Functional reach: Does it really measure dynamic balance?

262

Functional Reach: Does It Really Measure Dynamic Balance? Mm-a Wernick-Robinson,

MS, PT, NCS, David E. Krebs, PhD, P2; Marie M. Giorgetti, MS, PT, NCS

ABSTRACT. Wernick-Robinson M, Krebs DE, Giorgetti MM. Functional reach: does it really measure dynamic balance? Arch Phys Med Rehabil 1999;80:262-9. Background: Functional reach (FR) is a new clinical measurement intended to assessdynamic balance. The purposes of this study were (1) to measure the mean FR distance in healthy elders compared with individuals with known balance impairments, (2) to analyze the extent to which FR measures dynamic balance, and (3) to describe movement strategies used during FR. Methods: Thirteen healthy elders and 15 individuals with vestibular hypofunction (VH) were tested during FR and free gait. Whole body kinematic and kinetic data including the center of gravity (CG) and center of pressure (CP) using 11 body segments and two force plates, respectively, were collected. Results: There was no difference in FR distance between healthy elders and individuals with VH. FR distance was not correlated to lateral stability measures, but was related to anterior-posterior postural control measures of FR (r = .69 to .84) in both groups. Although FR distance strongly correlated with maximum moment arm during FR in both groups, the correlations were not as strong when the subjects were then classified by movement strategy. The mean moment arm during FR was significantly less than that of free gait. Conclusions: These data suggest FR does not measure dynamic balance; healthy elders and balance-impaired individuals with vestibular dysfunction attained the same FR distance and did so without increasing the moment arm during or at the end of FR. Recording the strategy used during FR, however, may provide other valuable information necessary in addressing balance control. Clinical implications of assessing movement strategy are discussed. 0 1999 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation

B

ALANCE REQUIRES interactions that include the vestibular, visual, proprioceptive, musculoskeletal, and cognitive systems. The majority of falls in the elderly occur during locomotion yet decreasedperformance on nonlocomotor, standing-still clinical balance tests are used to predict falls.’ Functional reach (FR) is a fairly new clinical measurement intended to detect dynamic balance impairments in elders2-5and, more recently, in patients with vestibular hypofunction (VH).6 FR is From the Biomotion Laboratory (Ms. Wemick-Robinson, Dr. Krebs), the Physical Therapy Department (Ms. Wemick-Robinson, Ms. Giorgetti), and the Institute of Health Professions Graduate Program in Physical Therapy (Dr. Krebs. Ms. Giorgetti). Massachusetts General Hosoital. Boston. MA. Submitted for publication’ March 10, i998. Accepted in revised form July 6, 1998. Supported in part by gmnts from the National Institute on Aging (NIH ROl AG 1256i;NIHROlAG 11255, andNIHP50AG11669) andNIDRRH133G3004. No commercial patty having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to Mara Wernick-Robinson MS, PT, NCS, Massachusetts General Hospital, PT Services, WACC 128, 15 Pa&man Street, Boston, MA02114. 0 1999 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003-9993/99/8003-4924$3.00/O

