Differences between physical therapist ratings, self-ratings, and posturographic measures when assessing static balance exercise intensity.

Introduction: In order for balance therapy to be successful, the training must occur at the appropriate dosage. However, physical therapist (PT) visual evaluation, the current standard of care for intensity assessment, is not always effective during telerehabilitation. Alternative balance exercise intensity assessment methods have not previously been compared to expert PT evaluations. The aim of this study was therefore to assess the relationship between PT participant ratings of standing balance exercise intensity and balance participant self-ratings or quantitative posturographic measures. Methods: Ten balance participants with age or vestibular disorder-related balance concerns completed a total of 450 standing balance exercises (three trials each of 150 exercises) while wearing an inertial measurement unit on their lower back. They provided per-trial and per-exercise self-ratings of balance intensity on a scale from 1 (steady) to 5 (loss of balance). Eight PT participants reviewed video recordings and provided a total of 1,935 per-trial and 645 per-exercise balance intensity expert ratings. Results: PT ratings were of good inter-rater reliability and significantly correlated with exercise difficulty, supporting the use of this intensity scale. Per-trial and per-exercise PT ratings were significantly correlated with both self-ratings (r = 0.77-0.79) and kinematic data (r = 0.35-0.74). However, the self-ratings were significantly lower than the PT ratings (difference of 0.314-0.385). Resulting predictions from self-ratings or kinematic data agreed with PT ratings approximately 43.0-52.4% of the time, and agreement was highest for ratings of a 5. Discussion: These preliminary findings suggested that self-ratings best indicated two intensity levels (i.e., higher/lower) and sway kinematics were most reliable at intensity extremes.

Thinking Beyond the Device: An Overview of Human- and Equity-Centered Approaches for Health Technology Design.

A shift in the traditional technocentric view of medical device design to a human-centered one is needed to bridge existing translational gaps and improve health equity. To ensure the successful and equitable adoption of health technology innovations, engineers must think beyond the device and the direct end user and must seek a more holistic understanding of broader stakeholder needs and the intended context of use early in a design process. The objectives of this review article are (a) to provide rationale for the need to incorporate meaningful stakeholder analysis and contextual investigation in health technology development and biomedical engineering pedagogy, (b) to review existing frameworks and human- and equity-centered approaches to stakeholder engagement and contextual investigation for improved adoption of innovative technologies, and (c) to present case studyexamples of medical device design that apply these approaches to bridge the gaps between biomedical engineers and the contexts for which they are designing.

Automatic ML-based vestibular gait classification: examining the effects of IMU placement and gait task selection.

BACKGROUND: Vestibular deficits can impair an individual's ability to maintain postural and/or gaze stability. Characterizing gait abnormalities among individuals affected by vestibular deficits could help identify patients at high risk of falling and inform rehabilitation programs. Commonly used gait assessment tools rely on simple measures such as timing and visual observations of path deviations by clinicians. These simple measures may not capture subtle changes in gait kinematics. Therefore, we investigated the use of wearable inertial measurement units (IMUs) and machine learning (ML) approaches to automatically discriminate between gait patterns of individuals with vestibular deficits and age-matched controls. The goal of this study was to examine the effects of IMU placement and gait task selection on the performance of automatic vestibular gait classifiers. METHODS: Thirty study participants (15 with vestibular deficits and 15 age-matched controls) participated in a single-session gait study during which they performed seven gait tasks while donning a full-body set of IMUs. Classification performance was reported in terms of area under the receiver operating characteristic curve (AUROC) scores for Random Forest models trained on data from each IMU placement for each gait task. RESULTS: Several models were able to classify vestibular gait better than random (AUROC > 0.5), but their performance varied according to IMU placement and gait task selection. Results indicated that a single IMU placed on the left arm when walking with eyes closed resulted in the highest AUROC score for a single IMU (AUROC = 0.88 [0.84, 0.89]). Feature permutation results indicated that participants with vestibular deficits reduced their arm swing compared to age-matched controls while they walked with eyes closed. CONCLUSIONS: These findings highlighted differences in upper extremity kinematics during walking with eyes closed that were characteristic of vestibular deficits and showed evidence of the discriminative ability of IMU-based automated screening for vestibular deficits. Further research should explore the mechanisms driving arm swing differences in the vestibular population.

Rating of perceived difficulty scale for measuring intensity of standing balance exercises in individuals with vestibular disorders.

