STRENGTHS AND LIMITATIONS OF THIS STUDY
-
Enhancing usual care with supported self-management will increase its credibility.
-
Blinding outcome assessors to participant group assignment, blinding participants to the study hypotheses and randomised assignment options and a two-stage consent process will mitigate expectancy effects.
-
Multisite recruitment from seven concussion clinics in three Canadian provinces, complemented by recruitment from emergency departments, will improve generalisability.
-
The primary outcome measure is self-reported, which is vulnerable to bias.
-
Pragmatic aspects of the study design (eg, minimal oversight of graded exposure therapy providers) may be a threat to internal validity.
Introduction
Up to half of adults continue to report disabling symptoms 6 months after mild traumatic brain injury (mTBI; also referred to as a concussion).1 Persistent symptoms are associated with reduced work productivity2 and quality of life.3 4 Psychological variables are among the strongest risk factors for persistent symptoms and disability after mTBI,5 more so than objective indicators of injury severity (eg, loss of consciousness at impact)6 7 or structural brain damage (eg, on diffusion tensor imaging).8 Fear avoidance behaviour, a coping style in which people avoid or escape from activities and situations that they expect will exacerbate their symptoms, may be an especially important and modifiable risk factor for chronic disability.9 10
In the hours to days after mTBI, physical and mental exertion as well as sensory stimulation (eg, light, noise and motion) often transiently exacerbate symptoms,11 so avoidance of triggering activities and situations may be adaptive. Accordingly, clinical practice guidelines recommend relative rest for the first 24–48 hours after mTBI.12 However, prolonged avoidance may become maladaptive. People who do not begin resuming their usual pre-injury activities can become increasingly prone to symptom flare-ups, more deconditioned and more convinced that their condition is serious and permanent, reinforcing fear avoidance.13 14 Considerable evidence now supports fear avoidance as an important mediator leading to chronic symptoms and disability after mTBI.9 15–20 In a pilot randomised controlled trial (RCT),21 we demonstrated that graded exposure therapy (GET) selectively reduced fear avoidance behaviour after mTBI. It is not known whether reduced fear avoidance behaviour translates to improved mTBI outcomes.
It would be helpful to know not only if GET is effective for persistent post-concussion symptoms, but whether, and for whom, it is more effective than alternative treatment options. Accumulating evidence indicates that prescribed aerobic exercise hastens recovery in adolescents with sport-related mTBI.22–25 Extrapolating from these research findings, prescribed exercise has begun to appear in clinical practice guidelines for adults with mTBI of all causes.26 27 The mechanisms by which exercise improves post-concussion symptoms are unclear. Prominent theories include the autonomic nervous system and cerebral blood flow regulation.28 However, reduced fear avoidance behaviour may also be a mechanism of therapeutic benefit from prescribed aerobic exercise in mTBI.29 The present study will add to the evidence base for this intervention in adults.
This study will evaluate the efficacy of GET for reducing persistent symptoms following mTBI. It has two primary aims: (1) To determine whether GET is more effective than usual care and (2) to identify for whom GET is the most effective treatment option, by evaluating whether baseline fear avoidance moderates differences between GET, usual care and prescribed aerobic exercise. We hypothesise that GET will outperform usual care more as baseline fear avoidance behaviour increases. We further hypothesise that prescribed aerobic exercise will perform at least as well as GET in patients with relatively low baseline levels of fear avoidance behaviour. Nevertheless, we expect that GET will achieve better treatment outcomes than prescribed aerobic exercise in patients with relatively high baseline levels of fear avoidance behaviour because: (1) GET involves exposure to a broader range of avoided activities than aerobic exercise; (2) high fear avoidance may limit engagement with prescribed aerobic exercise; (3) fear avoidance beliefs are idiosyncratic and aerobic exercise is not a universally valued goal; and (4) GET explicitly attempts to change dangerous beliefs and associated avoidant coping behaviours, whereas prescribed aerobic exercise may modify these variables implicitly and indirectly and therefore less potently. Our findings will guide evidence-based, personalised care after mTBI. A secondary study aim is to compare prescribed aerobic exercise to usual care, adding to the evidence base for this intervention in adults.
