pi is 0003999303008906

7/30/2019 Pi is 0003999303008906

1/9

Tai Chi Improves Standing Balance Control Under Reduced orConflicting Sensory Conditions

William W. Tsang, MPhil, Vivian S. Wong, MSc, Siu N. Fu, MPhil, Christina W. Hui-Chan, PhD

ABSTRACT. Tsang WW, Wong VS, Fu SN, Hui-Chan CW.Tai Chi improves standing balance control under reduced orconflicting sensory conditions. Arch Phys Med Rehabil 2004;85:129-37.

Objective: To investigate the effects of long-term Tai Chipractice on balance control when healthy elderly Tai Chi prac-titioners stood under reduced or conflicting somatosensory,visual, and vestibular conditions, as compared with healthyelderly nonTai Chi practitioners and young subjects.

Design: Cross-sectional study.Setting: University-based rehabilitation center in Hong

Kong.Participants: Twenty elderly Tai Chi practitioners (mean

experience standard deviation, 7.27.2y) were compared

with 20 elderly nonTai Chi practitioners and 20 young,healthy university students.Interventions: Not applicable.Main Outcome Measures: The amplitude of anteroposte-

rior body sway under different somatosensory, visual, andvestibular conditions was measured using computerized dy-namic posturography, whereby subjects underwent 6 combina-tions of visual and support surface conditions.

Results: The Tai Chi practitioners had significantly betterbalance control than the nonTai Chi subjects in the visual andvestibular ratios, but not in the somatosensory ratio. Further-more, there were no significant differences in any of these 3sensory ratios when the Tai Chi practitioners were comparedwith those of the young, healthy subjects.

Conclusions: Long-term Tai Chi practice improved balance

control in the elderly population when there was an increasedreliance on the visual and vestibular systems during stance. Ofparticular interest is that our elderly Tai Chi practitionersattained the same level of balance control performance as didyoung, healthy subjects when standing under reduced or con-flicting somatosensory, visual, and vestibular conditions.

Key Words: Balance; Rehabilitation; Tai Chi. 2004 by the American Congress of Rehabilitation Medi-

cine and the American Academy of Physical Medicine andRehabilitation

FALLS ARE A MAJOR CAUSE of morbidity and mortalityin the elderly population.1 In fact, 30% of community-dwelling older subjects over the age of 65 were reported tohave fallen at least once a year.2 Studies3-5 have suggested that

deterioration in balance control in the elderly can affect bothstatic postural and dynamic movement equilibrium, which maylead to falls. Preventive measures to improve balance controlare therefore needed to minimize falls among the elderly pop-ulation.

Tai Chi has been practiced by millions of elderly Chinese forthe past 300 years. Initially, it was practiced as a fighting formthat emphasized strength, balance, flexibility, and speed. Overtime, it evolved into a soft, slow, and gentle form of exercisethat is practiced by people of all ages.6 It is a common beliefthat practicing Tai Chi can improve a persons mental andphysical status.7,8 Some qualitative reports on the benefits ofTai Chi began to be published in China in the mid 1970s andin the Western literature in the 1980s.9,10 These initial investi-

gations focused on joint range, cardiorespiratory, and meta-bolic responses.9-11

Studies of Tai Chi and balance control began in the 1990s. In1992, Tse and Bailey12 were the first to evaluate the influenceof Tai Chi on balance control. In a cross-sectional study, theyfound that elderly people with more than 1 year of Tai Chipractice had better balance control than their sedentary coun-terparts in right and left leg standing with the eyes open. In2000, Hong et al13 found that practitioners with more than 10years of Tai Chi experience could maintain single-leg standingwith their eyes closed for a significantly longer period thannonTai Chi practitioners.

In 1993, the Atlanta Frailty and Injuries: Cooperative Stud-ies of Intervention Techniques (FICSIT) group evaluated 2

types of interventions, namely, Tai Chi and computerized bal-ance training, with an education group serving as the control.14

After 15 weeks of intervention, subjects in the Tai Chi grouponly had reduced their fear of falling and the risk of multiplefalls by 47.5%, when compared with the education group.15

However, these findings contradicted the objective balancemeasurements of the FICSIT groups subsequent study in1997,16 which showed a significantly reduced maximum swayamplitude after toe-up perturbations in the computerized bal-ance training group, but not in the Tai Chi group. The use oftoe-up perturbation as an outcome in a Tai Chi study wascriticized by other researchers as being either inappropriate17 ornot challenging enough to the balance control system.18

In 2001, Wong et al18 compared the sway amplitude ofelderly Tai Chi practitioners with an active elderly group, when

standing under 6 combinations of conditions: visual (eyesopen, eyes closed, sway-referenced) and support surface (fixed,sway-referenced). They found that the Tai Chi practitionersswayed significantly less than did the control group in the morechallenged conditions, when the somatosensory input was off-set by absent or conflicting visual input. However, it is notknown whether the improvement is comparable to that realizedin young, healthy subjects standing under similar conditions.

Therefore, in this study, we investigated the effect of Tai Chion balance control in people standing under reduced or con-flicting somatosensory, visual, and vestibular conditions. Byusing computerized dynamic posturography (CDP), we com-pared the balance control of young, healthy subjects with that

From the Department of Rehabilitation Sciences, The Hong Kong PolytechnicUniversity, Hong Kong.

Presented in part at the International Society for Postural and Gait Researchmeeting, June 2327, 2001, Maastricht, The Netherlands.

Supported by the Hong Kong Polytechnic University (research grant).No commercial party having a direct financial interest in the results of the research

supporting this article has or will confer a benefit upon the author(s) or upon anyorganization with which the author(s) is/are associated.

Reprint requests to Christina W. Hui-Chan, PhD, Dept of Rehabilitation Sciences, HongKong Polytechnic University, Hung Hom, Kowloon, e-mail: [email protected].

