Balance performance, force production, and activity levels in noninstitutionalized men 60 to 90 years of age.

Balance Performance, Force Production, and Activity Levels in Noninstitutionalized Men 60 to 90 Years of Age We often take balance for granted. Our ability to maintain an upright posture is an intricate and dynamic process. An essential feature of this process is the coordination of muscle activity. Lane [1] defined balance as a function requiring the constant adjustment of muscle activity and joint position to retain the body weight over the base of support. The ability to adjust muscle activity to maintain balance depends on the visual, vestibular, auditory, somatosensory, and motor systems. [2] Disorders involving any of these systems can cause disequilibrium.

The elderly are among those at greatest risk for disequilibrium. [3-5] One third to one half of the US population aged 65 years and older experience one or more falls in a year. [6] Falls, and the immobility that often results, are among the most common reasons for medical intervention in the elderly. [7] Although not every fall in the elderly results in an injury, repeated falls are a major cause of morbidity in the aged and are responsible for approximately half of the accidental injuries in the elderly. [8] Additionally, the psychological damage attributable to loss of self-esteem and fear of falling can be severely debilitating and can create risks of falling and self-protective immobility. [5]

A decrease in muscle mass, especially of the back extensors and proximal lower extremity muscles, is prevalent in the elderly. [9,10] Isometric and isokinetic force and torque measurements have been shown to decrease with increasing age. [11,12] Decreased muscle strength may be due to primary aging of the muscle or may be the result of secondary changes induced in the muscle by aging of the nervous, vascular, and endocrine systems. [13] At the cellular level, alterations in the muscle fiber composition and cross-sectional area occur. These effects of age are compounded by a reduction in mobility and decreased exercise. [11]

Although the relationship between muscle weakness in old age and equilibrium maintenance would seem to be closely related, the studies related to this topic are scarce and incomplete. Several studies have examined the physiological and anatomical effects of aging on muscle. [9-14] Other studies have attempted to determine the degree of equilibrium decline accompanying aging. [15-18] None of these studies examined the relationship between balance and muscle strength in an elderly population. Briggs et al [18] obtained descriptive balance data from elderly female subjects; however, descriptive information regarding balance test performance in elderly men is not available. The purpose of this study was to collect and analyze balance and hip flexor, extensor, and abductor force data on healthy men 60 to 90 years of age. A descriptive profile of balance and muscle performance was obtained from these data, and the relationship between balance and muscle force was quantified. The effect of activity level on balance performance and torque production also was examined. The following hypotheses were generated:

1. Balance time and force production would have a positive correlation.

2. A difference in balance time and force production would exist between subjects who are active as opposed to those who are inactive.

3. Balance time and force production would have a negative correlation with age.

Method

Subjects

Fifty-four noninstitutionalized male subjects, aged 60 to 90 years (X = 71.2, s = 6.3), participated in this descriptive study. The subjects' weight ranged from 124 to 230 lb (*1) (X = 179.0, s = 23.7). Fifty subjects were right-leg dominant, and 4 were left-leg dominant. Subjects were recruited from Birmingham (Ala) area churches, senior centers, country clubs, nutrition centers, and The University of Alabama at Birmingham community. All participants were independent in their activities of daily living. Participants also were able to ambulate without an assistive device. Volunteers with a history of Parkinson's disease, cerebrovascular accident, multiple sclerosis, middle ear disease, joint replacement, or heart disease were not asked to participate. Each subject was asked to sign a voluntary consent form prior to testing.

Location

The testing was conducted at various locations throughout the city convenient to the participants. Approval to use a testing site was obtained from the appropriate personnel at each facility. Each testing site had a linoleum or tile floor covering. A sturdy table was used during strength testing.