Methods and analyses
This protocol is written following the SPIRIT (Standard Protocol Items: Recommendations for Interventional Trials) statement30 (online supplemental appendix 1).
Supplemental material
Study design
In this multisite RCT with three arms, participants will be randomly assigned (in a 1:2:2 ratio) to receive enhanced usual care, GET or prescribed aerobic exercise. The study procedures are shown in figure 1. The outcome assessors will be blinded to participant group assignment. Participants will be blinded to study hypotheses and randomised assignment options (ie, treatment arms). We will achieve this blinding with two strategies. First, we will enhance usual care to reduce the likelihood that participants will identify it as a ‘control’ condition. Second, we will use a two-stage informed consent process. Participants will first consent to complete the baseline and outcome assessments and to be randomised to a ‘remote-delivered behavioural intervention for concussion’ (stage 1). After randomisation, they will complete a second consent form that describes only the intervention they were randomly assigned to (stage 2). Blinding of study therapists is not possible.
Overview of study procedures.
The study is registered on ClinicalTrials.gov. Any protocol amendments will be submitted to our institutional review board for approval and posted to ClinicalTrials.gov. The first participant was enrolled on 9 August 2022. We anticipate meeting our recruitment target by December 2024.
Setting and participants
Participants will be recruited from seven concussion clinics in three Canadian provinces: British Columbia (GF Strong Adult Concussion Service; Fraser Health Concussion Clinic), Alberta (Calgary Brain Injury Program and affiliated clinics), Ontario (Hull-Ellis Concussion and Research Clinic at UHN-Toronto Rehab, Sunnybrook Traumatic Brain Injury Clinic, St. Michael’s Hospital Head Injury Clinic, Canadian Concussion Centre at Toronto Western Hospital). Participants will learn about the study from their concussion clinic provider or delegate. All potential participants will be invited to complete a consent to contact form, allowing a research assistant from their province to contact them and assess their eligibility by telephone. In British Columbia and Alberta, this primary recruitment method will be complemented by contacting potentially eligible individuals who were evaluated in local emergency departments for mTBI: those meeting criteria for probable or possible mTBI on review of their chart will be mailed a letter of invitation with a link to the consent to contact form, and will receive a phone call to learn more about the study and determine eligibility (see inclusion and exclusion criteria).
Inclusion criteria are: (1) age 18–70 years; (2) sustained an mTBI according to the American Congress of Rehabilitation Medicine diagnostic criteria31 between 1 and 24 months ago, as assessed by a structured interview; (3) fluency in English; (4) access to a computer, tablet or smartphone with internet capability; (5) ≥3 moderate–severe symptoms on the Rivermead Post-concussion Symptom Questionnaire (RPQ).32 Exclusion criteria are: (1) medical contraindication to aerobic exercise (cardiac disease, chest pain with exertion, acute bone/joint/soft tissue injury aggravated by exercise); (2) change in cardiac medication within the last month, such as beta blockers, calcium channel blockers or ivabradine; (3) involved in personal injury litigation for index mTBI33 34; (4) currently pregnant or plan to become pregnant within the next 3 months; (5) severe/unstable medical or psychiatric condition that could worsen over the next year. Having pre-existing or co-occurring mood or anxiety disorders will not be an exclusion criterion, as these conditions are common (>30%) in concussion clinical patients and can be difficult to disentangle from persistent post-concussion symptoms. Participants will be permitted to access care outside of our study. Non-study care will be tracked but will not affect eligibility.