0003-9993/04/8501-7873$30.00/0doi:10.1016/j.apmr.2003.02.002

129

Arch Phys Med Rehabil Vol 85, January 2004

7/30/2019 Pi is 0003999303008906

2/9

of 2 groups of elderly (aged 60y) who were or were not TaiChi practitioners.

METHODS

Participants

Twenty young healthy subjects (12 men, 8 women; meanage standard deviation [SD], 21.51.6y) were recruited. Allwere university students who had exercised regularly for atleast 2 hours a week. Exclusion criteria were the presence ofinner ear problems, dizziness, long-term medication, a historyof injury within 1 year before the study, and a history of anorthopedic operation or a neurologic disease.

Forty community-dwelling elderly subjects (20 per group),aged 60 or older, participated in the study. Twenty Tai Chipractitioners (10 men, 10 women; mean age, 70.75.1y) were

recruited from the Tai Chi clubs of 3 Hong Kong social centersfor the elderly. All had practiced Tai Chi (Ng style) at least 3times a week (each session was 1h) for more than 1 year(mean Tai Chi experience, 7.27.2y). The nonTai Chi sub-jects (8 men, 12 women; mean age, 67.84.5y) were recruitedfrom several community centers for the elderly. They had noexperience with Tai Chi, although some took morning walks ordid stretching exercises. All 40 elderly subjects were ambula-tory, independent in their activities of daily living, and couldcommunicate and follow the testing procedures. Exclusioncriteria were the presence of symptomatic cardiovascular dis-eases at a moderate exertion level, poorly controlled hyperten-sion or symptomatic orthostatic hypotension, a stroke diagno-sis, Parkinsons disease, or other neurologic disorder,peripheral neuropathy of the lower extremities, crippling ar-

thritis, and metastatic cancer. In addition, subjects with ahistory of falls in the past 12 months were excluded.

Clinical evaluation of the elderly subjects included (1) ageneral health questionnaire, (2) Mini-Mental State Examina-tion19 (MMSE), (3) a physical activity level questionnaire, and(4) assessment of handgrip strength. The Guttman HealthScale20 was used as a general health questionnaire to screen outsubjects according to the exclusion criteria. The questionnairehas 3 sections, from which the functional health status can beassessed. The MMSE is a measure of cognitive ability devel-oped by Folstein et al in 1975.19 The Chinese version of the testwas validated21 and used in the present study. The scale rangesfrom 0 to 30, with a score below 24 suggesting cognitive

dysfunction.19 Elderly subjects who scored below 24 wereexcluded from this study. The physical activity level question-naire was a modified version of the Minnesota Leisure TimePhysical Activity Questionnaire.22,23 It was designed to evalu-ate subjects energy expenditure in leisure-time physical activ-ities such as performing household tasks and engaging insports. They were categorized into 3 levels according to theirmetabolic equivalents (METS) status: light (intensity 4.0METS), moderate (intensity from 4.0 METS to 5.5 METS),and heavy activities (intensity5.5 METS). Handgrip strengthprovides an objective assessment of the subjects general levelof muscle strength.24 Handgrip strength is statistically signifi-cantly associated with the scores achieved on the physicalactivity questionnaire.25-27 Its measurement has the advantageof avoiding recall bias that might occur from the question-naire.27 A Jamar hydraulic dynamometera was used to test the

maximum handgrip strength of both hands of each subject. Thesecond handle position of the dynamometer was chosen, asrecommended by the American Society of Hand Therapists. Inother words, we used both subjective and objective methods26

to compare the physical activity levels of the Tai Chi andnonTai Chi practitioners.

The project was approved by the ethics committee of theHong Kong Polytechnic University, and informed consent wasobtained from all subjects.

Posturography Test Procedures

Proper integration of the information from somatosensory,visual, and vestibular receptors is necessary to generate appro-priate balance responses.28,29 In this connection, CDP is aclinical tool used to measure a subjects ability to use somato-

sensory, visual, and vestibular inputs to coordinate the motorresponses that are necessary to maintain standing balance.29-31

We used a SMART EquiTest CDP machineb in this study.The Sensory Organization Test (SOT) was conducted to assessthe balance control of the elderly subjects. A detailed descrip-tion of the testing procedure has been published32,33 and istherefore only highlighted here. Subjects stood without shoeson the support platform, while wearing a security harness toprevent falls. They were instructed to stand quietly with theirarms at their sides with their eyes looking forward. During theSOT, subjects were exposed to 6 combinations of visual andsupport surface conditions. Each subject underwent 3 consec-utive trials for each of the 6 conditions (see fig 1). In conditions

Fig 1. The SOT for 6 different sensory conditions. Modified from Nashner.32

130 EFFECT OF TAI CHI ON BALANCE CONTROL, Tsang


7/30/2019 Pi is 0003999303008906

3/9

1, 2, and 3, subjects stood on a fixed platform under 3 visualconditions eyes open, eyes closed, and eyes open in a sway-referenced visual surround. In conditions 4, 5, and 6, subjectsstood on a sway-referenced platform under the same 3 visualconditions.

Data Reduction

The dual forceplate of the EquiTest has 4 built-in forcetransducers. The vertical forces captured by the transducerswere used to calculate the center of pressure (COP) of thestanding subject. We used an equilibrium quotient to compare

the 3 groups.31,32,34 The equilibrium quotient is a percentagethat compares the maximum anteroposterior extent of a sub-jects sway to theoretical limits of stability (maximum peak-to-peak12.5).35,36

An equilibrium quotient of 100 is equivalent to no sway,whereas a score of 0 represents a loss of balance, that is,exceeding the 12.5 estimated maximum range of sway. Theequilibrium quotient data were used to analyze the balancecontrol under different somatosensory, visual, and vestibularconditions. We used the sensory ratios that were proposed byNashner32 (fig 2). The somatosensory ratio compared the equi-

Fig 2. Sensory analysis ratios. Modified from Nashner.32

131EFFECT OF TAI CHI ON BALANCE CONTROL, Tsang


7/30/2019 Pi is 0003999303008906

4/9

librium quotients of condition 2 with those of condition 1. Itquantified the extent of stability loss when the subjects closedtheir eyes while standing on a fixed support platform. Thevisual ratio compared the equilibrium quotients of condition 4with those of condition 1. This ratio quantified the extent ofstability loss when the normally dominant somatosensory inputwas disrupted by sway referencing of the support surface withsubjects eyes open. The vestibular ratio compared the equi-librium quotients of condition 5 with those of condition 1. Thisratio reflected the relative reduction in stability when absentvisual and inaccurate somatosensory inputs occurred simulta-neously, forcing subjects to rely primarily on vestibular inputfor balance control.

