Relationship Among Balance Impairments, Functional Performance, and Disability in People With Peripheral Vestibular Hypofunction

Key Words: Balance, Disability, Dizziness Handicap Inventory, Posturography, Vestibular hypofunction.

Background and Purpose. Physical therapy interventions are often based on assumed relationships among impairments, functional performance, and disability. The purposes of this study were (1) to describe balance impairments, functional performance, and disability in subjects with unilateral peripheral vestibular hypofunction (UVH) and bilateral peripheral vestibular hypofunction (BVH), (2) to examine the relationship among these factors, and (3) to determine whether disability can be explained by commonly used tests of balance and functional performance. Subjects. Participants were 85 subjects (mean age = 62.5 years, SD = 16.5) with UVH (n = 41) or BVH (n = 44) diagnosed by vestibular function tests and clinical examination. Methods. Each subject completed the Dizziness Handicap Inventory (DHI) to obtain a measure of disability. Functional performance was measured with a modified Timed Up & Go Test (TUG). Balance impairments were measured with computerized posturography and balance tests. Descriptive statistics, correlational analyses, and stepwise regressions were performed. Results. Subjects with BVH had poorer balance but similar TUG scores and perceived levels of disability, as compared with subjects with UVH. Weak to moderate correlations existed among balance measurements, TUG scores, and DHI scores. Balance impairments and TUG scores together explained 78% of the variance in DHI scores of the subjects with BVH, whereas balance impairments alone explained 13% of the variance in DHI scores of the subjects with UVH. Conclusion and Discussion. Balance impairments and functional performance appear to be more closely related to disability in individuals with BVH as compared with those with UVH. Clinical tests of balance impairments and functional performance appear to be useful in explaining disability. [Gill-Body KM, Beninato M, Krebs DE. Relationship among balance impairments, functional performance, and disability in people with peripheral vestibular hypofunction. Phys Ther. 2000;80:748-758.]

Many physical therapy interventions, including vestibular rehabilitation, are based on assumed relationships among impairments, functional performance, and disability. However, the presence or strength of such relationships is largely unknown, indicating the need to investigate these relationships across different groups of patients.(n1) In the disablement model first described by Nagi(n2) and later refined by others,(n3) impairments are defined as organ-level abnormalities, and functional limitations refer to restrictions in performing usual activities of daily living. Disability refers to the inability to perform or limitation in performing roles and tasks expected of an individual within the sociocultural and physical environments. Not all impairments lead to functional limitations or disability.(n2) Both Jette(n1) and Guccione(n4, n5) suggested that a linear relationship may not always exist between impairments and disability because these relationships may be quite complex and influenced by personal characteristics such as age, sex, income, educational level, comorbidities, and motivation. With respect to vestibular rehabilitation, performance of activities that focus on reducing impairments in vestibulo-ocular function (gaze stability) and vestibulospinal function (balance) are major components of the intervention. Thus, one assumption underlying much of vestibular rehabilitation is that impairments in balance are related to the functional limitations and disability reported by individuals with reduced vestibular function (or vestibular hypofunction) who experience persistent disequilibrium.

Four studies have addressed the question of whether or not self-perception of disability in patients with vestibular disorders is related to impairments in balance. In all 4 studies, the Dizziness Handicap Inventory (DHI)(n6) or a screening version of the DHI (DHI-S)(n7) was used to measure self-perception of disability. The DHI is a 25-item questionnaire designed to measure the self-perceived disabling effects caused by vestibular system damage. Internal consistency, reliability, and face validity of measurements obtained with the DHI have been established,(n6, n8) and the DHI has been used to measure the effectiveness of vestibular rehabilitation.(n9) High internal consistency for the total DHI score (Alpha = .89) and the DHI subscale scores (Alpha = .72-.85) has been established by Jacobson and Newman.(n6) They administered the DHI to 63 patients with varied vestibular disorders (mean age = 49.4 years, SD = 18.5, range = 21-82) who were consequently referred for vestibular function tests. Test-retest reliability of the DHI scores was high (r = .97, P < .0001) in 14 subjects with unspecified vestibular disorders (mean age = 45 years, SD = 13.5, range = 26-71) who completed the DHI on 2 consecutive days. Total DHI scores were lower in 106 consecutive patients who were referred for vestibular function tests (mean age = 48 years, SD = 15.8, range = 15-85) and who reported more frequent bouts of dizziness and unsteadiness, supporting the face validity of the DHI scores.(n6)

Three of the studies limited measurement of balance to posturography. Jacobson and colleagues(n8) reported a moderate relationship (r = .39-.42) between the DHI and sensory organization measures of balance (posturography) and a weak relationship (r = .11-.12) with vestibulometric measures (tests for vestibulo-ocular function). However, more than 77% of the variance in DHI scores remained unexplained by the balance impairment measures. In contrast, Robertson and colleagues(n10) reported that there was essentially no relationship between disability and computerized posturography tests and that women and individuals with bilateral vestibular disorders have higher levels of disability than men or individuals with unilateral vestibular damage. These studies(n8, n10) included subjects with vestibular and nonvestibular causes of dizziness and imbalance, making it difficult to generalize or apply these findings to individuals with peripheral vestibular disorders.