Consent and baseline assessment
When a participant is determined to be eligible and agrees to participate in a baseline assessment, they will be sent a stage 1 consent form (online supplemental appendix 2) via email to read, review and sign electronically before attending the baseline assessment. The baseline assessment will be a single in-person session lasting approximately 120 min. It will include the following assessments to confirm eligibility, stratify randomisation and tailor the exercise prescription:
Supplemental material
-
Test of Memory Malingering (TOMM).35 The TOMM is a performance validity test that resembles a recognition memory task. Failure on this test is commonly interpreted as reflecting low effort or intentional underperformance, and is a risk factor for low engagement in the treatment process.36 Participants who fail the TOMM based on established cut-offs37 will be considered ineligible and will not be randomised.
-
Questionnaires, including the FAB-TBI Fear Avoidance Behaviour after Traumatic Brain Injury Questionnaire (FAB-TBI). The FAB-TBI includes 16 items measuring fear avoidance beliefs and behaviour. The items are rated from 0 (strongly disagree) to 3 (strongly agree). The FAB-TBI was developed10 and validated29 in adults with mTBI, and clinical normative data, stratified by sex, were derived for adults presenting to concussion clinics.38 Randomisation will be stratified by FAB-TBI scores at baseline, using the sex-stratified 50th percentile (Rasch total score of 22.63 for women and 21.65 for men) of this clinical reference sample.38
-
Buffalo Concussion Bike Task (BCBT).39 The BCBT is used to assess exercise tolerance after concussion. It involves pedalling at a fixed cadence and increasing resistance until the participant experiences significant symptom worsening or reaches exhaustion. It will be used to set the initial aerobic exercise prescription40 for participants randomised to that treatment arm (see Interventions).
Randomisation
Participants whose eligibility is confirmed in the baseline assessment will be randomised to usual care, GET or prescribed aerobic exercise. The randomisation lists will be generated by a person independent of the research team and uploaded to the trial’s Research Electronic Data Capture (REDCap) project, concealed from the research team. The baseline data will be entered into the REDCap project by a research assistant to confirm eligibility and determine the stratum, which will reveal the next treatment allocation. After allocation, participants will be asked to sign the second consent form (stage 2; online supplemental appendix 2) for participating in the intervention they were randomised to and fill in the Credibility and Expectancy Questionnaire measuring the expectancy about the intervention.41
Interventions
To improve the uniformity of early post-injury education (Standard 2.6 from the Ontario Neurotrauma Foundation Guidelines),42 participants in all study arms across the sites will receive a link to a website that provides education and symptom self-management strategies for patients with mTBI/concussion (MyGuide: Concussion; https://concussion.vch.ca/).
Enhanced usual care
We will minimally enhance usual care to protect the integrity of participant blinding and to make this study arm more appealing, engaging and credible, to mitigate drop-out and expectancy effects.43 Usual care will be enhanced through email support for MyGuide: Concussion. Email support will be provided by research assistants under the supervision of a clinical psychologist over 12 weeks. Research assistants will email participants on a prespecified schedule, using scripted messages to inquire about their use of MyGuide: Concussion. Participants will also be able to ask questions about the MyGuide: Concussion content and receive highly templated responses. If participants ask questions unrelated to the study intervention or that cannot be answered by a templated response, they will be redirected to speak with their family physician.
GET
GET will be delivered by a psychology provider (psychologist or trainee under the supervision of a registered psychologist) over 12 individual (1:1) secure videoconference (Zoom) sessions (50 min per week). The content is manualised (summary of the session content in table 1). The core active ingredient is graded in vivo exposure to foster habituation and challenge beliefs that the avoided activities are dangerous. Homework involves planned exposure exercises in the home and community to support generalisation. The therapist and participant collaboratively identify activities that the patient values but is currently avoiding because of concerns about symptom provocation or other anticipated negative consequences (eg, stress, embarrassment, poor performance). In vivo exposure exercises may involve cognitively demanding activity (eg, mental math), physical exertion (eg, treadmill jogging), sensory stimulation (eg, scenes with visual motion) or other feared situations (eg, meeting new people). Goals for progression are set and refined collaboratively using a Subjective Units of Discomfort Scale (0–100), adapted from anxiety disorder treatment protocols.44 Consistent with GET protocols for other health conditions,45 46 GET will also involve strategies for decreasing somatic hypervigilance and catastrophising. GET providers will have access to a private online group discussion board where they ask questions or post challenging case scenarios and receive peer consultation. GET sessions will be audio-recorded for later fidelity assessment. A randomly selected 20% of recorded sessions will be audited to measure therapists’ adherence to the treatment manual.