BACKGROUND: A method for prescribing the difficulty or intensity of standing balance exercises has been validated in a healthy population, but requires additional validation in individuals with vestibular disorders. OBJECTIVE: This study validated the use of ratings of perceived difficulty for estimation of balance exercise intensity in individuals with vestibular disorders. METHODS: Eight participants with a confirmed diagnosis of a vestibular disorder and 16 healthy participants performed two sets of 16 randomized static standing exercises across varying levels of difficulty. Root Mean Square (RMS) of trunk angular velocity was recorded using an inertial measurement unit. In addition, participants rated the perceived difficulty of each exercise using a numerical scale ranging from 0 (very easy) to 10 (very difficult). To explore the concurrent validity of rating of perceived difficulty scale, the relationship between ratings of perceived difficulty and sway velocity was assessed using multiple linear regression for each group. RESULTS: The rating of perceived difficulty scale demonstrated moderate positive correlations RMS of trunk velocity in the pitch (r = 0.51, p < 0.001) and roll (r = 0.73, p < 0.001) directions in participants with vestibular disorders demonstrating acceptable concurrent validity. CONCLUSIONS: Ratings of perceived difficulty can be used to estimate the intensity of standing balance exercises in individuals with vestibular disorders.

Reducing Slip Risk: A Feasibility Study of Gait Training with Semi-Real-Time Feedback of Foot-Floor Contact Angle.

Slip-induced falls, responsible for approximately 40% of falls, can lead to severe injuries and in extreme cases, death. A large foot-floor contact angle (FFCA) during the heel-strike event has been associated with an increased risk of slip-induced falls. The goals of this feasibility study were to design and assess a method for detecting FFCA and providing cues to the user to generate a compensatory FFCA response during a future heel-strike event. The long-term goal of this research is to train gait in order to minimize the likelihood of a slip event due to a large FFCA. An inertial measurement unit (IMU) was used to estimate FFCA, and a speaker provided auditory semi-real-time feedback when the FFCA was outside of a 10-20 degree target range following a heel-strike event. In addition to training with the FFCA feedback during a 10-min treadmill training period, the healthy young participants completed pre- and post-training overground walking trials. Results showed that training with FFCA feedback increased FFCA events within the target range by 16% for "high-risk" walkers (i.e., participants that walked with more than 75% of their FFCAs outside the target range) both during feedback treadmill trials and post-training overground trials without feedback, supporting the feasibility of training FFCA using a semi-real-time FFCA feedback system.

Preliminary Study of Vibrotactile Feedback during Home-Based Balance and Coordination Training in Individuals with Cerebellar Ataxia.

Intensive balance and coordination training is the mainstay of treatment for symptoms of impaired balance and mobility in individuals with hereditary cerebellar ataxia. In this study, we compared the effects of home-based balance and coordination training with and without vibrotactile SA for individuals with hereditary cerebellar ataxia. Ten participants (five males, five females; 47 ± 12 years) with inherited forms of cerebellar ataxia were recruited to participate in a 12-week crossover study during which they completed two six-week blocks of balance and coordination training with and without vibrotactile SA. Participants were instructed to perform balance and coordination exercises five times per week using smartphone balance trainers that provided written, graphic, and video guidance and measured trunk sway. The pre-, per-, and post-training performance were assessed using the Scale for the Assessment and Rating of Ataxia (SARA), SARAposture&gait sub-scores, Dynamic Gait Index, modified Clinical Test of Sensory Interaction in Balance, Timed Up and Go performed with and without a cup of water, and multiple kinematic measures of postural sway measured with a single inertial measurement unit placed on the participants' trunks. To explore the effects of training with and without vibrotactile SA, we compared the changes in performance achieved after participants completed each six-week block of training. Among the seven participants who completed both blocks of training, the change in the SARA scores and SARAposture&gait sub-scores following training with vibrotactile SA was not significantly different from the change achieved following training without SA (p>0.05). However, a trend toward improved SARA scores and SARAposture&gait sub-scores was observed following training with vibrotactile SA; compared to their pre-vibrotacile SA training scores, participants significantly improved their SARA scores (mean=−1.21, p=0.02) and SARAposture&gait sub-scores (mean=−1.00, p=0.01). In contrast, no significant changes in SARA scores and SARAposture&gait sub-scores were observed following the six weeks of training without SA compared to their pre-training scores immediately preceding the training block without vibrotactile SA (p>0.05). No significant changes in trunk kinematic sway parameters were observed as a result of training (p>0.05). Based on the findings from this preliminary study, balance and coordination training improved the participants' motor performance, as captured through the SARA. Vibrotactile SA may be a beneficial addition to training regimens for individuals with hereditary cerebellar ataxia, but additional research with larger sample sizes is needed to assess the significance and generalizability of these findings.