Procedure

Prior to testing, the subjects responded orally to a questionnaire regarding their activity level and general health. Each subject's weight (obtained with a Counselor digital scale (*2)) and lower limb dominance were determined and recorded. Subjects were asked to pantomime kicking a ball, stamping out a fire, and stepping up a step. Leg dominance was listed as the leg with which the subject performed at least two of the three activities. [19] The order of testing balance and force was assigned randomly. A digital stopwatch (*3) was used to measure balance performance to the nearest hundredth of a second.

The procedure for testing balance was modified from the procedure used by Briggs et al [18] in their study of 73 elderly women. Balance was tested using the sharpened Romberg test (SR) and the one-legged stance test (OLST). The order for performing these two tests was assigned randomly; however, the right OLST and left OLST were not separated by the SR in the randomization process. Four test sequences were possible: 1) SR, right OLST, left OLST; 2) SR, left OLST, right OLST; 3) right OLST, left OLST, SR; and 4) left OLST, right OLST, SR. In their study, Briggs et al [18] found no significant difference in balance performance between the shoes-on and shoes-off conditions; therefore, the subjects in our study were tested wearing their shoes. Each test was performed first with eyes open, then with eyes closed. Prior to testing, the examiner (BDI) demonstrated the position to assume for each of the balance tests. A subject who requested help to assume a testing position was allowed to use the investigator's arm to steady himself prior to starting the timed trials. Close guarding to prevent falls as a result of loss of balance was provided. The subject was given a 1- to 2-minute rest period between the SR and the OLST. The 2-minute rest period was used for those subjects who indicated they needed it.

The SR was performed standing in a heel-to-toe position with the dominant foot behind the nondominant foot. The subject began testing with his arms by his sides, but was allowed to move his arms during the testing period. Timing was begun after the subject had assumed the proper position and indicated that he was ready to begin. Timing was stopped if the subject moved his feet from the given position, opened his eyes on the eyes-closed trials, or reached the maximum balance time of 60 seconds. Three trials were performed if a maximum balance time was not reached in either of the first two trials. The first and the best (ie, longest) balance times of the recorded trials were used for analysis.

The OLST was performed in the standing position with the subject's arms by his side. The subject was allowed to move his arms during the testing period. Timing was begun when the subject raised the other leg. Timing was stopped if the subject moved the foot he was standing on, touched the suspended foot to the ground, used the suspended foot for support on the weight-bearing limb, opened his eyes on the eyes-closed trials, or reached the maximum balance time of 30 seconds. Three trials were performed if the maximum balance time was not reached in either of the first two trials. Both legs were tested, and the first and best balance times of the recorded trials for each leg were used for analysis.

A quantitative muscle tester (*4) (QMT) was used to measure the force of the hip flexors, extensors, and abductors. The QMT is a strain-gauge, hand-held dynamometer that provides a digital readout of force data in tenths of pounds. Studies have shown the QMT to be a reliable tool for clinical muscle testing. [20]

Hip flexors were measured in the sitting position with the hip and knee flexed to 90 degrees. [21] The QMT was placed on the thigh 4 in (*5) proximal to the knee joint line, and the subject was instructed to lift up into the manually stabilized instrument as hard as he could. A metal tape measure was used to measure and mark the point of application of the dynamometer prior to testing. Hip extensor force was measured with the subject positioned prone and his knee flexed to 90 degrees. [21] The QMT was placed on the thigh 4 in proximal to the lateral tibial plateau. Hip abductors were measured in the prone position with the thigh laterally (externally) rotated and abducted slightly. The subject was asked to grasp the edges of the table. The QMT was placed laterally on the thigh 4 in proximal to the lateral tibial plateau. The subjects were instructed and verbally encouraged to contract the appropriate muscle group until a maximal effort was perceived by the investigator, and then told to relax; the best of two trials was used for analysis.

The distance from the greater trochanter to the site of placement of the QMT was measured for each testing procedure. This distance was multiplied by the peak force value to calculate each subject's torque value for hip flexion, extension, and abduction.