Arch

Phys

Med

Rehabil

Vol 80, March

1999

the maximum distance an individual can reach forward while maintaining a fixed base of support in standing.2 Little research exists that validates FR and its mechanisms of challenging balance. Duncan and colleagues2 report FR is moderately associated with anterior-posterior (AP) center of pressure (CP) excursion (Y = .71). Recently, however, Maki and colleagues7 showed that fallers exhibited larger amplitudes of CP displacement in the medial-lateral (ML) direction compared with nonfallers; the AP amplitude showed less consistent evidence for fall association. Preferred gait speed, an accepted objective measure of functional ability in elders,* has been found to be moderately related to FR (Y = .71) in community-dwelling elders4 but poorly related to FR (v = .35) in nursing home residents.5 Duncan and colleagues2 reported FR to be precise and to have good test-retest and interobserver reliability. Giorgetti and colleagues,9however, reported only moderate reliability among two experienced examiners. Weiner and coworkersto found that FR, gait speed, and mobility skills changed significantly in a group of male veterans undergoing rehabilitation, but not in a control group. The correlation of baseline FR with change in FR was poor (Y = .38). Whether rehabilitation improves FR could not be answered by this study, perhaps because FR strategy was not accounted for. Duncan and colleagues2 proposed that FR be used as a dynamic measure of balance irrespective of movement strategy chosen or standardized during the measurement. A possible limitation of prior studies, however, is that movement strategy used in FR was not described. For example, the patient can flex the hips and plantarflex the ankles, similar to hip strategy kinematics,1’~i2 or primarily use ankle plantarflexion with minimal hip flexion, similar to ankle strategy kinematics.11~*2 Identifying the strategy used may be helpful in assessmentand in treatment planning.6,13,14Measurement of balance should identify not only the relative success of maintaining equilibrium, but also the efficiency of movement strategies used to achieve the equilibrium posture.14 Patients with partial or complete vestibular loss may rely on an ankle strategy for balance even when a task requires a hip strategy.15-t7Herdman and colleagues, r* by contrast, found an increased use of hip strategies in patients with bilateral vestibular loss compared to normal subjects. Some patients with vestibular pathologies may rely more heavily on a hip strategy to control their center of gravity (CC) when not required to do SO.~~,~O Balance can be estimated from CG or CP displacements and can be maintained when the horizontal CG projection is far from the CP, or even when the CG lies outside the base of support,21 such as in gait. 22 The whole body CG-CP moment arm is the CG-CP separation distance in the AP and the ML plane. 23 As the moment arm increases, static equilibrium decreases, and the person must exert greater torque and thus greater balance to stay upright. 6,23Therefore, we define balance as the ability to react to destabilizing forces quickly and efficiently so as to regain stability.r4 Although FR may displace the CG from the CP and increase the moment arm, it may be possible to reach forward without increasing the moment arm, perhaps by using what appears to be a hip strategy. Healthy individuals without complaints of impaired balance and with no reported history of falls serve as a powerful control

FUNCTIONAL

REACH,

263

Wernick-Robinson

group. If FR is to be used as a single screening tool to predict falls in the elderly,3 than the use of elderly as a control group seems logical to compare with a group of subjects who have balance impairments and have fallen. We chose to compare FR with walking speed because walking speed has been shown to correlate strongly with other measures of balance.‘.5,7,8Movement strategy during FR will be described as the whole body movement strategy, primarily identifying the trunk and lower extremity kinematics and whole body kinetics. The purposes of this study were (1) to measure the mean FR distance in healthy elders compared with individuals with known balance impairments; (2) to analyze the extent to which FR measures dynamic balance; and (3) to describe movement strategies used during FR. We hypothesized that (1) FR distance differs between healthy elders and people with VH; (2) FR and the whole body moment arm during FR are highly correlated; (3) FR and free gait velocity are highly correlated; and (4) subjects with no measurable vestibular function use a volitional hip strategy during FR. METHODS Subjects The sample consisted of 28 subjects: 13 healthy elders, age 74.53 -I: 6.83yrs, and 15 individuals with nonacute VH, age 58.97 i 18.49yrs (table 1). The healthy elders were randomly selected from Medicare lists and were part of a larger study. Their primary care physicians agreed for them to be eligible for this larger study based on the criteria that they had no musculoskeletal, neurologic, or cardiovascular conditions that would limit their ability to participate in a standing, progressive resistive exercise program. In addition to having no discernible neuromusculoskeletal deficits by physical examination, all healthy elder subjects had no reported falls and could run on a treadmill during an exercise tolerance testing using a modified Bruce protocol. Thus, healthy elder subjects had intact dynamic balance, including being able to move on a moving treadmill without falling. Because the modified Bruce protocol was used only as a screening test for the healthy elders, data from this test are not reported. The subjects with VH were diagnosed by an otoneurologist based on electronystagmography, sinusoidal vertical axis rotation (SVAR), visual-vestibular interaction rotation tests, and posturography. 24 All subjects had abnormal vestibulo-ocular tests and falls on Equitest posturography sensory organization tests 5 and/or 6. Subjects classified as having “no measurable vestibular function” had decreased caloric responses bilaterally and decreased gains (CO.1 gains at .05Hz) during SVAR testing. In addition, all subjects reported impaired balance for at least 2 months and had requested vestibular rehabilitation, although none had yet had rehabilitation at the time of testing.

