Abstract
BACKGROUND:
Vestibular rehabilitation (VR) is a commonly employed treatment method for disorders of dizziness and imbalance. Access to a clinic for rehabilitation appointments can be challenging for a person experiencing dizziness. Telehealth may offer a comparable alternative to clinic-based VR for some patients.
OBJECTIVE:
The objective of this study was to determine the efficacy of telehealth-based VR compared to traditional clinic-based VR, as measured with the Dizziness Handicapped Inventory (DHI) in a retrospective sample of patients with vestibular conditions.
METHODS:
This is a retrospective, multi-institutional review from May 2020 to January 2021. Three study groups were analyzed: a telehealth group, a hybrid group, and a clinic based control group. Treatment efficacy was measured using the DHI. A repeated measures ANCOVA was performed to compare changes between the groups and across timepoints.
RESULTS:
The repeated measures ANCOVA was not significant for the interaction of groups (control, telehealth, and hybrid) by time (pre and post) (p > 0.05). However, there was a significant main effect for time (pre and post) (p < 0.05). Specifically, all groups improved DHI scores from pre to post treatment with mean differences of control: 31.85 points, telehealth: 18.75 points, and hybrid: 21.45 points.
CONCLUSION:
Findings showed that in-clinic, telehealth, and hybrid groups demonstrated a decrease in DHI scores, indicating self-reported improvements in the impact of dizziness on daily life. Continued research is recommended to explore the efficacy of using telehealth in assessing and treating vestibular conditions.
Introduction
Dizziness and vertigo affects approximately 15–20% in large population-based studies [1]. One study estimates that vestibular complaints compromise 14.3% of family practice visits per year in the United States for all adults aged 65 and older [2]. The use of healthcare services (emergency care, hospital stays, primary visits, and specialist consultations) is high among these patients and imposes an economic burden on the healthcare system [3, 4]. Many individuals report that dizziness negatively affects their quality of life. The growing body of evidence related to physical therapy for vestibular disorders suggests that people improve faster with vestibular physical therapy versus no intervention [5]. Whitney et al concluded that “Persons with vestibular disorders may develop psychological dysfunction and fear of movement the longer they experience vestibular symptoms. Early intervention is advised and customized programs appear to be superior to a generic exercise program” [5].
Since the development and implementation of Cawthorne [6] and Cooksey’s [7] vestibular rehabilitation (VR) in the 1940’s, many studies, including a Cochrane Review, have shown VR to be safe and effective in treating peripheral vestibular disorders [8]. Additional studies have shown that VR results in significantly improved Dizziness Handicap Inventory (DHI) scores [9–13]. Giray et al. [11] examined 41 patients with chronic unilateral vestibular dysfunction and found statistically and clinically significant reduction in DHI scores, from 64 to 22 (65.6%). Multiple other peer-reviewed studies have also shown similar improvements in DHI scores after in-person VR [10, 14]. The 25-item DHI was developed to evaluate the self-perceived handicapping effects imposed by a vestibular system disease. It measures the functional, emotional, and physical effects of dizziness and unsteadiness in individuals greater than or equal to 19 years of age [9]. VR utilizes both head and eye movements with specific protocols, as well as more individualized treatment programs, to challenge the balance system and improve the body’s ability to compensate, reduce symptoms of vertigo, improve gaze stabilization, and gain postural control [5].
Traditionally, VR treatments are conducted by providers in person. The frequency and duration of treatment vary based on individual needs. For some patients, however, this traditional VR setting can be impractical due to geographical availability of a competent provider, cost, and debilitating symptoms which prevent safe driving or even leaving the house. Several systematic reviews and meta-analyses have evaluated the efficacy of telehealth rehabilitation after common orthopedic procedures and have found it to be similar or superior when compared to the in-clinic setting [15, 16]. Other authors have seen similar success with telehealth rehabilitation when treating common musculoskeletal conditions [17]. More recent research shows that real-time telehealth is effective and comparable to in-clinic methods of healthcare delivery for the improvement of physical function and pain in a variety of musculoskeletal conditions [17]. The literature, however, is limited when evaluating the efficacy of telehealth rehabilitation for neurologic or vestibular impairments.
