Age-related balance changes in hearing-impaired children.

Problems of the deaf have been defined primarily in terms of communication deficits.(1) Although the communication deficit is the major impediment associated with significant hearing loss, other physical problems are associated with deafness. Balance deficit, which may interfere with normal motor development and sensory integration, is often observed in deaf children.(2-7) Damage to portions of the vestibulocochlear nerve, the presumed cause of sensorineural deafness, may include not only damage to the cochlear apparatus but also damage to the vestibular afferents. Damage to vestibular afferent fibers is one possible explanation of the balance deficit.

Early research on motor function indicated that, when deaf children were compared with children with normal hearing, the deaf children showed a deficit only in balance ability.(3-5) Morsh(3) used the Dunlap Balancing Board to compare balance in deaf and normal-hearing children of high school and early college age. The Dunlap Balancing Board consists of a square platform resting centrally on ball bearings. The board tilts through an adjustable arc of motion and completes an electrical circuit during the period that balance is maintained. Morsh tested the balance of each blindfolded subject as the board tilted through three different arcs (ie, 3 deg, 11 deg, and 19 deg). Based on the time the electrical circuit was complete, he found that the deaf children were superior to the normal-hearing children balancing through a 3-degree excursion, but that the normal hearing children were far superior to the deaf children at both 11 and 19 degrees.(3) Long(4) had subjects walk heel-to-toe on a 3.81-cm-wide (1.5-inwide) balance beam, and he recorded the mean number of successful steps on the beam for the three trials. Ten successful steps was the maximum number permitted. Long concluded that balance in deaf children was inferior to that in children with normal hearing because the deaf children took fewer successful steps. Boyd(5) tested static and dynamic equilibrium in 8-, 9-, and 10-year-old boys using an adaptation of the Oseretsky scales. Although his method was not fully described, Boyd reported differences in static balance between deaf and normal-hearing boys at all ages and significant differences in dynamic balance between the deaf and normal-hearing boys of 9 and 10 years of age.

Later research by Lindsey and O'Neal,(6) which included items from Boyd's adaptation of the Oseretsky scales, showed that 8-year-old deaf children were far inferior to age-matched normal-hearing children in tasks involving both static and dynamic balance, Lindsey and O'Neal used 6 tasks to test for static balance and 10 tasks to test for dynamic balance. The results of their study conflicted with those of Boyd's earlier study regarding the 8-year-old population. This conflict could have been due to Lindsey and O'Neal not limiting their balance tasks to those used by Boyd, but using additional items from the Meeting Street School Screening Test7 and from tests used by Cratty(8) and Touwen and Prechtl.(9) The relationship between gender and balance has been examined by a variety of investigators.(2-4,6,10,11) Myklelbust,(2) Morsh,(3) and Long(4) each concluded that balance of deaf male subjects was superior to balance of deaf female subjects. They used either standing balance or beam walking as tests in their studies. Morsh(3) found that when subjects were blindfolded, the male superiority vanished. Later studies by Lindsey and O'Neal(6); Carlson,(10)who used the Brace Motor Ability Test; and Potter and Silverman," who used tests of standing balance for subjects with eyes open and closed, demonstrated no gender difference in balance skills.

Although several investigators(5,6,12) have reported impaired balance in deaf children, others have emphasized that not every deaf child appears to have a balance deficit. Potter and Silverman" suggest that children who are born deaf are less likely to have balance dysfunction than those who acquire deafness, usually secondary to meningitis, in infancy or early childhood. A lack of universal findings regarding balance deficits and deafness suggests that some deaf children simply do not have a deficit, whereas others may have the deficit but use other sensory systems to compensate. Crowe and Horak(l3) used a very sophisticated system for testing vestibular function in a group of hearingimpaired children prior to testing motor proficiency. They found a group of hearing-impaired children with normal vestibular function who perfon-ned a series of motor tasks as well as, or better than, control subjects. The hearing-impaired children with reduced vestibular function had poor results on the Balance subtest of the Bruininks-Oseretsky Test of Motor Proficiency (BOTMP).(13) There are inconsistencies in the literature regarding the improvement of balance ability with age in deaf children. Myklebust(l4) found that, as the age of deaf subjects increased, there was an increase in their ability to successfully complete a rail-walking test.(15) Morsh(3) described balance improvement in deaf subjects up to age 9 years, after which no clear improvement could be identified. Carlson,'O using the Brace Motor Ability Test, showed no significant improvement in balance in deaf children older than 7 years compared with younger children.

