A study of the clinical test of sensory interaction and balance.

Key Words: Balance, Equilibrium, Posture, Tests and measurements, Vestibular system.

Successful performance of some daily life tasks, such as reading the titles of books on a shelf, requires the ability to maintain an upright position. For this reason, many physical therapists are concerned with their patients' ability to perform this motor skill. We will refer to the ability to maintain an upright position during quiet standing as "balance." Force platforms, electro-myography, and motion analysis systems have all been used for assessment of balance.[1] These sophisticated systems, however, are expensive and often impractical for use by a therapist in a typical hospital or private practice. They require considerable floor space, special training, and computers. These resources may be unavailable to the clinician who would like to be able to test patients' balance, but who lacks funds to purchase sophisticated equipment or who must carry equipment from place to place. Therefore, a simpler, less expensive, valid, and reliable test is needed.

The Clinical Test of Sensory Interaction and Balance (CTSIB) is a timed test that was developed for systematically testing the influence of visual, vestibular, and somatosensory input on standing balance.[2] This test is inexpensive, requires minimal equipment, and is currently in use by some clinicians. Conditions 1, 2, and 3 involve standing on the floor with eyes open, eyes closed, and wearing a visual-conflict dome. The dome provides a sensory conflict by depriving the subject of peripheral vision and introducing a sway-referenced image. Use of the conflict dome results in a discrepancy between vestibular input stimulated by postural sway and visual flow.[1] Thus, conditions 2 and 3 should examine different aspects of sensory organization of visual information that may require different postural adjustments.[3] That is, condition 2 examines how well subjects maintain balance in the absence of any vision, and condition 3 examines how well subjects maintain balance when vision is present but that information conflicts with vestibular information.

Conditions 4, 5, and 6 involve standing on foam and repeating the visual conditions described for conditions 1 through 3. For each condition, the length of time the subject can maintain standing and the amount of body sway that occurs are assessed.

Although norms for this test have been established for children and young adults, norms for performance on the CTSIB have not been established for older adults and patients with vestibular deficits.[4,5] Currently, the CTSIB requires performance of at least one trial on each of the six conditions. Some investigators have questioned whether the eyes-closed and the visual/vestibular-conflict conditions produce different performance. Billek-Sawhney(5) found no differences on measures of duration and sway amplitude in the eyes-closed and visual-conflict conditions in neurologically asymptomatic (AS) young adults. No studies in the literature address these issues in older adults and individuals with vestibular disorders.

It is unclear whether patients improve on repeated trials on the CTSIB, because different investigators[4,6] have used different methods for calculating performance times and amount of sway. Physical therapists have advocated the use of balance retraining that involves repeated exposure to different sensory conditions when standing,[7] but no studies have distinguished between motor performance on this test and motor learning as a result of practice. Determining the need for repeated trials during assessment would be useful. These findings suggest that the CTSIB should be studied further. Previous findings suggest that therapists using the CTSIB should expect performance on this test to vary with respect to age and health status.[6,8-10] This study had several goals: (1) to determine whether healthy adults of different ages had different timed balance scores on the CTSIB, (2) to determine whether healthy subjects performed differently on each of the six conditions, (3) to learn whether individuals diagnosed with vestibular disorders performed differently than AS subjects, and (4) to determine whether subjects' performance improved over trials.

Method

Subjects

Pilot data were collected from 22 senior physical therapy students (9 men, 13 women), aged 20 to 24 years ([[bar]X]=21.3, SD=0.85). Subjects in the experiment were divided into four groups. Groups 1, 2, and 3 each comprised 15 AS subjects. Group 1 comprised 5 men and 10 women, aged 25 to 44 years ([[bar]X]=39.3, SD=5.5). Group 2 comprised 4 men and 11 women, aged 45 to 64 years (X=52.1, SD=6.2). Group 3 comprised 1 man and 14 women, aged 65 to 84 years [[bar]X]=75.1, SD=5.9). No subjects were obese. Subjects were screened for major health problems, and only individuals with no history of "dizziness," balance disorders, or recent orthopedic problems were included. Subjects in groups 1 and 2 were recruited from among the physical therapy students, staff, and faculty at the Medical College of Ohio, Toledo, Ohio. Group 3 subjects were retired elderly people living in the community.