Graded exposure therapy intervention session content
Aerobic exercise
The prescribed aerobic exercise intervention will be modelled after a protocol used in other trials.47 Participants will be asked to complete 30 min of aerobic exercise on 5 days/week for a 12-week period. Participants select the mode (eg, swimming, jogging, bicycling) and location of exercise (eg, outdoors, a gym or community centre, at home). The initial exercise intensity target will be based on the outcome of the BCBT39 (see table 2). Participants in this condition will be provided with a wrist-worn Fitbit Charge 5 to track their heart rate and asked to always wear it through the 12-week period, except when charging. Following the progression protocol recommended by Leddy et al,48 participants who tolerate aerobic exercise at their target intensity level (ie, do not experience symptom exacerbation) will be prompted to increase their target heart rate by 3–5 beats/min every 2 weeks. Participants who become able to exercise vigorously for 30 min on 5 days/week to a level of 90% of their predicted HRmax (calculated) without symptom exacerbation will continue to exercise at this intensity without further progression.
Algorithm for initial exercise prescription
Outcomes
The outcome assessment will occur within 14–18 weeks post-baseline assessment, after completing the 12-week treatment phase. The primary outcome will be completed over videoconference (Zoom) by a research assistant blinded to treatment allocation. At the end of this call, the secondary and tertiary outcomes will be distributed for remote completion via an online REDCap survey.
Primary outcome
The primary outcome will be post-concussion symptom severity, measured with the RPQ49 total score (ie, global symptom burden). The RPQ is the only outcome measure designated as ‘core’ in the National Institute of Neurological Disorders and Stroke Common Data Elements for both TBI and sport-related concussion. The RPQ queries the current severity of 16 common symptoms over the past week and correlates strongly with measures of disability and quality of life.3 50 The RPQ has high test–retest reliability (0.89).
Secondary outcomes
The FAB-TBI will be analysed as a total score converted into a Rasch score.29 The World Health Organization Disability Assessment Schedule (WHODAS) 2.0 is a measure of global disability. It incorporates 12 items covering domains of cognition, self-care, interpersonal relations, mobility, community participation and work/school. The WHODAS showed reliability and validity in samples with mTBI.51 The total score will be converted to Rasch scores for analysis.52 The Generalised Anxiety Disorder-753 measures anxiety severity symptoms. It has been validated as a screening tool for anxiety disorders and Posttraumatic Stress Disorder.53 54 It contains seven items with responses from 0 (not at all) to 3 (nearly every day). We will analyse the total score (a sum of responses to all items). The Personal Health Questionnaire-955 measures depression severity. It has been validated as a screening tool for major depression disorder in the general population and after TBI.55–57 It contains nine items with responses from 0 (not at all) to 3 (nearly every day). We will analyse the total score (a sum of responses to all items). The Post-Concussion Functional Scale-Screen is a component of the Concussion REcovery Questionnaire,58 for which participants are asked to identify the top five most challenging activities since their mTBI and rate how long they can currently perform each before their symptoms start or worsen.
Weekly measures
Participants will also complete a weekly REDCap survey (15 min) during the 12–16 weeks from the baseline to outcome assessments to track their non-study care (eg, physiotherapy visits), a single-item short form of the fear avoidance behaviour scale59 and the International Physical Activity Questionnaire (short form)60 to report how much time they spent doing moderate and vigorous physical activity over the previous week.
