Explanation for the choice of comparators {6b}
IVE. The IVE experimental comparator incorporates a portable device with micro-electronics that allows the training to be unassisted and performed at home. The device consists of a head unit and base unit (Figure 1). The head unit contains inertial sensors to measure the instantaneous 3-D orientation of the head in space at 250 Hz, and an integrated circuit mirror to dynamically control the position of a laser target in space. The base unit consists of a touch screen interface that allows users to calibrate and set the device, in addition to recording compliance [18].
VPT. The VPT control group will participate in gaze stability training per the current standard of care, which involves a set of active head rotation exercises to be performed daily [17].
Intervention description {11a}
For Aim I, we will randomize participants to the control (VPT) or experimental (IVE) group. For the following six weeks, participants in each group will be seen weekly for progression of their gaze and gait stability exercises. Gaze stability will be for a total of 15 minutes per day for each group. After six weeks, participants will enter the washout stage for 6 weeks and not perform any explicit rehabilitation. Following the washout, participants will cross-over into the other treatment group and complete another 6 weeks of IVE or VPT rehabilitation (Fig. 2).
For Aims II and III, we will also randomize participants initially to the IVE or VPT group with a washout period and crossover to the other intervention. As with Aim I, each group from Aim II and Aim III will participate in rehabilitation at home and will visit the clinic once per week for modification of the gaze stability exercises. However, participants in Aims II and III will have a shortened participation and washout. The duration for participation in Aims II and III is reduced from Aim I based on our desire to determine ideal dosing for gaze stability training considering burden and compliance. Additionally, the expectation for most patients undergoing vestibular rehabilitation is that 6 weeks is a typical time to expect change (per currently clinical practice guidelines, CPG) [17].
In Aim II, neither IVE or VPT groups will complete their balance and gait exercises until the washout period begins, Fig. 3. Each group will perform daily gaze stability exercises: VPT control group will follow the recommended CPG (15 minutes), members in the IVE group will use the IVE device (15 minutes). Participants will perform the gaze stability exercises-only for 3 weeks, before entering the washout stage for 3 weeks and not perform any explicit rehabilitation. Once they enter washout, they will begin gait/balance home exercises. Once the washout is complete, they will cross-over into the other group for gaze stability training and continue with gait/balance exercise and progression (Fig. 3).
For Aim III, we seek to identify a minimum effective dose of gaze stability training. To do this, IVE and VPT participants will each complete their respective gaze stability training every other day. Both groups will complete the same gait/balance exercises from the start of the Aim. Excluding the duration, the flow of Aim III is similar to Aim I (Figure 2).
For all Aims, the intensity of the exercise program will be determined at the baseline visit based on the validated findings of the Dizziness Handicap Inventory, Gait Speed, and Functional Gait Assessment (Fig. 4). Scores on each of these assessments will be summed for a composite impairment score. The composite score will inform the development of the exercise program to address the severity of impairment.
Criteria for discontinuing or modifying allocated interventions {11b}
Participants reporting neck pain would be examined during a clinical visit to assess whether the exercises are being performed as described in the protocol. If the participant develops severe neck pain (sudden severe sustained lasting 24 hours) that does not resolve within 48 hours, the participant would be referred to their treating physician for follow-up care. Medication will be prescribed for symptomatic relief as deemed appropriate by the participant’s treating physician. If the severe neck pain persists 3 days after treatment, the intervention would be discontinued. Modifications to interventions or discontinuation of study interventions will be documented on a protocol deviation form.
Strategies to improve adherence to interventions {11c}
In addition to the exercise program, we will provide a weekly paper exercise completion grid for participants to record exercise completion. The exercise completion grid will also serve as a reminder to complete the exercises. We will also use the information provided in the exercise grid to track compliance. We will provide the participant with a new exercise grid at each visit. We have used this method with success to assist completion of weekly exercises and track compliance in prior studies [19].