Group 4 comprised 17 patients (7 men, 10 women), aged 30 to 87 years X = 59.8, SD = 18.9), diagnosed with vestibular disorders by a board-certified otolaryngologist specializing in vestibular disorders. In addition to the clinical evaluation, diagnostic tests included computerized harmonic acceleration tests of the vestibuloocular reflex, optokinetic nystagmus, ocular pursuit and saccades, and caloric tests, during which eye movements were recorded with electrooculography. Patients' diagnoses included benign paroxysmal positional vertigo, vestibular neuronitis, cupulolithiasis, labyrinthitis, and vestibular disorder of idiopathic origin. Their initial complaints included vertigo, disequilibrium, and blurred vision. The physician referred all patients for physical therapy at the Medical College Hospital. All subjects gave informed consent before participating in this study.

Equipment

The materials for this test included a 40.64x40.64x7.62-cm piece of medium-density Sunmate[*] foam,(2) a visual/vestibular-conflict dome made from a Chinese lantern attached to a plastic sun visor, and a stopwatch. (A sun visor is a hatless brim attached to an elastic band covered in terry cloth, which holds the brim over the forehead to shade the eyes.) We used a sun visor that could be detached from its elastic band, so that different bands could be used with each subject. Between test sessions, the elastic bands were washed. This detail eliminated any concerns subjects might have had about hygiene. The dome was constructed according to the description by Shumway-Cook and Horak(2) so that the subject saw a fixation point, a large black cross, centered in visual field. The total cost for materials was approximately $40. Because the materials needed for this test are inexpensive, even clinics with small budgets can afford to obtain the necessary equipment.

Procedure

All subjects were tested for three trials on each of the six conditions, in stocking feet. The conditions were (1) quiet standing on the floor, looking straight ahead; (2) quiet standing on the floor with eyes closed; (3) quiet standing on the floor wearing the conflict dome; (4) quiet standing on the foam, with eyes open; (5) quiet standing on the foam, with eyes close; and (6) quiet standing on the foam wearing the conflict dome. Between trials for conditions 4 through 6, the foam was turned over and rotated 90 degrees to prevent the foam from wearing unevenly over many trials. Subjects rested between trials, for 30 to 60 seconds, to eliminate the confounding effect of fatigue.

Pilot work showed no effect of order of conditions, with subjects who understood the nature of the task, when conditions 1 through 3 were given before or after conditions 4 through 6. Similarly, no differences were found when the orders of conditions 2 and 3 and conditions 5 and 6 were reversed. Performance is known to be affected by the performer's level of understanding of the skill.[11] Although the postural control aspect of the test is presumably automatic, assuming the correct position of the feet and hands could require some practice to understand the nature of the task. Therefore, the experimental paradigm was always administered using conditions 1 through 3 first to give the subjects the idea of the position required. For that reason, condition 1, the least complicated condition, always preceded all other conditions, and condition 4 always preceded conditions 5 and 6.

Prior to testing in each condition, the investigator demonstrated the task. The test was administered with the conditions in the same order each time. For all conditions, the subject was instructed to stand quietly, with arms comfortably across the waist, feet together, for as long as possible, up to 30 seconds. This period of time had been specified in the original description of the test.2 The instructions given by all investigators were standardized and were changed slightly for each condition.

Prior to starting the test, subjects were told that each trial would last for up to 30 seconds. For condition 1, the investigator told the subject, "Stand with your feet together, hands across your waist, and look straight ahead.

Do this until I tell you to stop." The instructional set for the other conditions included the instruction to "close your eyes" for condition 2 and "Now I'd like you to wear this hat and look at the cross" for condition 3. For conditions 4 through 6, subjects were given the same instruction regarding visual conditions and were also asked to stand on the center of the foam. The length of time the subject could maintain balance was recorded. A trial was terminated when the subject's arms or feet changed position. For subjects who were able to perform all 18 trials for 30 seconds, the test took approximately 20 minutes. All subjects were tested in a quiet, well-lighted room with a linoleum floor.

Data Analysis

The data were analyzed using the mean time performing the test, over the three trials for each condition. Test-retest reliability and interrater reliability were determined using Pearson Product-Moment Correlation

Coefficients. Differences among experimental groups were determined using an analysis of variance (ANOVA) for repeated measures. Significant differences were then subjected to post hoc Tukey's tests. Differences between AS subjects and age-matched vestibularly impaired subjects were examined with t tests using the Bonferroni correction.

