Journal of Gerontology: MEDICAL SCIENCES 1991, Vol. 46. No. 3, M69-76
Copyright 1991 by The Geromological Society of America
Postural Stability and Associated Physiological Factors in a Population of Aged Persons Stephen R. Lord,1 Russell D. Clark,2 and Ian W. Webster1 'School of Community Medicine, The University of New South Wales, Australia. 2 St. Vincents Hospital, Sydney.
O
NLY a few published studies have compared stability in aged persons with impairment in somatosensory, visual, and vestibular functions. Brocklehurst et al. (1) found a significant inverse relationship between sway and vibration sense in women aged 75-84 but not in women aged 65-74 or 85+ years. They found no correlation between sway and joint position sense but acknowledged this lack of association may have reflected the imprecision of their test of proprioception. MacLennan et al. (2) found a significant association between vibration sense and sway in women aged 75-84 years but no association in women aged 65-74 or in men aged 65-74 or 75-84 in their study on 311 aged persons. They also measured proprioception in the toes but did not correlate this with body sway. Era and Heikkinen (3) found positive associations between sway and vibration thresholds in men aged 31-35 years and 51-55 years but not in men aged 71-75 years. Lichtenstein et al. (4) found that poor near-visual acuity was associated with increased areas of sway, and that increased body mass was associated with decreased velocity of sway. Increased areas of sway were also associated with slower reaction time and poor hearing, but these associations were removed after controlling for age and near-visual acuity. In a recent study, Manchester et al. (5) found that stability in elderly persons was significantly decreased under conditions in which peripheral vision was occluded and ankle somatosensation was limited, that is, when foveal vision and vestibular input are the only senses left unaltered. In another study on the relative contributions of vision peripheral sensation and vestibular sense to stability, Ring et al. (6) found sway increased on deprivation of vision and
alteration of proprioception due to foot pressure sensory change in 39 persons aged 17 to 79 years. They also found that when both forms of sensation were altered concurrently, sway increased markedly. There is general agreement that postural control involves many sensory and motor systems, and a number of investigators have found age-related declines in visual, vestibular, and sensorimotor functions (7-14). However, little work has been done on assessing the contribution of the demonstrated age-related declines in these systems on the overall decline in postural control that occurs with age. It is an important issue as to whether impairments in these systems contribute to the increased incidence of falls in elderly persons. In this study we examined the relationships between several sensory and motor factors and measures of postural stability in 95 persons living in a hostel for the aged. We developed a battery of simple, noninvasive tests to provide reliable measures of specific "postural control systems" that can be administered to elderly subjects in one brief session. By using this test battery, we were able to assess concurrently sensory and motor factors involved in postural control as well as three global measures of postural stability. Our model of postural control is outlined in Chart 1. The tests were selected so that functioning in the body systems in the model could be assessed. METHODS
Description of the sample. — The sample comprised 95 persons from a hostel for the aged in Sydney, Australia. The hostel housed 124 residents. Of the nonparticipants, 4 were M69
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
A battery of 13 visual, vestibular, sensorimotor, and balance tests was administered to 95 elderly persons (mean age 82.7 years) to examine the relationships between specific sensorimotor functions and measures of postural stability. When subjects stood on a firm surface, increased body sway was associated with poor tactile sensitivity and poor joint position sense. When subjects stood on a compliant surface (which reduced peripheral sensation) with their eyes open, increased body sway was associated with poor visual acuity and contrast sensitivity, reduced vibration sense, and decreased ankle dorsiflexion strength as well as reduced joint position sense. Increased body sway with eyes closed on the compliant surface was associated with poor tactile sensation, reduced quadriceps and ankle dorsiflexion strength, and increased reaction time. Poor performance in two clinical measures of postural stability was associated with reduced sensation in the lower limbs as measured by joint position sense, tactile sensitivity and vibration sense, reduced quadriceps and ankle dorsiflexion strength, and slow reaction times. The prevalence of vestibular impairments was high in this group, but vestibular function was not significantly associated with sway under any of the test conditions. The results suggest that reduced sensation, muscle weakness in the legs, and increased reaction time are all important factors associated with postural instability. An analysis of the percentage increases in sway under conditions where visual and peripheral sensation systems are removed or diminished, compared with sway under optimal conditions, indicated that peripheral sensation is the most important sensory system in the maintenance of static postural stability.
M70
LORDETAL.
Chart 1. Model of Postural Control Body systems that play important roles in postural control include:
Sensory systems
^
Vision Touch Proprioception Vibration sense Vestibular sense Muscle strength Neuromuscular control
Central nervous system
Integration of the sensory and motor factors
Body sway and the tests of static and dynamic balance provide summary measures or indices of postural stability.
ill, 5 were absent (on holidays, etc), and 20 declined to take part. The participation rate of available residents (excluding those ill and absent) was 82.6%. The subjects were aged between 59 and 97 years; the age and sex distribution is shown in Table 1. Most of the subjects were generally independent in such activities of daily living as dressing, bathing, and toileting. Forty percent had personal care assistance on a daily basis. Residents had their own "motel style" rooms and used a common dining room. Most residents (66%) left the hostel for varying periods every day, and a number regularly used public transport. Twenty-one residents (22%) used walking sticks or frames. All residents who were willing to take part in the study were included in the analysis. The fact that the subjects met the criteria for hostel living indicated that they were relatively independent and mobile. A history of stroke, Parkinson's disease, or hip fracture was accepted provided the subject could undertake the test program. The 13 tests were administered by the first author, and individual testing was completed within 1 hour. Sensory Systems Visual acuity was measured binocularly using a standard Snellen scale with subjects wearing their best correction (6). Acuity was measured in units derived from the logarithm of the minimum visual angle resolvable at a test distance of 4 meters. Contrast sensitivity was assessed using the Melbourne Edge Test (MET) — a test specifically designed for screening purposes (8). The MET is a non-grating contrast sensitivity chart. The test measures the contrast threshold for a single luminance profile edge, which is an aperiodic stimulus. The edge contrast threshold chart contains 20 circular test patches (25 mm diameter). The test presents a series of edges of reducing contrast with variable orientation as the identifying feature. A number under each test patch is the contrast sensitivity of that edge in decibels where dB = - 10 log Contrast. The test uses a four-alternative forced choice method of presentation. The edges are presented in the orientations: horizontal, vertical, 45° left, and 45° right. A key card containing the four possible edge angles is provided for
Age Group
Men
Women
Total
59-69 70-79 80-89 90-97
0 4 9 3 82.8 (7.4)
2 22 46 9 82.7(6.5)
2 26 55 12
Mean (SD)
82.7(6.6)
subject instruction. The lowest contrast patch correctly identified was recorded as the subject's contrast sensitivity. Touch thresholds were measured with a SemmesWeinstein Pressure Aesthesiometer, which is a modification of the calibrated hairs used by von Frey (15). The aesthesiometer set contained 20 nylon monofilaments of equal length (38mm) ranging in diameter from 0.06 to 1.14 mm. The force (gm) required to bend each monofi lament was precalibrated and ranges from 0.0045 to 447 gm. Pressure in gm was converted to log,0 0.1 mg, yielding a scale of approximately equal intensity intervals between filaments. The aesthesiometer filaments were applied to the center of the lateral malleolus of the ankle of the dominant leg. Subjects had their eyes closed during the procedure. The ascending and descending method of limits was used to determine tactile thresholds. Joint position sense was measured using an apparatus based on a design by De Domenico and McCloskey (16). Subjects (with eyes closed) attempted to simultaneously place the big toe of the right foot on the right side of a perspex sheet (60cm x 60cm x lcm) and the big toe of the left foot on the corresponding position on the left side of the sheet. The perspex sheet was mounted vertically on the floor. Errors in matching the two toes were measured by reading from a protractor inscribed on the sheet. Subjects had two practice trials, then five experimental trials. The subject's score was the mean error in matching the two toes measured in degrees. Vibration sense was measured using an electronic device that drove a 12-centimeter loudspeaker which generated a 200 cycle/sec vibration. The vibration was transferred to the tibial tuberosity of the knee via a rubber stopper lcm in diameter. The pressure or weight of the apparatus was held constant by a spring-loading mechanism. Intensity of the vibration was read in microns of motion perpendicular to the body surface. Three readings in the ascending mode and three readings in the descending mode were made. Vestibular function was assessed by two methods: the Vestibular Stepping Test and the Vestibular Optical Stability Test. The Vestibular Stepping Test measured subjects' ability to remain stationary and oriented in the one plane while "walking on the spot" with the eyes closed for a period of 1 minute (17,18). The test measurements included distance walked from initial starting point (cm), angle of body rotation (degrees), and sum of distance walked and degrees rotation. To minimize case losses, that is, the loss of individuals from the analysis because they could not complete the task, those who could undertake the test for half or more of the test period were included in the study. Their test scores — distance walked and degrees rotation — were extrapo-
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
Motor systems
Table 1. Age and Sex Distribution of the Sample
POSTURAL STABILITY IN THE ELDERLY
Motor Systems Quadriceps strength was measured by placing a strap around the subject's dominant (stronger) leg. The strap was connected to the central leg of a chair so that when the subject (seated on the chair) attempted to extend his leg he extended a spring gauge giving a measure of maximal quadriceps strength. The subject had three experimental trials, and the greatest extension of the spring gauge was recorded (in kg). Strength was corrected for body size by dividing the measure by the height of the subject. Ankle dorsiflexion strength was measured by having the subject place the foot of his dominant (stronger) leg on a foot rest. A strap (which protruded through the foot rest) was placed over the top of the foot just proxin.al to the commencement of the little toe. The strap was connected to the base of the foot rest so that when the subject (seated on the chair) attempted to raise the front of his foot (while keeping the heel placed on the foot rest) he extended a spring gauge giving a measure of maximal ankle dorsiflexion strength. The subject had three experimental trials, and the greatest extension of the spring gauge was recorded (in kg). Strength was corrected for body size by dividing the measure by the height of the subject.