In a study describing the development and preliminary testing of the DHI-S (a modified version of the DHI that consists of 10 of the original 25 questions), Jacobson and Calder(n7) found that scores indicating abnormal posturography were associated with higher DHI-S scores in 26 subjects with unspecified diagnoses referred for balance evaluation. Mann and colleagues(n12) compared DHI scores with measurements obtained for timed standing on one leg and functional reach(n11) in subjects with vestibular hypofunction and reported no relationship between DHI scores and measurements obtained for either balance test. However, Wernick-Robinson and colleagues(n13) recently reported that functional reach was not a valid measure of balance in people with vestibular hypofunction because elderly subjects without disability and subjects with vestibular hypofunction were able to reach the same distance during the test and did so without increasing the distance between the center of gravity and the center of pressure (moment arm). No studies have been performed that investigate the relationship among the various elements of disablement in individuals with vestibular disorders using commonly performed tests of balance and functional performance.

The purposes of this study, therefore, were (1) to describe balance impairments, functional performance, and disability in subjects with UVH and in subjects with BVH, (2) to examine the relationships among measures of balance impairments, functional performance, and disability, and (3) to determine whether disability can be explained by commonly used tests of balance and a measure of functional performance in these subjects. Clarification of the relationship among impairments, functional performance, and disability in individuals with vestibular disorders may serve to identify the specific type of balance impairment that is most related to an individual's functional performance and disability as well as to examine some of the assumptions underlying vestibular rehabilitation.

The study population consisted of 85 patients (mean age = 62.5 years, SD = 16.5, range = 20.3-92.2) with UVH or BVH referred for vestibular rehabilitation to the Massachusetts General Hospital (MGH) in Boston, Mass. Subjects were recruited from the Massachusetts Eye and Ear Infirmary (MEEI) otology and otolaryngology services, other MEEI services, neurology clinics, and staff at MGH and other area hospitals. Of the 123 patients initially screened, 85 patients (54 female and 31 male) satisfied the inclusion and exclusion criteria and elected to participate. The duration of subjects by type of vestibular hypofunction and the distribution of subjects by age, sex, and etiology of vestibular hypofunction are presented in Table 1. The subjects with UVH had a mean age of 60.3 years (SD = 17.4, range = 20.3-92.2), and the subjects with BVH had a mean age of 64.5 years (SD = 15.5, range = 27.7-90.3). The etiology of vestibular hypofunction varied among the subjects.

Subjects had either reduced UVH (n = 41) or reduced BVH (n = 44), as demonstrated by vestibular function tests and clinical examination. Each subject was examined by a neurotologist and by use of posturography and vestibular function tests at the MEEI, including caloric stimulation and sinusoidal vertical axis rotation testing with electronystagmography. The diagnosis of UVH was made if asymmetrical or insufficient responses during rotation testing and reduced unilateral response (>/= 30%) to caloric stimulation were found. Subjects with BVH demonstrated severely reduced (>/= 2.5 standard deviations below the norm) gains of the vestibulo-ocular reflex during rotation testing at low-frequency rotation (0.01-0.1 Hz) and small to absent eye movement with caloric stimulation. All subjects reported unsteadiness while walking but were able to rise from a chair and ambulate independently. All subjects had stable symptoms for at least 2 months prior to study entry. Subjects were excluded if they had evidence of other diseases or conditions that might impair balance such as a diagnosed central nervous system condition, Meniere disease, perilymphatic fistula, benign paroxysmal vertigo, orthopedic deficits, systemic disease, sensory loss, or severe visual impairment. Vestibular suppressant medications, if used, were discontinued at least 7 days prior to participation in the study. Informed consent was obtained from all subjects.

Subjects were then referred for physical therapy and underwent an examination that included history taking and clinical examination of vestibulo-ocular function, balance, and functional performance. Functional performance was assessed by performing a modified version of the Timed Up & Go Test (TUG)(n14) and by qualitative analysis of gait, locomotion, and transfers. The DHI also was administered to each subject.