Outcomes {12}
The primary outcome measure will be VOR gain (eye velocity/head velocity) as measured using the video-head impulse test (vHIT). The vHIT affords us the ability to identify other mechanisms of gaze stability (i.e., compensatory saccade), thus we will also capture metrics related to compensatory saccade frequency and latency. We will assess inter-trial correlations for both VOR gain and compensatory saccade latency. Secondary outcome measures include compensatory saccade metrics and those that capture data in the participative and impairment domains. Participative domain measures include the Dizziness Handicap Inventory (DHI), the Activities-Specific Balance Confidence Scale (ABC), the Neurobehavioral Symptom Inventory (NSI), and the Patient Global Impression of Change (PGIC). Finally, a one-time characterization of hearing and vestibular function will be obtained at enrollment.
Participative Measures
The DHI measures a participant’s perception of how the dizziness impacts their life. The index considers the emotional, physical, and functional aspects of their quality of life. Clinical significance is defined as a decrease in the DHI of either 18 points or 42% from the pre-treatment level [19]. VPT is known to reduce DHI score [20–22].
The ABC is a self-reported instrument that asks participants to rate their confidence performing 16 activities of daily living [23,24]. Scores range from 0 indicating no confidence to 100 indicating complete confidence in the participant’s ability to perform the task without losing their balance. Scores < 67% indicates a risk for falling and accurately classifies people who fall 84% of the time [25]. The ABC has excellent test-retest reliability (r = 0.92) and VPT is known to improve the ABC score [26].
The NSI collects data on the symptoms experienced after an mTBI. We will follow the Department of Defense suggestion that an individual experiencing a minimum of 20% score improvement from baseline is considered significant [27].
The PGIC is a commonly used method to quantify the amount of improvement the patient believes has occurred since beginning treatment (i.e., intervention). The PGIC asks one question that is rated on a seven-point Likert scale. PGIC values of 6 or more correlate best with actual change [28].
Impairment Measures
Impairment domain measures include the dynamic visual acuity test (DVA), the instrumented stand and walk test (ISAW), vertical and torsional alignment nulling (VAN and TAN), and the following clinical measures that we will instrument using inertial measurement units (9 degrees of freedom sensors)
a. Functional gait assessment (FGA)
b. Gait speed
c. Tandem walk
d. Modified clinical test for sensory integration of balance (mCTSIB)
The DVA test will measure visual acuity during active impulse head rotations. Participants will wear a headband with rate sensor attached that triggers a flashing letter once head velocity crosses a predetermined threshold. Participants will generate a single head rotation to the right and left (impulses, separately) with performance being assessed from the size of the smallest letter the participant can identify [19].
The ISAW uses wireless sensors to measure a 30 second stand, 7-meter walk, followed by a 180-degree turn. ISAW can distinguish soldiers with mTBI from healthy controls by their longer duration to turn (p < 0.001), their increased number of steps to complete a turn (p < 0.001), and their decreased peak velocities during the turn (p = 0.003) [28].
During vertical alignment nulling and torsional alignment nulling (VAN, TAN) the participant views one red and one blue line on the tablet screen through color-matched red and blue filters. The tests measure perception of ocular alignment and are reliable and valid in conditions of otolith dysfunction [30].
The FGA is an 8-item scale that was developed to determine fall risk in older adults. Participants are scored on a 4-level ordinal score while they perform various ambulation tasks. Scores < 20 indicate increased risk of falling in older adults.
The Patrol Exertion Multitask Test (PEMT) involves significant cognitive demands of situational awareness, memory, and decision making under the physical stress of moderate exertion. The PEMT is a 12-minute test where participants view a virtual patrolling scenario while responding to intermittent and unpredictable reaction timing cues. During this, they report visual clarity and perceived exertion. The PEMT has excellent interrater reliability for the assessment of cognitive (ICC 0.97), visual (ICC 0.99) and exertional (ICC 0.98) domains and can distinguish active duty service members (SM) with mTBI from healthy control service members based on scores of visual clarity and reaction time [31].
The Modified Clinical Test of Sensory Interaction on Balance (mCTSIB) tasks participants to stand quietly (arms folded) while sway and time are measured under four conditions: 1) eyes open firm surface; 2) eyes closed firm surface; 3) eyes open unstable surface (foam); and 4) eyes closed unstable surface (foam). The best possible timed scores are 30 seconds for each item or 120 seconds total score.