while fixating a point slightly below eye level at a distance of 3 meters. Testing was performed on a firm surface (a small pile carpeted floor) and on a piece of foam rubber (1 m by 1 m by 15 cm thick) with the subject standing in the center. The same test was repeated on both surfaces with the subject's eyes closed. The foam rubber was used to reduce proprioceptive input from the ankles and cutaneous inputs from the soles of the feet so that subjects would be required to rely on visual and vestibular cues to maintain a steady stance. Four testing conditions were employed: condition A — firm surface, eyes open; condition B — firm surface, eyes closed; condition C — compliant surface, eyes open; and condition D — compliant surface, eyes closed. Maximal sway in the anterior-posterior and lateral directions and total sway (number of square millimeter squares traversed by the pen) in the 30-second periods were recorded for the four test conditions. Figure 1 shows an example of sway tracings for the four conditions in a woman aged 83 years. To minimize case losses, those who could undertake the test for half or more of the test periods were included in the study. Their test scores — number of square millimeter squares traversed by the pen — were extrapolated accordingly. For example, an individual who recorded a sway measure of 100 mm in 20 seconds was coded as having swayed 150 mm. Those who could not undertake the test for 15 seconds were not entered into the analysis. The Static Balance Test measured the ability of subjects to maintain balance while standing on a firm and a compliant 1. EYES OPEN (FLOOR)
3. EYES OPEN (FOAM)
(CONDITION A)
(CONDITION C)
Reaction Time Reaction time was assessed using a simple reaction time paradigm, using a light as the stimulus and depression of a switch (by the hand) as the response. Subjects had 10 practice trials and 10 experimental trials. Reaction time was recorded in milliseconds. 2. EYES CLOSED (FLOOR)
Postural Stability Measures Sway. — Sway was measured using a swaymeter that measured displacements of the body at the level of the waist. The device consisted of a rod attached to the subject at waist level by a firm belt. The rod was 40 cm in length and extended behind the subject. A sheet of graph paper (with a millimeter square grid) was fastened to the top of an adjustable height table that was positioned behind the subject. The height of the table was adjusted so that the rod was in a horizontal plane and the tip of a pen (attached to the end of the rod) could record the movements of the subject on the graph paper. The subject was instructed to stand as motionless as possible for a period of 30 seconds (feet 5 cm apart)
(CONDITION B)
4. EYES CLOSED (FOAM) (CONDITION D)
Figure 1. Body sway under four test conditions in a woman aged 83 years. Source: University of NSW and Teaching Hospitals.
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
lated accordingly. For example, an individual who walked one meter from the mark and rotated 30 degrees in 45 seconds was coded as having walked 133 cm and rotating 40 degrees. Those who could not undertake the test for 30 seconds were not entered into the analysis. The second vestibular function measure was a new test developed to measure vestibulo-optical stability. In this test, subjects' visual acuity while walking on a treadmill was compared to their visual acuity when at rest. It is suggested that any difference in acuity is due to the vestibular system being unable to correct for head movements. Any difference in the log (minimum angle resolvable) for the two test conditions was used as the measure of vestibular optical stability.
M71
M72
LORDETAL.
Test Reliability Most of the tests have been assessed for test-retest reliability in pilot studies. In these studies the tests were administered to 37 subjects aged 63 to 92 years of age on two occasions 6 months apart. The reliability coefficients for the test measures plus 95% confidence intervals are shown in Table 2. Total sway with the eyes open and eyes closed, quadriceps strength, visual acuity, and the sum of distance and angle recorded in the vestibular stepping test had reliability coefTable 2. Test-Retest Reliability Coefficients (n = 37)
Sway — total eyes open Sway — total eyes closed Proprioception Vibration sense Visual acuity Touch Reaction time Vestibular stepping distance Vestibular stepping angle Vestibular sum (distance + angle) Vestibular optical stability Quadriceps strength *p< .01.
Correlation
CI
0.81** 0.73** 0.46** 0.69** 0.73** 0.48** 0.63** 0.51** 0.63** 0.75** 0.17 0.75**
0.66-0.90 0.53-0.85 0.16-0.68 0.47-0.83 0.53-0.85 0.18-0.70 0.38-0.79 0.22-0.72 0.38-0.79 0.56-0.86 -0.16-0.47 0.56-0.86
ficients greater than 0.70. Vibration sense, reaction time, vestibular stepping distance, and vestibular stepping angle had reliability coefficients between 0.50 and 0.70. The remaining tests—joint position sense, touch, and vestibular optical stability — showed moderate to poor test-retest reliability. The reliability of the vestibular optical stability test will be further evaluated. Five tests (that were not used in the pilot study) were included in the battery. Ankle dorsiflexion assessment was included because of a (then new) published finding by Whipple et al. (19) that reduced ankle dorsiflexion strength appears to be an important factor underlying poor balance. They found that ankle dorsiflexion strength was the most impaired of the four main leg muscle groups in a group of fallers compared with controls. Body sway on foam was included in an attempt to assess sway when subjects had reduced ankle support. It was hypothesized that the foam would reduce peripheral sensation, and compel subjects to rely more on visual and vestibular senses. Static balance was included to identify subjects who had difficulty maintaining balance on the floor and on the foam while undertaking the sway tests. Dynamic balance was included because it was noted in the pilot study that although the distance or angle moved in the vestibular stepping test was not associated with instability or falls, stability while performing this test appeared to be associated with other measures of balance, and that performance by those with a history of falls appeared to be worse than by those who had not fallen. Reliability for these measures is still to be determined. The Melbourne Edge Test was included because it was hypothesized that poor contrast sensitivity may be associated with instability and falls. This test, specifically designed for clinical settings, had only just become available. The reliability of the Melbourne Edge Test has been reported elsewhere (8). Data Analysis All test scores, with the exception of scores for static and dynamic balance, were coded as continuous variables. Grades in the tests of static and dynamic balance were collapsed into three groups before assessing associations with other sensory and motor factors. For static balance: grades 1 to 3 were collapsed into one group (poor balance), grade 4 was recoded as a second group (moderate balance) while grade 5 made up the third group (good balance). For dynamic balance, grades 1 and 2 was recoded as one group (poor balance), grades 3 and 4 were recoded as a second group (moderate balance), and grade 5 made up the third group (good balance). The 10 specific measures of sensorimotor systems were correlated with the measures of body sway (under the four test conditions), using partial correlation analysis to control for age. Multiple regression analysis was used to assess the associations between individual sensory and motor test measures and the two clinical measures of stability (static and dynamic balance). In these analyses, the sensorimotor factors were treated as dependent variables, and age was forced into the equation before including the stability measures (which were recoded as dummy variables). It was then
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
surface. Subjects were classified into 5 grades while undertaking the sway tests: "Grade 1" — unable to maintain balance for any period without support on a firm surface (a small pile carpeted floor) with eyes open; "Grade 2 " — unable to maintain balance on the floor for 30 seconds (eyes open or closed); "Grade 3 " — capable of maintaining balance on the floor for 30 seconds (eyes open and eyes closed) but unable to maintain balance on a soft compliant surface (a piece of foam lm by lm by 15 cm thick) without support for any period of time; "Grade 4 " — capable of performing the tests on the floor but unable to maintain stability on the foam for the full 30 seconds (eyes open or closed), and "Grade 5 " — capable of maintaining balance while standing on the floor and the foam (eyes open and closed) for 30-second periods without difficulty. Grades in this test were given while subjects were undertaking the sway tests. The Dynamic Balance Test was an assessment of the subject's ability to maintain balance while performing the vestibular stepping test — walking on the spot for 1 minute with the eyes closed. Subject's performance was classified into 5 grades. "Grade 1" indicated that the subject's balance was so poor that the test could not be administered; "Grade 2 " indicated that the subject could start the test but could not maintain balance for the full minute or required assistance from the experimenter to stay upright to complete the test; "Grade 3 " indicated that the subject could perform the test without requiring assistance but faltered or had to stop walking on two or more occasions to maintain balance; "Grade 4 " indicated that the subject could perform the test but faltered or had to stop walking on one occasion to keep balance; and "Grade 5 " indicated that the subject could perform the task with no difficulty maintaining good balance.
POSTURAL STABILITY IN THE ELDERLY
RESULTS
Table 3 shows the number of residents who could complete each test, and the means and standard deviations for the test measure scores. The missing data are due to certain subjects not complying with the test procedures or not being able to undertake the tests because of impaired balance.
Table 3. Mean Values for the Test Variables (SD)
Visual acuity* Met (dB log contrast) Proprioception (degrees difference) Touch (log10 0.1 milligrams pressure) Vibration senset Vestibular optical stability* Vestibular stepping distance§ Vestibular stepping angle|| Vestibular sum (Distance + Angle) Quadriceps strength/height (kg/cm x 100) Ankle strength/height (kg/cm x 100) Reaction time (milliseconds) Sway — eyes openil Sway — eyes closedH Sway on foam — eyes openil Sway on foam — eyes closedil Static balance Dynamic balance
N
Mean (SD)
95 94 93 94 93 57 73 73 73 94 95 93 94 94 73 67 95 95
3.85 (6.73) 16.3(4.0) 2.6 (2.0) 4.14 (0.41) 38.7 (29.2) 0.21 (0.27) 116 (76) 52 (57) 168(93) 8.49(3.24) 2.62(0.83) 310(140) 112 (63) 143 (111) 229 (113) 338 (137) 2.41 (0.77) 3.59 (1.90)
*Smallest visual angle correctly reported at 4 meters (minutes). tMicrons of motion perpendicular to body surface. ^Difference in smallest visual angle correctly reported at 4 meters on treadmill and smallest visual angle correctly reported at 4 meters while seated (minutes). ^Distance moved from initial position in 1 minute (centimeters). || Angle of rotation in 1 minute (degrees). 11 Millimeter squares traversed by pen on swaymeter in 30 seconds.