Statistical Analysis

To ensure data reliability, an intraclass correlation coeffi-cient (ICC3,3) was used to assess the test-retest reliability of theSOT in the elderly subjects. Paired ttests were used to comparethe equilibrium quotient of condition 1 with those of thesubsequent conditions, that is, conditions 2 to 6 in session 1, toassess the standing balance control under reduced or conflictingsensory conditions. One-way analysis of variance (ANOVA)was used to compare the age, height, and weight among the 3

groups, whereas gender was compared by using chi-square test.For the comparison between the 2 elderly groups, independentttests were conducted for the MMSE and the handgrip strengthtest, whereas chi-square test was used for the comparisons ofphysical activity level and hand dominance.

Because there was 1 independent variablesubject group(young subjects, elderly Tai Chi, nonTai Chi practitioners)and 3 dependent variablessomatosensory, visual, and vestib-ular ratioswe used multivariate analysis of variance to com-pare the balance control among the 3 groups. If we found astatistically significant difference in the multivariate tests, uni-variate tests were conducted for each of the sensory ratios. Posthoc analysis by using Bonferroni adjustment was conducted ifa significant difference was found in the univariate test of thesensory ratios. The significance level () in all the statistical

analyses was set at .05.

RESULTS

Test-Retest Reliability of the SOT

Twelve elderly subjects (4 men, 8 women, mean age687.4y) participated in the test-retest reliability study. Aftermeasuring their equilibrium quotients in each of the 6 condi-tions of the SOT, the testing procedures were repeated 1 weekafter the first testing session. The ICC3,3 was calculated for all6 conditions (table 1). These values ranged from .72 (condition3) to .93 (condition 5), which indicated moderate to goodcorrelation.37 The lower bound value of the 95% confidenceinterval (CI) of the ICC for condition 3 was considered low.This might be a result of the small sample size in the reliability

test. Because this particular score was not used in the calcula-tion of the 3 sensory ratios we used, the result of the test-retestreliability with a CI ranging from .37 to .98 was acceptable.

Data from the reliability test showed that the equilibriumquotients of conditions 2 to 6 were statistically significantlylower than the equilibrium quotient in condition 1 (all P.01after Bonferroni adjustment; table 1). Note that the decrease inthe equilibrium quotient was very small when the elderlysubjects stood on a fixed support platform with their eyesclosed, and with sway-referenced visual surround. However,the equilibrium quotients all decreased substantially when thesubjects stood on a sway-referenced support for the 3 visualconditions.

Differences Among Subject Groups

Table 2 shows a comparison of age, height, weight, andgender among the 3 groups. One-way ANOVA showedstatistically significant differences between the young sub-jects and the 2 elderly groups in age and height, but nosignificant differences between the practitioners and non-practitioners of Tai Chi. The difference in height betweenthe young and the elderly did not affect comparisons of the

equilibrium quotients and the sensory ratios because thesway angle was used in their calculation. There were nostatistically significant differences in weight and genderamong the 3 groups.

Table 3 provides further comparisons of the demographicdata of the 2 elderly groups. Independent ttests showed that the2 groups were comparable in MMSE scores and handgripstrength; all 40 subjects had MMSE scores above 24, indicatingthat they had no cognitive impairment.20,21 Physical activitylevels and hand dominance were also found to be similar, byusing the chi-square test and the Fisher exact test, respectively.As determined by the Guttman questionnaire, none of theelderly were limited in any of their activities. The similaritiesin these variables did not confound any differences between the

2 groups.

Comparison of Balance Control Using the SOT

The multivariate tests indicated an overall statistically sig-nificant effect across the 3 sensory ratios for the 3 groups(P.041). The univariate tests showed that there were statis-tically significant differences in the visual and vestibular ratios,but not in the somatosensory ratio (table 4).

Post hoc analysis was not conducted for the somatosensoryratio because no statistically significant difference was found inthe univariate test (table 4). Both the young and the elderlysubjects achieved similar somatosensory ratios (table 4) whenthe equilibrium quotients of condition 2 were compared withthose of condition 1 (fig 2).

For the visual ratio, analysis was conducted by using Bon-ferroni adjustment, because a statistically significant differencewas found in the univariate test (table 4). The nonTai Chipractitioners scored a statistically significantly lower visualratio (mean ratio, .73.14) than the young subjects (meanratio, .85.16; P.005) and the Tai Chi practitioners (meanratio, .82.11; P.049). However, no statistically significantdifference was found between the young subjects and the TaiChi practitioners (P.361). The decrease in the visual ratiosfor the young, the Tai Chi, and the nonTai Chi subjects were15%, 18%, and 27%, respectively, when the equilibrium quo-tients of condition 4 were compared with those of condition 1(table 4, fig 2).

Similarly, post hoc tests were conducted by using Bonferroniadjustment when a statistically significant difference was foundin the univariate test for the vestibular ratio (table 4). ThenonTai Chi practitioners scored a statistically significantlylower vestibular ratio (mean ratio, .58.17) than the youngsubjects (mean ratio, .67.15; P.033) and the Tai Chi prac-titioners (mean ratio, .67.09; P.033). However, no statisti-cally significant difference was found between the young sub-jects and the Tai Chi practitioners (P.996). The decrease inthe vestibular ratios for the young, the Tai Chi, and the nonTaiChi subjects were 33%, 33%, and 42%, respectively, when theequilibrium quotients of condition 5 were compared with thoseof condition 1 (table 4, fig 2).