Participant timeline {13}
Figure 4 illustrates the enrollment, interventions, and assessments of the INVENT VPT trial. The primary outcome is VOR gain as measured by the vHIT. Secondary outcomes include participative and impairment measures. During the enrollment/baseline visit, participants will receive an initial VOR characterization and complete assessments for determining exercise treatment intensity. Participants will be randomized to IVE-washout period- VPT or VPT-washout period-IVE. The washout period varies as do the timing of follow-up assessments by aim.
Sample size {14}
We have based our estimate of sample size on data published in healthy adults (21–58 years old) using the IVE paradigm, whose age reflects our target patient population [32]. In that study, active VOR gain pre-adaptation was 0.92 ± 0.18 and active VOR gain post adaptation was 1.11 ± 0.22. Therefore, using the mean difference between pre- and post-adaptation of 0.19 with a standard deviation 0.18, we would need a minimum of 9 participants to be able to reject the null hypothesis with probability (power) of 0.8, presuming α < 0.05 (PS Power and Sample Size Calculations, version 3.1.2). Presuming a 25% attrition, we will recruit at least 12 participants for each of the IVE-VPT and VPT-IVE patient cohorts (aims). We aim to enroll 24 service members at the FBCH site and 24 civilians at the JHU Outpatient Center for each of Aims I - III.
Recruitment {15}
Participants will be recruited via a combination of methods that include chart review, in person recruitment during clinic visit and from queries garnered using flyers posted in clinics.
Recruitment at the FBCH
In 2017, 226 SM with mTBI were evaluated for mTBI at FBCH. In the past, researchers from FBCH have had success recruiting 2–3 SM/month with mTBI for participation in studies (personal communication, Stephanie Beauregard PT). We intend to screen ~ 10 patients per month for the INVENT VPT, with the intentions of enrolling 2–3 participants per month.
Recruitment at Johns Hopkins University School of Medicine (JHU)
At the JHU Outpatient Center, the principal investigator (PI) of the INVENT VPT routinely sees approximately 20 patients per month with dizziness and balance disorders, ~ 120/year. Of those 20 patients per month, roughly 20% have a peripheral vestibular hypofunction. Therefore, we intend to screen a minimum of 16 patients per month to obtain the required number of enrolled participants with vestibular hypofunction, over the duration of the study.
Assignment of interventions: allocation
Sequence generation {16a}
Random assignments to VPT-IVE or IVE-VPT will be allocated on a 1:1 ratio via a computer-generated randomization schedule stratified by clinical center and utilizing random permuted blocks of random sizes. Block sizes are not disclosed.
Concealment mechanism {16b}
Participants will be assigned to VPT-IVE or IVE-VPT via an online central randomization program in REDCap that is maintained by the Data Coordinating Center. The random assignment is released to the study coordinator within the REDCap randomization program after the baseline study visit has been completed and the participant has been confirmed as eligible for the trial. The random assignments for the INVENT VPT will be maintained by the Data Coordinating Center and will not be accessible to the study investigators and clinical center personnel.
Implementation {16c}
Patients who provide informed consent and are deemed eligible for the INVENT VPT trial will be randomized to IVE-VPT or VPT-IVE.The allocation sequence will be generated by the Data Coordinating Center. Study coordinators will enroll participants after confirming eligibility via a screening form. The coordinator will enter specific baseline assessment data into a computerized randomization program generated in REDCap and the assignment will be provided after these data are entered and confirmed by the study coordinator. The sequence of allocations will not be disclosed to the study site and will be maintained by the Data Coordinating Center.
Assignment of interventions: Blinding
Who will be blinded {17a}
Due to the nature of the interventions, study participants and the research physical therapists administering the interventions are unable to be blinded/masked to treatment assignment. The study coordinator is unmasked to treatment assignment as they are involved in the randomization process. The research physical therapist will not be involved in the measurement of any trial outcomes. The research audiologist will be masked to treatment assignment and will conduct the auditory and vestibular function tests. A member of the research team other than the research physical therapist/audiologist will conduct measurement of all trial outcomes. Participants and unmasked research personnel will be advised not to share treatment assignments during the follow-up visits. INVENT VPT study forms that disclose treatment allocation will be entered by the study coordinator and these study forms will be kept separate from the participant’s medical records.