Table 4 shows the associations expressed as partial correlation coefficients between the individual sensorimotor system measures and the four body sway measures. Poor tactile sensitivity and joint position sense were associated with sway on the floor (conditions A and B). Increased body sway, eyes open, on the foam (condition C) was associated with poor visual acuity, poor contrast sensitivity, decreased joint position sense, decreased vibration sense, and reduced ankle dorsiflexion strength. Increased body sway, eyes closed, on the foam (condition D) was associated with high touch thresholds, reduced quadriceps and ankle dorsiflexion strength, and increased reaction time. Vestibular function measures were not associated with sway under any of the four test conditions, even though the prevalence of vestibular impairment was high in this group. Fifty-one of the 73 subjects (69%) who could perform the stepping test recorded an abnormal result, that is, they walked a distance greater than 1 meter or rotated more than 45 degrees (18). Similarly, 32 of the 57 subjects (56%) who could perform the vestibular walking test recorded an abnormal result, that is, their visual acuity while on the treadmill was worse than when seated. The two clinical tests of postural stability, the static balance test and the dynamic balance test, have not been previously described. These two tests of stability were significantly associated with total sway. The static balance test had a Spearman correlation coefficient of —0.26 (p < .01) with total sway eyes open, and - 0 . 2 0 (p < .05) with total sway eyes closed. The dynamic balance test had a Spearman correlation coefficient of - 0 . 2 8 (p < .01) with total sway eyes open, and — 0.28(p < .01) with total sway eyes closed. The two clinical measures were also significantly associated (Spearman r = 0.71, p < .01). Thirty-four residents (35.8%) received an abnormal rating in the static balance test (Grades 1-4), that is, they were unable to maintain balance on the foam rubber with eyes Table 4. Correlations Between Sensorimotor Systems and the Four Sway Measures Controlling for Age Sensorimotor system Visual acuity Contrast sensitivity
Sway EO Sway EC Floor 1 Floor 2 0.08 -0.02
0.01 0.07
Proprioception Touch Vibration sense
0.22* 0.17* 0.06
Vestibular stepping test Distance Angle Sum (Distance + Angle) Vestibular optical stability
0.01 0.10 0.02 0.07
0.03 -0.09 -0.08 0.17
Quadriceps strength/height Ankle strength/height
0.09 0.18
Reaction time
0.09
0.25* 0.27** 0.15
Sway EO Foam 3
Sway EC Foam 4
0.28** -0.26*
-0.03 -0.05
0.23* 0.19 0.23* 0.16 -0.02 0.12 0.17
0.05 -0.17 -0.17 0.13
0.14 0.15
-0.12 -0.24*
-0.23* -0.25*
0.08
0.16
Note. 1 - Sway EO (eyes open) on floor (Condition A) 2 - Sway EC (eyes closed) on floor (Condition B) 3 - Sway EO (eyes open) on foam (Condition C) 4 - Sway EC (eyes closed) on foam (Condition D)
*p< .05;**p< .01.
0.13 0.25* -0.01
0.24*
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
possible to assess the change in the multiple correlation coefficient (or multiple r2) when the effects of the stability measures were added to the equation, that is, after controlling for the effects of age. The change in r2 was tested for significance using an F-test. Multiple regression analysis was also used when analyzing other multivariate associations. In these analyses, the measures of sway were the dependent variables, and the sensory and motor measures were independent (or "predictor") variables. Age was forced into the regression equations, before other variables were included using forward selection. Discriminant function analysis, which is a statistical technique which highlights the variables that are most important in the identification of groups, was used to determine which postural control variables discriminated between the subjects with poor, moderate, and good stability as measured by the static and dynamic balance tests. In the discriminant function procedure, age was included as a possible "predictor" variable. For variables with right skewed distributions (such as the vestibular stepping test scores and the joint position sense, quadriceps strength, reaction time and sway measures), logs of variables were analyzed. The data were analyzed using the SPSSX computer package (20).
M73
M74
LORDETAL.
Table 5. Subjects' Performance in the Two Clinical Tests of Stability Static Balance n Grade 1 Grade 2 Grade 3 Grade 4 Grade 5
Table 7. Sensorimotor "Predictor" Variables Identified by the Multiple Regression Analysis
Dynamic Balance n
(%)
(%)
1(1.1) 2(2.1)
21 (22.1) 23 (24.2)
19(20.0) 12(12.6) 61 (64.2)
17(17.9) 9 (9.5) 25 (26.3)
Sensorimotor system Visual acuity Contrast sensitivity
Static Balance
Dynamic Balance 0.01 0.03
Proprioception Touch Vibration sense
0.03 0.02 0.19* 0.13** 0.16**
0.14** 0.24** 0.06
Vestibular stepping test Distance Angle Sum (Distance + Angle) Vestibular optical stability
0.01 0.11 0.08 0.06
0.03 0.01 0.05 0.21**
Quadriceps strength/height Ankle strength/height
0.16** 0.12*
0.14** 0.06
Reaction time
0.03
0.11**
tAssociations are presented in terms of change in Multiple R that result when adding the stability measure to a multiple regression equation after controlling for the effects of age. *Change in r2 significant atp < .05. **Change in r2 significant at p < .01.
closed for 30 seconds. Seventy residents (73.3%) received an abnormal rating in the dynamic balance test (Grades 1-4), that is, they were unable to maintain good balance while walking on the spot for one minute with eyes closed. Grades for the subjects in both tests are shown in Table 5. Table 6 shows the associations between the sensorimotor system measures and scores in the static and dynamic balance tests. Poor scores on the static balance test were associated with reduced tactile sensitivity, joint position and vibration sense, and reduced quadriceps and ankle dorsiflexion strength. Poor scores in the dynamic balance test were significantly associated with reduced tactile sensitivity, reduced sense, poor vestibular optical stability, reduced quadriceps strength, and increased reaction time. Table 7 shows the sensory and motor system variables that were included in the multiple regression equations for the four sway measures (used as dependent variables). Beta weights for each independent variable included in the regression equations and the multiple correlation coefficients are also shown. Beta weights are the coefficients of the independent variables included in the regression equation expressed in a standardized (z score) form. As the units of each measure have been standardized, the beta weights give an indication of the relative importance of each variable in
Beta
Multiple/?
Sway (EO) on floor
Proprioception Age
0.222 -0.109*
0.24
Sway (EC) on floor
Proprioception Touch Age
0.263 0.283 •0.071*
0.37
Sway (EO) on foam
Visual acuity Age
0.290 0.063*
0.31
Sway (EC) on foam
Ankle strength Age
-0.254 •0.091*
0.26
Note. EO = Eyes open; EC = eyes closed. *Significance of Beta weight >.05, i.e., not significant.
explaining the variance in the dependent variable. (However, it should be noted that as the beta weights are affected by correlations among the independent variables they do not in an absolute sense reflect the importance of the various independent variables.) An analysis of residuals indicated that the assumptions for the regression procedures were not violated. Discriminant function analysis was used to determine which postural control variables discriminated between subjects with poor stability, moderate stability, and good stability in the static and dynamic balance tests. For static balance, the following variables: proprioception, vibration sense, and quadriceps strength discriminated significantly between the three groups, as indicated by the final Wilk's lambda of 0.75 and a canonical correlation for the discriminant function of 0.49. The standardized canonical correlation coefficients for the first function (which accounted for 96% of the total between-groups variability) were 0.49 for proprioception, 0.55 for vibration sense, and — 0.76 for quadriceps strength. For dynamic balance, the following variables: proprioception, quadriceps strength, reaction time, and age discriminated significantly between the three groups, as indicated by the final Wilk's lambda of 0.73 and a canonical correlation for the discriminant function of 0.50. The standardized canonical correlation coefficients for the first function (which accounted for 93% of the total between-groups variability) were 0.43 for proprioception, - 0 . 5 5 for quadriceps strength, 0.38 for reaction time, and 0.51 for age. Mean amount of sway condition A (eyes open on the floor) for the 67 subjects who could complete all four sway tests (conditions A-D) was 105 mm. Compared with sway under condition A, sway in condition B (eyes closed on the floor) increased 1.30 times. In this condition vision is absent and peripheral sensation and vestibular sense are present. Sway in condition C (eyes open on the foam) increased 2.58 times. In this condition, peripheral sensation is reduced and vision and vestibular sense are present. Sway in condition D (eyes closed on the foam) increased 4.11 times. Under this condition, vestibular sense is present, vision is absent, and peripheral sensation is reduced. If the assumption is made that sway on the floor is the optimal measure for stability where all senses are available,
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
Table 6. Associationst Between the Sensorimotor Systems and the Clinical Stability Measures Controlling for Age
"Predictor" Variables
Dependent Variables
POSTURAL STABILITY IN THE ELDERLY
DISCUSSION
Cross-sectional studies of this type suffer a number of inherent limitations. Foremost is the problem of uncertainty of causal directions in any associations uncovered. The study also has the complication of multiple measurement and influence. The associations and multiple regression results reported in this article will require validation, as some of the associations may have happened by chance. Other problems and limitations include biases on the part of observer and
Chart 2. The Relative Contributions of Vision, Peripheral Sensation, and Vestibular Sense to the Maintenance of Postural Stability Compared with sway (eyes open on floor) [A]: Sway (eyes closed on floor) [B] is increased 130.2% Sway (eyes open on foam) [C] is increased 258.4% Sway (eyes closed on foam) [D] is increased 410.5% Using the following assumptions: (B-A)/B = Contribution by vision (C-A)/C = Contribution by peripheral sensation (A/D) = Contribution by vestibular sense The contributions of vision, peripheral sensation, and vestibular sense to postural stability can be calculated:* Vision 22%t Peripheral sensation 58% Vestibular sense 20% 'Calculated using results of the 67 subjects who could complete all four sway tests. fContributions are rounded so as to add to 100%. The percentages are broad estimates only.