A Pilot Study Comparing the Effects of Concurrent and Terminal Visual Feedback on Standing Balance in Older Adults.

While balance training with concurrent feedback has been shown to improve real-time balance in older adults, terminal feedback may simplify implementation outside of clinical settings. Similarly, visual feedback is particularly well-suited for use outside the clinic as it is relatively easily understood and accessible via ubiquitous mobile devices (e.g., smartphones) with little additional peripheral equipment. However, differences in the effects of concurrent and terminal visual feedback are not yet well understood. We therefore performed a pilot study that directly compared the immediate effects of concurrent and terminal visual feedback as a first and necessary step in the future design of visual feedback technologies for balance training outside of clinical settings. Nineteen healthy older adults participated in a single balance training session during which they performed 38 trials of a single balance exercise including trials with concurrent, terminal or no visual feedback. Analysis of trunk angular position and velocity features recorded via an inertial measurement unit indicated that sway angles decreased with training regardless of feedback type, but sway velocity increased with concurrent feedback and decreased with terminal feedback. After removing feedback, training with either feedback type yielded decreased mean velocity, but only terminal feedback yielded decreased sway angles. Consequently, this study suggests that, for older adults, terminal visual feedback may be a viable alternative to concurrent visual feedback for short duration single-task balance training. Terminal feedback provided using ubiquitous devices should be further explored for balance training outside of clinical settings.

Retention Effects of Long-Term Balance Training with Vibrotactile Sensory Augmentation in Healthy Older Adults.

Vibrotactile sensory augmentation (SA) decreases postural sway during real-time use; however, limited studies have investigated the long-term effects of training with SA. This study assessed the retention effects of long-term balance training with and without vibrotactile SA among community-dwelling healthy older adults, and explored brain-related changes due to training with SA. Sixteen participants were randomly assigned to the experimental group (EG) or control group (CG), and trained in their homes for eight weeks using smart-phone balance trainers. The EG received vibrotactile SA. Balance performance was assessed before, and one week, one month, and six months after training. Functional MRI (fMRI) was recorded before and one week after training for four participants who received vestibular stimulation. Both groups demonstrated significant improvement of SOT composite and MiniBESTest scores, and increased vestibular reliance. Only the EG maintained a minimal detectable change of 8 points in SOT scores six months post-training and greater improvements than the CG in MiniBESTest scores one month post-training. The fMRI results revealed a shift from activation in the vestibular cortex pre-training to increased activity in the brainstem and cerebellum post-training. These findings showed that additional balance improvements were maintained for up to six months post-training with vibrotactile SA for community-dwelling healthy older adults.

Effects of Concurrent and Terminal Visual Feedback on Ankle Co-Contraction in Older Adults during Standing Balance.

This preliminary investigation studied the effects of concurrent and terminal visual feedback during a standing balance task on ankle co-contraction, which was accomplished via surface electromyography of an agonist-antagonist muscle pair (medial gastrocnemius and tibialis anterior muscles). Two complementary mathematical definitions of co-contraction indices captured changes in ankle muscle recruitment and modulation strategies. Nineteen healthy older adults received both feedback types in a randomized order. Following an analysis of co-contraction index reliability as a function of surface electromyography normalization technique, linear mixed-effects regression analyses revealed participants learned or utilized different ankle co-contraction recruitment (i.e., relative muscle pair activity magnitudes) and modulation (i.e., absolute muscle pair activity magnitudes) strategies depending on feedback type and following the cessation of feedback use. Ankle co-contraction modulation increased when concurrent feedback was used and significantly decreased when concurrent feedback was removed. Ankle co-contraction recruitment and modulation did not significantly change when terminal feedback was used or when it was removed. Neither ankle co-contraction recruitment nor modulation was significantly different when concurrent feedback was used compared to when terminal feedback was used. The changes in ankle co-contraction recruitment and modulation were significantly different when concurrent feedback was removed as compared to when terminal feedback was removed. Finally, this study found a significant interaction between feedback type, removal of feedback, and order of use of feedback type. These results have implications for the design of balance training technologies using visual feedback.