Procedure for unblinding if needed {17b}
If unmasking of the research technologist is deemed necessary, the clinical center investigator will notify the Data Coordinating Center and note the reason for unmasking on the data collection form. Investigators will be encouraged to maintain masking unless there is reason to believe that unmasking the research technologist is deemed to be in the best interest of the participant’s safety.
Data collection and management
Plans for assessment and collection of outcomes {18a}
Plans for assessment and collection of outcomes are discussed in the Outcomes section.
Training and certification of study personnel
All study personnel responsible for recruiting, enrolling, collecting, and entering INVENT VPT data will receive role specific training on the study procedures to include virtual and in person training sessions, written examinations, and direct observation sessions. Study procedures, including the follow-up schedule, data collection, completion of the data collection forms, and the data entry and data quality query protocol will be discussed during live and recorded presentations. Personnel who satisfactorily complete all INVENT VPT training and receive passing scores on written examinations/direct observation sessions will be certified to participate in the INVENT VPT and will receive a certification ID to be used as a personnel identifier on all study documents.
Plans to promote participant retention and complete follow-up {18b}
Study coordinators will develop a schedule for contacting participants by email, telephone, or text message (depending on the participant’s preference for contact) to remind them of upcoming study visits and to remind them to continue their study-assigned rehabilitation and other related procedures while at home.
We will also share results of testing, including the video head impulse test reflecting the integrity of the vestibulo-ocular reflex measures, with participants at the conclusion of INVENT VPT trial participation and include comparisons with age matched controls.
Data management {19}
Clinical center personnel who have been certified for INVENT VPT data entry will be provided access to the REDCap data system that will be used for data entry and access to the data for their center only. We have designed a double data entry system with automated range and missing data checks upon entry. Inconsistencies will be further checked after entry via a series of programs that are designed to detect specific within form and cross form errors. Data quality query reports will be distributed biweekly to each clinical center for resolution. The study coordinator will be responsible for making corrections to the data system based on the queries that are distributed. The REDCap data system records all entries to maintain an audit trial of changes to the data system. Each clinical center will be responsible for responding to all queries and returning their responses to the Data Coordinating Center (DCC) for review.
Confidentiality {27}
Participants who agree to be screened for eligibility will be assigned a study ID that will be used on all data collection forms. All data will be associated with the anonymous number and personal information will no longer be used. All data are stored on secured password protected media. There will be a master document that links identifiable information with the anonymous identifier. This master document will only be accessible to the study coordinator. All participant study documents will be stored in a locked cabinet in a location with controlled access.
Plans for collection, laboratory evaluation and storage of biological specimens for genetic or molecular analysis in this trial/future use {33}
Biological specimens are not collected in this trial.
Statistical methods
Statistical methods for primary and secondary outcomes {20a}
The primary outcome for determining treatment efficacy is VOR gain, which is measured weekly for each subject in each intervention period. To verify randomization between sequences, baseline demographics and characteristics of vestibular injury will be compared across arms. The primary analysis will follow the intent to treat principle, where patient outcomes are analyzed following how they were randomized and not adjusted or excluded based on adherence to assigned treatment. To determine efficacy for the primary outcome, a generalized linear mixed model will be used with VOR gain as the dependent factor [33]. This model will be used to compare VOR gain between interventions and to estimate period and carryover effects. With substantial wash out periods for each aim, we do not expect there to be substantial differences in the carryover effects between intervention sequences. We will verify there are no differences in the carryover effects by testing for an interaction between treatment and time. If this interaction is statistically significant at α = 0.05, we will focus our analysis on patient outcomes in the first period, and disregard outcomes in the following period. We will adjust this model for site as well as demographic and vestibular injury related factors that are imbalanced between arms at baseline. All statistical tests will be conducted as two-sided for an α of 0.05.
For secondary analyses, we will examine compensatory saccade metrics, which is also measured at each weekly visit similarly to VOR gain. Similar to our analysis of the primary outcome, we will examine this secondary outcome with a generalized linear mixed model, estimating carryover effects by testing for an interaction between period and intervention. We will also adjust these comparisons of VOR gain and compensatory saccade metrics for site and for factors that are imbalanced at baseline. Analyses with greater than 25% loss to follow-up or incompleteness for other reasons will use multiple imputation [33]. All analyses will be conducted using the SAS software version 9.4 (SAS Institute Inc, Cary, NC, USA).