subject, unintended order effects in the administration of the tests, and variable effort extended by the subjects when undertaking the tests. The approach used in this study to factors associated with postural control assumes that the group effects represent the population and that people with decreased stability have the same underlying predisposing influences. It is acknowledged that it is also possible there are different influences in different people (such as the effects of a wide variety of chronic diseases and illnesses), and that the statistical evidence for the group cannot test whether some subjects had one or more other major influences. The results of the present study revealed a number of significant associations between the measures of stability and specific postural control systems. Increased sway (when standing on a firm surface) was associated with poor tactile sensitivity, joint position sense and reaction time, but not with the visual, vestibular function, or strength measures. The findings that vestibular sense showed no association with sway is in accord with the findings of Nashner (21). He found that the otoliths play no role in the initial detection of body sway. Brocklehurst et al. also reported no correlation between vestibular sense (as measured by response to a slow tilt) and sway (1). These findings agree with previous studies that have found that increased sway in elderly persons is associated with a loss of sensory input from the lower limbs, and that visual and vestibular systems appear to be secondary in the maintenance of posture under normal conditions. When the subjects were placed in a situation that provided reduced support (when standing on a compliant surface), other postural control systems were found to be associated with total sway. Increased body sway with eyes open on the foam was associated with poor visual acuity and contrast sensitivity and reduced quadriceps and ankle dorsiflexion strength. These findings suggest that, in this situation with reduced proprioceptive input from the ankles, subjects are compelled to rely on other sensory and motor factors. Under this condition, subjects with poor vision and strength demonstrated greatly increased sway. When vision was denied and peripheral sensation and ankle support were reduced (eyes closed on the foam), increased body sway was associated with high touch thresholds, reduced quadriceps and ankle dorsiflexion strength, and increased reaction time. In this condition — vision removed and ankle support reduced — subjects with poor tactile sensitivity, muscle strength, and reaction time demonstrated greatly increased sway. The finding that reduced ankle dorsiflexion strength is associated with decreased stability is in accord with the findings by Whipple et al. (19). They found that ankle dorsiflexion strength was significantly impaired in elderly fallers compared with nonfallers and suggested that the inability to initiate appropriate ankle dorsiflexion strength may contribute to instability and falls (especially to the rear). The multiple regression analyses revealed that the main sensorimotor factor contributing to balance under normal conditions (standing on a firm surface with eyes open or closed) is sensation in the lower limbs. Under more adverse conditions (standing on foam — where this sensation is reduced), vision and strength play major roles.
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
it is possible to estimate the relative contributions of vision, peripheral sensation, and vestibular sense to the maintenance of postural stability. For example, if stability is twice as bad with a sense removed, then that sense must contribute half the information for stability. Again, if stability is three times as bad with a sense removed, then that sense must contribute two-thirds the information for stability. The results revealed that sway was increased 1.3 times in condition B (eyes closed on the floor) compared with condition A (eyes open on the floor), that is, stability has been reduced by a factor of 1/1.3. Vision, therefore, must contribute 0.3/1.3 (or 23%) to the maintenance of postural stability. Similarly, as sway is increased 2.58 times in condition C (eyes open on the foam) compared with condition A (eyes open on the floor), stability has been reduced by a factor of 1/ 2.58. Peripheral sensation, therefore, must contribute 1.58/ 2.58 (or 61%) to the maintenance of postural stability. Finally, as sway is increased 4.11 times in condition D (eyes closed on the foam) compared with condition A (eyes open on the floor), stability has been reduced by a factor of 1/4.11. Vestibular sense, therefore, must contribute 1/4.11 (or 24%) to the maintenance of postural stability as it is the only sense left unaltered. Chart 2 shows calculations for determining the relative contributions of these systems. Correcting these percentages, so that the total is 100%, reveals the contribution by vision as 21.3%, the contribution by peripheral sensation as 56.3% and the contribution by vestibular sense as 22.4%.
M75
M76
LORD ETAL.
ACKNOWLEDGMENTS
Address correspondence and reprint requests to Stephen R. Lord, School of Community Medicine, University of NSW, P.O. Box 1, Kensington, N.S.W., 2033, Australia.
REFERENCES
1. Brocklehurst JC, Robertson D, James-Groom P. Clinical correlates of sway in old age-sensory modalities. Age Ageing 1982; 11:1-10. 2. MacLennan WJ, Timothy JI, Hall MRP. Vibration sense, proprioception and ankle reflexes in old age. J Clin Bxp Gerontol 1980;2:159-71. 3. Era P, Heikkinen E. Postural sway during standing and unexpected disturbance of balance in random samples of men of different ages. J Gerontol 1985 ;40:287-95. 4. Lichtenstein MJ, Shields SL, Schiavi R, Burger MC. Clinical determinants of biomechanics platform measures of balance in aged women. J AmGeriatrSoc 1988;36:996-1002. 5. Manchester D, Woollacott M, Zederbauer-Hylton N, Mann O. Visual, vestibular and somatosensory contributions to balance control in the older adult. J Gerontol Med Sci 1989;44:M118-27. 6. Ring C, Nayak USL, Isaacs B. The effect of visual deprivation and proprioceptive change on postural sway in healthy adults. J Am Geriatr Soc 1989;37:745-9. 7. Pitts DG. The effects of aging on selected visual functions. In Sekjuler R, Kline DW, Dismukes K. Aging in human visual functions. New York: Alan R. Liss, 1982. 8. Verbaken JH, Johnston AW. Population norms for edge contrast sensitivity. Am J Optom Physiol Opt 1986;63:724-32. 9. Kaplan FS, Nixon JE, Reitz M, Rindfleish L, Tucker J. Age-related changes in proprioception and sensation of joint position. Acta Orthop Scand 1985;56:72-4. 10. Larsson L. Morphological and functional characteristics of the aging skeletal muscle in man. A cross-sectional study. Acta Physiol Scand 1978 (Suppl 457): 1-36. 11. Mulch G, Petermann W. Influence of age on results on vestibular function tests. Ann Otol Rhinol Laryngol 1979;88 (Suppl 56 No 2): 1-17. 12. Thornbury JM, Mistretta CM. Tactile sensitivity as a function of age. J Gerontol 1981;36:34-9. 13. Welford AT. Motor performance. In: Birren JE, Schaie KW. Handbook of the psychology of aging. New York: Van Nostrand Reinhold, 1977. 14. Whanger AD, Wang HS. Clinical correlates of the vibratory sense in elderly psychiatric patients. J Gerontol 1974;29:39-45. 15. Semmes J, Weinstein S, Ghent L, Teuber H. Somatosensory changes after penetrating brain wounds in man. Cambridge, MA: Harvard University Press, 1960. 16. De Domenico G, McCloskey DI. Accuracy of voluntary movements at the thumb and elbow joints. Exp Brain Res 1987;65:471—8. 17. Fukuda T. The stepping test — two phases of the labyrinthine reflex. Acta Otolaryngol 1959;50:95-108. 18. Kosoy J. The oto-neurologic examination. Acta Otolaryngol 1977;Suppl 343:1-95. 19. Whipple RH, Wolfson LI, Amerman PM. The relationship between knee and ankle weakness to falls in nursing home residents: an isokinetic study. J AmGeriatrSoc 1987;35:13-20. 20. SPSS Inc. SPSSX users guide. Chicago: McGraw Hill, 1983. 21. Nashner LM. A model describing vestibular detection of body sway motion. Acta Otolaryngol 1971 ;72:429-36. Received January 9, 1990 Accepted April 16, 1990
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
Reduced sensation and strength in the lower limbs and increased reaction time were also associated with the two clinical measures of stability — static and dynamic balance. One of the measures of vestibular function, vestibular optical stability, was also associated with performance in the dynamic balance test. These clinical measures of balance appear to be summary measures of stability. In spite of the significant associations uncovered, it is apparent that most of variance in the measures of sway and balance is left unexplained. This is in agreement with other published work (1-4). It may be that important components are missing from the model of postural control and/or that there is a high degree of inherent variability in the test measures. In addition, the model assumes that the systems that contribute to stability do so in an additive way, and provides no allowance for interactions and feedback mechanisms. However, despite the limitations of the approach, the associations found provide interesting insights into the complex process of postural control. The results suggest that reduced sensation and muscle weakness in the legs and increased reaction time are important factors associated with postural instability. The findings that sway increases with alterations to visual and peripheral sensation, and especially when both are altered concurrently, are in agreement with the findings of Manchester et al. (5)andRingetal. (6). The analysis of the percentage increases in sway under conditions where visual and peripheral sensation systems are removed or diminished, compared with sway under optimal conditions, indicated that peripheral sensation is the most important sensory system in the maintenance of static postural stability. The contributions defined (peripheral sensation — 56%, vision — 21%, and vestibular sense — 22%) are broad estimates and are intended to provide relative measures only. As peripheral sensation was only reduced and not removed, the contribution defined above is most certainly an underestimate of the role of peripheral sense in the maintenance of stability, and an overestimate of the vestibular component. Relationships between these sensorimotor factors and falls will be pursued in further research.