Intrarater and interrater reliability for obtaining force data with the QMT were established for the investigator prior to data collection. An intraclass correlation coefficient (ICC[2,1]) [22] was used to assess intrarater reliability (right and left hip flexors, ICC = .95; right and left hip extensors, ICC = .94; right and left hip abductors, ICC = .88; n = 7). Interrater reliability was assessed by comparing force values obtained by the investigator with those obtained by an expert in the field. the ICC(2,1) [22] was computed for each of the muscle groups (right and left hip flexors, ICC = .76; right and left hip extensors, ICC = .74; right and left hip abductors, ICC = .74; n = 7).

Descriptive data regarding each subject's self-estimate of activity were obtained. Because the percentage of subjects who estimated their activity level to be sedentary or mildly active was relatively small, those categories were combined, as Group 1 (n = 11), for statistical analysis; Group 2 (n = 25) consisted of those subjects who considered themselves to be moderately active; and Group 3 (n = 18) consisted of those subjects who considered themselves to be very active.

Data Analysis

Data were analyzed using the statistical analysis system (SAS-PC) computer program. [23] Descriptions of central tendency and variability of the data were included in the statistical analysis. The subject's torque measurement was divided by his body weight to normalize the torque data. Two Pearson product-moment correlation coefficients (r) were calculated for the mean normalized torque value correlated, one with the first balance time and the second with the best balance time. Each subject's age was correlated with each of his torque values to determine the relationship between age and torque production. The Pearson product-moment method of correlation also was used to determine the relationship between age and torque production. The same statistical procedure was used to test the relationship between the subjects' age and balance times. Two one-way analyses of variance (ANOVAs) were performed to test the effect of activity level on torque and on balance. Significant F ratios were tested using least-squares means for the post hoc analysis. A stepwise linear multiple-regression analysis was performed to determine the predictive value of the force variables in determining balance performance. An alpha level of .05 was used as the criterion for statistical significance.

Results

Descriptive statistics on the balance performance for the total sample are shown in Table 1. The mean balance times for the best trial of the SR were 54.70 seconds (s = 14.92) with eyes open and 24.62 seconds (s = 24.00) with eyes closed. The mean performance values for the best trial of the OLST ranged from 21.92 seconds (s = 10.21) in the left-leg--eyes-open condition to 4.93 seconds (s = 4.14) in the left-leg--eyes-closed condition. Subjects stood for a slightly longer period of time during the left OLST than they did during the right OLST with their eyes open. Mean balance performance on the OLST, with eyes closed, was slightly better on the right than on the left.

The subjects produced the greatest amount of torque with their hip flexors (Tab 2). The standard error of measurement for the torque measurements ranged from 0.59 to 1.02 kg.m for the muscle groups tested (Tab 2). No significant difference was observed between right and left mean torque values. Coefficients of determination ([R.sup.2]) indicated 7.8% to 39.5% of the variance in the balance times was accounted for by the torque values. Correlations between the normalized torque values and balance times of the best trials of the eyes-closed SR were significant (p [is less than] .05) for the bilateral hip extensor and right hip abductor muscle groups (Tab 3). Correlations for the best trials of the OLST were significant (p [is less than] .05) for each muscle group tested (Tab 3).

Stepwise linear multiple-regression analysis revealed that normalized torque values of each muscle group tested along with their interactions were significant predictors of balance performance in the OLST when combined; however, when analyzed separately, they were not good predictors (Tab 4). The coefficients of determination obtained from the regression analysis indicated the torque of the hip flexors, extensors, and abductors accounted for 15.6% (right leg) and 14.4% (left leg) of balance maintenance during the OLST in the eyes-open condition. When the effects of vision were eliminated by closing the eyes, the predictive value of these variables increased to 25.9% (right leg) and 36.1% (left leg). The stepwise linear multiple-regression analysis results for the SR were not significant.

