Efficacy of integrating vestibular rehabilitation and cognitive behaviour therapy in persons with long-term dizziness in primary care– a study protocol for a randomised controlled trial

Background: Dizziness is a common complaint, and the symptom often persists, together with additional complaints. A treatment combining of Vestibular Rehabilitation (VR) and Cognitive Behaviour Therapy (CBT) is suggested. However, further research is necessary to evaluate the efficacy of such intervention. The objective of this paper is to present the design of a randomised controlled trial aiming at evaluating the efficacy of an integrated treatment of VR and CBT on dizziness, physical function, psychological complaints and quality of life in persons with persistent dizziness. Methods/Design: The randomised controlled trial is an assessor-blinded block randomised parallel-group design, with a 6-and 12- month follow-up. The study includes 125 participants from Bergen (Norway) and surrounding areas. Included participants present with persistent dizziness lasting at least 3 months, triggered or exacerbated by movement. All participants receive a one-session treatment (Brief Intervention) with VR before being randomised into a control-group or an intervention-group. The intervention-group will further be offered an eight-session treatment integrating VR and CBT. The primary outcomes in the study are the dizziness handicap inventory and preferred gait velocity. Discussion: Previous studies combining these treatments have been of varying methodological quality, with small samples, and long-term effects have not been maintained. In addition, only the CBT has been administered in supervised sessions, with VR offered as home-exercises. The current study focuses on the integrated treatment, sufficiently powered sample size, and a standardised treatment program evaluated by validated outcomes using a standardised assessment protocol. Trial registration: The study was registered at www.clinicaltrials.gov (NCT02655575) on January 14th 2016.

facilitates cognitive restructiuring (e.g. reduce avoidance behaviour) (7,19). A recent Randomised Controlled Trial (RCT) providing just three sessions of Cognitive Behaviour Therapy (CBT) for panic anxiety found reduced dizziness, handicap and use of safety behaviours in persons with chronic subjective dizziness (20). Furthermore, the positive changes were maintained at 6-month follow up (21), but the outcomes only focused on psychological complaints, and no objective outcomes were assessed.
As both VR and CBT have shown positive effects on persons with dizziness, the combination of VR and CBT seems to be an appropriate treatment approach for persistent dizziness, and a few studies have investigated this combination (22)(23)(24)(25). The effects of the combined VR and CBT treatments are reported to be reduced dizziness related handicap (23)(24)(25), improved walking (23), and reduced anxiety and depression (25). A systematic review on psychotherapy in dizziness found a small and clinically relevant effect on dizziness, but no effect on anxiety and depression (26). However, the included studies had small sample sizes (19 to 31 participants) (23)(24)(25), no random allocation (25), no standardised CBT treatment manual (25), and improvements found in the short-term were not maintained as long-term effects (22). Further as none of the combined treatments included a focus on musculoskeletal complaints, there is a need to further develop the treatment combining VR and CBT, also targeting the musculoskeletal aspects, and afterward the effects of the program must be tested in an RCT. A recent feasibility study integrating VR and CBT with an additional focus on musculoskeletal effects showed that such treatment approach was feasible and safe (Kristiansen et al, 2019-in review).
Therefore, this treatment is now ready to be evaluated in an RCT.

Study objectives
The aim of the randomised controlled trial is to evaluate short-(6 months) and long-term (12 months) efficacy of an integrated treatment of VR and CBT in persons with persistent dizziness. It is hypothesised that persons receiving the additional VR-CBT program will show superior reduction in self-reported dizziness related handicap in addition to increased preferred gait velocity compared with persons receiving BI-VR alone.

Methods/design Study design
The study is a prospective assessor-blinded block-randomised controlled trial, with a parallel group-design, with a 6-and 12-months follow-up. The protocol conforms with the recommendations from the EQUATOR network (27), using the Standard Protocol Items:

Recommendations for Interventional Trails (SPIRIT) checklist and the Consolidated
Standard of Reporting Trials (CONSORT) guidelines when reporting the results (28,29).
The SPIRIT checklist is available in additional file 1.