Interim analyses {21b}
No interim analyses are planned for the INVENT VPT trial.
Methods for additional analyses (e.g. subgroup analyses) {20b}
This is addressed in the statistical methods section.
Methods in analysis to handle protocol non-adherence and any statistical methods to handle missing data {20c}
This is addressed in the statistical methods section.
Plans to give access to the full protocol, participant level-data and statistical code {31c}
Investigators engaged in Department of Defense sponsored research must share participant-level data and the study protocol to the Federal Interagency Traumatic Brain Injury Research Informatics System (FITBIR). De-identified INVENT VPT data will also be available upon request to the principal investigator.
Oversight and monitoring
Composition of the coordinating center and trial steering committee {5d}
The INVENT VPT Data Coordinating Center (DCC) is responsible for assuring that the proposed study design and methods are statistically sound; that the investigators and study personnel adhere to the protocol; that valid and reliable data are collected and integrated; and that the accumulated data are summarized and reported as required to monitor study progress and promote and assess data quality. The DCC is composed of a director, data programmer, and biostatistician. A trial steering committee has not been formed for this trial.
Specific tasks are given by phase of the INVENT VPT Trial. Some of the operations at the DCC are designed to facilitate operations at the clinical centers.
DCC responsibilities during this period are:
-
To assist with the development of the study design, and to estimate sample size requirements.
-
To outline the data collection methods, data management methods and operations, and plans for data reporting.
-
To outline quality assurance and monitoring procedures.
-
To draft, critique, and revise major portions of the manual of procedures and assure adherence to the protocol.
-
To evaluate currently available hardware and software for data collection, management, and analysis.
-
To design the training session for the INVENT VPT.
Protocol Refinement and Implementation Phase
This phase begins when funding is awarded and ends when patient enrollment is initiated or shortly thereafter. DCC responsibilities during this period are:
-
To test and refine data collection methods in REDCap.
-
To complete assigned portions of the manual of procedures and edit for consistency and ease of reference.
-
To implement the central data management system and test procedures and operations, including tasks such as acquisition, installation, and implementation of necessary hardware and software, writing data definition tables and edit tables for data items to be collected, etc.
-
To develop materials for training and certifying clinical center staff in INVENT procedures.
-
To design aids to assist with the management of all INVENT centers, including reports to clinical centers, etc.
-
To update INVENT data collection forms and distribute electronic copies to the clinical centers.
-
To undertake site initiation visits and to ensure timely startup.
Enrollment and Follow-Up Phases
Data Management Responsibilities:
-
To receive clinical data from INVENT VPT clinical center staff via REDCap.
-
To edit all data for completeness, accuracy, and consistency and to resolve anomalous data with clinical center staff.
-
To revise data entry screens as required.
-
To prepare periodic reports concerning enrollments, completeness of scheduled follow-up interviews, database quality, etc.
Data Analysis and Reporting Responsibilities:
-
To employ appropriate statistical methods for analysis and summarization of accumulated data based on the recommendations of the Study Biostatistician.
-
To undertake final data analyses for INVENT VPT primary manuscripts.
-
To maintain the data files in a secure manner to assure their integrity.
-
To back up data files to assure that data are not lost.
-
To maintain a program of regular monitoring of the quality of the INVENT VPT database.
Composition of the data monitoring committee, its role and reporting structure {21a}
The interventions included in the INVENT VPT are considered minimal risk so a data monitoring committee will not be included in this trial. An independent medical monitor will review all adverse events that are reported during the conduct of the INVENT VPT trial and assess the potential that the event may be attributed to the study interventions or participation in the trial. The medical monitor’s findings will be reported to the IRB.
Adverse event reporting and harms {22}
Some of the tests may cause symptoms of neck pain, dizziness, nausea and/or motion sickness. The participant will be asked to report any of these symptoms to the investigator. The participant or the investigator may stop testing at any time. These symptoms usually resolve once testing has stopped. Infrequently, motion sickness symptoms may persist for as long as a few hours.