The self-report questionnaire revealed the majority of subjects considered themselves to be either moderately active or very active. The mean balance time and standard deviation for each test condition were determined for each of the three activity-level groups (Tab 5). The one-way ANOVA revealed statistical significance (p [is less than] .05) for the effects of activity level on balance performance in two test conditions (Tab 6). Post hoc analysis (least-squares means test) revealed significant differences (p [is less than] .05) in the SR in the eyes-closed condition and in the OLST in the right-leg--eyes-open condition between Groups 1 and 3 and between Groups 2 and 3. The one-way ANOVA showed activity level had a significant effect on normalized torque values for all muscle groups except right hip extensors (p [ .07) (Tab 7). The post hoc analysis revealed a significant contrast between those subjects who were inactive to moderately active and those who considered themselves to be very active for all normalized torque values.

Torque production for all muscle groups, except the right hip abductors, showed a significant negative relationship with increasing age (Tab 8). Balance performance for all test conditions, except the SR in the eyes-open condition, also showed a significant negative relationship with increasing age (Tab 9).

Discussion

Hypothesis 1

The results of this study indicate a positive relationship between balance time and torque production for certain test conditions. The correlation between balance times achieved during the OLST and torque production were significant for each muscle group, whereas the SR in the eyes-closed condition only showed a significant correlation with the bilateral hip extensors and right hip abductors. The nonsignificant results of the SR in the other test conditions may be explained by the fact that 87% of the men balanced the full 60 seconds with their eyes open and 18.5% balanced the full 60 seconds with their eyes closed.

For most of the subjects tested, maintaining the SR position, especially in the eyes-open testing condition, was not a challenge. Sixty-nine percent of the women tested by Briggs et al [18] were able to maintain the SR position the entire 60 seconds as compared with 87% of the subjects in our study, indicating our sample of men was able to perform better on the SR with their eyes open than a similar sample of women. The data obtained in our study indicate that when testing elderly healthy male subjects, the OLST gives a better assessment of the force-balance relationship than does the SR.

Comparing the SR and OLST findings on women in the study of Briggs et al [18] with this study's SR and OLST findings reveals similar mean balance times. Although gender does not appear to have a major influence on balance performance with the SR and the OLST, it is not possible to draw conclusions without performing a statistical analysis. The question of gender differences on balance performance was beyond the scope of this study.

The results of the stepwise linear multiple-regression analysis indicate that torque production of the hip flexors, abductors, and extensors is a predictor of balance performance on the OLST. The coefficients of determination indicate muscle strength is a better predictor of OLST balance performance with the eyes closed than with the eyes open.

The decrement in balance times and force production with increasing age, together with the relationship between these two variables, suggest the importance of maintaining hip muscle force capabilities in the elderly. This suggestion is supported by the work of Woollacott et al, [24] who observed a proximal-to-distal muscle recruitment pattern in response to postural perturbation among elderly adults.

Hypothesis 2

The results of this study indicate a significant difference in the torque production of an elderly man who exercises five or six times a week versus a man who exercises less frequently or not at all. Thirteen of the subjects who estimated their activity level to be very active play 2 hours of tennis at least five times a week. The majority of the men who estimated their activity level to be either mildly or moderately active exercised by walking or doing yard work. In general, the activity level of the subjects studied may have been above average for this age group because of the low percentage of sedentary subjects and the high percentage of very active subjects recruited. A likely explanation for this bias toward increased activity levels is that most of the subjects were recruited from country clubs and recreation centers. Another possible explanation for the relatively high percentage of active elderly men recruited for this study is that sedentary men are less likely to venture outside of their homes and are therefore difficult to recruit.