The mandated lockdown and quarantine requirements early in the COVID-19 pandemic challenged many aspects of the healthcare system, and in particular, limited in-clinic visits with patients. COVID-19 response policies forced many healthcare systems to change how they conduct patient care because of the risk associated with face to face encounters. Medicare and many other third party payers began reimbursing telehealth as an alternative means for providers to continue services. During this time period clinicians and patients faced two dichotomous challenges: the risk of infection in a clinic setting, and the need to quickly learn the technological skills to use telehealth platforms. Much of the non-emergent patient care interactions were suspended in the early pandemic, providing the opportunity to do this study. Many healthcare providers continued to use telehealth technology after lockdown restrictions were lifted because this provided an option for a care plan for those who were fearful of contracting COVID-19 in public places. The purpose of this study was to determine if virtual VR treatment would have comparable outcomes to traditional treatment in a clinic setting.
To date, few studies directly compare traditional in-clinic VR to telehealth VR. A 2017 study from England and a 2019 follow-up study from the Netherlands found that automated, internet-based vestibular rehabilitation was clinically effective and safe to treat adults aged 50 and above with a chronic vestibular syndrome [18, 19]. In 2018, Crane et al developed and implemented a self-directed computer software-based adaptive vestibular rehabilitation (AVR) program on four patients diagnosed with unilateral vestibular hypofunction [20]. They found that this novel program showed clinically and statistically significant decreases in vestibular symptoms after 4 weeks of therapy, based on DHI scores [20]. More recent research suggests that real-time telehealth appears to be effective and comparable to traditional methods of healthcare delivery for the improvement of physical function and pain in a variety of musculoskeletal conditions [17]. This study aims to compare the DHI outcome measure for patients receiving VR through telehealth, those in a traditional in-clinic VR setting, and a hybrid group who had a combination of telehealth and traditional in-clinic VR.
Materials and methods
This is a retrospective, multi-institutional review from May 2020 to January 2021. Institutional review board approval was obtained on 01/28/2021 for the retrospective review of all participants who underwent both in-clinic and telehealth VR during this period and completed the DHI pre and post treatment. The telehealth platform used was HIPAA compliant, secure Zoom video conferencing. Zoom is an electronic communication permitting secure, two-way, real-time interactive communication between a physical therapist and a patient who are not in the same location. Demographic data, including age and gender, were obtained for all participants, along with subjective history details such as fall history and relevant comorbidities. No identifying information was collected or recorded in this study. Patients were included in the study if they had a history of dizziness and could complete both the pre and post DHI questionnaire. Three study groups were analyzed. Patients who completed all treatment sessions in a clinic for traditional VR were labeled as the control group. Patients who completed their treatment sessions entirely through means of telehealth were labeled as the experimental group. Finally, patients who completed a combination of in-clinic and telehealth treatment sessions were placed in the hybrid group.
Both participating licensed vestibular physical therapists (S.C. & M.K.) in this study have completed the Emory University Vestibular Competency Course and are familiar with the evidence supporting efficacy of available treatments as well as the limitations within current research. Both therapists used similar equipment for in-clinic visits, including infrared goggles, balance foam pad, head mounted laser, Snellen eye chart, and gaze stability supplies. Each conducted customized treatment plans for each patient according to their presenting symptoms and diagnosis. Each patient received care from the same qualified vestibular physical therapist throughout the duration of their treatment.
The physical therapists initially assessed each patient’s balance, gaze stability, and postural control, as well as other relevant contributing factors, using standard tests and measures to establish the individual’s impairments limiting participation in the home and with community activities. Realistic and individualized treatment goals were established to address their function with daily activities and to decrease their vestibular symptoms. The DHI was used to assess participants’ symptoms and perceived disability before and after their individualized rehabilitation programs.
Each treatment session included an assessment of progress toward patient-specific goals, a detailed discussion regarding changes in symptoms, and the patient’s perspective on their functional gains. In addition, each session included supervised exercises, progression or modification of the home program, and correction or re-instruction of any incorrect or suboptimal techniques.
Treatments in all groups included, but were not limited to, gaze stability training, oculomotor exercises, habituation strategies, cervical exercises and techniques, postural stability exercises, somatosensory up-weighting, canalith repositioning maneuvers, and gait training activities. The therapists chose a variety of interventions based on the participants’ needs. These were similar across all research groups. Each patient received comprehensive education related to the vestibular system pathology, along with lifestyle habits and strategies which are known to enhance overall health and neuroplasticity. The total number and frequency of visits for each patient was determined based on presenting symptoms, diagnosis, and response to therapy. All three groups were instructed to perform regular daily home exercises designed by the physical therapist, and specific to the patient’s impairments, home environment, and daily routine.