Copyright 1991 by The Geromological Society of America
Postural Stability and Associated Physiological Factors in a Population of Aged Persons Stephen R. Lord,1 Russell D. Clark,2 and Ian W. Webster1 'School of Community Medicine, The University of New South Wales, Australia. 2 St. Vincents Hospital, Sydney.
O
NLY a few published studies have compared stability in aged persons with impairment in somatosensory, visual, and vestibular functions. Brocklehurst et al. (1) found a significant inverse relationship between sway and vibration sense in women aged 75-84 but not in women aged 65-74 or 85+ years. They found no correlation between sway and joint position sense but acknowledged this lack of association may have reflected the imprecision of their test of proprioception. MacLennan et al. (2) found a significant association between vibration sense and sway in women aged 75-84 years but no association in women aged 65-74 or in men aged 65-74 or 75-84 in their study on 311 aged persons. They also measured proprioception in the toes but did not correlate this with body sway. Era and Heikkinen (3) found positive associations between sway and vibration thresholds in men aged 31-35 years and 51-55 years but not in men aged 71-75 years. Lichtenstein et al. (4) found that poor near-visual acuity was associated with increased areas of sway, and that increased body mass was associated with decreased velocity of sway. Increased areas of sway were also associated with slower reaction time and poor hearing, but these associations were removed after controlling for age and near-visual acuity. In a recent study, Manchester et al. (5) found that stability in elderly persons was significantly decreased under conditions in which peripheral vision was occluded and ankle somatosensation was limited, that is, when foveal vision and vestibular input are the only senses left unaltered. In another study on the relative contributions of vision peripheral sensation and vestibular sense to stability, Ring et al. (6) found sway increased on deprivation of vision and
alteration of proprioception due to foot pressure sensory change in 39 persons aged 17 to 79 years. They also found that when both forms of sensation were altered concurrently, sway increased markedly. There is general agreement that postural control involves many sensory and motor systems, and a number of investigators have found age-related declines in visual, vestibular, and sensorimotor functions (7-14). However, little work has been done on assessing the contribution of the demonstrated age-related declines in these systems on the overall decline in postural control that occurs with age. It is an important issue as to whether impairments in these systems contribute to the increased incidence of falls in elderly persons. In this study we examined the relationships between several sensory and motor factors and measures of postural stability in 95 persons living in a hostel for the aged. We developed a battery of simple, noninvasive tests to provide reliable measures of specific "postural control systems" that can be administered to elderly subjects in one brief session. By using this test battery, we were able to assess concurrently sensory and motor factors involved in postural control as well as three global measures of postural stability. Our model of postural control is outlined in Chart 1. The tests were selected so that functioning in the body systems in the model could be assessed. METHODS
Description of the sample. — The sample comprised 95 persons from a hostel for the aged in Sydney, Australia. The hostel housed 124 residents. Of the nonparticipants, 4 were M69
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
A battery of 13 visual, vestibular, sensorimotor, and balance tests was administered to 95 elderly persons (mean age 82.7 years) to examine the relationships between specific sensorimotor functions and measures of postural stability. When subjects stood on a firm surface, increased body sway was associated with poor tactile sensitivity and poor joint position sense. When subjects stood on a compliant surface (which reduced peripheral sensation) with their eyes open, increased body sway was associated with poor visual acuity and contrast sensitivity, reduced vibration sense, and decreased ankle dorsiflexion strength as well as reduced joint position sense. Increased body sway with eyes closed on the compliant surface was associated with poor tactile sensation, reduced quadriceps and ankle dorsiflexion strength, and increased reaction time. Poor performance in two clinical measures of postural stability was associated with reduced sensation in the lower limbs as measured by joint position sense, tactile sensitivity and vibration sense, reduced quadriceps and ankle dorsiflexion strength, and slow reaction times. The prevalence of vestibular impairments was high in this group, but vestibular function was not significantly associated with sway under any of the test conditions. The results suggest that reduced sensation, muscle weakness in the legs, and increased reaction time are all important factors associated with postural instability. An analysis of the percentage increases in sway under conditions where visual and peripheral sensation systems are removed or diminished, compared with sway under optimal conditions, indicated that peripheral sensation is the most important sensory system in the maintenance of static postural stability.
M70
LORDETAL.
Chart 1. Model of Postural Control Body systems that play important roles in postural control include:
Sensory systems
^
Vision Touch Proprioception Vibration sense Vestibular sense Muscle strength Neuromuscular control
Central nervous system
Integration of the sensory and motor factors
Body sway and the tests of static and dynamic balance provide summary measures or indices of postural stability.
ill, 5 were absent (on holidays, etc), and 20 declined to take part. The participation rate of available residents (excluding those ill and absent) was 82.6%. The subjects were aged between 59 and 97 years; the age and sex distribution is shown in Table 1. Most of the subjects were generally independent in such activities of daily living as dressing, bathing, and toileting. Forty percent had personal care assistance on a daily basis. Residents had their own "motel style" rooms and used a common dining room. Most residents (66%) left the hostel for varying periods every day, and a number regularly used public transport. Twenty-one residents (22%) used walking sticks or frames. All residents who were willing to take part in the study were included in the analysis. The fact that the subjects met the criteria for hostel living indicated that they were relatively independent and mobile. A history of stroke, Parkinson's disease, or hip fracture was accepted provided the subject could undertake the test program. The 13 tests were administered by the first author, and individual testing was completed within 1 hour. Sensory Systems Visual acuity was measured binocularly using a standard Snellen scale with subjects wearing their best correction (6). Acuity was measured in units derived from the logarithm of the minimum visual angle resolvable at a test distance of 4 meters. Contrast sensitivity was assessed using the Melbourne Edge Test (MET) — a test specifically designed for screening purposes (8). The MET is a non-grating contrast sensitivity chart. The test measures the contrast threshold for a single luminance profile edge, which is an aperiodic stimulus. The edge contrast threshold chart contains 20 circular test patches (25 mm diameter). The test presents a series of edges of reducing contrast with variable orientation as the identifying feature. A number under each test patch is the contrast sensitivity of that edge in decibels where dB = - 10 log Contrast. The test uses a four-alternative forced choice method of presentation. The edges are presented in the orientations: horizontal, vertical, 45° left, and 45° right. A key card containing the four possible edge angles is provided for
Age Group
Men
Women
Total
59-69 70-79 80-89 90-97
0 4 9 3 82.8 (7.4)
2 22 46 9 82.7(6.5)
2 26 55 12
Mean (SD)
82.7(6.6)
subject instruction. The lowest contrast patch correctly identified was recorded as the subject's contrast sensitivity. Touch thresholds were measured with a SemmesWeinstein Pressure Aesthesiometer, which is a modification of the calibrated hairs used by von Frey (15). The aesthesiometer set contained 20 nylon monofilaments of equal length (38mm) ranging in diameter from 0.06 to 1.14 mm. The force (gm) required to bend each monofi lament was precalibrated and ranges from 0.0045 to 447 gm. Pressure in gm was converted to log,0 0.1 mg, yielding a scale of approximately equal intensity intervals between filaments. The aesthesiometer filaments were applied to the center of the lateral malleolus of the ankle of the dominant leg. Subjects had their eyes closed during the procedure. The ascending and descending method of limits was used to determine tactile thresholds. Joint position sense was measured using an apparatus based on a design by De Domenico and McCloskey (16). Subjects (with eyes closed) attempted to simultaneously place the big toe of the right foot on the right side of a perspex sheet (60cm x 60cm x lcm) and the big toe of the left foot on the corresponding position on the left side of the sheet. The perspex sheet was mounted vertically on the floor. Errors in matching the two toes were measured by reading from a protractor inscribed on the sheet. Subjects had two practice trials, then five experimental trials. The subject's score was the mean error in matching the two toes measured in degrees. Vibration sense was measured using an electronic device that drove a 12-centimeter loudspeaker which generated a 200 cycle/sec vibration. The vibration was transferred to the tibial tuberosity of the knee via a rubber stopper lcm in diameter. The pressure or weight of the apparatus was held constant by a spring-loading mechanism. Intensity of the vibration was read in microns of motion perpendicular to the body surface. Three readings in the ascending mode and three readings in the descending mode were made. Vestibular function was assessed by two methods: the Vestibular Stepping Test and the Vestibular Optical Stability Test. The Vestibular Stepping Test measured subjects' ability to remain stationary and oriented in the one plane while "walking on the spot" with the eyes closed for a period of 1 minute (17,18). The test measurements included distance walked from initial starting point (cm), angle of body rotation (degrees), and sum of distance walked and degrees rotation. To minimize case losses, that is, the loss of individuals from the analysis because they could not complete the task, those who could undertake the test for half or more of the test period were included in the study. Their test scores — distance walked and degrees rotation — were extrapo-
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
Motor systems
Table 1. Age and Sex Distribution of the Sample
POSTURAL STABILITY IN THE ELDERLY
Motor Systems Quadriceps strength was measured by placing a strap around the subject's dominant (stronger) leg. The strap was connected to the central leg of a chair so that when the subject (seated on the chair) attempted to extend his leg he extended a spring gauge giving a measure of maximal quadriceps strength. The subject had three experimental trials, and the greatest extension of the spring gauge was recorded (in kg). Strength was corrected for body size by dividing the measure by the height of the subject. Ankle dorsiflexion strength was measured by having the subject place the foot of his dominant (stronger) leg on a foot rest. A strap (which protruded through the foot rest) was placed over the top of the foot just proxin.al to the commencement of the little toe. The strap was connected to the base of the foot rest so that when the subject (seated on the chair) attempted to raise the front of his foot (while keeping the heel placed on the foot rest) he extended a spring gauge giving a measure of maximal ankle dorsiflexion strength. The subject had three experimental trials, and the greatest extension of the spring gauge was recorded (in kg). Strength was corrected for body size by dividing the measure by the height of the subject.