Aniansson et al [13] examined isometric and isokinetic quadriceps femoris muscle torque in healthy 70-year-old subjects (40 men, 32 women). They found an inverse relationship between isometric and isokinetic quadriceps femoris muscle torque and previous occupational physical activity. Isokinetic torque in the knee extensor and flexor muscles was lowest in the men with a high level of previous occupational physical activity. The authors explained this surprising result by noting that each of the subjects had been retired at least 5 years. They suggested that the attitude of manual workers to physical activities is less positive after they retire than that of sedentary workers. [13] Our study examined current activity level and did not look at previous occupation. Our results showed significantly higher force production among men who are very active versus those who are less active. The findings of a later study conducted by Aniansson et al [12] concur with our results. In that study, muscle function in 75-year-old men and women was examined over a 5-year period. [12] The authors found isokinetic strength to be highest in men with a high level of physical activity. [12]

The results of our study suggest the decrement of muscle strength observed may be associated with inactivity; therefore, maintaining a physically active life style is important for the elderly. In another study, Aniansson and Gustoffson [25] showed that physical training in 64- to 79-year-old men (n = 5), with an emphasis on quadriceps femoris muscle strengthening, can increase quadriceps femoris muscle torque up to 22% and reverse type II fiber atrophy.

A more objective assessment of activity level as well as information describing the types of activities in which the subjects are involved would be useful in further defining the relationship between activity level, torque production, and balance. Longitudinal studies of healthy men involved in a fitness program may also yield information regarding the changes with age in balance and torque production.

Hypothesis 3

Previous studies have shown that balance time decreased with increasing age. [15,16,18] The results from this study concur with these results and showed a significant negative relationship between age and balance times. Balance times showed a significant decrease (p [is less than] .05) with age for all testing conditions, except the SR in the eyes-open condition. As noted previously, the lack of a significant negative relationship between balance performance on the SR in the eyes-open condition and age most likely was due to the majority of the subjects reaching the maximum balance time of 60 seconds.

A statistically significant decrease in torque production as a correlate of age also was observed in this study. This result is consistent with those reported by Serratrice et al [9] and Aniansson and colleagues. [12,13] Only the torque produced by the right hip abductors did not show a significant relationship. A possible explanation for the absence of a significant negative relationship between right hip abductor torques and age is that several subjects played tennis on a regular basis and were right-handed. The gluteus medius muscle of the dominant limb is used for lateral stabilization while swinging the tennis racket, which may explain why the expected relationship was not observed. Although many of these relationships were weak, collectively they accounted for a large portion of the variance and thus can serve as predictors of balance when considered together.

Understanding the relationship between muscle torque production and balance has clinical relevance for patients with balance deficits. Specific strengthening programs and other strategies for the prevention of falls and the promotion of a physically active life style can be formulated based on the findings of this study and other related studies.

Conclusions

Balance performance and force production have a positive relationship (ie, as force production increases, balance performance improves). The torques of the hip flexors, extensors, and abductors--when considered collectively--can serve to predict balance performance on the OLST. Age is correlated negatively with balance performance and with force production. The effect on force production, however, seems to be diminished by increased activity level. Studies of the efficacy of strengthening hip musculature relative to balance performance in patients with balance dysfunction are needed.

References

[1] Lane RE. Physiotherapy in the treatment of balance problems. Physiotherapy. 1969;55:405-408

[2] Mumenthaler M; Appenzeller O, trans. Neurologic Differential Diagnosis. New York, NY: Thieme-Stratton Inc; 1985:61, 108

[3] Nickens H. Intrinsic factors in falling among the elderly. Arch Intern Med. 1985;145:1089-1093

[4] Louis M. Falls and their causes. J Gerontol Nurs. 1983;9:143-156

[5] Tideiksaar R, Kay A. What causes falls? a logical diagnostic procedure. Geriatrics. 1986;41:32-46

[6] Campbell AJ, Reinken J, Allan B, et al. Falls in old age: a study of frequency and related clinical factors. Age Ageing. 1981;10:264-270

[7] Kataria MS. Fits, Faints, and Falls in Old Age. Boston, Mass: MTP Press Ltd; 1985:1, 19-24