Follow-up loss
In addition to enhancing usual care with email-supported education (see Interventions), we will minimise follow-up loss by (1) allowing for remote outcome data collection, at times that are convenient for participants who are working (eg, evenings), (2) scheduling email reminders supplemented by phone calls from the study coordinator as needed, (3) collecting alternate contact information at the time of enrolment, (4) offering a financial incentive (e-gift card), (5) for participants who are difficult to reach or schedule (estimated 5–10%), we will offer an abbreviated outcome assessment that includes only the primary outcome (5 min).
Data management
All study data will be collected and managed using REDCap hosted on the Faculty of Medicine instance at the University of British Columbia. REDCap (Research Electronic Data Capture)61 62 is a secure, web-based software platform designed to support data capture for research studies. To ensure confidentiality, participants will be assigned a unique study number and only this number will be used on any research-related information collected during the study. Data will be downloaded to and accessed from a secure server at the University of British Columbia. Data can be available on reasonable request to the principal investigator after the primary trial analysis has been completed.
Sample size and statistical power
The target recruitment is 220 patients. With a randomisation ratio of 1:2:2, this will yield 44 participants assigned to usual care and 88 each to GET and prescribed aerobic exercise. Based on simulating 1000 replications of the trial, assuming 15% attrition and alpha=0.05 (two-sided) for all comparisons, we will have 99% power to detect a difference of ≥4 points on the primary outcome between GET and usual care (Aim 1) and 83% power to detect the hypothesised treatment effect modification by baseline fear avoidance behaviour (Aim 2). We will also have 80% power to detect a difference between the prescribed aerobic exercise versus usual care groups.
Planned statistical analysis
To address both aims, we will employ linear-mixed effects models. Point and interval estimates of between-group differences will be calculated using linear contrasts within models. To assess the robustness of results to missing observations, multiple imputations of missing RPQ outcome scores will be computed using a regression model with baseline RPQ and FAB-TBI score, gender and treatment group. All primary analyses will be by intention-to-treat, that is, including all randomised individuals in their assigned treatment groups. To investigate treatment effect modification (Aim 2), we will include group-specific smooth functions for baseline fear avoidance in the model. Using smooth functions will allow flexible modelling of the covariate-outcome relationship. A model with a common smooth will be compared with the model incorporating group-specific smooths using a generalised likelihood ratio test. Between-group differences as a function of fear avoidance will be illustrated graphically via the prediction matrix.63 No interim analyses are planned. Prior to the close of the recruitment a senior biostatistician will develop a detailed statistical analysis plan in collaboration with the principal investigator.
Safety and adverse effects
Serious adverse events are not anticipated. None occurred in our initial case series involving exposure to post-traumatic headache triggers,64 our pilot trial of GET21 or in prior studies of home-based exercise interventions after mTBI.65 66 GET involves having patients intentionally engaging in activities that elicit their symptoms. Activity-related symptom exacerbations after mTBI are not dangerous.11 Some participants might disclose suicidal thoughts on questionnaires. This will trigger a standardised distress protocol (approved by our ethics board for multiple prior studies) where participants are triaged using the Columbia-Suicide Severity Rating Scale-Screener.67 We formed a Data and Safety Monitoring Board (DSMB) comprised of a physician and a psychologist independent of the study. Participants will be asked during the outcome assessment if they have been harmed in any way by study participation. Those who indicate that they have will be asked to provide details about the nature, precipitants and severity of this potential harm. Any serious adverse events will be reported to the DSMB within three business days, and all other adverse events will be reported to the DSMB quarterly. The DSMB will classify adverse events on three dimensions: severity, expectedness and potential relatedness to the study intervention. Within 30 days of receiving a report of adverse events, the DSMB must issue a recommendation that the study can continue without changes, or that changes to or termination of the study are required to protect participant safety.