Results

Both test-retest and interrater reliability were high (r=.99, P<.0l). These measures were taken with the pilot group. Interrater reliability was evaluated by having two investigators at a time test five subjects simultaneously, using identical digital stopwatches. Values were rounded to the nearest half second. Test-retest reliability was tested by having the same investigator test five subjects twice.

The ANOVA showed no significant differences among groups for conditions 1 through 3. That is, all subjects could stand on the floor for 30 seconds with eyes open, eyes closed, and wearing the conflict dome, for all three trials.

The results were somewhat different for conditions 4, 5, and 6 eyes open, eyes closed, and wearing the conflict dome, respectively, while standing on the foam). As shown in Figure 1, and confirmed with the ANOVA, subjects in groups 1, 2, and 3 performed condition 4 for significantly longer than condition 5 (F[16,34] = 11.35, P<.001). This difference stemmed from differences across conditions among the older subjects. Group 3 showed a performance decrement on conditions 5 and 6, and their scores on conditions 5 and 6 did not differ significantly. Performance of the subjects in group 4 was slightly different. As with the AS subjects, their performance on condition 4 was significantly better than on condition 5, but unlike AS subjects their performance on condition 6 was significantly poorer than on condition 4. Figure 1 shows that their scores on conditions 5 and 6 did not differ significantly.

When the data were evaluated by conditions (Fig. 2), no significant differences were found between the scores of groups 1 and 2, on any conditions. On condition 4 (eyes open on the foam), no differences were found among scores of subjects in groups 1, 2, and 3. Subjects in group 4, however, scored significantly lower than subjects in the other groups (F[16,34] = 2.12, P<.04). On condition 5 (eyes closed on the foam), group 3 had scores significantly lower than those of groups 1 and 2, but not significantly different from those of group 4. Although group 3 performed significantly worse than groups 1 and 2 on condition 6 (foam and dome), the scores of group 3 were significantly higher than those of group 4, as shown in Figure 2. As the figure indicates, both groups 3 and 4 had considerable variability in their scores.

When subjects were matched by age and mean scores were compared with related-measures t tests, no differences were found between AS and vestibularly impaired subjects on condition 4. Figure 3 shows that groups 1, 2, and 3 performed significantly better than group 4 on conditions 5 and 6 (condition 5: t[161=4.17, P<.001; condition 6: t[161=5.58; P<.001). Under these conditions, vision was eliminated or conflicted with vestibular input. Group 4 also had much more variability than did the other groups. In general, these data show that older AS subjects (group 3) and subjects with vestibular impairments (group 4) had consistently lower scores on conditions 5 and 6.

The data were also analyzed by trials to look for a practice effect. Groups 1 and 2 had no significant differences across trials for any condition. Similarly, for groups 3 and 4, no differences were found across trials on condition 4. On conditions 5 and 6, however, groups 3 and 4 had differences between scores on trials 1 and 2 F[3,58,8,241 =7.22, P<.001) (Fable). Some subjects in both groups improved their performance on both conditions. The importance of this finding, however, is unclear because both of these groups had such great variability in their scores. The means and standard deviations of these data are shown in the Table.

Discussion

The CTSIB is an inexpensive, easily administered test of standing balance that is useful in the typical physical therapy clinic. These results suggest that it can be modified for even easier administration. Because no differences were found among groups on the first three conditions, these conditions could be eliminated for subjects with only peripheral vestibular disorders. Norre et al(12) and other investigators (CA Blatchly, SL Whitney, and JMRF Furman; unpublished data) have reported differences in conditions 1 and 2, but these differences may be attributed to the use of different foot positions in each study, as well as to the presence of central neurological problems.

Asymptomatic older adults can perform condition 4 as well as younger subjects, although people with vestibular disorders cannot. Therefore, this condition may serve as a useful baseline, particularly when assessing older patients with vestibular disorders and other balance problems.

On condition 5, both the older group and the vestibularly impaired group performed more poorly than did the younger AS groups. These findings are consistent with those of previous work.(10) The vestibularly impaired group performed at the same level as the older AS group, regardless of age. On the measure reported in this study, younger subjects with vestibular impairments performed as if they were older people.