Unfortunately, few previous studies systematically compared age-related changes in balance ability of deaf children with those of a similar population of normal-hearing children. Carlson(10) and Heath(15) each used a standardized test instrument to determine age-related changes in balance. Each standardized instrument had a flaw, however, that raises questions about the validity of the studies. Carlson(10) used a balance test that was standardized and published 44 years prior to his 1971 study. We believe that a significant improvement in physical characteristics, including height and weight, may have occurred during the years between the development of the instrument (1927) and its use by Carlson.(10) In addition, we believe that, because of societal trends, girls were more likely than boys to be physically active in recent years when compared with 1927. Myklebust(l4) studied 203 deaf subjects, aged 5 to 21 years, and noted progressive improvement in balance until the age of 18 years. Myklebust, however, used a standardized test developed by Heath,(15) who used US Army soldiers as the standardization population. We question the use of data obtained on an adult male sample as a standard against which to compare the abilities of children and teenagers of both sexes.

The purpose of this study was to compare the scores on a standardized balance test of three age groups of deaf children with those of a sample of normal-hearing children on whom the balance test was standardized. Our specific research hypotheses were: 1. When compared with normal hearing children, deaf children would have a deficit in balance. 2. The balance deficit in deaf children would diminish with age. 3. There would be no difference in balance ability between deaf male and female children.

Method

Subjects

A sample of convenience was obtained at the Pennsylvania School for the Deaf (Philadelphia, Pa). Informed consent was obtained from a parent or guardian of each participating child. Inclusion criteria for the subjects included a sensorineural hearing loss of >/=65 dB and normal intelligence (a score of 80 or higher on a standard test of intelligence). We did not determine the etiology of hearing loss. Exclusion criteria included any neuromuscular or musculoskeletal condition, legal blindness, developmental delay, or learning disability as identified from school records. The children were divided into three groups according to age. Group 1 consisted of 10 children, aged 4.5 to 6.5 years; group 2 consisted of 8 children, aged 8 to 10 years; and group 3 consisted of 10 children, aged 12.5 to 14.5 years. We attempted to use age groups that span the 4- to 14-year age range of the standardization population for the BOTMP.

Equipment

We used the Balance subtest of the BOTMP and a stopwatch. The Balance subtest of the BOTMP measures both static and dynamic balance.

Procedure

We administered the Balance subtest of the BOTMP in a room free from distractions. Before testing, the BOTMP pretest for arm and leg preference was administered to each subject. We asked each subject to kick a ball twice to determine his or her preferred leg, which was used for the first three test items. The following eight items comprise the Balance subtest of the BOTMP:

1. Subject stands on the preferred foot on a line drawn on the floor while looking at a target on the wall. Both hands are on the hips, and the free (nonpreferred) leg is flexed at the knee.

2. Subject stands on the preferred foot on a balance beam while looking at a target on the wall. Both hands are on the hips, and the free leg is flexed at the knee.

3. Subject stands as in item 2, except with eyes closed. (In items 1-3, the trial is stopped after 10 seconds and the time is then recorded. The trial is stopped before 10 seconds if the subject touches the free leg to the floor, drops the free leg below a 45' angle, hooks the free leg behind the supporting leg, or shifts the supporting foot out of place.)

4. Subject walks forward on a line on the floor using a normal stride. Both hands are on the hips.

5. Subject walks forward on a balance beam using a normal stride. Both hands are on the hips. (Items 4 and 5 are scored with a maximum of six steps. If the subject places one foot or both feet completely off the line or beam prior to six steps, the test is stopped and the number of successful steps is recorded.)

6. Subject walks forward on a line on the floor with a heel-to-toe gait. Both hands are on the hips.

7. Subject walks forward on a balance beam with a heel-to-toe gait. Both hands are on the hips. (Items 6 and 7 are scored with a maximum of six correct steps. A step is incorrect if one foot or both feet are placed completely off the line or beam, the heel of the front foot fails to touch the toe of the rear foot, or the toe of the rear foot is moved forward to touch the heel of the front foot.)

8. Subject walks on a balance beam using a normal stride. Both hands are on the hips. Subject steps over a wand held by the examiner above the beam, at a height just below the knee. (The trial is recorded as a failure if the subject touches the stick firmly, swings the leg around the stick, or steps off the beam.)