Table

1: Descriptive Healthy

Range Age

Characteristics by Group Elder Group (n = 13) Mean

of Subjects

Range

(n = 15) Mean

(SD)

(yrs

+ mo)

Height(m) FR (cm) Gait velocity (cmkec)

66.33-88.0 1.52-1.80 24.26-38.64 85.31-137.8

74.53

(6.83)

1.64 (.I) 30.81 (4.48) 107.9

(16.48)

27.67-81.67 1.52-l .80 9.35-46.85 55.89-125.98

Kinematic II camerasa embedded two force

All subjects could stand, rise from a chair, and walk without assistance.All subjects signed a written informed consent. Instrumentation The system used to measure the variables (eg, CG, CP, and moment arm) during the tasks of FR and gait is described in detail elsewhere.23*25Briefly, whole body kinematic and kinetic data were collected using arrays on both feet, shanks, thighs, arms, pelvis, trunk, and head and force plates (fig 1). The sampling rate for all data was 150Hz. This measurement technique, and the automatic error detection routines within the software, produced accuracies of < 1.Omm and
Categorized

VH Group (SD)

Fig 1. Full body data acquisition system, sag&al view. and kinetic data were collected based on four SELSPOT (detected the activity of light emitting diodes that were in each of the arrays attached to 11 body segments) and plates, respectively.

58.97

(18.49)

1.67 (.I) 29.76 (10.54) 97.69

(23.96)

Procedure FR. Barefoot subjects stood with one foot on each force plate, with feet parallel and a comfortable distance apart. The subjects flexed the shoulder (of the dominant arm), with elbow fully extended, to 90”. Verbal directions. without demonstration, were: “Reach forward as far as you can, doing whatever you wish except taking a step. Begin reaching when I say go.” The tester said, “1, 2, ready, go.” Data collection began when the tester said “ready” and continued for 5 seconds. Subjects did one practice trial and two test trials. The starting position of the subject and the verbal instructions were similar to those Arch

Phys

Med

Rehabil

Vol 80, March

1999

264

FUNCTIONAL

REACH,

used by Duncan and colleagues2; however, there was no yardstick affixed to the wall. A trial was discarded if the subject reached down (eg, the hand moved down toward the ground), whereby the arm was not in the horizontal plane parallel to the floor. Gait. Subjects were asked to “walk at your normal pace as if you are taking a brisk walk in the park.” Subjects did 2 walking trials, 3 seconds each, with at least 3 strides completed before data were collected. Data Analysis FR distance was defined as the horizontal translational displacement of the dominant arm between the beginning and end of FR. The beginning of FR was defined as occurring at .02 second (just after the tester said “ready”) and the end of FR at 4.98 seconds, to avoid initial and end condition artifacts. The highest FR value of the two FR trials was used for data analysis. Whole body AP and ML CG displacement was calculated between the beginning and end of FR. The total range of CG displacement, which may have occurred any time during FR, was also calculated. Whole-body AP CP displacement was calculated similarly to the CG variable. The CG-CP separation, or whole body moment arm, at the end of FR and the maximum value of the moment arm during FR were analyzed. Accuracy of the CG estimate is about lcm.23 Movement strategy during FR was operationally defined by sagittal hip, knee, and ankle kinematics based on the crosscorrelation between sagittal hip and ankle position at the end of FR. A negative cross-correlation between hip and ankle motion (eg, hip flexion and ankle plantarflexion) with a minimum of 20” of hip flexion and 5” of ankle plantarflexion, bilaterally, was operationally defined as a volitional “hip strategy.” Crosscorrelations greater than -0.1 were classified as “other strategy.” These other strategies included: (1) ankle plantar flexion with less than 20” hip flexion and greater than 5” plantarflexion; (2) trunk rotation in the transverse plane; and (3) hip, knee, and ankle flexion similar to a squat. The highest average preferred gait velocity of two walking trials was used for data analysis. Gait velocity was defined by the average change in CG displacement over time. The maximum whole-body moment arm during gait was also calculated.23 Descriptive statistics characterized the sample and repeated measures analysis of variance (ANOVA) compared the difference between FR distance trials 1 and 2. An unpaired t test compared the difference in FR distance and postural control measures between healthy elder and VH groups, and between hip strategy and other strategies. The association between FR score and other variables was tested using Pearson’scorrelation coefficients. Partial correlations were used to account for the effect of the subject’s age, height, or group on these associations. RESULTS FR Distance and Subject Characteristics FR distance for all subjects ranged from 9.35 to 46.85cm (mean = 30.25cm). The mean FR distance for the healthy elder and VH groups were 30.81 2 4.48cm and 29.76 t l0..54cm, respectively (table 1). There was no difference in mean FR distance (p > 0.7) between healthy elder and VH subjects. There was no association between subject age and FR distance. A moderate correlation between height and FR distance was found in the VH group (Y = .63, p 5 .05), but there was not a significant correlation of height and FR in the healthy elders. Arch