7/30/2019 Pi is 0003999303008906

5/9

DISCUSSION

Sensory Organization Test

Quantitative posturography has been used by several re-searchers to study the ability of elderly subjects to use somato-sensory, visual, and vestibular information to control their bodysway when standing under reduced or conflicting sensory con-

ditions.29,31,38 Results from our reliability study showed that thedata obtained under the 6 sensory conditions were repeatablewhen the same elderly subjects were tested 1 week apart, withICCs ranging from .72 to .93 (table 1).

In this study, the COP data, as captured by the force plat-form, were used to estimate the center of gravity (COG) fromwhich the equilibrium quotient was obtained. Benda et al39

investigated the biomechanical relation between the COP andCOG, and pointed out that these 2 parameters are not dynam-ically, but only statically, interchangeable. The group recom-mended that the COG position, as estimated from COP data,should be interpreted with caution when unstable patients aretested.39 Although we recruited young university students andhealthy elderly subjects, precaution is advised in the interpre-tation of our results.

Studies have shown that the SOT can differentiate betweenelderly fallers from nonfallers. In 1997, Ho40 found that among15 elderly fallers and 15 nonfallers, the fallers exhibited sig-nificantly greater sway in conditions 4 to 6. In 2001, Wall-mann41 used the SOT to compare 10 elderly fallers with 15nonfallers (age 60y) and found that the fallers swayed sig-nificantly more in condition 4. In the same year, Girardi et al42

used both dynamic posturography and electronystagmography

tests to predict the risk of falls in an elderly population. Boththe SOT and limits of stability test proved to be more sensitivethan the electronystagmography test for predicting falls inelderly subjects.

Differences Among Groups: Effects of Aging and of Tai

Chi Practice

Our study revealed statistically significant differences in thevisual and vestibular ratios (table 4) among the differentgroups. Three main findings emerged when all study subjectswere tested while standing under reduced or conflicting sensoryconditions.

First, the nonTai Chi practitioners exhibited more sway andattained statistically significantly lower visual and vestibularratios than did the young subjects (P.005, P.033, respec-tively). This agrees with previous findings that older subjectssway significantly more than young subjects, when there is anincreased reliance on the visual and vestibular inputs, with theother sensory inputs reduced and/or distorted.29,43,44 Such in-crease in sway is thought to be attributable to the differentdegrees of deterioration that can occur with aging in the so-matosensory, visual, and vestibular systems responsible for

balance control.3,45,46Second, the Tai Chi practitioners swayed significantly less

and achieved significantly higher visual and vestibular ratiosthan did the nonTai Chi practitioners (P.049, P.033, re-spectively). These results suggest that practicing Tai Chi for7.27.2 years improved balance control in the elderly popu-

Table 1: Test-Retest Reliability of the SOT in Elderly Subjects

SOT

Conditions

EQ (%)

ICC3,3 95% CI

Decrease of EQ (%) in

Session 1 vs Condition 1 P

Session 1

(n12)

Session 2

(n12)

Condition 1 94.31.6 93.81.5 .81 .37.95 0.0

Condition 2 92.51.5 92.42.1 .84 .47.95 1.9 .001*

Condition 3 90.44.1 90.62.9 .72 .08.92 4.1 .005*

Condition 4 69.016.9 73.416.5 .90 .66.97 26.8 .000*

Condition 5 54.216.6 58.016.2 .93 .77.98 42.5 .000*

Condition 6 44.921.7 51.016.9 .87 .56.96 52.4 .000*

NOTE. Data are mean SD.Abbreviations: CI, confidence interval; EQ, equilibrium quotient.*Denotes significant difference at P.05/5 or P.01 using Bonferroni adjustment.

Table 2: Comparison of Age, Height, and Weight Among Young,Elderly Tai Chi, and NonTai Chi Subjects

Young Subjects

(n20)

Tai Chi Subjects

(n20)

NonTai Chi

Subjects

(n20) P

Age (y) 21.51.6 70.75.1 67.84.5 .000*

Height (m) 1.660.09 1.540.08 1.590.09 .000*

Weight (kg) 54.810.3 56.210.7 61.18.7 .122

Gender (M/F) 12/8 10/10 8/12 .449

NOTE. Values are mean SD.Abbreviations: F, female; M, male.*Denotes significant difference at P.001 using 1-way ANOVA.Denotes significant difference at P.001 between young andelderlysubjects determined by post hoc analysis using Bonferroni adjust-ment.Denotes significant difference at P.05 between young and elderlynonTai Chi subjects determined by post hoc analysis using Bonfer-roni adjustment.

Table 3: Characteristics of Tai Chi and NonTai Chi Subjects

Tai Chi Subjects

(n20)

NonTai Chi

Subjects

(n20) P

MMSE score 28.31.8 28.12.0 .681

HandgripLeft (kg) 24.67.0 24.27.2 .877

Right (kg) 26.57.8 27.28.7 .790

Physical activity levels .574

Light (4 METS) 1 0

Moderate (5.5 METS) 15 15

Heavy (5.5 METS) 4 5

Hand dominance 1.0*

Left 2 1

Right 18 19

NOTE. Values are mean SD or n.*Determined by the Fisher exact test, because the 2 cells had anexpected count of less than 5.



7/30/2019 Pi is 0003999303008906

6/9

lation. Such positive findings when subjects stood under morefunctional contexts with moving visual surround and supportplatform might explain why the Tai Chi practitioners studiedby Wolf et al15 had decreased fears of falling and a decreasedrisk of multiple falls. Our results are similar to those of Wonget al,18 who also showed that elderly Tai Chi practitionersachieved a better equilibrium score under conditions 5 and 6(fig 1); however, we provide new information on the visual andvestibular ratios. Note also that the preliminary findings werepublished in the same year47 as those of Wong.18 Thus, 2independent studies from 2 different regions produced similarresults in the same year.