Balance impairments were measured with commonly used balance tests and sensory organization measures of posturography. The balance tests were administered by 1 of 3 physical therapists who were trained to administer these tests using standardized test procedures. Timed tests of standing with feet together and eyes closed, standing on foam with eyes open and closed, tandem standing with eyes open, and unilateral standing with eyes open and closed were included in the test battery. Tandem walking along a straight line with eyes open and closed, and walking while turning the head from side to side were also measured. Due, in part, to modifications in test protocol during the 5-year period of data collection, not all subjects completed each balance test. All tests were performed with shoes removed and without any assistive gait devices. A digital stopwatch was used for all timed measurements.

Standing with feet together. Subjects were asked to stand on the floor with their feet together, and their arms folded across their chest. The test was performed with the eyes closed for a maximum of 60 seconds. Subjects who were unable to perform the test for the maximum time on the first trial were allowed a second trial. The average time across the trials performed was used for data analysis.

Standing on foam. Subjects were asked to stand with their feet together and their arms folded across their chest on a 7.62-cm-thick, high-density foam cushion measuring 35.56 by 30.48 cm. The subjects performed the test with their eyes open, and then with their eyes closed for a maximum of 30 seconds. Subjects who were unable to perform the test for 30 seconds on the first trial were allowed a second trial. The average time across the trials performed was used for data analysis.

Tandem standing. Foot dominance was established by asking subjects which foot they would likely swing to kick a ball. Subjects were asked to stand with the non-swing foot placed behind the swing foot along a straight line and with their arms folded across their chest. They performed the test with their eyes open for a maximum of 60 seconds. Subjects who were unable to perform the test for 60 seconds on the first trial were allowed a second trial. The average time across the trials performed was used for data analysis.

Unilateral standing. Subjects were asked to stand on the swing leg with their arms folded across their chest. The performed the test with their eyes open and then with their eyes closed for a maximum of 30 seconds. Subjects who were unable to perform the test for 30 seconds on the first trial were allowed a second trial. The average time across the trials performed was used for data analysis.

Tandem walking. Subjects were asked to walk heel to toe along a 15-cm straight line marked on the floor for a maximum of 10 steps with their eyes open and then with their eyes closed. The subjects' arms were positioned by their sides. The number of steps performed consecutively before deviating from the marked line was recorded. Subjects who were unable to perform 10 steps on the first trial were allowed a second trial. The average number of steps across the trials performed was used for data analysis.

Walking with head rotation. Subjects were asked to walk at their typical pace in a straight path on a marked 0.61-m-wide, 6.1-m-long pathway while turning their head approximately 45 degrees side to side every third step. Gait during walking with head rotation along the marked path was timed. The trial was considered invalid if the subject walked outside of the marked path. Each subject performed this test 2 times, and the average time across the 2 trials was used for data analysis for subjects who could complete the test.

Posturography. The sensory organization component of posturography (Equitest(*) NeuroCom International Inc, 9570 SE Lawnfield Rd, Clackamas, OR 97015.) also was used to measure balance impairment. Each subject stood with a standardized foot placement on 2 strain gauge platforms that measured horizontal and vertical forces. The platform and visual surround were computer controlled. A computer program allowed measurement of body sway through estimation of the body's center of gravity from measurements of platform center of pressure. Subjects wore a parachute harness to prevent falling.

In Sensory Organization Test (SOT) conditions 1 through 3, the platform was fixed and the visual surround was stable, absent (eyes closed), or sway referenced with the subject's anterior-posterior body sway.(n15) In SOT conditions 4 through 6, the platform was sway referenced with the subject's anterior-posterior body sway and the visual surround was stable, absent, or sway referenced with the subject's body sway.(n15) Each sensory subtest was 20 seconds long and repeated 3 times in succession except for SOT conditions 1 through 3, which were performed only once each if performance during the initial trial was considered normal. An "equilibrium score" was obtained for each SOT trial. This score is based on the assumption than an individual without known balance disorders can exhibit 12.5 degrees of anterior-posterior sway without falling.(n15) The equilibrium score is determined by comparing the angular difference between the subject's calculated maximum displacement of the center of gravity during each trial and the subject's theoretical maximum displacement. The score is expressed as a percentage between 0% and 100%, with 100% indicating maximum stability (no sway) and 0% indicating maximum sway or a step from the platform.(n15) The average equilibrium score of the trials performed for each SOT condition was used for this study. Test-retest reliability of SOT scores using the average of 3 trials has not been examined in people with vestibular hypofunction, but it has been described as poor to good (intraclass correlation coefficients [ICC] = .26-.68) in noninstitutionalized older adults, with SOT conditions 5 and 6 demonstrating the best reliability (ICC = .64-.68).(n16)

Functional performance was tested with a modified version of the TUG.(n14) The TUG has been previously described as a measure of functional mobility for older individuals, and measurements obtained with the TUG correlate well with measurements of gait speed.(n14) Each subject was seated in a chair with armrests at one end of the 6.1-m-long, 0.61-m-wide marked pathway. The subject was instructed to stand up, walk at his or her usual pace down the pathway to the end of the walkway, turn around, walk back to the chair, and sit down. Timing began when the tester said "go" as part of the final instructions "Ready, set, go." Timing ended when the subject's buttocks touched the chair upon sitting down at the end of the test. Each subject performed the modified TUG 2 times, once turning to the right at the end of the walkway and to sit back down in the chair, and once turning to the left at these points in the test.