Settings and location
Participants will be recruited through general practitioners, physiotherapists, Ear-Nose and Throat specialists, and information through newspapers and the social media.
Participants will be recruited from the region in and around Bergen, Norway. Blinded baseline and follow-up testing and one-session VR intervention (BI-VR) will be conducted at the Western Norway University of Applied Sciences (HVL). Group treatment (VR-CBT) will be offered at HVL, as well as at selected physiotherapy clinics in Bergen. The group treatment will be led by physiotherapists trained in the treatment protocol (please see below).

Participants
Eligible participants must meet all the following inclusion criteria: Working age  years) with acute onset of dizziness and with symptoms lasting at least 3 months, and the dizziness has to be triggered/ worsened by head movements, Participants will be excluded if they meet one or more of the following exclusion criteria: known non-vestibular reason for dizziness, diagnosed with fluctuating vestibular diseases (e.g. Ménière's disease), scheduled for treatment of/ have had treatment for benign paroxysmal positional vertigo (BPPV) within one month, fast head movements are contraindicated (e.g. whiplash associated injuries, osteoporosis of the neck ), presentation of severe/ terminal pathology (cancer, psychiatric diagnosis), participated in grouptherapy for dizziness within the past year, inadequate Norwegian language proficiency (verbal and written), or unable to attend test and treatment locations.

Procedure
Eligible participants will initially be screened by a telephone interview followed by further screening at HVL. Participants fulfilling the inclusion and exclusion criteria who are willing to participate will be asked to sign an informed consent. The first meeting comprises screening and baseline testing and is scheduled to last two hours. During baseline testing the included participants will complete physical tests and questionnaires. Following baseline testing the participants attend a one-hour treatment session (BI-VR), and afterwards randomisation to either the intervention group or the control group. Follow-up testing are scheduled 6 and 12 months after inclusion (Table. 1) comprising online or paper versions of questionnaires (completed separately from testing) and objective outcomes. Both follow-up tests are scheduled to last up to one hour.

Interventions
Brief Intervention Vestibular Rehabilitation (BI-VR) All the participants receive Brief Intervention Vestibular Rehabilitation (BI-VR), a singlesession treatment based on elements from traditional VR (30,31), but adapted to a single session in line with a brief intervention model developed for patients with low back pain (32). The purpose of the treatment is to give the participant the understanding that movement is the key factor in improving symptoms and that dizziness rarely is related to serious illness. BI-VR comprises examination, information regarding the vestibular system, what causes dizziness, advice related to specific findings, and supervision in selected standardised VR exercises. All participants are encouraged to stay active, and provoke dizziness in line with established recommendations (14,33).

Control group
Participants allocated to BI-VR only, will be encouraged to do the prescribed exercises on their own. The BI-VR physiotherapist will call twice during a four-month period, to encourage compliance with the home-exercises and answer questions that may arise.
Participants in the intervention group will be invited to attend an additional structured group-treatment program integrating vestibular rehabilitation and cognitive behaviour therapy (VR-CBT). The VR-CBT manual was developed through collaboration between researchers, physiotherapists and clinical psychologists. The treatment offers eight weekly two-hour sessions with five to eight participants in each group, with the aim of addressing both the physical and psychological challenges of persistent dizziness. The CBT approach is based on previous findings indicating that treatment for panic disorders can also be efficacious for persons with chronic dizziness (20). CBT focuses on the vicious cycle between somatic anxiety symptoms elicited by the "fight or flight" response, the catastrophic misinterpretations of these and other bodily symptoms, and the resulting safety-and avoidance behaviour (34). The VR comprises habituation, gaze stability and balance exercises, with body awareness promoted throughout, in addition to guided relaxation (33,35). The exercises may be individually adapted, by for instance adjustments in base of support, speed of movement, and environmental conditions. All sessions will have elements of both VR and CBT, however, the first three sessions mostly emphasise CBT, while the subsequent five sessions mostly emphasise VR. This set up allows the participants to practice exercises in a safe environment, and provides opportunities to reflect on dizziness, safety and avoidance behaviours that may occur.
Participants are further asked to carry out and register home exercises following the treatment sessions, and daily VR exercises are introduced from session three onwards. A brief description of the VR-CBT manual is presented in Table 2.