Third, our findings further show that the elderly Tai Chipractitioners achieved the same level of balance performanceas the young healthy subjects (P.361 for visual ratio, P.996for vestibular ratio). In other words, these findings suggest thatpracticing Tai Chi for 7.27.2 years enabled the elderly sub-jects to achieve performance in balance control similar to thatof the young control subjects.

How is Tai Chi related to balance control? The practice ofTai Chi requires constant shifting between double-stance and

single-stance, a requirement that constantly challenges the bal-ance control system to maintain the subjects center of masswithin the base of support.48 Such a repetitive demand forbalance control during Tai Chi practice of more than 1 yearprobably explains why our Tai Chi practitioners attained bettervisual and vestibular ratios than did our nonTai Chi elderlysubjects.

Balance and the Visual System

The contribution of vision to balance control has been welldocumented. The Romberg quotient, often used to describe thevisual contribution to stabilizing posture, is a ratio betweenbody sway values recorded during standing with the eyes openand closed. The lower the quotient, the more effect vision hason postural stabilization. Pyykko et al49 reported that, in old age

groups (age80y), the Romberg quotient was .48, whereas thevalue at ages ranging from 50 to 60 years was .78. This findingsuggests that vision is important in maintaining balance inelderly subjects.

However, with aging, there is a decrease in visual acuity,restriction of the visual field, increased susceptibility to glare,and poorer depth perception.3,50,51 These deficiencies may re-sult in a longer delay before the visual system alerts the centralnervous system to a potential fall; they could also affect thevisual sampling necessary to accurately assess the rate anddirection of the falls and the time to landing.50 As a result, anolder persons use of visual information for balance controlcould be diminished, thus increasing his/her chances of falling.

As discussed, investigators29,43,44,52 using the SOT havefound that elderly subjects swayed more than younger subjectsunder conditions of reduced or conflicting sensory information.Elderly subjects also had greater difficulty when the visualinput was manipulated to make it inappropriate to the posturaltask, and when both proprioception and visual inputs werereduced or absent. We found that Tai Chi practitioners had asignificantly higher visual ratio (mean ratio, .82.11) than thenonpractitioners (mean ratio, .73.14; P.049). This was de-noted by less sway of the Tai Chi practitioners when there wereerror signals arising from the somatosensory system (condition4: eyes open with sway-referenced support; fig 1). Even moreinteresting is that the visual ratio attained by the Tai Chipractitioners was comparable to that of the young healthysubjects (mean ratio, .85.16; P.361). As reported by Ho,40

the mean visual ratio of the elderly fallers was .65, which waslower when compared with our elderly Tai Chi and nonTaiChi practitioners (respectively, 26% and 12% lower). Thus,subjects who sway more when there are error signals arisingfrom the somatosensory system may have a higher tendency tofall.

Many Tai Chi forms require practitioners to focus their eyeson their hand movement through head and/or trunk rotation. Inthis connection, Szturm et al53 found that the left-right differ-ences in the vestibulo-ocular reflex gain of patients withchronic peripheral vestibular dysfunction could be reduced byan exercise program of eye and head movements. Some com-ponents of this exercise program, such as focusing on a sta-tionary visual target and tracking a moving visual target withthe head either fixed or moving, were similar to those thatoccur during Tai Chi practice. However, the mechanismthrough which Tai Chi practice can improve balance controlthrough the visual system is not known, and further investiga-tion is needed.

Balance and the Vestibular System

Practicing Tai Chi involves head movements that will stim-ulate the vestibular system, and repeated vestibular stimulationover time may facilitate balance control in the practitioners. Inthis study, the Tai Chi practitioners attained a higher vestibularratio than the nonTai Chi elderly under condition 5 (P.033),when the subjects stood with their eyes closed on a sway-referenced support surface. Condition 5 was designed so thatsubjects had to rely more heavily on the vestibular system tomaintain balance.36,54 Interestingly, the vestibular ratioachieved by the elderly Tai Chi practitioners was similar to thatof the young adult subjects, and no statistically significantdifference was found between them (P.996). In this connec-tion, the mean vestibular ratio of the elderly fallers was .51 in

Table 4: SOT: Comparison of Balance Control Under Different Sensory ConditionsAmong Young, Elderly Tai Chi, and NonTai Chi Subjects

Young Subjects

(n20)

Tai Chi Subjects

(n20)

NonTai Chi Subjects

(n20) P

Somatosensory ratio .98.03 (2%) .98.02 (2%) .98.03 (2%) .827

Visual ratio .85.16 (15%) .82.11 (18%) .73.14 (27%) .015*

Vestibular ratio .67.15 (33%) .67.09 (33%) .58.17 (42%) .049*

NOTE. Values are mean SD. The figures in parentheses denote the decrease of sensory ratio as a percentage when compared with condition 1.*Denotes significant difference at P.05 between the young and the elderly subjects by using univariate tests.Denotes significant difference at P.01 between young and elderly nonTai Chi subjects determined by post hoc analysis using Bonferroniadjustment.Denotes significant difference at P.05 between elderly Tai Chi and elderly nonTai Chi subjects determined by post hoc analysis usingBonferroni adjustment.Denotes significant difference at P.05 between elderly nonTai Chi subjects with young and elderly Tai Chi subjects determined by post hocanalysis using Bonferroni adjustment.



7/30/2019 Pi is 0003999303008906

7/9

Hos study,40 which was 31% lower than that of our Tai Chipractitioners and 14% lower when compared with our nonTaiChi practitioners (table 4).

The effect of aging on the vestibular system has been inves-tigated in posturography studies,29,31,38 in which the elderlysubjects tested swayed significantly more in conditions 5 and 6than did the young subjects (fig 1). Poorer balance control inthe elderly subjects could be attributed to age-related structuraldegeneration. For example, Rosenhall and Rubin55 reported a40% reduction in the sensory cells of the vestibular apparatusin subjects more than 70 years of age.