Phys

Med

Rehabil

Vol 80, March

1999

Wernick-Robinson

Table

2: Mean

Control

Healthy

Measures of FR (cm)

Postural

Elderly

Range

MAFR MAMAX BOS

Control

Measure

Values

Group

VH Group

Mean (SD)

Mean (SD)

.70-5.50

3.13

(1.44)

.69-6.75

2.93-6.21 17.07-33.31

4.32 24.52

(1.25) (5.14)

1.72-8.24

2.65

4.39 (1.87)

(1.74)

16.34-42.36 .34-9.27 1.18-9.86

26.86 (7.34) 4.36 (3.18) 5.75 (2.98)

APCG APCGMAX

.43-8.89 2.19-8.89

4.02 (2.88) 4.60 (2.03)

MLCG APCP

.41-4.98 1.15-11.85

2.43 (1.52) 6.35 (3.42)

.03-5.48 .66-13.89

2.42 6.23

(1.95) (4.23)

APCPMAX

3.42-12.5

7.87 (3.14)

2.89-16.64

10.15

(3.61)

Abbreviations: MAFR. moment arm at end of maximum moment arm during FR; BOS, distance between CG of feet); APCG, AP CG FR; APCGMAX, maximal AP CG displacement CG displacement at end of FR; APCP, AP CP FR; APCPMAX, maximal AP CP displacement

FR (4.98seck MAMAX. base of support (ML displacement at end of during FR; MLCG, ML displacement at end of during FR.

There was not a significant correlation between age and FR in either the healthy elder group (Y = -.28) or VH group (r = -.23). FR Distance and Postural Control Measures There was no difference in the postural control measures between healthy elders and VH subjects (table 2). The mean maximum moment arm values generated during FR for the healthy elder and VH subjects were 4.32 5 1.25cm and 4.39 + 1.87cm, respectively. In the VH group, FR moderately correlated with maximum moment arm during FR and the moment arm at the end of FR (Y = .69 and Y = .70, respectively); however, there were no significant correlations of the same measures in the healthy elder group (table 3). Partial correlations controlling for the effect of the subject’s height, age, or group did not change the strength of the correlations in either group (table 4). Both groups had a moderate correlation between FR and whole body AP CG displacement during FR, but ML CG displacement did not correlate significantly. Movement Strategies Used During FR To identify if a particular movement strategy was adopted or avoided by people with vestibular dysfunction, FR was classified by strategy (table 5). The mean FR of the 21 subjects who used a volitional hip strategy was 29.47 F 8.04cm, compared with 32.6 % 8.75cm for the 7 subjects who used other strategies. No single characteristic (eg, age or group) described Table

3: Pearson’s Correlation Coefficients With Postural Control Measures

Postural MAFR

Control

Measures

of FR

(cm)

MAMAX (cm) APCG (cm) APCGMAX (cm) MLCG (cm) APCP (cm) APCPMAX (cm) Gait Velocity (cmkec)

-.3

of FR Distance of FR

HE

VH

.3 (9) .36 (9) .74t (9) .47 (9)

.70* (13) .69+ (12)

(9) .76* (9)

.03 (12) .84* (IO) .76* (IO)

.70+ (9) .08 (13)

.72* .70*

(13) (13)

.39 (15)

Number of subjects used in correlations are shown in parentheses. Abbreviations: MAFR, moment arm at end of FR (4.98sec); MAMAX, maximum moment arm during FR; APCG, AP CG displacement at end of FR; APCGMAX, maximal AP CG displacement during FR; MLCG, ML CG displacement at end of FR; APCP, AP CP displacement at end of FR; APCPMAX, maximal AP CP displacement during FR. * ps.01. + ps .05.

FUNCTIONAL

Table

4: Partial

Postural

Correlations

Control

Measures

of FR Distance Measures of FR of FR

MAFR MAMAX Abbreviations: MAFR, moment maximum moment arm during * ps .Ol. + ps.05.

arm FR.