On condition 6, although the elderly AS subjects had lower scores than did their younger counterparts, they were better able to perform this condition than subjects with vestibular disorders. These data suggest that a score of 20 seconds on conditions 4, 5, and 6 with the feet together is within normal limits for older subjects. Condition 6 may also be useful in discriminating between older people with and without vestibular disorders. When a therapist suspects a vestibular disorder in a patient without such a diagnosis, these data may help the physical therapist make a referral to the appropriate physician for evaluation. Such a difference may also be useful in reassessing patients after a course of physical therapy.

Older AS and vestibularly impaired subjects tended to show higher scores with successive trials on the two conditions in which vision was eliminated or not useful. This finding suggests that these subjects may have used an unsuccessful movement strategy initially, but were able to modify their motor plans with practice. This finding may indicate that these subjects took longer than younger AS subjects to understand the motor requirements of the task. Because the subjects with vestibular lesions had more variability than other subjects, and because those subjects did improve over trials, it might be useful to administer this test using three or more trials and take the mean of those trials.

This study examined subjects' ability to maintain quiet upright standing when sensory inputs were systematically altered. Measures of sway could provide further insight into performance abilities of individuals in different age groups, but that issue was not examined in this study because observing sway in the clinic would have required two observers or more sophisticated, expensive equipment. one purpose of this study was to make it easy for a single physical therapist to administer this test.

Conclusions

Although the CTSIB does not specify the exact nature of a subject's balance problem, it is useful in differentiating between individuals with and without vestibular disorders. The test is also useful for obtaining data about patients' performance before and after therapy, and thus in documenting the efficacy of treatment, for the benefit of third-party payers. Because the CTSIB is inexpensive, it is a useful option for clinics in which expensive dynamic posturography testing equipment is unavailable, but where the therapists still need objective data about balance.

Acknowledgments

We thank Rebecca Koch, PT, and Millicent Branch, PT, for their assistance.

References

[1] Horak FB. Clinical measurement of postural control in adults. Phys Ther. 1987;67: 1881-1885. [2] Shumway-Cook A, Horak FB. Assessing the influence of sensory interaction on balance: suggestion from the field. Phys Ther. 1986;66: 1548-1550. [3] Nashner LM, McCollum G. The organization of human postural movements: a formal basis and experimental synthesis. Bebav Brain Sci. 1985;8:135--172. [4] Crowe TK, Deitz JC, Richardson PK, Atwater SW. Interrater reliability of the pediatric clinical test of sensory interaction for balance. Physical and Occupational Therapy in Pediattics. 1990;10:1--27. [5] Biliek-Sawhney B. Clinical and Objective Assessment of postural Stability. Pittsburgh, Pa: University of Pittsburgh; 1990. Thesis. [6] Di Fabio RP, Badke MB. Relationship of sensory organization to balance function in patients with hemiplegia. Phys Ther. 1990;70: 542-548. [7] Toal Tangeman P, Wheeler J. inner ear concussion syndrome: vestibular implications and physical therapy treatment. Topics in Acute Care Trauma Rebabilitation. 1986;1:72-83. [8] Pyykko I, Aalto H, Hytonen M, et al. Effect of age on posture control. In: Amblard B, Berthoz A, Clarack E, eds. Posture and Gait: Development, Adaptation, and Modulation. Amsterdam, the Netherlands: Elsevier; 1988: 95-104. [9] Straube A, Botzel K, Hawken M, et al. Postural control in the elderly: differential effects of visual, vestibular and somatosensory input. in: Amblard B, Berthoz A, Clarack E, eds. Posture and Gait: Development, Adaptation, and Modulation. Amsterdam, the Netherlands: Elsevier; 1988:105-114. [10] Woollacott MH. Aging, posture control and movement preparation. In: Woollacott MH, Shumway-Cook A, eds. Posture and Gait Across the Lifespan. Columbia, SC: University of South Carolina Press; 1989:155-175. [11] Gentile AM. A working model of skill acquisition with application to teaching. Quest. 1972; 17:3-23. [12] Norre ME, Forrez G, Beckers A. Vestibular habituation training and posturography on benign paroxysmal positioning vertigo. ORLJ Otorbinolaryngol Relat Spec. 1987;49:22-25.