In studies on the rehabilitation outcome of peripheral ves-tibular disorders, specific vestibular exercises have producedbetter balance performance than nonspecific general condition-ing exercises. Horak et al56 examined the effectiveness of aspecific vestibular rehabilitation program, a general condition-ing exercise program, and a vestibular suppressant medicationin reducing symptoms in patients with chronic peripheral ves-tibular disorders. Although their sample size was small and theage range was wide (18 60y), patients in the vestibular reha-bilitation groups showed improvement in their equilibriumquotients, especially in conditions 5 and 6. The vestibularrehabilitation program included practicing balance during func-

tional activities such as walking with head turns and pivotturning or with altered surface or visual inputs. Horak sug-gested that these activities may improve patients ability to usethe remaining vestibular inputs to achieve better balance con-trol.

Exercise used in vestibular rehabilitation emphasizes headand trunk rotations, changes in the base of support, and fixationof the eyes on a target. It also involves turning while walking,making large circles, and gradually making smaller and smallerturns. It is well known that adaptation of the vestibulo-ocularsystem is context specific and that adaptation of the vestibularsystem is frequency dependent. Thus, during vestibular reha-bilitation, patients are instructed to perform head movementexercises at different frequencies and with different head po-sitions to achieve optimal effects.57 This is similar to Tai Chi,

which also involves many head and body rotations, changes inthe base of support, and fixation of the eyes on the hand duringits practice.

Vestibular adaptation is affected by voluntary motor control.Mental effort will help to improve the gains.57 Patients investibular rehabilitation are encouraged to concentrate on thetask and not to be distracted by conversation and other activ-ities.58 Tai Chi also requires the subject to practice with arelaxed body and a calm but concentrated mind.6 Last, balanceexercises for vestibular adaptation emphasize progression bydecreasing the base of support, changing the head or armposition, manipulating sensory cues, and moving from static todynamic activities. Tai Chi also starts with a simple and stableposture and then gradually progresses to more difficult trunkand head rotations and from weight shifting to single-leg stand.

The base of support keeps changing, and different arm move-ments are also involved.6,59 All these motions show similaritiesto the vestibular rehabilitation protocols recommended by var-ious investigators. This may explain why the Tai Chi practi-tioners had better balance performance than the nonTai Chigroup when there was an increased reliance on the vestibularsystem. Further investigation of the relation between Tai Chipractice and vestibular function is warranted.

Other Considerations for Future Studies

Because we used an observational, cross-sectional studydesign, we cannot establish a causal relation between thepractice of Tai Chi and the improvement of balance control

when standing under changing sensory conditions. Prospective,randomized control studies should be conducted to delineatethe training effect of Tai Chi on the somatosensory, visual, andvestibular systems that are responsible for balance control.Indeed, by using a prospective study, we had shown that 2months of intensive Tai Chi training was sufficient to improvethe balance control of the elderly subjects, in terms of (1) lessbody sway when standing under conditions requiring an in-

creased reliance on the vestibular system and (2) more smoothmovement to different targeted positions when standing withsubjects limits of stability.60

We used the SOT to compare only 1 perspective of balancecontrol among the 3 groups. It focuses on subjects ability touse somatosensory, visual, and vestibular inputs in coordinat-ing motor responses that are necessary to maintain standingbalance; it does not reflect the full complexity of balancecontrol required during the performance of functional tasks.More specifically, research findings from Krebs and col-leagues61,62 have shown that posturography findings do notnecessarily correlate with the more commonly used functionaltests. Krebs61 investigated the correlation between the SOTwith gait and clinical tests of timed gait and standing in patientswith peripheral vestibular hypotension and found only 7 of 18

comparisons to be statistically significant. In another study,Krebs62 correlated the SOT with 88 functional performancetests to monitor the change of balance control in patients withperipheral vestibular hypofunction who had received 6 to 8weeks of vestibular rehabilitation. Only 7 statistically signifi-cant correlations were detected. It should be noted that, al-though the SOT examines the control of standing balance underreduced or conflicting sensory conditions, functional tests oftenrequire voluntary control of movement with time and speedcontrol demands that are beyond the control of standing bal-ance alone. Furthermore, functional tests are conducted undernormal sensory conditions, whereas the SOT requires a subjectto retain standing balance despite conflicting sensory condi-tions (ie, to rely on the vestibular system when vision is absent,despite receiving error signals from the somatosensory system,

as in condition 5). Therefore, our findings in this study onstanding balance should not be generalized to functional per-formance. Along this line of thought, we are studying the motorcontrol of elderly Tai Chi practitioners in stair descent, as TaiChi practice involves a lot of eccentric leg muscle work that isrequired to control the body with respect to gravity force.

In this study, we investigated only 1 type of exercise, TaiChi. Other exercises, such as a balance training program, havebeen documented by others to have an effect on balance con-trol.44,63 We have also initiated a study to compare the speci-ficity of Tai Chi training in improving balance control with anequivalent active exercise group.

CONCLUSION

We investigated possible effects of Tai Chi on balancecontrol by examining subjects ability to use somatosensory,visual, and vestibular information to control their body swaywhile standing under 6 different sensory conditions when so-matosensory or visual inputs were either absent or inaccurate.Our results indicate that practicing Tai Chi for 7.27.2 yearsimproved the control of stance under reduced or conflictingsensory conditions in an elderly population. Elderly Tai Chipractitioners showed better balance performance than nonTaiChi practitioners who were similar in age and gender, whenthere was an increased reliance on the visual and vestibularsystems. An important finding is that practicing Tai Chi for7.27.2 years enabled the elderly subjects to achieve a level of



7/30/2019 Pi is 0003999303008906

8/9

balance performance comparable to that of young healthysubjects under our present experimental paradigm.