while fixating a point slightly below eye level at a distance of 3 meters. Testing was performed on a firm surface (a small pile carpeted floor) and on a piece of foam rubber (1 m by 1 m by 15 cm thick) with the subject standing in the center. The same test was repeated on both surfaces with the subject's eyes closed. The foam rubber was used to reduce proprioceptive input from the ankles and cutaneous inputs from the soles of the feet so that subjects would be required to rely on visual and vestibular cues to maintain a steady stance. Four testing conditions were employed: condition A — firm surface, eyes open; condition B — firm surface, eyes closed; condition C — compliant surface, eyes open; and condition D — compliant surface, eyes closed. Maximal sway in the anterior-posterior and lateral directions and total sway (number of square millimeter squares traversed by the pen) in the 30-second periods were recorded for the four test conditions. Figure 1 shows an example of sway tracings for the four conditions in a woman aged 83 years. To minimize case losses, those who could undertake the test for half or more of the test periods were included in the study. Their test scores — number of square millimeter squares traversed by the pen — were extrapolated accordingly. For example, an individual who recorded a sway measure of 100 mm in 20 seconds was coded as having swayed 150 mm. Those who could not undertake the test for 15 seconds were not entered into the analysis. The Static Balance Test measured the ability of subjects to maintain balance while standing on a firm and a compliant 1. EYES OPEN (FLOOR)
3. EYES OPEN (FOAM)
(CONDITION A)
(CONDITION C)
Reaction Time Reaction time was assessed using a simple reaction time paradigm, using a light as the stimulus and depression of a switch (by the hand) as the response. Subjects had 10 practice trials and 10 experimental trials. Reaction time was recorded in milliseconds. 2. EYES CLOSED (FLOOR)
Postural Stability Measures Sway. — Sway was measured using a swaymeter that measured displacements of the body at the level of the waist. The device consisted of a rod attached to the subject at waist level by a firm belt. The rod was 40 cm in length and extended behind the subject. A sheet of graph paper (with a millimeter square grid) was fastened to the top of an adjustable height table that was positioned behind the subject. The height of the table was adjusted so that the rod was in a horizontal plane and the tip of a pen (attached to the end of the rod) could record the movements of the subject on the graph paper. The subject was instructed to stand as motionless as possible for a period of 30 seconds (feet 5 cm apart)
(CONDITION B)
4. EYES CLOSED (FOAM) (CONDITION D)
Figure 1. Body sway under four test conditions in a woman aged 83 years. Source: University of NSW and Teaching Hospitals.
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
lated accordingly. For example, an individual who walked one meter from the mark and rotated 30 degrees in 45 seconds was coded as having walked 133 cm and rotating 40 degrees. Those who could not undertake the test for 30 seconds were not entered into the analysis. The second vestibular function measure was a new test developed to measure vestibulo-optical stability. In this test, subjects' visual acuity while walking on a treadmill was compared to their visual acuity when at rest. It is suggested that any difference in acuity is due to the vestibular system being unable to correct for head movements. Any difference in the log (minimum angle resolvable) for the two test conditions was used as the measure of vestibular optical stability.
M71
M72
LORDETAL.
Test Reliability Most of the tests have been assessed for test-retest reliability in pilot studies. In these studies the tests were administered to 37 subjects aged 63 to 92 years of age on two occasions 6 months apart. The reliability coefficients for the test measures plus 95% confidence intervals are shown in Table 2. Total sway with the eyes open and eyes closed, quadriceps strength, visual acuity, and the sum of distance and angle recorded in the vestibular stepping test had reliability coefTable 2. Test-Retest Reliability Coefficients (n = 37)
Sway — total eyes open Sway — total eyes closed Proprioception Vibration sense Visual acuity Touch Reaction time Vestibular stepping distance Vestibular stepping angle Vestibular sum (distance + angle) Vestibular optical stability Quadriceps strength *p< .01.
Correlation
CI
0.81** 0.73** 0.46** 0.69** 0.73** 0.48** 0.63** 0.51** 0.63** 0.75** 0.17 0.75**
0.66-0.90 0.53-0.85 0.16-0.68 0.47-0.83 0.53-0.85 0.18-0.70 0.38-0.79 0.22-0.72 0.38-0.79 0.56-0.86 -0.16-0.47 0.56-0.86
ficients greater than 0.70. Vibration sense, reaction time, vestibular stepping distance, and vestibular stepping angle had reliability coefficients between 0.50 and 0.70. The remaining tests—joint position sense, touch, and vestibular optical stability — showed moderate to poor test-retest reliability. The reliability of the vestibular optical stability test will be further evaluated. Five tests (that were not used in the pilot study) were included in the battery. Ankle dorsiflexion assessment was included because of a (then new) published finding by Whipple et al. (19) that reduced ankle dorsiflexion strength appears to be an important factor underlying poor balance. They found that ankle dorsiflexion strength was the most impaired of the four main leg muscle groups in a group of fallers compared with controls. Body sway on foam was included in an attempt to assess sway when subjects had reduced ankle support. It was hypothesized that the foam would reduce peripheral sensation, and compel subjects to rely more on visual and vestibular senses. Static balance was included to identify subjects who had difficulty maintaining balance on the floor and on the foam while undertaking the sway tests. Dynamic balance was included because it was noted in the pilot study that although the distance or angle moved in the vestibular stepping test was not associated with instability or falls, stability while performing this test appeared to be associated with other measures of balance, and that performance by those with a history of falls appeared to be worse than by those who had not fallen. Reliability for these measures is still to be determined. The Melbourne Edge Test was included because it was hypothesized that poor contrast sensitivity may be associated with instability and falls. This test, specifically designed for clinical settings, had only just become available. The reliability of the Melbourne Edge Test has been reported elsewhere (8). Data Analysis All test scores, with the exception of scores for static and dynamic balance, were coded as continuous variables. Grades in the tests of static and dynamic balance were collapsed into three groups before assessing associations with other sensory and motor factors. For static balance: grades 1 to 3 were collapsed into one group (poor balance), grade 4 was recoded as a second group (moderate balance) while grade 5 made up the third group (good balance). For dynamic balance, grades 1 and 2 was recoded as one group (poor balance), grades 3 and 4 were recoded as a second group (moderate balance), and grade 5 made up the third group (good balance). The 10 specific measures of sensorimotor systems were correlated with the measures of body sway (under the four test conditions), using partial correlation analysis to control for age. Multiple regression analysis was used to assess the associations between individual sensory and motor test measures and the two clinical measures of stability (static and dynamic balance). In these analyses, the sensorimotor factors were treated as dependent variables, and age was forced into the equation before including the stability measures (which were recoded as dummy variables). It was then
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
surface. Subjects were classified into 5 grades while undertaking the sway tests: "Grade 1" — unable to maintain balance for any period without support on a firm surface (a small pile carpeted floor) with eyes open; "Grade 2 " — unable to maintain balance on the floor for 30 seconds (eyes open or closed); "Grade 3 " — capable of maintaining balance on the floor for 30 seconds (eyes open and eyes closed) but unable to maintain balance on a soft compliant surface (a piece of foam lm by lm by 15 cm thick) without support for any period of time; "Grade 4 " — capable of performing the tests on the floor but unable to maintain stability on the foam for the full 30 seconds (eyes open or closed), and "Grade 5 " — capable of maintaining balance while standing on the floor and the foam (eyes open and closed) for 30-second periods without difficulty. Grades in this test were given while subjects were undertaking the sway tests. The Dynamic Balance Test was an assessment of the subject's ability to maintain balance while performing the vestibular stepping test — walking on the spot for 1 minute with the eyes closed. Subject's performance was classified into 5 grades. "Grade 1" indicated that the subject's balance was so poor that the test could not be administered; "Grade 2 " indicated that the subject could start the test but could not maintain balance for the full minute or required assistance from the experimenter to stay upright to complete the test; "Grade 3 " indicated that the subject could perform the test without requiring assistance but faltered or had to stop walking on two or more occasions to maintain balance; "Grade 4 " indicated that the subject could perform the test but faltered or had to stop walking on one occasion to keep balance; and "Grade 5 " indicated that the subject could perform the task with no difficulty maintaining good balance.
POSTURAL STABILITY IN THE ELDERLY
RESULTS
Table 3 shows the number of residents who could complete each test, and the means and standard deviations for the test measure scores. The missing data are due to certain subjects not complying with the test procedures or not being able to undertake the tests because of impaired balance.
Table 3. Mean Values for the Test Variables (SD)
Visual acuity* Met (dB log contrast) Proprioception (degrees difference) Touch (log10 0.1 milligrams pressure) Vibration senset Vestibular optical stability* Vestibular stepping distance§ Vestibular stepping angle|| Vestibular sum (Distance + Angle) Quadriceps strength/height (kg/cm x 100) Ankle strength/height (kg/cm x 100) Reaction time (milliseconds) Sway — eyes openil Sway — eyes closedH Sway on foam — eyes openil Sway on foam — eyes closedil Static balance Dynamic balance
N
Mean (SD)
95 94 93 94 93 57 73 73 73 94 95 93 94 94 73 67 95 95
3.85 (6.73) 16.3(4.0) 2.6 (2.0) 4.14 (0.41) 38.7 (29.2) 0.21 (0.27) 116 (76) 52 (57) 168(93) 8.49(3.24) 2.62(0.83) 310(140) 112 (63) 143 (111) 229 (113) 338 (137) 2.41 (0.77) 3.59 (1.90)
*Smallest visual angle correctly reported at 4 meters (minutes). tMicrons of motion perpendicular to body surface. ^Difference in smallest visual angle correctly reported at 4 meters on treadmill and smallest visual angle correctly reported at 4 meters while seated (minutes). ^Distance moved from initial position in 1 minute (centimeters). || Angle of rotation in 1 minute (degrees). 11 Millimeter squares traversed by pen on swaymeter in 30 seconds.