(TABLE 2 INSERTED APPROXIMATELY HERE)
Physiotherapists One physiotherapist experienced in VR, and trained in the BI-VR protocol, will run all BI-VR sessions. Six physiotherapists delivering the VR-CBT treatment will attend a competency course to before leading the treatment. The competency course contains the principles of VR-CBT, the elements of the treatment manual, and training of practical skills related to the manual, as described in the feasibility study (Kristiansen et al, 2019-in review). After each of the first two treatment sessions a clinical psychologist and the principal investigator will be available for support and guidance, without unblinding participant allocation.

Data collection and follow up
Three assessors (principal investigator, project lead, and one research assistant) are involved in the blinded data collection at baseline and follow up adhering to the standardised test protocol. In addition, the assessors will practise together before and during data collection, in order to unify performance and interpretation of the outcome measures. The principal investigator will perform the majority of the data collection. Table 3 describes the outcomes that will be collected at the various stages in the study.

Primary outcome measures
The primary outcome measures are the Dizziness Handicap Inventory (DHI) and preferred gait velocity. The DHI is a questionnaire developed in order to assess the impact of dizziness on quality of life (36). It is translated into Norwegian and has shown satisfactory test-retest reliability (37). Preferred gait velocity is assessed using 6-meter timed gait test, with one additional meter at each end allowing acceleration and deceleration (38).
The test has been found reliable in healthy adults (39), as well as in persons with vestibular disorders (38).

Secondary outcome measures
The secondary outcomes include dizziness severity, psychological complaints, fatigue, subjective health complaints, standing balance, walking, strength, flexibility, and general quality of life.
The patient reported outcomes is used to evaluate dizziness severity using the Norwegian version (40) of the Vertigo Symptom Scale short form (VSS) (41). In addition, the Hospital Secondary objective measures include standing balance (sway measured during the four conditions in the modified clinical test for sensory interaction and balance (mCTSIB)) (50) using balance trainer BTG4 (HUR health, Norway), walking (fast velocity and with dual task (51)), visual acuity (clinical dynamic visual acuity test (CDVA) (52)), and grip strength (53). Musculoskeletal aberrations is registered using four elements from the Global Physiotherapy Examination (GPE) (17) and Head movement induced dizziness (38) measured using the numeric rating scale (HmDizz). The Patient Global Impression of Change questionnaire (PGIC) (54) is used to evaluate perceived improvement at follow-up testing at 6 and 12 months (TABLE 3 INSERTED APPROXIMATELY HERE) Demographic data and other measurements In addition to the outcomes, information regarding gender, age, work status, medication, and activity level will be gathered. The VR-CBT physiotherapists will register attendance to sessions and reasons for absence and collect the home-exercise registrations.
Satisfactory compliance to VR-CBT will be defined as minimum 75% attendance to VR-CBT sessions (6/8 sessions), and minimum 80% completion of the exercise diary for homeexercises, where 100% completion is defined by reporting exercises and following a walking program five times per week. Satisfactory compliance in the control-group will be defined as completion of at least one telephone call with the BI-VR physiotherapist.

Sample size and power considerations
The study is designed as a randomised controlled trial comparing two groups (BI-VR, and BI-VR with VR-CBT). To obtain a clinical important group difference in DHI of 11 points (37) with a significance level of 0.05 and a power of 80%, 47 participants will be required per group. To obtain a clinical important change in preferred gait velocity of 0.1 m/s (55) with a significance level of 0.05 and a power of 80%, 36 participants will be required in each group. In order to ensure power of both primary outcomes at least 47 participants are selected as basis for the sample size needed in the study. The final sample size is set at 125 participants, allowing for approximately 35% drop-out, based on drop-outs in the feasibility study (Kristiansen, 2019 -in review), and in previous studies (18,20,56,57).

Randomization and concealment of allocation
The participants are block-randomised in groups of 16, and randomly assigned to BI-VR followed by VR-CBT (intervention group) or BI-VR alone (control group). Group allocation is performed using a random number generator and is presented on a folded paper, in a concealed envelope. The principle investigator is blinded from group allocations. The envelopes are stored in a locked cupboard only accessible to the BI-VR physiotherapist handing out the allocation envelopes. After group allocation the VR-CBT participants will be contacted by the project lead regarding the first VR-CBT appointment.