Acknowledgments We thank Dr. Rhonda Scudds, Department ofRehabilitation Sciences, the Hong Kong Polytechnic University, forher statistical advice, and Bill Purves for his English editorial advice.

REFERENCES1. Wolter LL, Studenski SA. A clinical synthesis of falls intervention

trials. Top Geriatr Rehabil 1996;11:9-19.2. Tinetti ME, Williams TF, Mayewski R. Risk factors for falls

among elderly persons living in the community. Am J Med1988;80:429-34.

3. Horak FB, Shupert CL, Mirka A. Components of postural dys-control in the elderly: a review. Neurobiol Aging 1989;10:727-38.

4. Patla A, Frank J, Winter D. Assessment of balance control in theelderly: major issues. Physiother Can 1990;42:89-97.

5. Wolfson L, Whipple R, Derby C, et al. Balance and strengthtraining in older adults: intervention gains and tai chi mainte-nance. J Am Geriatr Soc 1996;44:498-506.

6. Tsao SW. In-depth analysis of Taijichuan. Hong Kong: ChineseUniv Pr; 1995.

7. Jin PT. Efficacy of Tai Chi, brisk walking, mediation, and readingin reducing mental and emotional stress. J Psychosom Res 1992;36:361-70.

8. Kutner NG, Barnhart HX, Wolf SL, McNeely E, Xu T. Self-reportbenefits of Tai Chi practice by older adults. J Gerontol B PsycholSci Soc Sci 1997;52:242-6.

9. Zhuo D, Shephard RJ, Plyley MJ, Davis GM. Cardiorespiratoryand metabolic responses during Tai Chi Chuan exercise. CanJ Appl Sport Sci 1984;9:7-10.

10. Deusen JV, Harlowe D. The efficacy of the ROM dance programfor adults with rheumatoid arthritis. Am J Occup Ther 1987;41:90-5.

11. Lan C, Lai JS, Wong MK, Yu ML. Cardiorespiratory function,flexibility, and body composition among geriatric Tai Chi Chuanpractitioners. Arch Phys Med Rehabil 1996;77:612-6.

12. Tse SK, Bailey DM. Tai Chi and postural control in the wellelderly. Am J Occup Ther 1992;46:295-300.

13. Hong YL, Li JX, Robinson PD. Balance control, flexibility, and

cardiorespiratory fitness among older Tai Chi practitioners. Br JSports Med 2000;34:29-34.14. Wolf SL, Kutner NG, Greenwood R, McNeely E. The Atlanta

FICSIT study: two exercise interventions to reduce frailty inelders. J Am Geriatr Soc 1993;41:329-32.

15. Wolf SL, Barnhart HX, Kutner NG, McNeely E, Coogler C, Xu T.Reducing frailty and falls in older persons: an investigation of TaiChi and computerized balance training. J Am Geriatr Soc 1996;44:487-97.

16. Wolf SL, Barnhart HX, Ellison GL, Coogler CE, Atlanta FICSITGroup. The effect of Tai Chi Quan and computerized balancetraining on postural stability in older subjects. Frailty and Injuries:Cooperative Studies on Intervention Techniques. Phys Ther 1997;77:371-81.

17. Horak FB. Discussion of The effect of Tai Chi Quan and com-puterized balance training on postural stability in older subjects.Phys Ther 1997;77:382-3.

18. Wong AM, Lin YC, Chou SW, Tang FT, Wong PY. Coordinationexercise and postural stability in elderly people: effect of Tai ChiChuan. Arch Phys Med Rehabil 2001;82:608-12.

19. Folstein MF, Folstein SE, McHugh PR. Mini-mental state. Apractical method for grading the cognitive status of patients for theclinician. J Psychiatr Res 1975;12:189-98.

20. Rosow I, Breslau N. A Guttman health scale for the aged. JGerontol 1966;21:556-9.

21. Chiu HF, Lee HC, Chung WS, Kwong PK. Reliability and validityof the Cantonese version of Mini-Mental Status Examinationapreliminary study. J Hong Kong Coll Psychiatr 1994;4:25-8.

22. Taylor HL, Jacobs DR, Schucker B, Knudsen J, Leon AS, De-backer G. A questionnaire for the assessment of leisure timephysical activities. J Chronic Dis 1978;31:741-55.

23. Van Heuvelen MJ, Kempen GI, Ormel J, Rispens P. Physicalfitness related to age and physical activity in older persons. MedSci Sports Exerc 1998;30:434-41.

24. Maki BE, Holliday PJ, Tropper AK. A prospective study ofpostural balance and risk of falling in an ambulatory and inde-pendent elderly population. J Gerontol 1994;49:M72-84.

25. Cooper C, Baker DJ, Wickham C. Physical activities, musclestrength, and calcium intake in fracture of the proximal femur inBritain. BMJ 1988;297:1443-6.

26. Sandler RB, Burdett R, Zaleskiewicz M, Sprowls-Repcheck C,Harwell M. Muscle strength as an indicator of the habitual level ofphysical activity. Med Sci Sports Exerc 1991;23:1375-81.

27. Rantanen T, Parkatti T, Keikkinen E. Muscle strength accordingto level of physical exercise and educational background in mid-dle-aged women in Finland. Eur J Appl Physiol 1992;65:507-12.

28. Anacker SL, Di Fabio RP. Influence of sensory inputs on standingbalance in community-dwelling elders with a recent history offalling. Phys Ther 1992;72:575-84.

29. Wolfson L, Whipple R, Derby CA, et al. A dynamic posturogra-phy study of balance in healthy elderly. Neurology 1992;42:2069-75.

30. Hamid MA, Hughes GB, Kinney SE. Specificity and sensitivity ofdynamic posturography. A retrospective analysis. Acta Otolaryn-gol 1991;481(Suppl):596-600.

31. Camicioli R, Panzer VP, Kaye J. Balance in the healthy elderly:

posturography and clinical assessment. Arch Neurol 1997;54:976-81.