Commentary

The development of effective methods for assessing and treating adults with vestibular deficits is a prominent issue for physical therapists and occupational therapists involved with "vestibular rehabilitation." The article by Cohen et al provides a vehicle for the kind of dialogue that is needed about this important topic. I would consider their study preliminary, however, in view of several issues related to the broad generalization of their results, the inconsistency of age-matched comparisons, the recommendation to delete various aspects of the Clinical Test of Sensory Interaction on Balance (CTSIB), and the absence of a documented relationship between stance duration and functional status in patients with vestibular impairments.

Generalization of Findings

The primary conclusion reported by Cohen and colleagues was that the CTSIB "... is useful in differentiating between individuals with and without vestibular disorders." I believe that this conclusion is potentially misleading for several reasons:

1. Subjects with and without active vertigo have equivalent scores on tests of sensory interaction acquired with posturography.[1] The conditions used for evaluating balance with posturography and the CTSIB are essentially the same. Posturography, however, incorporates a force platform and a visual enclosure that can be referenced to spontaneous displacements of the subject's center of force.[2] Posturography provides a more sensitive measure of balance compared with the CTSIB because manipulation of the sensory environment is precisely controlled and equilibrium scores are derived from vertical floor reaction forces. It is unlikely, therefore, that the CTSIB will identify sensory integration deficits in many patients with vertigo, because more sensitive measures do not detect deficits related to this symptom.

2. Subjects with compensated (chronic) unilateral peripheral vestibular impairments often have normal balance responses when tested with a neutral head position during posturography.[3 ]The sensitivity of posturography for identifying abnormalities related to chronic vestibular lesions appears to be increased when patients are positioned with their head tilted contralateral to the side of the lesion.[3 ]The protocol described by Cohen et al did not include variations in head position. The CTSIB, therefore, may fail to identify many patients with compensated unilateral vestibular impairment, partly because head position was not varied during the test procedure.

3. Sensory integration deficits in patients with vestibular impairment of idiopathic origin (VIIO) often cannot be distinguished from balance responses in healthy subjects.4 Cohen and colleagues included subjects with vestibular deficits of unknown origin together with those who had localized vestibular lesions (group 4). This procedure could mask the normal balance response associated with VIIO in some conditions, because stance times were averaged for the entire patient group. When the CTSIB is applied in the field to a given patient with a vestibular deficit of undetermined origin, I would not expect this test to consistently identify sensory integration deficits.

These three examples illustrate that the sensitivity of the CTSIB may be dependent on the diagnosis and the patient's particular symptoms. Combining the types of patients described in the examples above with those who have localized vestibular lesions or acute vestibular deficits might conceal abnormalities that are typically not detectable using the CTSIB. The point is that the authors' conclusion might have been inappropriately generalized to all patients with vestibular impairment because "vestibular disorders" were treated as a single entity.

Inconsistencies in Age-Matched Comparisons

The use of "aged-matched" comparisons between healthy subjects and those with vestibular disorders was not consistently applied throughout the study. Postural instability is known to increase with advancing age beyond the second decade of life.[5-7] Therefore, when measuring balance across groups that are not equivalent with respect to age, groups with younger subjects would be expected to have longer stance times, whereas groups with older subjects would be expected to have lower stance durations.[8] The comparisons of stance duration between the patient group and the remaining three groups of healthy subjects (Fig. 2 in the article) were not corrected for age differences. Based on the analysis associated with Figure 2, the authors assert that (1) subjects with vestibular disorders (group 4) did not perform as well in condition 4 (stance on foam with eyes open) compared with healthy subjects (group 3), and (2) "the vestibularly impaired group performed at the same level as the older AS [asymptomatic] group, regardless of age" in condition 5 (stance on foam with eyes closed).

These assertions directly contradict the "age-matched" results reported by the authors in Figure 3. Specifically, the subjects with vestibular disorders showed equivalent stance duration compared with age-matched healthy subjects in condition 4 and were found to have lower stance duration compared with the age-matched healthy subjects in condition 5.