With

Postural

REACH,

Control

Age

Height

Group

.59”

.60+

.57”

.61*

.50+

.62*

at end of FR (4.98sec);

MAMAX,

Other Findings A repeated-measures ANOVA revealed high trial-to-trial reliability (no significant difference between the first and second FR trials) for FR distance (p > .99) and for the moment arm values (p > .80). Therefore, the highest FR score of the two trials was analyzed. Because regression analysis showed that height was not a strong predictor of FR distance (u* = .29, p < .Ol), and only one correlation changed after height was partialled out (FR and maximum moment arm [from r = .61 to r = .50]), FR was not normalized to the subject height for data analysis.

5: Descriptive

Characteristics,

Hip Strategy Range

Categorized

by Strategy

(n = 21)

Other Strategy

Mean (SD)

Range

(n = 7) Mean (SD)

Age (yrs + mo) Height(m) FR (cm) Gait velocity (cmkec)

27.67-88.0 1.52-1.80 9.35-41.06 55.89-137.79

6: Mean

69.24 (15.66) 1.64 LOS)

30.58-71.25 1.52-1.80

57.06 (14.98) 1.69(.11)

29.47

20.37-46.85

32.6

105.91-125.98

117.58

(8.04)

97.40 (21.83)

(8.75) (7.78)

Values

MAFR MAMAX

APCPMAX

of Postural Control Grouped by Strategy

Measures

Hip Strategy

Postural Control Measures of FR (cm)

APCGMAX MLCG APCP

Gait Velocity Preferred gait velocity did not correlate significantly with FR distance in the healthy elder or VH groups (table 3).

Table

Table

BOS APCG

the subjects who used other strategies. Eleven subjects with VH (73%) used a hip strategy; five of these 11 subjects (43%) had no measurable vestibular function. There was no significant difference in FR (p = .39) when subjects were divided into the two strategy groups. Table 6 outlines the mean values of the postural control measures of FR for subjects grouped by strategy. There was no significant difference in the postural control measures between the strategy groups. Among the subjects who used other strategies, there were strong correlations between FR and maximum moment arm during FR and between FR and moment arm at the end of FR (u = .95, and r = .97, respectively; p < .05) (table 7). Also; there were significant, strong correlations between FR and the maximum AP CG and CP displacement among the subjects who used other strategies. This relationship was not found in the ML direction. The subject who reached the greatest distance (46.85cm) had no measurable vestibular function. He used a squat strategy (fig 2) with postural control measures above the mean. Another subject chose a trunk rotation strategy but reached only 33.65cm with AP, but not ML, postural control measures below the mean. In comparison, two subjects who used the hip strategy (fig 3) generated approximately the same AP postural control measures but had a difference of 7.41cm in their FR distance (figs 4 and 5).

265

Wernick-Robinson

Range

Other

Mean (SD)

.7-5.5

of FR, When

Range

2.72

(1.31)

.69-6.75

1.72-6.62 16.34-42.36

4.34 25.68

(4.34) (6.58)

2.6-8.24 17.6-33.3

.34-8.89 1.29-8.89

3.67 4.87

(3.68) (2.24)

.77-9.27 1.18-9.86

.03-5.48 1.15-11.85

2.38 6.47

(1.78) (3.22)

.16-4.98 .66-13.89

3.42-12.5

8.77

(2.80)

2.9-16.64

Strategy Mean (SD)

2.84

(2.69)

4.45 (2.60) 25.72 (6.07) 6.07 (3.46) 6.67 (3.67) 2.58 (1.79) 5.61 (5.96) 10.19

(5.91)

Abbreviations: MAFR, moment arm at end of FR (4.98sec); MAMAX, maximum moment arm during FR; BOS, base of support (ML distance between CG of feet); APCG, AP CG displacement at end of FR; APCGMAX, maximal AP CG displacement during FR; MLCG, ML CG displacement at end of FR; APCP, AP CP displacement at end of FR; APCPMAX, maximal AP CP displacement during FR.