Self-perception of disability was measured using the DHI.(n6) Although the DHI was originally described as a measure of "handicap," we use the term "disability" here because the meaning of "handicap" as described by the developers of the DHI is equivalent to "disability." The DHI was completed by each subject, or, if the subject was unable to clearly read the questions, it was completed by verbal questioning with the physical therapist. Each self-completed DHI was reviewed with the subject for accuracy and completion. The DHI is a questionnaire that contains 9 items related to emotional well-being, 9 items related to functional activities, and 7 items related to physical activities. Each question is answered "yes," "sometimes," or "no," indicating how problematic or difficult each item is for the subject due to the underlying symptoms of dizziness or unsteadiness. A "yes" response is scored as 4 points, a "sometimes" is scored as 2 points, and a "no" response is scored as 0 points. Possible scores on the DHI range from 0 (indicating no self-perception of disability) to 100 (indicating marked self-perception of handicap). Scores for each of the DHI subscales were determined, with scores on the DHI emotional and functional subscales ranging from 0 to 36 and scores on the DHI physical subscale ranging from 0 to 28, as described by Jacobson and colleagues.(n8)

Prior to data analysis, a composite score for standing balance was determined by summing the scores of the 6 timed standing tests. We thought that such a composite measure of standing balance might be useful because there is no single gold standard for measuring standing balance and because each of the standing balance tests measures slightly different aspects of pastural control. There are no data, however, on the reliability of these summated scores or on whether they reflect standing balance better than the individuals scores do. Scores on the SOT portion of posturography testing for conditions 1 through 6 were determined by averaging the scores of the available trials for each condition. Summed scores for average performance on the SOT for conditions 1 through 6 and conditions 4 through 6 were also determined. The 2 modified TUG scores (turning to the right and turning to the left) were compared using a t test to determine whether they differed from each other. Because there was no difference between these scores, they were averaged for subsequent data analysis to create one modified TUG score for each subject. The DHI emotional, physical, and functional subscale scores were standardized to relative subscale scores (0-100) to facilitate subscale comparisons.

The mean, standard deviation, and range were determined for all variables. Chi-square analysis was performed to determine whether there was a difference in the distribution of men and women between the subjects with UVH and the subjects with BVH. We performed t tests to determine whether DHI scores differed by sex and whether duration of vestibular hypofunction differed between the 2 groups. Hotelling's t test was performed to determine whether there were overall differences in the variables between the subjects with UVH and the subjects with BVH. Multiple t tests with a Bonferroni correction factor (P < .0021) were performed to determine whether there were any differences between the 2 groups on the balance test, SOT, modified TUG, and DHI scores. Pearson product moment correlation coefficients were calculated to determine the relationships among the balance test, SOT, modified TUG, and DHI scores for the 2 groups. Only linear relationships were considered. We decided that Pearson correlation coefficients below .25 were considered weak, values from .26 to .50 suggested a fair relationship, values from .51 to .75 indicated moderate relationships, and values of .76 or higher were considered strong. A stepwise regression analysis was performed to determine the extent to which sex, age, and balance impairments (as measured by the balance tests and posturography) could explain functional performance (as measured by the modified TUG) for the 2 groups. A second regression analysis was performed to determine the extent to which the balance impairment and functional performance could explain disability, as measured by the DHI.

The means, standard deviations, and ranges of the balance impairment and modified TUG scores for the subjects with UVH and the subjects with BVH are presented in Table 2. The DHI subscale and total scores are illustrated graphically in the Figure. Chi-square analysis revealed no difference in the distribution of men and women between the 2 groups. Hotelling's t test showed that there were overall differences in impairments, functional performance, and disability between subjects with UVH and the subjects with BVH (P Explanation of Disability and Functional Performance

Sixty-four percent of the variance in DHI scores in the subjects with BVH was accounted for by 2 measures of balance impairment related to standing balance (Tab. 3). When modified TUG scores were added to the regression equation, 78% of the variance in DHI scores was explained. Similarly, scores on 2 measures designed to assess standing balance, when combined with the modified TUG scores, explained 71% of the total variance in the DHI physical subscale scores. The combined standing balance measure scores accounted for a smaller amount of the variance in the DHI emotional subscale scores, and the modified TUG scores explained 42% of the variance of the DHI functional subscale scores. For the subjects with UVH, single balance impairment measures accounted for a small amount of the variance (12%-14%) in the DHI emotional and functional subscale scores and in the total DHI scores. Age and scores on a single standing balance impairment measure together explained 31% of the variance in the DHI physical subscale scores. Unlike for the subjects with BVH, the modified TUG scores of the subjects with UVH did not account for any portion of the variance in the DHI subscale scores or in the total DHI scores.