Deletion of Stance Conditions

I would urge some caution in following the authors' recommendation to delete conditions 1 through 3 (quiet standing on a fixed, noncompliant surface with eyes open, eyes closed, and wearing a visual-conflict dome). This recommendation was based on the finding that subjects with vestibular disorders had normal stance times when tested in these conditions. Specific diagnostic groups were not studied by Cohen and colleagues, and the laterality or symmetry of vestibular deficits was not reported. Their results do not agree with the findings of previous studies that did focus on specific vestibular deficits. Black and Nashner[9] showed that patients with "pure" benign paroxysmal positional nystagmus had significant abnormalities in condition 3 (stance on a fixed surface with sway-referenced vision). Norre and Forrez[10] used information from several test conditions including conditions 1 and 2 (stance on a fixed surface with eyes open and eyes closed) to classify the extent of abnormal sensory interaction on balance in patients with Meniere's disease and paroxysmal positional vertigo. Kantner et al4 found that patients with peripheral or central vestibular deficits had greater instability in conditions 1 through 3 compared with healthy subjects. Clinical tests of balance that include conditions with eyes open, eyes closed, and visual stabilization during stance on a fixed and compliant surface, therefore, provide more complete information about the underlying cause of balance dysfunction than do tests with these sensory conditions deleted.

Functional Significance of the CTSIB

The authors suggested that the CTSIB could be used for "documenting the efficacy of treatment...." In order to support this claim, I believe that the relationship between stance duration and some measure of functional status should have been reported. The relevance of stance time as a functional outcome measure for patients with vestibular impairment has not been established in the context of the CTSIB. Stance time is directly correlated with functional status in healthy elders8 and with functional level following stroke.[11] Single-leg stance time has been used as one index to measure the effectiveness of vestibular rehabilitation,[12] but this procedure did not involve visual stabilization or stance on a compliant surface. The outcome measure used by Cohen and colleagues requires further study in order to understand the functional significance of the CTSIB when used with subjects who have vestibular deficiencies. it is my hope that their article will serve as a stimulus for additional clinical research to address this issue. Richard P Di Fabio, PbD, PT Associate Professor and Director of Graduate studies Program in Physical Therapy Department of Physical Medicine and Rehabilitation and Department of otolaryngology University of Minnesota PO Box 388, UMHC 420 Delaware St SE Minneapolis, MN 55455

References

[1] Goebel JA, Paige GD. Dynamic posturography and calorie test results in patients with and without vertigo, Otolaryngol Head Neck Surg. 1989;100:553-558. [2] Black FO, Wall C, Nashner LM. Effects of visual and support surface orientation references upon postural control in vestibular deficit subjects. Acta Otolaryngol (Stockh). 1983; 95:199-210. [3] Barin K, Seitz CM, Welling DB. Effect of head orientation on the diagnostic sensitivity of posturography in patients with compensated unilateral lesions. Otolaryngol Head Neck Surg. 1992;106:355-362. [4] Kantner RM, Rubin AM, Armstrong CW, Cummings V. Stabilometry in balance assessment of dizzy and normal subjects. Am J Otoglaryngol 991;12:196-204. [5] Kollegger H, Baumgartner C, Wober C, et al. Spontaneous body sway as a function of sex, age, and vision: posturographic study in 30 healthy adults. Eur Neurol. 1992;32:253-259. [6] Horak FB, Shupert CL, Mirka A. Components of postural dyscontrol in the elderly: a review. Neurobiol Aging. 1989;10:727-738. [7] Sheldon JH. The effect of age on the control of sway. Geront Clin. 1963;5:129-138. [8] Anacker SL, Di Fabio RP. Influence of sensory inputs on standing balance in community-dwelling elders with a recent history of falling. Phys Ther. 1992;72:575-584. [9] Black FO, Nashner LM. Postural disturbance in patients with benign paroxysmal positional nystagmus. Ann Otol Rhinol Laryngol. 1984;93: 595-599. [10] Norre ME, Forrez G. Vestibulospinal function in otoneurology. ORL J Otorbinolaryngol Relat Spec. 1986;48:37-44, [11] Di Fabio RP, Badke MB. Relationship of sensory organization to balance function in patients with hemiplegia. Phys Ther. 1990;70: 542-548. [12] Horak FB, Jones-Rycewicz C, Black FO, Shumway-Cook A. Effects of vestibular rehabilitation on dizziness and imbalance. Otolaryngol Head Necksurg. 1992;106:175-180.