[8] Accident Facts. Chicago, Ill: National Safety Council; 1986:2, 15, 17

[9] Serratrice G, Rouh H, Aquaron R. Proximal muscular weakness in elderly subjects. J Neurol Sci. 1968;7:275-299

[10] Tomonaga M. Histochemical and ultrastructure changes in senile human skeletal muscle. J Am Geriatr Soc. 1977;25:125-131

[11] Larsson L, Grimby G, Karlsson J. Muscle strength and speed of movement in relation to age and muscle morphology. J Appl Physiol. 1979;46:451-456

[12] Aniansson A, Sperling L, Rundgren A, et al. Muscle function in 75-year-old men and women: a longitudinal study. Scand J Rehabil Med. 1983;90(suppl):92-102

[13] Aniansson A, Grimby G, Rundgren A. Isometric and isokinetic quadriceps muscle strength in 70-year-old men and women. Scand J Rehabil Med. 1980;12:161-168

[14] Hasselkus BR. Aging and the human nervous system. Am J Occup Ther. 1974;28:16-21

[15] Bohannon RW, Larkin PA, Cook AC, et al. Decrease in timed balance test scores with aging. Phys Ther. 1984;64:1067-1070

[16] Nayak US, Gabell A, Simmons M. Measurement of gait and balance in the elderly. J Am Geriatr Soc. 1982;30:516-520

[17] Wolfson LI, Whipple R, Amerman P. Stressing the postural response: a quantitative method for testing balance. Am Geriatr Soc. 1986;34:845-850

[18] Briggs RC, Gossman MR, Birch R, et al. Balance performance among noninstitutionalized elderly women. Phys Ther. 1989;69:748-756

[19] Harris AJ. Harris Tests of Lateral Dominance. New York, NY: The Psychological Corp; 1958:10

[20] Bohannon RW. Test-retest reliability of hand-held dynamometry during a single session of strength assessment. Phys Ther. 1986;66:206-209

[21] Kendall FP, McCreary EK. Muscles: Testing and Function. 3rd ed. Baltimore, Md: Williams & Wilkins; 1983:291-294

[22] Shrout PE, Fleiss J. Intraclass correlations: uses in assessing rater reliability. Psychol Bull. 1979;86:420-428

[23] Statistical Analysis System. Cary, NC: SAS Institute Inc; 1985

[24] Woollacott MH, Shumway-Cook A, Nashner LM. Aging and posture control: changes in sensory organization and muscle coordination. Int J Aging Hum Dev. 1986;23:97-112

[25] Aniansson A, Gustoffson E. Physical training in elderly men with special reference to quadriceps muscle strength and morphology. Clin Physiol. 1981;1:87-98

(*1) 1 lb = 0.4536 kg.

(*2) Model 2200, Brearley Co, 2107 Kishwaukee St, Rockford, IL 61108.

(*3) Model 1000, Heuer Time and Electronics Corp, 960 S Springfield Ave, Springfield, NJ 07081.

(*4) QMT hand-held dynamometer, Program in Physical Therapy, Washington University, St Louis, MO 63110.

(*5) 1 in = 2.54 cm.

B Iverson, MS, is Physical Therapist, Knapp Rehabilitation Center, Metropolitan Mount Sinai Hospital, Minneapolis, MN 55404. He was a student in the Division of Physical Therapy, The University of Alabama at Birmingham, when this study was completed in partial fulfillment of the requirements for his Master of Science degree. Address all correspondence to Mr Iverson at 1329 NE Lake St, Apt 428, Hopkins, MN 55343 (USA).

M Gossman, PhD, PT, is Professor and Director, Division of Physical Therapy, The University of Alabama at Birmingham, Rm B41, Birmingham, AL 35294.

S Shaddeau, MMSc, PT, is Associate Professor, Division of Physical Therapy, The University of Alabama at Birmingham.

M Turner, PhD, is Professor, Biostatistics and Biomathematics, The University of Alabama at Birmingham.