For the subjects with BVH, unilateral standing with eyes closed accounted for 23% of the variance in the modified TUG scores. For the subjects with UVH, tandem gait and standing with feet together and eyes closed, when the scores were combined, explained 42% of the variance in the modified TUG scores (contributing 28% and 14%, respectively). No SOT score accounted for any portion of the variance in the modified TUG scores for either group.

Pearson product moment correlation coefficients for the relationships that existed between age, balance impairment scores, and modified TUG scores for subjects with UVH are shown in Table 4. Increasing age was correlated with longer time to perform the modified TUG, whereas fair to moderate negative relationships existed between scores on several measures of balance impairment and scores on the modified TUG. For the BVH group, the only relationship was found between scores for unilateral standing with eyes closed and scores on the modified TUG (r = -.47). No relationships were found between the SOT scores and the modified TUG scores for either group.

Pearson product moment correlation coefficients for the relationships among balance impairment scores, modified TUG scores, and DHI scores are shown in Table 5. In the subjects with BVH, both the modified TUG scores and the combined standing balance impairment measure score had moderate relationships with all DHI subscale scores and the total DHI scores. Fair relationships were demonstrated between age and the DHI physical subscale scores as well as between SOT scores for condition 3 and DHI emotional subscale scores. In the subjects with UVH, fair relationships were found between scores on individual measures of balance impairment and the DHI total and subscale scores.

Individuals with reduced peripheral vestibular function report impairments (including complaints of nausea, oscillopsia [the illusion of movement in the environment], dizziness, and vertigo; unsteadiness during standing and walking; and motion intolerance(n17)), functional limitations (including difficulty with lower-extremity dressing, walking, driving, and other tasks that require bending or head movements(n18)), and disability (including the inability to work and reduced participation in social and leisure activities(n19)). In the only other published study examining the relationships with disability using the DHI in individuals with balance dysfunction, however, Jacobson and colleagues(n8) reported that scores on measures of balance impairment (SOT scores for condition 5) accounted for only 8% to 23% of the variance in DHI subscale and total scores, leaving a large proportion of the variance in self-reported disability unexplained. In contrast, we found that the tests designed to measure balance and the test created to measure function explained between 12% and 78% of the variance in DHI scores when considered in combination and between 12% and 46% of the variance in DHI scores when considered individually (Tab. 3).

There are several potential explanations for the differences in these results. First, we included only patients with peripheral vestibular hypofunction (Tab. 1) while the etiology of the balance impairment for the subjects in Jacobson and colleagues' study(n8) was not specified. The presence and strength of relationships between balance impairment and disability may differ across individuals with various underlying causes of their balance impairment and various forms of balance impairment. Posturography scores differed between the 2 groups subjects in our study, suggesting that there was a range of impairments in sensory organization among subjects. Second, there was a marked difference in the age of the subjects in the two studies. Our subjects were older (mean age = 62.5 years, SD = 16.5) than those in Jacobson and colleagues' study (mean age = 48.8 years, SD = 14.5). We do not know whether there were differences in perception of disability between subjects in the 2 studies because Jacobson and colleagues did not report raw DHI scores. Finally, we included more tests designed to measure balance impairment (including balance tests commonly performed in the clinic), as compared with Jacobson and colleagues, and we used a measure designed to reflect functional performance. We contend that our results suggest that the measures of balance and functional performance included in our study add to posturography in explaining and understanding the disability that results from peripheral vestibular hypofunction.

Although we found that more variance in DHI scores could be explained by the tests and measures performed in our study than by those performed by Jacobson et al,(n8) we also found marked differences in the amount of disability that could be accounted for between our 2 groups of subjects. For example, a large portion (78%) of the variance in total DHI scores could be explained for the subjects with BVH as compared with a very small portion (13%) for the subjects with UVH (Tab. 3). This trend was consistent across the DHI subscales, with 2 to 3 times more of the variance accounted for in the subjects with BVH than in the subjects with UVH for each of the subscales (Tab. 3). The 2 groups did not differ in DHI or modified TUG scores, so baseline differences in functional performance or level of disability between the groups were not a factor. Rather, our results suggest that balance impairments and functional performance are much more closely related to the disability that occurs in individuals with BVH, although other factors not measured in our study may contribute to the disability experienced by those with unilateral vestibular pathology. Such factors may include individual coping mechanisms, anxiety, or the strength of social support systems.(n10)