32. Nashner LM. Computerized dynamic posturography. In: JacobsonG, Newman C, Kartush J, editors. Handbook of balance functionand testing. St. Louis: Mosby Year Book; 1993. p 280-307.

33. Lepers R, Bigard AX, Diard JP. Posture control after prolongedexercise. Eur J Appl Physiol 1997;76:55-61.

34. Wolfson L, Whipple R, Judge J, Amerman P, Derby C, King M.Training balance and strength in the elderly to improve function.J Am Geriatr Soc 1993;41:341-3.

35. Nashner LM, McCollum G. The organization of human posturalmovements: a formal basis and experimental synthesis. BehavBrain Sci 1985;8:135-72.

36. Nashner LM. Practical biomechanics and physiology of balance.In: Jacobson G, Newman C, Kartush J, editors. Handbook ofbalance function and testing. St. Louis: Mosby Year Book; 1993.

p 261-79.37. Portney L, Watkins M. Foundations of clinical research: applica-tions to practice. East Norwalk: Appleton & Lange; 1993.

38. Whipple R, Wolfson L, Derby C, Singh D, Tobin J. Alteredsensory function and balance in older persons. J Gerontol 1993;48:71-6.

39. Benda BJ, Riley PO, Krebs DE. Biomechanical relationship be-tween center of gravity and center of pressure during standing.IEEE Trans Rehabil Eng 1994;2:3-9.

40. Ho ST. Balance performance in community-dwelling female el-derly fallers and non-fallers [dissertation]. Hong Kong: HongKong Polytechnic Univ; 1997.

41. Wallmann HW. Comparison of elderly nonfallers and fallers onperformance measures of functional reach, sensory organization,and limits of stability. J Gerontol A Biol Sci Med Sci 2001;56:M580-3.

42. Girardi M, Konrad HR, Amin M, Hughes LF. Predicting fall risks

in an elderly population: computer dynamic posturography versuselectronystagmography test results. Laryngoscope 2001;111:1528-32.

43. Shepard NT, Schultz A, Alexander NB, Gu MJ, Boismier T.Postural control in young and elderly adults when stance is chal-lenged: clinical versus laboratory measurements. Ann Otol RhinolLaryngol 1993;102:508-17.

44. Ledin T, Kronhed AC, Moller C, Moller M, Odkvist LM, OlssonB. Effects of balance training in elderly evaluated by clinical testsand dynamic posturography. J Vestib Res 1991;1:129-38.

45. Alexander NB. Postural control in older adults. J Am Geriatr Soc1994;42:93-108.

46. Hageman PA, Leibowitz JM, Blanke D. Age and gender effects onposture control measures. Arch Phys Med Rehabil 1995;76:961-5.



7/30/2019 Pi is 0003999303008906

9/9

47. Tsang WW, Wong V, Hui-Chan CW. A comparison of balancecontrol under different sensory conditions between elderly Tai Chiand non-Tai Chi practitioners. In: Duysens J, Smits-EngelsmanBC, Kingma H, editors. Control of posture and gait. Maastricht:ISPGR; 2001. p 382-7.

48. Winter DA. Human balance and postural control during standingand walking. Gait Posture 1995;3:193-214.

49. Pyykko I, Jantti P, Aalto H. Postural control in elderly subjects.Age Ageing 1990;19:215-21.

50. Stelmach GE, Worringham C. Sensorimotor deficit related topostural stability. Clin Geriatr Med 1985;1:679-94.

51. Woollacott MH. Age-related changes in posture and movement. JGerontol 1993;48:56-60.

52. Woollacott MH, Shumway-Cook A, Nashner LM. Aging andposture control: changes in sensory organization and muscularcoordination. Int J Aging Hum Dev 1986;23:97-114.

53. Szturm T, Ireland DJ, Lessing-Turner M. Comparison of differentexercise programs in the rehabilitation of patients with chronicperipheral vestibular dysfunction. J Vestib Res 1994;4:461-79.

54. Shepard NT. The clinical use of dynamic posturography in theelderly. Ear Nose Throat J 1989;68:940-57.

55. Rosenhall U, Rubin W. Degenerative changes in the human ves-tibular sensory epithelia. Acta Otolaryngol (Stockh) 1975;79:67-80.

56. Horak FB, Jones-Rycewicz C, Black FO, Shumway-Cook A.Effects of vestibular rehabilitation on dizziness and imbalance.

Otolaryngol Head Neck Surg 1992;196:175-80.

57. Herdman SJ. Balance rehabilitation: background, techniques, andusefulness. In: Jacobson GP, Newman CW, Kartush JM, editors.Handbook of balance function testing. St. Louis: Mosby YearBook; 1993. p 392-406.

58. Herdman SJ. Role of vestibular adaptation in vestibular rehabili-tation. Otolaryngol Head Neck Surg 1998;119:49-54.

59. Wolf SL, Coogler CE, Xu T. Exploring the basis for Tai ChiChuan as a therapeutic exercise approach. Arch Phys Med Rehabil1997;78:886-92.

60. Hui-Chan CW, Tsang WW. Two months of Tai Chi practice issufficient to improve balance control in the well elderly? Paperpresented at: International Conference on Tai Chi Chuan; 2001Dec 1-3; Hong Kong. p 22-3.

61. Evans MK, Krebs DE. Posturography does not test vestibulospinalfunction. Otolaryngol Head Neck Surg 1999;120:164-73.

62. ONeill DE, Gill-Body KM, Krebs DE. Posturography changes donot predict functional performance changes. Am J Otol 1998;19:797-803.

63. Rose DJ, Clark S. Can the control of bodily orientation be signif-icantly improved in a group of older adults with a history of falls?J Am Geriatr Soc 2000;48:275-82.

Suppliersa. Sammons Preston Rolyan, 4 Sammons Ct, Bolingbrook, IL 60440.b. NeuroCom International Inc, 9570 SE Lawnfield Rd, Clackamas,

OR 97015-9943.



pi is 0003999303008906

Documents