such as adaptive enrichment in which non-responding groups may be discontinued at pre-planned assessments and adjustments and sample size revised on accruing data. There is great value in the application of Natural History datasets to predict clinical outcomes in SCI, with propensity score matching, and work towards matching historical or synthetic digital twins to newly enrolled subjects. A particularly compelling clinical trial design is the “platform study” which is suitable for continuous testing of successive SCI therapeutics. In this process, a very substantial control group is created that is valid within the trial design and can increase statistical power. In summary, adaptive innovations are essential to refine future clinical trials to make them more efficient.
Lessons Learned from the NIH Pragmatic Trials Collaboratory Program for the Medical Rehabilitation Resource Network Scientific Retreat
1National Center for Complementary and Integrative Health, National Institutes of Health
The NIH Pragmatic Trials Collaboratory Program begun in 2012 as a NIH Common Fund Program with the goal of improving the national capacity to implement cost-effective, large-scale research studies that engage health care delivery organizations as research partners. The program has learned that pragmatic trials need to include design elements along a continuum from explanatory to pragmatic based on the hypothesis of the trial. In some cases, a more explanatory approach to assessing fidelity or follow up of participants is necessary to collect sufficient data to determine effectiveness. Factors that improve the chance of successfully conducting a pragmatic trial include active engagement from the health system partners; leveraging data in the electronic health record (EHR); integrating the intervention into the routine clinical workflow; and selecting health care system partners that serve diverse populations to enhance generalizability. The program expected there to be challenges in pragmatic trials with missing data in the EHR, staff turnover, decreased fidelity to the intervention, and evolving priorities and evolving health care systems. What the program did not expect was the degree and frequency of staff turnover; the amount of time needed to get and clean the EHR data; the importance of assessing fidelity during a pragmatic trial; and how strong the effects of change over time impacts certain study designs, such as the stepped wedge design. Many resources are available on the program’s website https://rethinkingclinicaltrials.org. In summary, the embedded pragmatic trial design is a good choice when the intervention has demonstrated efficacy in previous trials and can be integrated into clinical care; partnering health care systems are interested in the hypothesis; the outcomes are captured consistently in the EHR or with minimal burden; and the study will not need data or specimens captured at specific timepoints.