Individuals may also differ in the degree of severity of symptoms (such as dizziness or nausea) experienced. Questions on the DHI are answered by the individual with respect to problems or limitations caused by symptoms of dizziness or disequilibrium. Therefore, one individual may report disability related primarily to dizziness symptoms, whereas another individual may focus on role limitation or changes in emotional wellbeing related to poor balance. In addition, individuals with UVH may experience symptoms due to the asymmetry of their condition, whereas those with BVH may be less susceptible to these symptoms.(n20) These results suggest that it is important to separately consider subjects with UVH and subjects with BVH when analyzing the relationships among balance impairments, functional performance, and disability.

Measures of balance impairment explained the highest percentage of the variance in the DHI physical subscale scores, whereas modified TUG performance explained the highest percentage of the variance in the DHI functional subscale scores in the subjects with BVH (Tab. 3). This may be due to the nature of these tests and the questions contained in each of these DHI subscales. For example, the modified TUG accounted for 42% of the variance in the DHI functional subscale scores. This result seems logical because the modified TUG consists of several component activities (rising from and sitting down in a chair, walking, and turning) that are common to many functional activities. Variance in scores for unilateral standing with eyes open accounted for the single largest amount of variance (46%) for any DHI subscale. This finding is in contrast to the findings of Mann and colleagues,(n12) who reported no relationship between single-leg stance time (measured up to 300 seconds) and DHI scores in 28 patients with mixed peripheral vestibular disorders. This difference in results may be explained by the variance in the testing procedures (our unilateral standing test was stopped at 30 seconds) and by differences in vestibular pathology between subjects in the 2 studies. Our results for the subjects with BVH are in general agreement with those of Jacobson and colleagues,(n8) who reported that, among individual posturography tests, SOT condition 5 contributes the most to explaining disability (Tab. 3). This result seems logical because SOT condition 5 has vestibular inputs as the only remaining accurate source of sensory information.

For the subjects with BVH, a combination of the composite balance test scores, overall posturography score, and functional performance scores accounted for the largest amount of variance in the total DHI scores (78%) (Tab. 3). These results suggest that no single balance impairment measure is useful in explaining overall disability and that both tests of balance impairment and functional performance are important to include to obtain a more accurate estimate of the disability that results from BVH. Posturography tests explain a smaller portion of the variance in disability as compared with balance tests commonly performed in the clinic, but posturography tests do make a contribution beyond what balance measures alone contribute. This finding is particularly important because most prior studies have limited their measure of balance impairment to posturography.(n7, n8, n10)

We included posturography as a measure of balance impairment in our study to allow comparison with prior studies, even though there is controversy regarding its usefulness. Some authors(n20, n21) have suggested that posturography may not be a valid measure of vestibulospinal function based on its poor correlation with measures of stability during gait. Dobie(n22) suggested that posturography may not provide useful information beyond that provided by standard clinical tests, whereas Furman(n23) argued that posturography is a useful measure in that it provides an indication of the impact of balance impairments on daily activities. The strength of the relationship between posturography test scores and DHI scores found in our study (poor to fair) is slightly less than the moderate relationships reported by Jacobson and colleagues(n8) and in contrast to the results of Robertson and colleagues,(n10) who found no such relationship. These differences may be explained by variances in the study samples. Our subjects were older than the subjects in both prior studies,(n8, n10) had higher DHI scores compared with Robertson and colleagues' subjects(n10) (no raw data were reported by Jacobson and colleagues(n8)), and had more homogenous etiologies of balance dysfunction. Although Jacobson and Calder(n7) reported that abnormal SOT scores were associated with greater DHI-S scores, they did not compare actual equilibrium scores for each SOT condition as we did in our study, but rather classified SOT scores as "normal" or "abnormal." Differences in how the SOT data were analyzed among studies makes comparisons difficult.

We found that none of the posturography tests explained the variance in modified TUG scores for either the subjects with UVH or the subjects with BVH. In general, standing and walking balance test scores accounted for a small portion of the modified TUG scores (23%-42%). We expected that walking balance test scores would explain more of the variance in modified TUG scores than the standing balance test scores or the posturography test scores due to the intrinsic differences between standing and walking tasks. Standing tests of balance, such as those used in our study, require the individual to keep the center of gravity within the base of support, whereas walking requires movement of the center of gravity outside the base of support. Although variances in tandem gait scores contributed to explaining the variance in the modified TUG scores to a greater extent than the standing balance test scores and were more strongly related to the modified TUG scores (Tab. 4), the differences were small. This may have occurred because some balance tests were either very easy for most subjects to perform (standing with feet together and eyes closed, foam standing with eyes open) or very difficult (unilateral standing with eyes closed, tandem gait), leading to skewed data distributions for these variables. Of the balance tests included in our study, no single balance test was able to explain a large portion of functional performance across the 2 groups of subjects. This finding implies that different balance tests, rather than a few similar tests, may be most appropriate for use with individuals with different types of vestibular pathology. Alternately, it may be that the modified TUG is not the most appropriate test of functional performance to use with individuals with peripheral vestibular pathology.