Table 4 shows the associations expressed as partial correlation coefficients between the individual sensorimotor system measures and the four body sway measures. Poor tactile sensitivity and joint position sense were associated with sway on the floor (conditions A and B). Increased body sway, eyes open, on the foam (condition C) was associated with poor visual acuity, poor contrast sensitivity, decreased joint position sense, decreased vibration sense, and reduced ankle dorsiflexion strength. Increased body sway, eyes closed, on the foam (condition D) was associated with high touch thresholds, reduced quadriceps and ankle dorsiflexion strength, and increased reaction time. Vestibular function measures were not associated with sway under any of the four test conditions, even though the prevalence of vestibular impairment was high in this group. Fifty-one of the 73 subjects (69%) who could perform the stepping test recorded an abnormal result, that is, they walked a distance greater than 1 meter or rotated more than 45 degrees (18). Similarly, 32 of the 57 subjects (56%) who could perform the vestibular walking test recorded an abnormal result, that is, their visual acuity while on the treadmill was worse than when seated. The two clinical tests of postural stability, the static balance test and the dynamic balance test, have not been previously described. These two tests of stability were significantly associated with total sway. The static balance test had a Spearman correlation coefficient of —0.26 (p < .01) with total sway eyes open, and - 0 . 2 0 (p < .05) with total sway eyes closed. The dynamic balance test had a Spearman correlation coefficient of - 0 . 2 8 (p < .01) with total sway eyes open, and — 0.28(p < .01) with total sway eyes closed. The two clinical measures were also significantly associated (Spearman r = 0.71, p < .01). Thirty-four residents (35.8%) received an abnormal rating in the static balance test (Grades 1-4), that is, they were unable to maintain balance on the foam rubber with eyes Table 4. Correlations Between Sensorimotor Systems and the Four Sway Measures Controlling for Age Sensorimotor system Visual acuity Contrast sensitivity
Sway EO Sway EC Floor 1 Floor 2 0.08 -0.02
0.01 0.07
Proprioception Touch Vibration sense
0.22* 0.17* 0.06
Vestibular stepping test Distance Angle Sum (Distance + Angle) Vestibular optical stability
0.01 0.10 0.02 0.07
0.03 -0.09 -0.08 0.17
Quadriceps strength/height Ankle strength/height
0.09 0.18
Reaction time
0.09
0.25* 0.27** 0.15
Sway EO Foam 3
Sway EC Foam 4
0.28** -0.26*
-0.03 -0.05
0.23* 0.19 0.23* 0.16 -0.02 0.12 0.17
0.05 -0.17 -0.17 0.13
0.14 0.15
-0.12 -0.24*
-0.23* -0.25*
0.08
0.16
Note. 1 - Sway EO (eyes open) on floor (Condition A) 2 - Sway EC (eyes closed) on floor (Condition B) 3 - Sway EO (eyes open) on foam (Condition C) 4 - Sway EC (eyes closed) on foam (Condition D)
*p< .05;**p< .01.
0.13 0.25* -0.01
0.24*
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
possible to assess the change in the multiple correlation coefficient (or multiple r2) when the effects of the stability measures were added to the equation, that is, after controlling for the effects of age. The change in r2 was tested for significance using an F-test. Multiple regression analysis was also used when analyzing other multivariate associations. In these analyses, the measures of sway were the dependent variables, and the sensory and motor measures were independent (or "predictor") variables. Age was forced into the regression equations, before other variables were included using forward selection. Discriminant function analysis, which is a statistical technique which highlights the variables that are most important in the identification of groups, was used to determine which postural control variables discriminated between the subjects with poor, moderate, and good stability as measured by the static and dynamic balance tests. In the discriminant function procedure, age was included as a possible "predictor" variable. For variables with right skewed distributions (such as the vestibular stepping test scores and the joint position sense, quadriceps strength, reaction time and sway measures), logs of variables were analyzed. The data were analyzed using the SPSSX computer package (20).
M73
M74
LORDETAL.
Table 5. Subjects' Performance in the Two Clinical Tests of Stability Static Balance n Grade 1 Grade 2 Grade 3 Grade 4 Grade 5
Table 7. Sensorimotor "Predictor" Variables Identified by the Multiple Regression Analysis
Dynamic Balance n
(%)
(%)
1(1.1) 2(2.1)
21 (22.1) 23 (24.2)
19(20.0) 12(12.6) 61 (64.2)
17(17.9) 9 (9.5) 25 (26.3)
Sensorimotor system Visual acuity Contrast sensitivity
Static Balance
Dynamic Balance 0.01 0.03
Proprioception Touch Vibration sense
0.03 0.02 0.19* 0.13** 0.16**
0.14** 0.24** 0.06
Vestibular stepping test Distance Angle Sum (Distance + Angle) Vestibular optical stability
0.01 0.11 0.08 0.06
0.03 0.01 0.05 0.21**
Quadriceps strength/height Ankle strength/height
0.16** 0.12*
0.14** 0.06
Reaction time
0.03
0.11**
tAssociations are presented in terms of change in Multiple R that result when adding the stability measure to a multiple regression equation after controlling for the effects of age. *Change in r2 significant atp < .05. **Change in r2 significant at p < .01.
closed for 30 seconds. Seventy residents (73.3%) received an abnormal rating in the dynamic balance test (Grades 1-4), that is, they were unable to maintain good balance while walking on the spot for one minute with eyes closed. Grades for the subjects in both tests are shown in Table 5. Table 6 shows the associations between the sensorimotor system measures and scores in the static and dynamic balance tests. Poor scores on the static balance test were associated with reduced tactile sensitivity, joint position and vibration sense, and reduced quadriceps and ankle dorsiflexion strength. Poor scores in the dynamic balance test were significantly associated with reduced tactile sensitivity, reduced sense, poor vestibular optical stability, reduced quadriceps strength, and increased reaction time. Table 7 shows the sensory and motor system variables that were included in the multiple regression equations for the four sway measures (used as dependent variables). Beta weights for each independent variable included in the regression equations and the multiple correlation coefficients are also shown. Beta weights are the coefficients of the independent variables included in the regression equation expressed in a standardized (z score) form. As the units of each measure have been standardized, the beta weights give an indication of the relative importance of each variable in
Beta
Multiple/?
Sway (EO) on floor
Proprioception Age
0.222 -0.109*
0.24
Sway (EC) on floor
Proprioception Touch Age
0.263 0.283 •0.071*
0.37
Sway (EO) on foam
Visual acuity Age
0.290 0.063*
0.31
Sway (EC) on foam
Ankle strength Age
-0.254 •0.091*
0.26
Note. EO = Eyes open; EC = eyes closed. *Significance of Beta weight >.05, i.e., not significant.
explaining the variance in the dependent variable. (However, it should be noted that as the beta weights are affected by correlations among the independent variables they do not in an absolute sense reflect the importance of the various independent variables.) An analysis of residuals indicated that the assumptions for the regression procedures were not violated. Discriminant function analysis was used to determine which postural control variables discriminated between subjects with poor stability, moderate stability, and good stability in the static and dynamic balance tests. For static balance, the following variables: proprioception, vibration sense, and quadriceps strength discriminated significantly between the three groups, as indicated by the final Wilk's lambda of 0.75 and a canonical correlation for the discriminant function of 0.49. The standardized canonical correlation coefficients for the first function (which accounted for 96% of the total between-groups variability) were 0.49 for proprioception, 0.55 for vibration sense, and — 0.76 for quadriceps strength. For dynamic balance, the following variables: proprioception, quadriceps strength, reaction time, and age discriminated significantly between the three groups, as indicated by the final Wilk's lambda of 0.73 and a canonical correlation for the discriminant function of 0.50. The standardized canonical correlation coefficients for the first function (which accounted for 93% of the total between-groups variability) were 0.43 for proprioception, - 0 . 5 5 for quadriceps strength, 0.38 for reaction time, and 0.51 for age. Mean amount of sway condition A (eyes open on the floor) for the 67 subjects who could complete all four sway tests (conditions A-D) was 105 mm. Compared with sway under condition A, sway in condition B (eyes closed on the floor) increased 1.30 times. In this condition vision is absent and peripheral sensation and vestibular sense are present. Sway in condition C (eyes open on the foam) increased 2.58 times. In this condition, peripheral sensation is reduced and vision and vestibular sense are present. Sway in condition D (eyes closed on the foam) increased 4.11 times. Under this condition, vestibular sense is present, vision is absent, and peripheral sensation is reduced. If the assumption is made that sway on the floor is the optimal measure for stability where all senses are available,
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
Table 6. Associationst Between the Sensorimotor Systems and the Clinical Stability Measures Controlling for Age
"Predictor" Variables
Dependent Variables
POSTURAL STABILITY IN THE ELDERLY
DISCUSSION
Cross-sectional studies of this type suffer a number of inherent limitations. Foremost is the problem of uncertainty of causal directions in any associations uncovered. The study also has the complication of multiple measurement and influence. The associations and multiple regression results reported in this article will require validation, as some of the associations may have happened by chance. Other problems and limitations include biases on the part of observer and
Chart 2. The Relative Contributions of Vision, Peripheral Sensation, and Vestibular Sense to the Maintenance of Postural Stability Compared with sway (eyes open on floor) [A]: Sway (eyes closed on floor) [B] is increased 130.2% Sway (eyes open on foam) [C] is increased 258.4% Sway (eyes closed on foam) [D] is increased 410.5% Using the following assumptions: (B-A)/B = Contribution by vision (C-A)/C = Contribution by peripheral sensation (A/D) = Contribution by vestibular sense The contributions of vision, peripheral sensation, and vestibular sense to postural stability can be calculated:* Vision 22%t Peripheral sensation 58% Vestibular sense 20% 'Calculated using results of the 67 subjects who could complete all four sway tests. fContributions are rounded so as to add to 100%. The percentages are broad estimates only.