DISCUSSION FR Distance: Healthy Elders and Individuals With VH These data suggest that measurement of FR distance does not differentiate healthy elders from individuals with balance impairments; no difference in mean FR distance was found between healthy elders and people with VH. One discerning characteristic between our healthy elder and VH groups is that although the latter group was slightly younger than the healthy elders, the VH group was unstable and sought treatment for their gait and balance and vestibulo-ocular dysfunction. The healthy elder group, in comparison, had no complaints of balance deficits. Vestibular testing was not done in the healthy elder group because these subjects did not have vestibuloocular or vestibulospinal impairments, according to their physicians. The mean FR distance of our healthy elder subjects was similar to the mean FR distance among age-matched elders found by Duncan and colleagues,2 only slightly greater than the mean FR among frail elders4 The mean FR distance of the VH group is similar to the results of Mann and colleagues,27 who examined FR in people with vestibular dysfunction who were between 35 and 84 years old. Although Duncan and colleagues2 found a general trend toward a decreasein FR distance with age in older individuals, Mann and colleagues27 did not find a Table 7: Pearson’s Correlation Coefficients With Postural Control Measures, Grouped Postural Control Measures of FR (cm)

Hip Strategy

of FR Distance by Strategy Other

Strategy

MAFR MAMAX

.43 .48+

.97+ .95:

APCG APCGMAX

.60" .45

.81 .89*

MLCG APCP

.06 .74"

APCPMAX

.47

p.31 .so .96*

Abbreviations: MAFR, moment arm at end of FR (4.98sec); MAMAX, maximum moment arm during FR; BOS, base of support (ML distance between CG of feet); APCG, AP CG displacement at end of FR; APCGMAX, maximal AP CG displacement during FR; MLCG, ML CG displacement at end of FR; APCP, AP CP displacement at end of FR; APCPMAX, maximal AP CP displacement during FR. * ps.01. + p5 .05.

Arch

Phys

Med

Rehabil

Vol 80, March

1999

266

FUNCTIONAL

REACH,

Wernick-Robinson

Fig 2. Three-dimensional whole-body schematic representation (sagittal view) of a subject performing an “other strategy,” squatting, during FR. The “+‘I sign represents the whole-body center of gravity. The dark vet-ticaliine represents the vertical ground reaction force. The point where the vertical ground reaction force is applied to the ground is the CP The left side of the body is depicted by the solid line; the right side is the dashed line.

significant relationship between age and FR distance in people with vestibular dysfunction. Conflicting reports appear in the literature about the effects of aging on the vestibular system and the time at which changes might begin to occur.28-32For example, Peterka and coworkers30 reported that among 7- to 81-year-old persons there was no association between vestibular ocular reflex, measured by caloric testing, and age. Wall and associates,2ghowever, reported that among subjects 20 to 59 years old, the vestibular ocular reflex gain decreases with age. Baloh and colleagues31 compared normal subjects over 75 years old to young adults and reported a decrease in vestibular ocular reflex gain with age. Baloh and colleagues31 concluded that this decline in vestibular function may be a contributing factor to the common complaints of dizziness and disequilibrium in the elderly, yet their study described their subjects as “normal.” It was unclear what qualified the subjects as “normal.” Our healthy elders had no complaints of disequilibrium and were screened by a physician for neurologic pathology. Further study should address the comparison of FR distance of age-matched healthy individuals and individuals with VH. This analysis would further explain how useful FR distance is in differentiating those individuals with and without balance impairments. An alternative explanation for not finding a difference in FR

time

Moment

B

O

’

time

Functional

(5)

Arm

(sf

4

5

Reach

4ol------7

Fig 4. lime-history graphs of a healthy elderly subject’s postural control measures and arm displacement during FR, using a hip strategy. (A) AP CG (solid line) and CP (dashed line) displacement. (B) Whole body moment arm. At the start of FR (time .02sec) the subject is in upright posture, the CG is over the CP, and the moment arm is approximately 0. During FR, the CG and CP diverge and a moment arm is created to maintain dynamic stability. This subject generated a maximum moment arm of 3.02cm during FR and then reduced the moment arm to 0.7cm at the end of FR. (C) Right arm displacement. FR = 33.65cm.

Fig 3. Three-dimensional whole-body schematic subject with VH performing a hip strategy kinematics (hip flexion of ~20” and plantarflexion hip strategy. Symbols are as in figure 2.