Due to the lack of standardized telehealth protocols, as well as pandemic restrictions for medical practices at the time of this study, additional measures were implemented to ensure the safety of all patients within both settings. In order to approximate the same safety standards as in the clinic, both clinicians made the following adaptations for a virtual appointments: obtaining each patient’s emergency contact information, establishing a check-in procedure to verify a patient’s identity prior to start of care, discussing a customized emergency plan, and training the patient to use safety equipment within their home to improve safe mobility.
Each therapist used their experience and clinical judgment for evaluation and treatment in both clinic and virtual settings. Additional safety protocols were in place for virtual appointments with patients experiencing acute vestibular symptoms, positional vertigo, or any other characteristic which elevates the risk of injury. These included having a second person in the home, if possible, having the emergency contact person on call, finding the patient an urgent care facility, or encouraging an in-clinic evaluation if more appropriate. If the physical therapist determined during the initial screening of the telehealth visit that continuing the visit was either unsafe or that telehealth was not appropriate for the individual, the physical therapist encouraged an in-clinic session or referred the patient to a local provider. These circumstances along with other individual preferences and circumstances necessitated the creation of the hybrid group.
All patients were treated based on their presenting symptoms and diagnosis, with conditions including Benign Paroxysmal Positional Vertigo (BPPV), Bilateral and Unilateral Vestibular Loss (BVL, UVL), Vestibular Migraine (VM), Concussion or mild Traumatic Brain Injury (mTBI), Meniere’s Disease (MD), Persistent Postural Perceptual Dizziness (PPPD), Cervicogenic Dizziness and multisystem imbalance. Diagnoses represented in each group are located in Table 4. When appropriate, the clinicians involved in this study utilized the 2016 American Physical Therapy Association clinical practice guidelines specifically addressing VR for peripheral vestibular hypofunction [21]. These guidelines were later expanded in 2022 [22]. The most common treatments employed included postural control/somatosensory weighting techniques, gait training, static and dynamic vestibular and balance stabilization exercises, musculoskeletal treatments, patient education and lifestyle considerations to reduce anxiety with daily activities, and canalith repositioning maneuvers as needed.
Dizziness Handicap Inventory
The Dizziness Handicap Inventory (DHI) was designed and published by Jacobson and Newman in 1990 to quantify the effects of vestibular treatment and the impact of dizziness on everyday life [9]. This self-assessment questionnaire consists of 25 items focusing on three domains of everyday life: functional, emotional, and physical. Each item is scored by the patient with a “yes” (4 points), “sometimes” (2 points) or “no” (0 points). The DHI has been successfully validated, has an excellent test-retest reliability, and has since been used as a standard measurement tool in the assessment of vestibular treatment outcomes [9]. As a metric used in recent VR studies [20] clinically significant DHI score reduction is stated to be a decrease in DHI by 18 points or 42% from the pretreatment level [23]. Vereeck and colleagues concluded that functional balance tests involving locomotion correlate moderately with DHI scores, and that static balance measures such as the Romberg and single-leg stance, although more weakly, also have a correlation with the DHI [24]. Though the DHI tool is a self reported assessment, it has been found to be correlated with the Sensory Organization Test and the Head Shake-Sensory Organization Test in participants with unilateral vestibulopathy [25]. The correlation with these objective measures of balance and postural stability adds to the validity of the DHI as an outcome measure in this study.
Both functional independence and subjective symptoms are reflected in the DHI items. These factors both contribute to the quality of life reported by individuals with vestibular impairments.
In this study, the DHI was used to quantify treatment outcomes by analyzing patient-completed questionnaires before and after therapeutic intervention. The DHI has ease of completion in any setting, as well as high inter-rater reliability. All participants in the study were able to complete the DHI either in-clinic or via digital format.
Statistical analysis
Descriptive statistics were calculated for the DHI for each group (control, telehealth, and hybrid) for two time points (pre and post treatment). The independent variables were group and time, with the dependent variable DHI. A one-way analysis of variance (ANOVA) was conducted between groups for age to determine any age differences between groups. To determine if there was a difference, a Chi-Square test was conducted for gender and group (control, telehealth, and hybrid). A repeated measures analysis of covariance (ANCOVA) using a general linear model approach was performed with the between subjects factor group (control, telehealth, and hybrid), the within subjects factor time (pre and post treatment), and the covariate age for the dependent variable DHI. Post-hoc testing was conducted if appropriate. Alpha was set at 0.05 for all analyses. Cohen’s d effect sizes were calculated for each group between each time point.