To minimize risk of neck strain, severe neck pain, or other cervical injury due to head-on-body movements, all such movements are first self-generated by the participants. Passively received (i.e., delivered by the experimenter) head movements are kept within the participant’s demonstrated range of self-generated head movements. In keeping with standard clinical practice, participants with known cervical spine instability are not subjected to tests requiring rapid head-on-body movements. To minimize the risk of skin irritation, the skin of the neck will be examined for irritation before we apply electrodes for vestibular evoked myogenic potential testing.
Balance and gait testing have a small risk of falling. To ensure that participants do not fall, testing is conducted with investigators or clinical support staff placed around them to assist as needed. Medical risks, listing all procedures, their major and minor risks and expected frequency are always conveyed to the participants.
Any adverse events will be recorded on an adverse event form and reported within 24 hours to the principal investigator. The principal investigator will notify the IRB by the next business day.
Frequency and plans for auditing trial conduct {23}
Onsite and remote data monitoring will be completed via an export of key data from the RedCap database. Data monitoring of key data points will be completed by the DCC. The responsibilities for remote data monitoring at the DCC are as follows:
-
To develop and test a SAS program to monitor incoming data for inconsistencies based on cross form, within form, cross visit, range, consistency, and expected data entry checks.
-
To execute the SAS program that monitors incoming data for inconsistencies on a weekly basis during data collection.
-
To distribute data quality queries to Clinical Coordinators biweekly.
Data entered into the REDCap system by the Clinical Centers will also be monitored for out-of-range values, inconsistent data, and double data entry inconsistency and completeness.
Periodic in person site visits are necessary to ensure that there is standardization of procedures, that personnel have been trained adequately, that the clinical facilities meet standards, and that participants and their data are being managed as specified in the protocol. A standard agenda is used for monitoring visits. All sites are visited within the first three months of the initiation of participant recruitment and then again at 6 months after the commencement of participant recruitment. Other clinical center monitoring visits shall be conducted on an as-needed basis.
General areas of review during the site visit include:
-
Clinic staff, facilities and equipment.
-
Flow of participants through the clinic.
-
Up-to-date study documentation including the manual of procedures, data collection forms, and documentation confirming reports of serious adverse events to the local IRB and other regulatory documents.
-
Review of signed consent forms for 100% of participants.
-
Review of a sample (approximately 10%) of data collection forms for comparison with data in the REDCap database and source documents.
-
Observation of the study coordinator during a participant visit, if possible.
-
Storage and access to study participant files, including proper storage of signed consent forms and handling of data quality queries.
-
Discussion of individual participants with compliance problems that may be alleviated by the study coordinator or other study personnel.
-
Meeting with the Principal Investigator of the clinic to discuss any areas of concern.
The PI, study coordinator, and Site Co-Investigators receive a detailed report of the findings within 7 business days. Recommendations for remedial actions with a timeline for resolution are included, if necessary. The monitor will also inform the Study PI of any remedial actions recommended within 7 business days of the site visit and will submit a subsequent report to the PI summarizing resolution of all deficiencies within 14 business days of final resolution by the clinical center. Remote monitoring will be independent from the study investigators and the Department of Defense.
Plans for communicating important protocol amendments to relevant parties (e.g. trial participants, ethical committees) {25}
Protocol amendments, including any modifications to the eligibility criteria, study design, recruitment procedures, or analysis plan will be documented in a revised protocol with a new version number and date. These modifications will be submitted to the IRB and approved prior to distribution to the clinical centers. All protocol modifications, including those that do not significantly impact the conduct of the study, will be communicated to the INVENT VPT investigators and personnel via a policy and procedures memorandum.
Dissemination plans {31a}
It is expected that the PI of the INVENT VPT trial will be responsible for dissemination of the INVENT study methods and findings in appropriate locations of visibility (talks, journals). As the data collection develops and statistics are applied, it will be the PI’s responsibility to discuss authorship and order of authorship with the appropriate study team members. It is likely that many of the eventual manuscripts will include reference to the INVENT VPT study team participants, rather than listing each individual as an author. Ultimately, decision on authorship will be made via study team discussion and final decision by the PI.