Arch

Phys

Med

Rehabil

Vol 80, March

1999

representation during FR. The of ~5~) define

of a hip the

distance between the two groups may be that the severity of the vestibular dysfunction was not measured. There is no conclusive test that measures the severity of the vestibular dysfunction. Mann and colleaguesz7 examined FR distance and the Dizziness Handicap Inventory (DHI) score in people with VH and found a significant, moderate correlation of Y = .65, but

FUNCTIONAL

-4v

A0

REACH,

5 time

Moment

Functional

(s)

Arm

Reach

50-

Fig 5. Time-history graphs of a VH subject’s postural control measures and arm displacement during FR, using a hip strategy. (A) AP CG (solid line) and CP (dashed line) displacement. The AP CG and CP maximum displacement between this subject and the subject in figure 4 differed by 2.16cm and 0.6cm, respectively. (B) Whole body moment arm. This subject generated a maximum moment arm of 5.05cm during FR and then reduced the moment arm to 3.3cm at the end of FR. (C) Right arm displacement. FR = 41.06cm.

only when the DHI score was less than 50/100. Additionally, there was no relation between age and FR distance.27 FR Distance: Postural Control Measures The FR association with CP excursion during FR (r = .76) among the healthy elder group was similar to that found by Duncan and colleagues,* who reported a correlation of r = .7 1. In upright posture the CG is, on average, over the CP and the moment arm is approximately 0. To move the CG forward

267

Wernick-Robinson

during FR, a person must exert dynamic control by creating a moment arm to shift the CP behind the CG.22Therefore, only an FR strategy that produced a large moment arm would show dynamic postural control. Although we hypothesized that moment arm and FR distance would be associated, we found that some movement strategies did not increase the moment arm but produced the mean FR distance. Therefore, the type of movement strategy used during FR may be the more important component of the test than the distance of arm displacement if FR is to be used as a measure of dynamic postural control. There was no significant correlation between FR distance and moment arm values among the healthy elder group, suggesting that FR was not a dynamic postural control challenge in this group of healthy elders. We found no significant difference in postural control measures between the healthy elder and VH groups (p > .5), suggesting that FR distance alone cannot distinguish dynamic balance challenges from the other factors that may limit FR score. These factors include trunk and lower extremity range of motion and strength, willingness to risk loss of balance, or fear of falling. Successful treatment for people with impaired postural control depends on an understanding of the postural systems that are likely to be disordered by aging and pathology.33 There was greater age variability in the VH group that may contribute to the strong correlation of FR and moment arm seen in the VH group compared with the healthy elder group, because based on force plate data alone, younger subjects are more likely to approach their limits of stability.2,34,35Blaszczyk and colleagues35 reported that during leaning tasks, elderly subjects had a slower speed and smaller maximum voluntary excursion in the backward, but not forward, direction compared with young subjects. Therefore, the ability to return to upright from the end functional reach position or a backward reach should also be assessed.Furthermore, the decreased speed with which elders moved to their maximum position35 implies that the time to get to the end FR position could also be used as a measurement of postural control. Movement Strategies Used During FR Although FR is a potentially useful measurement tool that can place a patient in a position in which dynamic postural control is needed, the data show that some strategies (eg, hip strategy with only 5” of plantarflexion) used during FR were more static, rather than dynamic, defined by a low moment arm. If elderly subjects are not likely to approach their limits of stability, then they may develop compensatory movement strategies that are more static to reach forward during FR and that do not increase the moment arm. Our results show that subjects who used stable strategies could actually reach the average distance forward. The subject in figure 4 reached forward by flexing his hips, which is seemingly similar kinematics to a hip strategy but with a low moment arm particularly at the end of the FR. Trunk rotation during FR also generated a low moment arm. Therefore, compensatory strategies used by some of the subjects produced static, not dynamic, challenges to postural control. The correlation between FR strategy and moment arm among the subjects who used other strategies should be interpreted with caution because of the small number (n = 7) of people who used other strategies. Despite this small number, the strong correlation between FR and moment arm values suggests the need to describe the strategy used in FR as an important component for the test to be used as a measure of postural control. The nonspecific measure of arm displacement during FR does not help identify the source of the postural control deficit and, therefore, gives no new information about the cause Arch