Results
A total of 73 patients met selection criteria between May 2020 and January 2021. These 73 patients were obtained from two identified vestibular physical therapy offices and had pre and post DHI scores. There were 28 individuals in the control group, 24 in the telehealth group, and 21 in the hybrid group. Demographic statistics including age and gender differences between groups are located in Table 1. Those in the telehealth group who completed the entirety of their treatment virtually, on average, completed 4.76 treatment visits (range 1–10). The patients in the control group who completed the entirety of their treatment in-person at a vestibular clinic, on average completed 3.78 treatment visits (range 1–8). Lastly, the hybrid group that completed a combination of both in-person vestibular clinic and telehealth treatments, on average completed 2.13 vestibular clinic treatments (range 1–6) and 3.57 telehealth treatments (range 1–10).
Breakdown of study groups and demographic data
Breakdown of study groups and demographic data
The one-way ANOVA was significant for age between groups (F2,70 = 4.13, p = 0.02). There was not a significant difference between groups and sex (X2 = 0.93, p = 0.63). Therefore, age was included in further analyses. The repeated measures ANCOVA was not significant for the interaction of time by group (F2,69 = 2.76, p = 0.07). However, there was a significant main effect for time (F1,69 = 31.10, p < 0.001). Specifically, all groups had a significant improvement in DHI scores from pre to post treatment (all p values < 0.001 for each group). Descriptive statistics for DHI scores unadjusted and adjusted for age for each group between time points is located in Table 2. Furthermore, there were large effect sizes (Cohen’s d > 0.7) for each group between time points. This indicates that regardless of treatment method, all individuals improved scores pre to post treatment.
Descriptive statistics for unadjusted and age adjusted DHI scores for each group between timepoints. *p < 0.05 between timepoints
Acute and chronic vestibular conditions can impose significant stress on individuals and reduce quality of life, as evidenced by a patient’s inability to engage in everyday tasks, for example being in a car and navigating community spaces. As such, the availability for intervention in the home is crucial. Some studies have shown success with at-home VR treatments including online-based therapies and home exercise programs [18–20], but few papers exist on telehealth VR. The purpose of this study was to examine the validity of telehealth for this population, not only during this unique period in healthcare, but as a potential long term option for patients with vestibular conditions. This study shows that telehealth VR with a vestibular therapist results in a decrease in DHI scores after treatment similar to that of in-clinic VR. The similar decrease in DHI scores for all groups provide evidence needed to assure clinicians of the validity of treating vestibular patients using telehealth.
The group receiving VR via telehealth, saw a statistically and clinically significant reduction in DHI scores with an age adjusted mean of 37.27, reduced to 18.53 after treatment, a 18.74 score reduction (50.3%). The control group receiving traditional in-clinic VR saw a significant reduction in DHI scores. This group had an age adjusted mean pretreatment score of 47.89 which was reduced to 16.05 after treatment, a 31.84 score reduction (66.5%). Based upon data from Clendaniel and Goode, clinically significant DHI score reduction is a decrease in DHI by 18 points or 42% from the pretreatment level [23]. The control group in the current study had both a statistically significant and a clinically significant reduction in DHI scores. The score reductions in the control group were comparable to previous in-person VR studies [23] (Table 3). In the study by Crane et al. [20] which utilized an automated, online adaptive vestibular rehabilitation program in four patients with unilateral vestibular hypofunction, the researchers also found a statistically and clinically significant improvement in DHI scores. Other studies have examined the effectiveness of booklet-based education to provide VR to patients with chronic vestibular disorders, however, Yardley et al cited poor compliance as a limiting factor in this method [26]. The current study utilized the specialized knowledge of two competent vestibular therapists who provided patient-specific treatment plans live through telehealth for patients in their homes. Individualized home programs were provided to patients in all three groups.
Comparison of DHI data between the current study treatment arms and extracted data from previous studies
Comparison of DHI data between the current study treatment arms and extracted data from previous studies
Diagnoses represented in each treatment group
Researchers found the ability to deliver high-quality, patient-specific VR treatment through online telehealth platforms has advantages, the most notable being the access to VR when in-clinic services are not an option for an individual. Patients could complete the course of vestibular treatment without the challenges associated with leaving their home. Virtual appointments provided opportunities for patient feedback and questions, while allowing the therapists to provide periodic reassessment, deliver consistent education, make corrections to suboptimal techniques, and modify or advance daily home exercise programs.