Post-Drive Standing Balance of Vehicle Passengers Using Wearable Sensors: The Effect of On-Road Driving and Task Performance.

Postural sway has been demonstrated to increase following exposure to different types of motion. However, limited prior studies have investigated the relationship between exposure to normative on-road driving conditions and standing balance following the exposure. The purpose of this on-road study was to quantify the effect of vehicle motion and task performance on passengers' post-drive standing balance performance. In this study, trunk-based kinematic data were captured while participants performed a series of balance exercises before and after an on-road driving session in real-time traffic. Postural sway for all balance exercises increased following the driving session. Performing a series of ecologically relevant visual-based tasks led to increases in most post-drive balance metrics such as sway position and velocity. However, the post-drive changes following the driving session with a task were not significantly different compared to changes observed following the driving session without a task. The post-drive standing balance performance changes observed in this study may increase vulnerable users' risk of falling. Wearable sensors offer an opportunity to monitor postural sway following in-vehicle exposures.

Noninvasive Estimation of Hydration Status in Athletes Using Wearable Sensors and a Data-Driven Approach Based on Orthostatic Changes.

Dehydration beyond 2% bodyweight loss should be monitored to reduce the risk of heat-related injuries during exercise. However, assessments of hydration in athletic settings can be limited in their accuracy and accessibility. In this study, we sought to develop a data-driven noninvasive approach to measure hydration status, leveraging wearable sensors and normal orthostatic movements. Twenty participants (10 males, 25.0 ± 6.6 years; 10 females, 27.8 ± 4.3 years) completed two exercise sessions in a heated environment: one session was completed without fluid replacement. Before and after exercise, participants performed 12 postural movements that varied in length (up to 2 min). Logistic regression models were trained to estimate dehydration status given their heart rate responses to these postural movements. The area under the receiver operating characteristic curve (AUROC) was used to parameterize the model's discriminative ability. Models achieved an AUROC of 0.79 (IQR: 0.75, 0.91) when discriminating 2% bodyweight loss. The AUROC for the longer supine-to-stand postural movements and shorter toe-touches were similar (0.89, IQR: 0.89, 1.00). Shorter orthostatic tests achieved similar accuracy to clinical tests. The findings suggest that data from wearable sensors can be used to accurately estimate mild dehydration in athletes. In practice, this method may provide an additional measurement for early intervention of severe dehydration.

Automatically evaluating balance using machine learning and data from a single inertial measurement unit.

BACKGROUND: Recently, machine learning techniques have been applied to data collected from inertial measurement units to automatically assess balance, but rely on hand-engineered features. We explore the utility of machine learning to automatically extract important features from inertial measurement unit data for balance assessment. FINDINGS: Ten participants with balance concerns performed multiple balance exercises in a laboratory setting while wearing an inertial measurement unit on their lower back. Physical therapists watched video recordings of participants performing the exercises and rated balance on a 5-point scale. We trained machine learning models using different representations of the unprocessed inertial measurement unit data to estimate physical therapist ratings. On a held-out test set, we compared these learned models to one another, to participants' self-assessments of balance, and to models trained using hand-engineered features. Utilizing the unprocessed kinematic data from the inertial measurement unit provided significant improvements over both self-assessments and models using hand-engineered features (AUROC of 0.806 vs. 0.768, 0.665). CONCLUSIONS: Unprocessed data from an inertial measurement unit used as input to a machine learning model produced accurate estimates of balance performance. The ability to learn from unprocessed data presents a potentially generalizable approach for assessing balance without the need for labor-intensive feature engineering, while maintaining comparable model performance.

Automated Loss-of-Balance Event Identification in Older Adults at Risk of Falls during Real-World Walking Using Wearable Inertial Measurement Units.