Patient and public involvement
Two patient partners with lived experience of mTBI helped to shape our study design and patient-facing materials (eg, consent forms). The Canadian TBI Research Consortium (CTRC) facilitated the development of this proposal through public presentation and discussion at CTRC meetings and internal peer-review of the grant proposal.
Ethics and dissemination
All study procedures were approved by the University of British Columbia Clinical Research Ethics Board (H21-02605), University of Calgary Conjoint Health Research Ethics Board (REB22-0151), University Health Network Research Ethics Board—Panel D (Clinical Trials Ontario, Project ID: 3909). Operational approvals were obtained from the Vancouver Coastal Health Research Institute and the Provincial Health Services Authority. Informed consent will be obtained from all participants.
We will publish the primary trial results in a peer-reviewed journal. If GET is found effective, we will work with a publishing company to create a refined version of the GET treatment manual and present instructional workshops for clinicians. In collaboration with our patient partners we will prepare a lay summary of the research findings for our participants.
Discussion
How people cope with their symptoms following mTBI strongly influences their recovery. Fear avoidance behaviour is a potent but malleable risk factor for symptom chronicity10 and can be lowered with GET.21 In this study protocol, we described plans for evaluating whether GET effectively reduces persistent symptoms after mTBI compared with enhanced usual care and an active comparator (prescribed aerobic exercise). The study will clarify for whom GET is the optimal treatment, by examining whether fear avoidance behaviour at baseline is associated with differential responsiveness to GET. Our findings will enable better matching of mTBI patients to treatments based on their characteristics.
The study design is pragmatic with respect to the primary outcome, analysis and setting, but also has features of an explanatory trial (eg, highly manualised treatment content) to maximise attribution of group differences to the intervention content.68 Strengths of the study include a theoretical mapping of the experimental intervention to treatment targets with specific mechanisms of action,69 multisite recruitment to support generalisability, enhancements to usual care to make it more uniform and credible, blinding of outcome assessors and a two-stage consent process to mitigate expectancy effects. Limitations include a self-reported primary outcome measure, that is, susceptible to bias, and a lack of generalisability of results to populations from disadvantaged backgrounds, who may be under-represented.21 GET providers will work independently, without close oversight. This design feature will support the generalisability of the findings but could threaten internal validity. We will measure therapist adherence to the treatment manual.
Although not a primary aim, the trial will add to existing evidence regarding prescribed aerobic exercise (vs usual care) as a treatment for persistent symptoms after mTBI. The benefits of exercise after mTBI have been predominantly demonstrated in studies involving adolescents with sport-related mTBI.22–25 The efficacy of prescribed aerobic exercise in adults, non-athletes and after all-cause mTBI is not well established. We will include adult participants (aged 18–70 years) with no restrictions in terms of their pre-injury physical activity or cause of injury.
Current treatment for mTBI is primarily based on expert opinion and evidence extrapolated from studies of other health conditions (eg, randomised trials in primary migraine or insomnia). Broad-spectrum cognitive-behavioural interventions for persistent symptoms after mTBI generally have small effect sizes.70–72 Larger treatment effects may be achievable with cognitive-behavioural therapy protocols targeted to specific symptoms73 or psychological risk factors.21 GET targets a specific maladaptive coping style after mTBI—fear avoidance behaviour. The current trial is poised to produce new evidence-based treatment options for preventing and treating persistent symptoms after mTBI.
Ethics statements
Patient consent for publication
Acknowledgments
The authors thank patient partners Wenmei Zou and Nicole Howey for their input on methodological decisions and improvements to the participant-facing study materials. The authors thank Nishtha Parag, Katherine MacMillan, Melanie Vergeer, Cristina Campbell, Alison Wilson, Evan Foster, Monic Szczypinski, Tharshini Chandra, Shweta Aswani, David Murty, Artee Srivastava, Erica Reyvas and Elke McLellan for assistance with recruitment and data collection.
This post was originally published on https://bmjopen.bmj.com