Blinding
The principal investigator and assessors are blinded from group allocation and not involved in the treatment of the participants. Blinding of group allocation for VR-CBT physiotherapists and participants are not possible. However, both groups are informed that the optimal treatment is not known, and the study hypothesis is not presented. In order to ensure blinding of assessors the participants are encouraged not to reveal their allocation during testing.

Statistical analysis plan
The efficacy analysis is assessment of the between-group differences in changes in DHI score and preferred gait velocity at 6 and 12 months follow up. The analysis will use the intention to treat (ITT) principle, analysing all randomised participants independent of compliance and withdrawals. In the event of missing data two methods will be used. For missing single questions, the mean baseline value for the respected group will be assigned. If complete questionnaires or objective measures are missing a non-responder imputation will be used, including baseline data carried forward. Sensitivity analyses will be performed to study whether those who drop out differ from those who complete the required program.
An analysis of covariates (ANCOVA) will be used to analyse mean changes in continuous variables, and logistic regression for categorical variables. The model will include the respective dependent variable, in addition to fixed effects of group allocation, baseline value, age, gender and height.
The results will be expressed as a difference between the group means and 95% confidence intervals with associated p-values. All data analysis will be performed according to a pre-established statistical analysis plan and interpreted according to a consensus document signed by all authors. All analyses will be performed using Statistical Package for Social Sciences (SPSS) (version 25, IBM, New York, USA).

Interim analyses
Drop-out rates will be assessed in interim analyses to determine potential need for adjustment in sample-size in line with the power calculations.

Ethical considerations
Although participants may experience increased symptoms in the short-term, it is not anticipated that participation will cause any serious adverse events or harms. A recent feasibility study (Kristiansen et al, 2019-in review) has confirmed that the intervention BI-VR and VR-CBT is feasible and safe for persons with persistent dizziness. The study will follow the criteria and principles of the Declaration of Helsinki. It has been approved by The novelty of the current study is the integration of VR and CBT, with an additional focus on musculoskeletal complaints, as treatment for persons with persistent dizziness. The previous studies conducted CBT as supervised sessions (individually (25) or in small groups (23,24)) while VR exercises where administered as home exercises (24,25). The integration of the supervised VR-CBT sessions in the current RCT allows the participants to practice excises in a safe environment supervised by physiotherapists, and the exercises may be adapted to meet the level of complaints and capacity in each participant.
One possible limitation in the study, as with all exercise trials, is the inability to blind physiotherapists and participants to treatment allocation. Thus, only the testers can be blinded to group-allocation. Another possible limitations may include the specification of compliance criteria, which to our knowledge has not been done in this research area before, and there is no consensus for how much is sufficient. Also, a block-randomisation of 16 may be seen as a limitation as it may take time before group number is reached. In addition, confounding factors like for instance age, gender, duration of complaints and psychological factor may have an impact on primary and secondary outcomes. However, the randomised design is expected to distribute these variables equally between the allocation-groups.
The strength of the current study is the inclusion of relevant reliable main outcomes with a sufficiently powered sample size. In addition, the study utilizes standardised testing procedures, evaluating both short-and long-term efficacy. Another strength is the initial feasibility study conducted, showing that test procedures and interventions were feasible and sage for the present population (Kristiansen et al, 2019-in review). To our knowledge, the current RCT study is the largest study to date, combining VR and CBT to treat persons with persistent dizziness.
The aim of the current study is to evaluate the efficacy of BI-VR followed by VR-CBT, compared with BI-VR alone. If the treatment group improves more than the control group,   Chalder's Fatigue Questionnaire (CFQ) 13 items. The first 11 items are scored from 0 (better than usual) to 3 (much worse than usual), giving a score range of 0-33. The last two items rates duration and constancy of fatigue (46). Higher scores indicating more fatigued.
Patient Specific Functional Scale (PSFS) Registers up to 3 activities participants find difficult. In addition the level of difficulty is rated on an 11 point scale (48), where 0 maximum difficulty and 10 is no difficulty.

Secondary outcomes/ Physical tests
Dual-Task walking Similar walking protocol as for preferred gait velocity, with an added task of counting backwards by three out loud, while walking. Each trial was timed and the numbers of miscounts were documented. Mean velocity, and mistakes over two trials calculated.
Fast gait velocity (m/s) Similar protocol to preferred gait velocity, however, participants were asked to walk as fast as possible.