One investigator (MM) was responsible for testing items 1 through 4, and a second investigator OCS) was responsible for testing items 5 through

8. Subjects in group 1 were tested two at a time to decrease the children's apprehension about unfamiliar people and setting. All other subjects were tested individually. As recommended in the BOTMP handbook, subjects wore either sneakers or crepe-soled shoes without regard to the height of the shoe. All directions were explained to each subject via total communication, which involves speech, sign language, body language, facial expression, and demonstration.

To ensure that instructions were understood, each child was permitted one practice trial for each item. The entire battery of tests was administered once to each child.

Data Analysis

Static balance was determined by the number of seconds, up to a maximum of 10 seconds, the subject could perform each of the first three items. Dynamic balance was determined in items 4 through 7 by counting the number of steps, up to a maximum of six steps, taken during each item. item eight was rated as pass or fail. If the subject was unable to reach the maximum time or number of steps on the first trial of each item, a second trial was permitted. As stated in the directions for the BOTMP, the highest score of the two trials was used for analysis.

Raw scores were converted to point scores as described in the BOTMP manual. Point scores are used for the BOTMP in order to convert raw scores (ie, seconds on a beam versus steps on a beam) to a common set of values. To determine whether a balance deficit was present, we used a z test to compare the mean point score of each age group with the mean point score of the appropriate normative data from the BOTMP. A z score can be used to compare sample data with the standardization population data. Each raw score is expressed as a deviation from the mean of the normative data to form a z score.(16) Normative data for the BOTMP were developed from a standardization plan based on the 1970 US census. A total of 765 children were tested, with consideration made for equal numbers of boys and girls, equally divided age groups, racial proportions, community size, and geographic region. Gallahue(17) stated that motor proficiency, as measured by the performance of a given child on a given day, shows the BOTMP to be valid. In addition, the test discriminates well between nondisabled populations and those with learning disability and motor delay related to mental retardation.(17) Test-retest reliability coefficients (Pearson Product-Moment Correlation Coefficients [r]) for the Balance subtest of the BOTMP range from .46 for second-grade girls to .73 for second-grade boys, with a general mean of .56.

A one-way analysis of variance (ANOVA) was performed on the point scores of groups 1, 2, and 3 to determine whether balance ability was age related. A second one-way ANOVA was used to compare the differences between point scores of the subjects in our sample and the normative data from the BOTMP standardization population. This difference in point scores is indicative of the size of the balance deficit. We used a Student's t test for independent samples to determine whether there was a difference in mean point scores between the male and female subjects in the sample. We determined intratester reliability in a manner consistent with the BOTMP using a Pearson ProductMoment Correlation Coefficient. We tested each Balance subtest item on the same five subjects in two separate testing sessions. Because each item was scored by only one investigator, there was no need to determine intertester reliability.

Results

Intratester reliability testing showed perfect correlation r=1.00) for test items 1, 4, 5, and 8. item 3 had an intratester reliability value of -.33. items 2, 6, and 7, although not perfect, were so closely correlated and had such little variability that it was impossible to calculate a Pearson Product-Moment Correlation Coefficient. For items 2, 6, and 7, there was perfect agreement for four of the five subjects; the fifth subject had only a minimal difference between test and retest values, as shown in Table 1. Individual point scores for all subjects and corresponding normative data are shown in Table 2. The mean balance score for group 1 was 8.8 (SD=3.5) as compared with a normative value of 16.3 (SD=9.4) for the corresponding age group of normal-hearing children z=2.52, P<.05). Group 2 had a mean balance score of 19.3 (SD=4.5) versus a normative value of 24.5 (SD=4.0) z= -3.70, P<.Ol). Group 3 had a mean balance score of 20.3 (SD=4.7) compared with a normative value of 25.9 (SD = 2.6) z= - 6.80, P<.Ol).

The first one-way ANOVA showed a significant difference among mean balance scores for all three groups (Tab. 3). A Scheffe test for multiple comparisons showed significant differences in mean balance scores between groups 1 and 2 and between groups 1 and 3 P<.0l), but no significant difference was found between groups 2 and 3.

The second one-way ANOVA showed no significant difference among the differences between mean scores for our groups of subjects and corresponding normative data. That is, the size of the balance deficit remained constant, regardless of age (Figure). The t test for independent samples, used to compare balance scores of male and female subjects, indicated no gender-based difference in balance within each age group and no difference across the entire sample.