Author Response

We appreciate Dr Di Fabio's comments on our study of the Clinical Test of Sensory Interaction and Balance (CTSIB). We agree that this study was a preliminary examination of the issues related to this test. We did not intend to address the relationship between stance duration and functional status. We intended only to study the issue of timed performance in older asymptomatic adults and vestibularly impaired adults. We are pleased that our study has stimulated some discussion, and we hope that it will stimulate further research.

We agree that computerized, well-controlled posturography may be more sensitive and more accurate than the CTSIB. Many physical therapists, however, do not have sufficient funds in their budgets to buy such expensive test apparatuses. Because of the minimal expense in obtaining the materials for this test, we believe it is more accessible for physical therapists in small clinics, or in clinical environments in which posturography is not practical, such as home health care.

Therapists in these practice settings have the same needs for valid and reliable tests of balance as therapists in major medical centers or university settings. For these practitioners, the CTSIB is a good alternative test.

Di Fabio's comments about the generalizability of our findings are interesting, but we disagree about the meaning of at least one of his citations. He cited a reference that shows "subjects with and without active vertigo have equivalent scores on tests of sensory interaction acquired with posturography" and suggests that this finding implies that neither posturography nor the CTSIB can be used to distinguish between patients with and without vestibular disorders. in spite of the possibility that posturography alone may not show these differences, electromyographic studies demonstrated that patients with peripheral vestibular deficits showed response amplitudes that were significantly less than those of "normal" individuals and were correlated with extent of clinical vestibular deficit.[1 ]These findings suggest that individuals with vestibular disorders may indeed demonstrate shorter stance times on the CTSIB than "normal" individuals.

Our protocol involved no changes in head position because we tested the CTSIB using the original description in the literature.[2] We agree that tilting the head may alter the results of testing in patients with peripheral vestibular disorders. We also agree that future studies of subjects with different types of vestibular pathology should attempt to separate these groups based on type of pathology.

We agree that conditions 1, 2, and 3 standing on the floor with eyes open, eyes closed, and wearing a visual-conflict dome) should be deleted with caution. As we stated in our report, the only group for which we considered eliminating these conditions were those subjects with peripheral vestibular disorders. This group generally demonstrates less severe instability than individuals with mixed, central, or bilateral vestibular disorders.[3] The finding that Di Fabio cited by Kantner et al may be problematic because the algorithm used to calculate sway may have been incorrect (S Woolley, personal communication). These conditions might prove to be more useful when administered with the head tilted than with the head erect.

We agree that the relevance of stance time as a functional outcome measure has not been established for the CTSIB. If, however, stance time correlates well with functional status in asymptomatic elderly people, then improved stance time in the variety of sensory conditions on the CTSIB should correlate well with functional status in individuals who demonstrate initial instability secondary to vestibular deficits.

We agree that the issues in balance testing, including those raised by this study, are not yet settled. We hope that this dialogue stimulates other teams of academic and clinical therapists to look into this important issue. Helen Cohen, Edd, OTR Cathleen A Blatchly, PT Laurie L Gomhash, PT

References

[1] Allum JHJ, Keshner EA, Honeggar F. Organization of leg-trunk-head equilibrium movements in normals and patients with peripheral vestibular deficits. Prog Brain Res. 1988;76: 277-289. [2] Shumway-Cook A, Horak FB. Assessing the influence of sensory interaction on balance: suggestion from the field. Phys Ther. 1986;66: 1548-1550. [3] Blatchly CA. The Relationship Between Subjective Measures of dizziness and Objective Measures of balance. Pittsburgh, Pa: University of Pittsburgh; 1990. Master's thesis.

H Cohen, Edd, OTR, is Assistant Professor, Department of Otorhinolaryngology and Communicative Sciences, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030 (USA). She was Assistant Professor, Program in Rehabilitation Science, and Assistant Professor, Department of Otola Medical College of Ohio, Toledo, OH 43699, at the time of this study. Address all correspondence to Dr Cohen.

CA Blatchly, PT, is Assistant Professor, Program in Physical Therapy, Medical College of Ohio.

LL Combash, PT, is Physical Therapist, Medical College Hospital, and Clinical Lecturer, Department of Otolaryngology, Medical College of Ohio.

This study was approved by the Institutional Review Board of the Medical College of Ohio.

This article was submitted February 18, 1992, and was accepted January 29, 1993.