Loss-of-balance (LOB) events, such as trips and slips, are frequent among community-dwelling older adults and are an indicator of increased fall risk. In a preliminary study, eight community-dwelling older adults with a history of falls were asked to perform everyday tasks in the real world while donning a set of three inertial measurement sensors (IMUs) and report LOB events via a voice-recording device. Over 290 h of real-world kinematic data were collected and used to build and evaluate classification models to detect the occurrence of LOB events. Spatiotemporal gait metrics were calculated, and time stamps for when LOB events occurred were identified. Using these data and machine learning approaches, we built classifiers to detect LOB events. Through a leave-one-participant-out validation scheme, performance was assessed in terms of the area under the receiver operating characteristic curve (AUROC) and the area under the precision recall curve (AUPR). The best model achieved an AUROC ≥0.87 for every held-out participant and an AUPR 4-20 times the incidence rate of LOB events. Such models could be used to filter large datasets prior to manual classification by a trained healthcare provider. In this context, the models filtered out at least 65.7% of the data, while detecting ≥87.0% of events on average. Based on the demonstrated discriminative ability to separate LOBs and normal walking segments, such models could be applied retrospectively to track the occurrence of LOBs over an extended period of time.

Defining and characterizing task-shifting medical devices.

BACKGROUND: Task shifting could help address limited human resources available for the delivery of quality health care services in low-resource settings. However, the role of medical devices in supporting task shifting is not fully understood. This study aimed to 1) define "task-shifting medical devices" and 2) identify product characteristics to guide the design and development of task-shifting medical devices. A three-part survey questionnaire comprising open-ended, rank-ordering, and multiple-choice questions was disseminated to healthcare professionals worldwide. The survey included questions to capture stakeholders' general understanding of and preferences for task shifting in medicine and public health, and questions to define task-shifting medical devices and identify desirable product characteristics of task-shifting medical devices. RESULTS: Task-shifting medical devices were defined by respondents as "devices that can be used by a less specialized health worker". Aside from safe and effective, both essential characteristics for medical devices, easy to use was the most cited product characteristic for a task-shifting medical device. Responses also emphasized the importance of task-shifting medical devices to enable local agency, such as peer-to-peer training and local maintenance. Several additional frequently mentioned attributes included low cost, contextually appropriate, maintainable, capable of using an alternative power source, easy to understand, easy to learn, reusable, and easy to manage throughout its use cycle. CONCLUSION: This study defines and characterizes task-shifting medical devices based on healthcare professionals' responses. Ease of use was identified as the most important characteristic that defines a task-shifting medical device, alongside safe and effective, and was strongly associated with enabling peer-to-peer training and maintainability. The findings from this study can be used to inform technology product profiles for medical devices used by lower-level cadres of healthcare workers in low-resource settings.

Age Differences in Vestibular Brain Connectivity Are Associated With Balance Performance.

Visual and auditory brain network connectivity decline with age, but less is known about age effects on vestibular functional connectivity and its association with behavior. We assessed age differences in the connectivity of the vestibular cortex with other sensory brain regions, both during rest and during vestibular stimulation. We then assessed the relationship between vestibular connectivity and postural stability. A sample of seventeen young and fifteen older adults participated in our study. We assessed the amount of body sway in performing the Romberg balance task, with degraded somatosensory and visual inputs. The results showed no significant difference in balance performance between age groups. However, functional connectivity analyses revealed a main effect of age and condition, suggesting that vestibular connectivity was higher in young adults than older adults, and vestibular connectivity increased from resting state to stimulation trials. Surprisingly, young adults who exhibited higher connectivity during stimulation also had greater body sway. This suggests that young adults who exhibit better balance are those who respond more selectively to vestibular inputs. This correlation is non-significant in older adults, suggesting that the relationship between vestibular functional connectivity and postural stability differs with age.

Effects of long-term vestibular rehabilitation therapy with vibrotactile sensory augmentation for people with unilateral vestibular disorders - A randomized preliminary study.

BACKGROUND AND OBJECTIVE: This pilot study aimed to investigate the effects of incorporating vibrotactile sensory augmentation (SA) on balance performance among people with unilateral vestibular disorders (UVD). METHODS: Eight participants with UVD were recruited. Participants completed 18 balance training sessions across six weeks in a clinical setting. Four participants (68.1±7.5 yrs) were randomized to the experimental group (EG) and received trunk-based vibrotactile SA while performing the balance exercises, and four participants (63.1±11.3 yrs) were assigned to the control group (CG); CG participants completed the balance training without SA. Clinical and kinematic balance performance measures were collected before training; midway through training; and one week, one month, and six months after training. RESULTS: All participants, regardless of group, demonstrated improvements in a subset of the clinical or balance metrics immediately following completion of the balance training protocol. The EG showed significantly greater improvements than the CG for the Activities-specific Balance Confidence Scale and postural stability during the two standing balance exercises with head movements. The EG also had larger improvements than the CG for the Sensory Organization Test (SOT), Mini Balance Evaluations Systems Test, Gait Speed Test, Dynamic Gait Index, Functional Gait Assessment, and vestibular reliance metric calculated based on the SOT. CONCLUSIONS: Incorporating vibrotactile SA into vestibular rehabilitation programs may lead to additional benefits that may be retained up to six months after training compared to training without vibrotactile SA. A larger study is warranted to demonstrate statistical significance between the groups.