Discussion

Previous studies(3-6) clearly demonstrated a balance deficit in deaf children. in order to examine age-related changes in balance ability in deaf children, we first needed to determine whether the expected balance deficit was present in our sample. This deficit was found when our sample was divided into three age groups and their scores on the Balance subtest of the BOTMP were compared with normative data for similar age groups of the BOTMP standardization population. The difference between our subjects' scores and the normative data for the Balance subtest was greatest for group 1 and was less pronounced for groups 2 and 3. Group 1, the youngest children, achieved Balance subtest scores that were almost 50% lower than the mean score of the normal-hearing population of the same age in the standardization trials. Groups 2 and 3 each had scores that were approximately 20% lower than the mean score of the standardization population of the appropriate age. We did not have access to information regarding the etiology of deafness in our sample, nor were we able to accurately determine whether the vestibular system of each child was intact. Accordingly, we may have included some children who did not have vestibular dysfunction. In addition to the previously described possibility of vestibulocochlear dysfunction causing poor performance in tests of balance ability, a lack of complete comprehension of testing instructions by the deaf child may be an additional cause of poor performance on the Balance subtest. We attempted to minimize the lack of comprehension by using total communication and permitting one practice trial for each test item. Our results agree with the findings of Morsh,(3) who used a Dunlap Balancing Board for his balance studies of deaf children and teenagers. Morsh reported an improvement in balance ability in deaf subjects only until age 9 years, after which no clear improvement was seen.(3) We found a significant improvement in balance when comparing the scores of groups I and 2 and groups 1 and 3. No significant difference was found between the scores of groups 2 and 3. The change in balance scores between the youngest children (group 1) and both older groups suggests that, in our sample, balance ability improved until approximately 9 years of age, the mean age of group 2. The lack of a significant difference between the balance scores of groups 2 and 3 supports the finding of previous studies(3,7) that balance ability reaches a plateau around 9 years of age.

Our findings agree with those of previous studies that demonstrated a significant balance deficit in children who are deaf. Despite this substantial deficit, the age-related changes in balance in our sample of deaf children followed a pattern almost identical to, at their reduced level, the scores of the standardization population from the BOTMP (Figure). Our results show that, when comparing our groups' scores with those of age matched normal-hearing subjects from the BOTMP standardization population, the deficit remains constant throughout the period of balance maturation.

Results of previous studies,(1,11,17-20) indicate that, in order to compensate for balance deficits, deaf children use other sensory systems, such as proprioception, kinesthesia, and vision. Our finding that mean balance scores of deaf children were equidistant from scores of normal-hearing children across age groups (Figure) strongly suggests that, if compensation occurs, the degree of compensation is insufficient to diminish the balance deficit in the three age groups we examined. Our findings of a significant balance deficit in each of the age groups of deaf children tested strongly suggest to us a need for intervention prior to the time at which balance ability becomes mature. Two recent studies(2l,22) examined the effects on balance of an exercise regimen for young deaf children. Effgen(2l) used a force platform to measure static balance before and after a 10-day exercise program in 25 deaf children, whose mean age was 9 years. She found a significant improvement in static balance following the exercise regimen.(21) Lewis and colleagues(22) implemented a 6-week exercise program for 11 deaf children aged 6 through 8 years. Using the Balance subtest of the BOTMP, the authors found that the exercise regimen improved balance scores in the experimental group, but they found no change in a control group of deaf children who did not exercise.(22) Although the studies cited previously involved formal exercise regimens, the physical education teacher can consult with the physical therapist to develop an age-appropriate physical activity program (eg, running, jumping, gymnastics) aimed at improving balance ability. just as early intervention appears beneficial for children with Down syndrome, cystic fibrosis, and other disabling conditions, early intervention may help reduce the balance deficits in children who are deaf(23,24)

Gender is another factor that some investigators(2-4) believed affected balance. Deaf male subjects were thought to be superior in balance ability to deaf female subjects, according to studies conducted in the 1930s.(2-4) Some researchers,(6,10,11) however, found no difference in balance abilities of deaf male and female subjects. The discrepancy could be explained by the observation that the studies by Myklebust,(2) Morsh,(3) and Long4 were conducted in the 1930s, when females typically had more protected lives and were less physically active than males. Consequently, deaf females may not have had the same opportunities to develop balance ability as did their male counterparts. Our findings support those of the more recent studies. We found no significant difference between male and female subjects. An uneven distribution of male and female subjects occurred within each group; however, the distribution was close within the total sample (13 girls and 15 boys). We believed age-appropriate activities or exercise used to improve balance ability in deaf children need not differ for male and female subjects because neither gender is superior.