All of the subjects in our study were examined just prior to beginning a vestibular rehabilitation program, and even though the subjects had a range of durations of vestibular hypofunction, all subjects reported locomotor instability, indicating lack of compensation for their vestibular deficits. There were no differences in perceived disability between men and women in our sample. This finding is in contrast to that of Robertson and colleagues,(n10) who reported that women and had higher levels of perceived disability than men and that individuals with BVH had higher levels of perceived disability than individuals with UVH. There were also no differences between in functional performance or perceived disability between the 2 groups, even though the subjects with BVH had poorer balance and less remaining overall vestibular function than did the subjects with UVH. Our results suggest that individuals with poorly compensated UVH or BVH may have similar levels of disability if examined at least 2 months after the onset of their reduced vestibular function. What may distinguish individuals in the 2 groups is their response to treatment or eventual functional performance and level of disability. This suggestion should be examined in future studies.

Although the distribution of scores for the modified TUG and the DHI was not skewed, several of the balance impairment measures had a skewed distribution of scores. Consequently, relationships among balance, functional performance, and disability may exist beyond what is reported here if other (unmeasured) tests of balance had been examined. We tested only for linear relationships; it is also possible that nonlinear relationships exist.

This study is the first to demonstrate relationships among balance impairment, functional performance, and disability in individuals with vestibular disorders. The power of the regression analysis may have been limited due to the large number of variables. Some variables were close to being added into the regression equation, but fell just short of significance. With greater statistical power, some or all of these variable may have added to the explanation of disability.

Differences in balance impairments exist between individuals with UVH and BVH. There are also differences in the relationships among measures designed to reflect balance impairment, functional performance, and disability between these 2 groups of patients. For individuals with UVH, tests of balance impairment contribute more than measures of disability to explaining functional performance, and posturography scores do not contribute to explaining functional performance. By contrast, balance impairments (as measured by both balance tests performed in the clinic and posturography) and functional performance together explain a large component of disability in individuals with bilateral vestibular hypofunction. For individuals with UVH or BVH, tests of balance and measures of functional performance, in our view, are helpful to use in addition to posturography to explain disability. These findings lend some support to the underlying tenets of the vestibular rehabilitation intervention commonly used with these patients.(n24-n30) Further studies are needed to determine whether, following a course of vestibular rehabilitation, changes in balance impairment, functional performance, and disability are interrelated.

All authors provided concept/research design, writing, and data analysis. Ms Gill-Body and Dr Krebs provided data collection. Dr Krebs provided project management, fired procurement, subjects, facilities/equipment, institutional liaisons, clerical support, and consultation (including review of manuscript before submission). Leslie Portney, PT, PhD, also assisted with data analysis.

This study was approved by the Massachusetts General Hospital Subcommittee on Human Studies.

This work was supported in part by National Institute of Disability and Rehabilitation Research Grant H133G60045 and National Institutes of Health Grant R01AG11255.

This work was reported at the Annual Conference and Exposition of the American Physical Therapy Association; June 6, 1999; Washington, DC.

This article was submitted June 9, 1999, and was accepted March 21, 2000.

(*) NeuroCom International Inc, 9570 SE Lawnfield Rd, Clackamas, OR 97015.

GRAPH: Figure. Means and standard deviations for Dizziness Handicap Inventory (DHI) total scores and subscores for subjects with bilateral vestibular hypofunction (BVH) and subjects with unilateral vestibular hypofunction (UVH). EMODHI=DHI emotional subscale score, FUNCDHI=DHI functional subscale score, PHYSDHI=DHI physical subscale score, SUMDHI=total DHI score.

(n1) Jette AM. Outcomes research: shifting the dominant research paradigm in physical therapy. Phys Ther. 1995;75:965-970.

(n2) Nagi SZ. Some conceptual issues in disability and rehabilitation. In: Sussman MB, ed. Sociology and Rehabilitation. Washington, DC: American Sociological Association; 1965:100-113.

(n3) Brandt EN, Pope AM, eds (Committee on Assessing Rehabilitation Science and Engineering, Division of Health Sciences Policy, Institute of Medicine). Enabling America: Assessing the Role of Rehabilitation Science and Engineering. Washington, DC: National Academy Press; 1997.