subject, unintended order effects in the administration of the tests, and variable effort extended by the subjects when undertaking the tests. The approach used in this study to factors associated with postural control assumes that the group effects represent the population and that people with decreased stability have the same underlying predisposing influences. It is acknowledged that it is also possible there are different influences in different people (such as the effects of a wide variety of chronic diseases and illnesses), and that the statistical evidence for the group cannot test whether some subjects had one or more other major influences. The results of the present study revealed a number of significant associations between the measures of stability and specific postural control systems. Increased sway (when standing on a firm surface) was associated with poor tactile sensitivity, joint position sense and reaction time, but not with the visual, vestibular function, or strength measures. The findings that vestibular sense showed no association with sway is in accord with the findings of Nashner (21). He found that the otoliths play no role in the initial detection of body sway. Brocklehurst et al. also reported no correlation between vestibular sense (as measured by response to a slow tilt) and sway (1). These findings agree with previous studies that have found that increased sway in elderly persons is associated with a loss of sensory input from the lower limbs, and that visual and vestibular systems appear to be secondary in the maintenance of posture under normal conditions. When the subjects were placed in a situation that provided reduced support (when standing on a compliant surface), other postural control systems were found to be associated with total sway. Increased body sway with eyes open on the foam was associated with poor visual acuity and contrast sensitivity and reduced quadriceps and ankle dorsiflexion strength. These findings suggest that, in this situation with reduced proprioceptive input from the ankles, subjects are compelled to rely on other sensory and motor factors. Under this condition, subjects with poor vision and strength demonstrated greatly increased sway. When vision was denied and peripheral sensation and ankle support were reduced (eyes closed on the foam), increased body sway was associated with high touch thresholds, reduced quadriceps and ankle dorsiflexion strength, and increased reaction time. In this condition — vision removed and ankle support reduced — subjects with poor tactile sensitivity, muscle strength, and reaction time demonstrated greatly increased sway. The finding that reduced ankle dorsiflexion strength is associated with decreased stability is in accord with the findings by Whipple et al. (19). They found that ankle dorsiflexion strength was significantly impaired in elderly fallers compared with nonfallers and suggested that the inability to initiate appropriate ankle dorsiflexion strength may contribute to instability and falls (especially to the rear). The multiple regression analyses revealed that the main sensorimotor factor contributing to balance under normal conditions (standing on a firm surface with eyes open or closed) is sensation in the lower limbs. Under more adverse conditions (standing on foam — where this sensation is reduced), vision and strength play major roles.
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
it is possible to estimate the relative contributions of vision, peripheral sensation, and vestibular sense to the maintenance of postural stability. For example, if stability is twice as bad with a sense removed, then that sense must contribute half the information for stability. Again, if stability is three times as bad with a sense removed, then that sense must contribute two-thirds the information for stability. The results revealed that sway was increased 1.3 times in condition B (eyes closed on the floor) compared with condition A (eyes open on the floor), that is, stability has been reduced by a factor of 1/1.3. Vision, therefore, must contribute 0.3/1.3 (or 23%) to the maintenance of postural stability. Similarly, as sway is increased 2.58 times in condition C (eyes open on the foam) compared with condition A (eyes open on the floor), stability has been reduced by a factor of 1/ 2.58. Peripheral sensation, therefore, must contribute 1.58/ 2.58 (or 61%) to the maintenance of postural stability. Finally, as sway is increased 4.11 times in condition D (eyes closed on the foam) compared with condition A (eyes open on the floor), stability has been reduced by a factor of 1/4.11. Vestibular sense, therefore, must contribute 1/4.11 (or 24%) to the maintenance of postural stability as it is the only sense left unaltered. Chart 2 shows calculations for determining the relative contributions of these systems. Correcting these percentages, so that the total is 100%, reveals the contribution by vision as 21.3%, the contribution by peripheral sensation as 56.3% and the contribution by vestibular sense as 22.4%.
M75
M76
LORD ETAL.
ACKNOWLEDGMENTS
Address correspondence and reprint requests to Stephen R. Lord, School of Community Medicine, University of NSW, P.O. Box 1, Kensington, N.S.W., 2033, Australia.
REFERENCES
1. Brocklehurst JC, Robertson D, James-Groom P. Clinical correlates of sway in old age-sensory modalities. Age Ageing 1982; 11:1-10. 2. MacLennan WJ, Timothy JI, Hall MRP. Vibration sense, proprioception and ankle reflexes in old age. J Clin Bxp Gerontol 1980;2:159-71. 3. Era P, Heikkinen E. Postural sway during standing and unexpected disturbance of balance in random samples of men of different ages. J Gerontol 1985 ;40:287-95. 4. Lichtenstein MJ, Shields SL, Schiavi R, Burger MC. Clinical determinants of biomechanics platform measures of balance in aged women. J AmGeriatrSoc 1988;36:996-1002. 5. Manchester D, Woollacott M, Zederbauer-Hylton N, Mann O. Visual, vestibular and somatosensory contributions to balance control in the older adult. J Gerontol Med Sci 1989;44:M118-27. 6. Ring C, Nayak USL, Isaacs B. The effect of visual deprivation and proprioceptive change on postural sway in healthy adults. J Am Geriatr Soc 1989;37:745-9. 7. Pitts DG. The effects of aging on selected visual functions. In Sekjuler R, Kline DW, Dismukes K. Aging in human visual functions. New York: Alan R. Liss, 1982. 8. Verbaken JH, Johnston AW. Population norms for edge contrast sensitivity. Am J Optom Physiol Opt 1986;63:724-32. 9. Kaplan FS, Nixon JE, Reitz M, Rindfleish L, Tucker J. Age-related changes in proprioception and sensation of joint position. Acta Orthop Scand 1985;56:72-4. 10. Larsson L. Morphological and functional characteristics of the aging skeletal muscle in man. A cross-sectional study. Acta Physiol Scand 1978 (Suppl 457): 1-36. 11. Mulch G, Petermann W. Influence of age on results on vestibular function tests. Ann Otol Rhinol Laryngol 1979;88 (Suppl 56 No 2): 1-17. 12. Thornbury JM, Mistretta CM. Tactile sensitivity as a function of age. J Gerontol 1981;36:34-9. 13. Welford AT. Motor performance. In: Birren JE, Schaie KW. Handbook of the psychology of aging. New York: Van Nostrand Reinhold, 1977. 14. Whanger AD, Wang HS. Clinical correlates of the vibratory sense in elderly psychiatric patients. J Gerontol 1974;29:39-45. 15. Semmes J, Weinstein S, Ghent L, Teuber H. Somatosensory changes after penetrating brain wounds in man. Cambridge, MA: Harvard University Press, 1960. 16. De Domenico G, McCloskey DI. Accuracy of voluntary movements at the thumb and elbow joints. Exp Brain Res 1987;65:471—8. 17. Fukuda T. The stepping test — two phases of the labyrinthine reflex. Acta Otolaryngol 1959;50:95-108. 18. Kosoy J. The oto-neurologic examination. Acta Otolaryngol 1977;Suppl 343:1-95. 19. Whipple RH, Wolfson LI, Amerman PM. The relationship between knee and ankle weakness to falls in nursing home residents: an isokinetic study. J AmGeriatrSoc 1987;35:13-20. 20. SPSS Inc. SPSSX users guide. Chicago: McGraw Hill, 1983. 21. Nashner LM. A model describing vestibular detection of body sway motion. Acta Otolaryngol 1971 ;72:429-36. Received January 9, 1990 Accepted April 16, 1990
Downloaded from http://geronj.oxfordjournals.org/ at Tulane University Library, Serials Acquisitions Dept. on August 25, 2014
Reduced sensation and strength in the lower limbs and increased reaction time were also associated with the two clinical measures of stability — static and dynamic balance. One of the measures of vestibular function, vestibular optical stability, was also associated with performance in the dynamic balance test. These clinical measures of balance appear to be summary measures of stability. In spite of the significant associations uncovered, it is apparent that most of variance in the measures of sway and balance is left unexplained. This is in agreement with other published work (1-4). It may be that important components are missing from the model of postural control and/or that there is a high degree of inherent variability in the test measures. In addition, the model assumes that the systems that contribute to stability do so in an additive way, and provides no allowance for interactions and feedback mechanisms. However, despite the limitations of the approach, the associations found provide interesting insights into the complex process of postural control. The results suggest that reduced sensation and muscle weakness in the legs and increased reaction time are important factors associated with postural instability. The findings that sway increases with alterations to visual and peripheral sensation, and especially when both are altered concurrently, are in agreement with the findings of Manchester et al. (5)andRingetal. (6). The analysis of the percentage increases in sway under conditions where visual and peripheral sensation systems are removed or diminished, compared with sway under optimal conditions, indicated that peripheral sensation is the most important sensory system in the maintenance of static postural stability. The contributions defined (peripheral sensation — 56%, vision — 21%, and vestibular sense — 22%) are broad estimates and are intended to provide relative measures only. As peripheral sensation was only reduced and not removed, the contribution defined above is most certainly an underestimate of the role of peripheral sense in the maintenance of stability, and an overestimate of the vestibular component. Relationships between these sensorimotor factors and falls will be pursued in further research.