Conclusion

The results of this study indicate acceptance of our first hypothesis that a balance deficit exists in deaf children. In order to accept our second hypothesis, there should have been a decrease in the size of the balance deficit in either group 2 or group 3. The size of the balance deficit was virtually identical in all three groups, however, as shown in the Figure. Because the difference appears to be unrelated to age, we must reject our second hypothesis. Our findings enable us to accept the third hypothesis, because we found no significant difference in the balance ability of deaf male and female subjects.

Further research should examine a wider range of ages with large numbers of children to more completely define the age-related changes in balance in deaf individuals. More importantly, a true longitudinal study of maturation of balance ability would be extremely instructive, particularly if a distinction is made between children with and without vestibular dysfunction. Other studies should determine whether early intervention will reduce the deficit in balance ability in deaf children and when, in the course of balance development, intervention is most effective.

References 1 Mykiebust HR. Towards a new understanding of the deaf child. Am Ann Deaf 1953;98:345-357 2 Mykiebust HR. The Psychology of deafness. 2nd ed. New York, NY: Grune & Stratton Inc; 1964. 3 Morsh JE. Motor performance of the deaf Comparative Psychological Monograph. 1936;13:1-51. 4 Long J. Motor abilities of deaf children. In: Contribution to Education, No. 514. New York, NY: Columbia University Teacher's College; 1932:1-67. 5 Boyd J. Comparison of motor behavior in deaf and hearing boys. Am Ann Deaf. 1967;112:598-605. 6 Lindsey D, O'Neal J. Static and dynamic balance skills of eight-year-old deaf and hearing children. Am Ann Deaf. 1976;121:49-55, 7 Hainsworth PK, Siqueland ML. Early Identification of Children with Learning Disabilities: The Meeting Street School Screening Test. Providence, RI: Crippled Children and Adults of Rhode Island Inc; 1969, 8 Cratty Bj. Development Sequence of Perceptual Motor Tasks. Long Island, NY: Educational Activities Inc; 1967. 9 Touwen BCL, Prechtl HFR. Examination Of the Child with Minor Nervous Dysfunction. Philadelphia, Pa: JB Lippincott Co; 1970. 10 Carlson RB. Assessment of motor ability of selected deaf children in Kansas. Percept Mot Skills, 1972;34:303-305. 11 Potter CN, Silverman LN. Characteristics of vestibular function and static balance skills in deaf children. Phys Ther. 1984;64:1071-1075. 12 McCarron LT, Ludlow GC. Sensori-neural deafness and neuromuscular dysfunctions: considerations for vocational evaluation and job placement. Journal of Rehabilitation. 1981;47:59-62. 13 Crowe TK, Horak FB. Motor proficiency associated with vestibular deficits in children with hearing impairments. Phys Ther. 1988;68:1493-1499. 14 Myklebust HR. Significance of etiology in motor performance of deaf children with special reference to meningitis. Am j Psychol. 1946;59:249-258. 15 Heath SR. Clinical significance of motor defect, with military implications. Am J Psychol 1944;57:482-499. 16 Cronbach IJ. Essentials of Psychological Testing. 3rd ed. New York, NY: Harper & Row, Publishers Inc; 1970:95. 17 Gallahue D. Understanding Motor Behavior in Children. New York, NY: john Wiley & Sons Inc; 1982. 18 Kandel ER, Schwartz JH. Principles of Neural Science. 2nd ed. New York, NY: Elsevier Science Publishing Co Inc; 1985:584-595. 19 Padden DA. Ability of deaf swimmers to orient themselves when submerged in water. Research Quarterly. 1959;30:214-225. 20 Kaga K, Suzuki J, Morsh RR. Influence of labyrinthine hypoactivity on gross motor development of infants. Ann AIY Acad Sci. 1981;374:412-420. 21 Effgen SK, Effect of exercise program on the static balance of deaf children. Phys Ther. 1981;61:873-877. 22 Lewis S, Higham L, Cherry DB. Development of an exercise program to improve the static and dynamic balance of profoundly hearing-impaired children. Am Ann Deaf 1985:130;278-284. 23 Connolly BH, Morgan S, Russell FF. Evaluation of children with Down syndrome who participated in an early intervention program. Phys Ther. 1984;64:1515-1519. 24 Doershuk CF, Mathews LW, Tucker AS, Spector S. Evaluation of a prophylactic and therapeutic program for patients with cystic fibrosis. Pediatrics, 1965;36:675-688.