(n4) Guccione AA. Physical therapy diagnosis and the relationship between impairments and function. Phys Ther. 1991;71:499-504.

(n5) Guccione AA. Arthritis and the process of disablement. Phys Ther. 1994;74:408-414.

(n6) Jacobson GP, Newman CW. The development of the Dizziness Handicap Inventory. Arch Otolaryngol Head Neck Surg. 1990;116: 424-427.

(n7) Jacobson GP, Calder JH. A screening version of the Dizziness Handicap Inventory (DHI-S). Am J Otol. 1998;19:804-808.

(n8) Jacobson GP, Newman CW, Hunter L, Balzer GK. Balance function test correlates of the Dizziness Handicap Inventory. J Am Acad Audiol. 1991;2:253-260.

(n9) Cowand JL, Wrisley DM, Walker M, et al. Efficacy of vestibular rehabilitation. Otolaryngol Head Neck Surg. 1998;118:49-54.

(n10) Robertson DD, Ireland DJ. Dizziness Handicap Inventory correlates of computerized posturography. J Otolaryngol. 1995;24:118-124.

(n11) Duncan PW, Weiner DK, Chandler J, Studenski S. Functional reach: a new clinical measure of balance. J Gerontol. 1990;45:M192-M197.

(n12) Mann GC, Whitney SL, Redfern MS, et al. Functional reach and single leg stance in patients with vestibular disorders. J Vestib Res. 1996;6:343-353.

(n13) Wernick-Robinson MS, Krebs DE, Giorgetti MM. Functional reach: does it really measure dynamic balance? Arch Phys Med Rehabil. 1999;80:262-269.

(n14) Podsiadlo D, Richardson S. The timed "Up and Go": a test of basic functional mobility of frail elderly persons. J Am Geriatr Soc. 1991;39:142-148.

(n15) Data Interpretation Manual: EquiTest System. Clackamas, Ore: NeuroCom International Inc; 1991: chap 2.

(n16) Ford-Smith CD, Wyman JF, Elswick RK Jr, et al. Test-retest reliability of the sensory organization test in noninstitutionalized older adults. Arch Phys Med Rehabil. 1995;76:77-81.

(n17) Brandt T. Falls in vestibular and neurological disorders. In: Woollacott MH, Horak FB, eds. Posture and Gait: Control Mechanisms. Vol II. Portland, Ore: University of Oregon Books; 1992:1-3.

(n18) Cohen H. Vestibular rehabilitation reduces functional disability. Otolaryngol Head Neck Surg. 1992;107:638-643.

(n19) Yardley L, Todd AM, Lacoudraye-Harter MM, Ingham R. Psychosocial consequences of recurrent vertigo. Psychology Health. 1992;6:85-96.

(n20) Evans MK, Krebs DE. Posturography does not test vestibulospinal function. Otolaryngol Head Neck Surg. 1999;120:164-173.

(n21) O'Neill DE, Gill-Body KM, Krebs DE. Posturography changes do not predict functional performance changes. Am J Otol. 1998;19:797-803.

(n22) Dobie RA. Clinical forum: Does computerized dynamic posturography help us care for our patients? Am J Otol. 1997;18:108-112.

(n23) Furman J. Posturography: uses and limitations. Baillieres Clin Neurol. 1994;3:501-513.

(n24) Lacour M, Toupet M, Denise P, Christen Y. Vestibular Compensation: Facts, Theories and Clinical Perspectives. Amsterdam, the Netherlands: Elsevier; 1989.

(n25) Shumway-Cook A, Horak FB. Vestibular rehabilitation: an exercise approach to managing symptoms of vestibular dysfunction. Semin Hearing. 1989;10:194-204.

(n26) Norre ME, Beckers A. Vestibular habilitation training: exercise treatment for vertigo based upon the habituation effect. Otolaryngol Head Neck Surg. 1989;101:14-19.

(n27) Telian SA, Shepard NT, Smith-Wheelock M, Kemink JL. Habituation therapy for chronic vestibular dysfunction: preliminary results. Otolaryngol Head Neck Surg. 1990;103:89-95.

(n28) Gill-Body KM, Krebs DE, Parker SW, Riley PO. Physical therapy management of peripheral vestibular dysfunction: two clinical case reports. Phys Ther. 1994;74:129-142.

(n29) Shumway-Cook A, Horak FB. Rehabilitation strategies for patients with vestibular deficits. Neurol Clin. 1990;8:441-457.

(n30) Horak FB, Jones-Rycewicz C, Black FO, Shumway-Cook A. Effects of vestibular rehabilitation on dizziness and imbalance. Otolaryngol Head Neck Surg. 1992;106:175-180.