This retrospective study utilized a third group of patients, a hybrid group, who did not complete treatment in a single location. Patients were not required to remain in one treatment arm, and some found it favorable to utilize a combination of in-clinic treatments and virtual visits to complete their rehabilitation. This group had an age adjusted mean pretreatment score of 47.02 which was reduced to 25.58 after treatment, a 21.44 score reduction (45.6%). This group obtained a similar statistical difference to the other two groups. This finding offers patients and therapists more flexibility in treatment delivery while not compromising the efficacy of care.
Prior to the recent pandemic, physical therapists and other providers were beginning to explore telehealth as a possible delivery method to achieve similar outcomes as those from in-clinic services. Reasons a patient with vestibular symptoms might desire a telehealth appointment include distance to a competent provider, inability to drive due to symptoms and family responsibilities that limit leaving the home. The COVID-19 lockdowns accelerated that exploration as telehealth rapidly became more commonly used. The prevalence of vestibular conditions is significant, and outcomes for this population improve with early intervention. This study was conducted in order to determine if the outcomes of the telehealth interventions would be similar to in-clinic treatment outcomes. The results were similar and provide evidence in favor of continuing the use of telehealth with this population.
The results of this study are promising for clinicians and researchers interested in the use of telehealth for people with vestibular conditions. The study’s limitations, however, should be considered. An age discrepancy between groups may pose a potential bias. On average the telehealth group was approximately 10 years younger than the control group. While the age was accounted for statistically, self-selection resulted in differences in mean age of each group. Age may have been a variable that influenced the choice of treatment group.
A retrospective study has inherent limitations and a lower level of evidence when compared to a prospective design. Patients included in the current study were not randomized to a particular study arm. The treatment location was a joint decision made between the patient and the therapist. All patients with vestibular conditions, who completed pre and post DHI were placed in the most appropriate study group for their situation.
This study had a 26% drop out rate, leaving 73 patients who completed their treatment. We recognize that our sample size is small, giving it lower statistical power in detecting a true effect. There was no significant difference between groups with regard to dropout rate. Most of the dropout was due to lack of completion of the final DHI, however, there may have been some unknown variables associated with those who chose not to do a follow up DHI. The short time period offered by COVID-19 pandemic and low patient volume, produced our small sample size. The small sample size did not allow for any groupings based on specific vestibular diagnosis. We did not track or analyze according to treatment frequency and duration, diagnosis, or treatment protocols. Developing a study that looks at specific treatment protocols and dosage within a telehealth model would add depth to our findings. These authors speculate that there might be differences in outcomes with telehealth versus in-person intervention for specific vestibular diagnosis groups, and recommend this also be considered in future studies.
Additionally, this retrospective study was conducted during a unique time. The participants were treated from May 2020 to January 2021. The world was experiencing a pandemic with emergency measures new to the researchers and most of the participants. Various responses to this novel time period created opposing forces of unfamiliarity with technology versus health and safety concerns. This may have had an influence on the choice of venue and the treatment outcomes in all groups. State and national restrictions were in place, which may have in some way added variables such as decreased accessibility of public transportation, decisions to stay home following exposures, as well as possible unknown variables.
People were not shopping, going to social events, or traveling as often as pre-pandemic times. This may have affected the DHI scores. We believe these possible differences in DHI scores, however, would have been similar in all three study groups. Future studies could give clarity to these unknown variables.
Lastly, the DHI was the only outcome measure used. Despite its proven efficacy, the outcome of these study groups is only provided through this measure. The use of additional outcome measures could add strength to future research. This, as well as the limitations discussed above, provide interested researchers potential variations in future study of the validity of telehealth for vestibular conditions.
Conclusion
Research shows that initiating VR early in a care plan for individuals with dizziness can change the health trajectory for an individual [5]. Telehealth can allow access to competent providers for people with vestibular conditions. This study was conducted during the beginning of the global pandemic when telehealth became a vital platform for delivering healthcare. The authors of this study aimed to determine if similar outcomes would be found with the delivery of telehealth VR as compared to in-clinic or a hybrid model for VR. All three groups had statistically significant reduction in DHI scores after treatment. While using only one well-known and validated tool allowed for simplicity in this study, it is suggested that further research seek to establish standardized telehealth procedures and outcome measures for evaluation and treatment of patients with vestibular conditions. Although in-clinic VR will likely remain as the standard of care, this study suggests that VR with a vestibular physical therapist through a telehealth platform is an effective alternative for treating vestibular conditions.
Conflict of interest
None.