Deactivation of somatosensory and visual cortices during vestibular stimulation is associated with older age and poorer balance.

Aging is associated with peripheral and central declines in vestibular processing and postural control. Here we used functional MRI to investigate age differences in neural vestibular representations in response to pneumatic tap stimulation. We also measured the amount of body sway in multiple balance tasks outside of the MRI scanner to assess the relationship between individuals' balance ability and their vestibular neural response. We found a general pattern of activation in canonical vestibular cortex and deactivation in cross modal sensory regions in response to vestibular stimulation. We found that activation amplitude of the vestibular cortex was correlated with age, with younger individuals exhibiting higher activation. Deactivation of visual and somatosensory regions increased with age and was associated with poorer balance. The results demonstrate that brain activations and deactivations in response to vestibular stimuli are correlated with balance, and the pattern of these correlations varies with age. The findings also suggest that older adults exhibit less sensitivity to vestibular stimuli, and may compensate by differentially reweighting visual and somatosensory processes.

Vibrotactile display design: Quantifying the importance of age and various factors on reaction times.

Numerous factors affect reaction times to vibrotactile cues. Therefore, it is important to consider the relative magnitudes of these time delays when designing vibrotactile displays for real-time applications. The objectives of this study were to quantify reaction times to typical vibrotactile stimuli parameters through direct comparison within a single experimental setting, and to determine the relative importance of these factors on reaction times. Young (n = 10, 21.9 ± 1.3 yrs) and older adults (n = 13, 69.4 ± 5.0 yrs) performed simple reaction time tasks by responding to vibrotactile stimuli using a thumb trigger while frequency, location, auditory cues, number of tactors in the same location, and tactor type were varied. Participants also performed a secondary task in a subset of the trials. The factors investigated in this study affected reaction times by 20-300 ms (reaction time findings are noted in parentheses) depending on the specific stimuli condition. In general, auditory cues generated by the tactors (<20 ms), vibration frequency (<20 ms), number of tactors in the same location (<30 ms) and tactor type (<50 ms) had relatively small effects on reaction times, while stimulus location (20-120 ms) and secondary cognitive task (>130 ms) had relatively large effects. Factors affected young and older adults' reaction times in a similar manner, but with different magnitudes. These findings can inform the development of vibrotactile displays by enabling designers to directly compare the relative effects of key factors on reaction times.

Reliability and Validity of Ratings of Perceived Difficulty During Performance of Static Standing Balance Exercises.

BACKGROUND: Standardized instruments for measuring the intensity of balance exercises in clinical environments are lacking. OBJECTIVE: The objective of this study was to develop a method for quantifying the perceived intensity of standing balance exercises. DESIGN: A test-retest study design was used, with repeated evaluations within the same visit and between visits 1 week later. METHODS: Sixty-two participants who were healthy and 18 to 85 years old (with a mean age of 55 years [SD = 20 years]; 50% women) were enrolled. On each of 2 visits, they performed 2 sets of 24 randomized static standing exercises consisting of combinations of the following factors: surface, vision, stance, and head movement. Postural sway was measured with an inertial measurement unit, and ratings of perceived difficulty (RPD) were recorded using numerical and qualitative scales. The RPD scales were validated against the quantitative sway measures using a general linear model approach. The test-retest reliability of the RPD scales was examined using a weighted kappa coefficient. RESULTS: Both RPD scales were associated with postural sway measures with correlation coefficients > 0.6 for the whole sample. The test-retest reliability of the ratings varied considerably across the different balance exercises, and the highest weighted kappa values occurred for RPD scores on the numerical scale within the second visit, as moderate agreement was achieved in 18 of the 24 exercises. LIMITATIONS: The limitations are that the RPD scales need to be validated for other types of balance exercises and in individuals with balance disorders. CONCLUSIONS: The RPD scores correlated with the magnitude of postural sway, suggesting that they can be used as a proxy measure of perceived intensity of balance exercises.