Phys

Med

Rehabil

Vol 80, March

1999

268

FUNCTIONAL

REACH,

of the instability. Horak and coworkers33 recommend that to fully characterize a patient’s postural, clinicians should consider the mechanisms related to equilibrium control in anticipation of and during movement, locomotion, and changes in posture. To assess volitionally selected movement strategies, FR was not demonstrated to the subject. The choice of hip strategy during FR among the VH subjects supports previous research showing the use of hip strategies to control CG among people with diminished vestibular function15,1gT20 and with bilateral vestibular loss. l8 By contrast, Horak and coworkers33 showed people with complete bilateral vestibular loss did not use a hip strategy, suggesting that vestibular information is necessary in controlling equilibrium in a task that requires a hip strategy. By contrast, five of our subjects with no measurable vestibular function did employ a hip strategy. Documentation that a person with vestibular hypofunction was or was not able to employ a hip strategy provides information that may assist movement strategy training during volitional feedforward activities.33 FR and Gait Velocity The correlation between FR distance and free gait speed was poor, suggesting FR may not be a valid indicator of the dynamic stability needed during gait. The poor correlation (r = .08, p = .79) between FR and gait velocity in healthy elder subjects was similar to the results of Thapa and colleagues5 (Y = .35), but different from the correlation reported by Weiner and associates4 (r = .67). Moment arm during gaitz4 is approximately six times larger than during FR, suggesting that gait is more dynamically challenging than FR. The majority of falls in the elderly occur during locomotioni yet nonlocomotor clinical balance tests are used to predict falls.3,36 Recently, Cho and Kamen36 examined the ability of clinical balance measures to identify fallers from nonfallers. No significant correlation between the two groups was found using FR; however, gait velocity was significantly faster for normal older subjects than for the fallers. Further research is needed with a larger sample of subjects to evaluate moment arm during gait and other known dynamic measures of postural control to identify the magnitude of moment arm needed for a measure to be considered dynamic. Although our data are the first to examine the biomechanics by which FR might stress postural control, one limitation of our study was that the healthy elders were not perfectly agematched to the VH group. Further studies should compare the biomechanics of FR in young people with vestibular dysfunction to age-matched healthy controls and should compare FR in elders with confirmed vestibular dysfunction and symptoms of dizzines.s and dysequilibrium to age-matched healthy elders to identify if FR is a valid measure of dynamic balance. Continued research is also suggested to further analyze other biomechanical factors such as the velocity, ground reaction shear forces, and electromyogram during FR. CONCLUSION Our results suggest that FR distance (arm displacement) alone does not measure dynamic balance. An indispensable component of assessing postural control is the evaluation of movement strategies used to accomplish the functional task. Documenting the presence of a hip, ankle, or other strategy during FR would more qualitatively describe the reaching task. Determining movement strategy may also assist the clinician in assessing underlying impairments that may contribute to functional limitations. We suggest that in addition to measuring the forward displacement of the arm the movement strategy should Arch

Phys

Med

Rehabil

Vol 80, March

1999

Wernick-Robinson

be assessed if FR is used as one component of postural control. Other impairments such as decreased flexibility and strength and fear of falling may also contribute to FR performance. As hypothesized, FR distance strongly correlated with maximum moment arm, although the correlations were not as strong when the subjects were classified by movement strategy. If FR is to be used as a measure of dynamic balance, we recommend further research that includes comparing FR to other dynamic functional activities to determine if FR relates to dynamic components of postural control. Acknowledgments: We thank the Massachusetts General Hospital Biomotion Laboratory staff: Drs. Jose Ramirez, Simone Bortolami, and Chris McGibbon; and Kathy Price, Rita Popat, Niyom Luepongsak, and Dan Sommers for assistance with data collection. References 1. Tinetti ME, Douchette JT, Claus EB. The contribution of predisposing and situational risk factors to serious fall injuries. J Am Geriatric Sot 1995;43:1207-13. 2. Duncan PW, Weiner DK, Chandler J, Studenski SA. Functional reach: a new clinical measure of balance? J Gerontol Med Sci 1990;45:M192-7. 3. Duncan PW, Studenski S, Chandler J, Prescott B. Functional reach: predictive validity in a sample of elderly male veterans. J Gerontol Med Sci 1992;47:M93-7. 4. Weiner DK, Duncan PW, Chandler J, Studenski SA. Functional reach: a marker of physical frailty. .I Am Geriatr Sot 1992;40: 203-7. 5. Thapa PB, Gideon P, Fought RL, Konnicki M, Ray WA. Comparison of clinical and biomechanical measures of balance and mobility in elderly nursing home residents. J Am Geriatr Sot 1994;42:493-500. 6. Gill-Body KM, Krebs DE. Locomotor stability problems associated with vestibulopathy: assessment and treatment. Phys Ther Pratt 1994;4:232-45. I. Maki BE, Holliday PJ, Topper AK. A prospective study of postural balance and risk of falling in an ambulatory and independent elderly population. J Gerontol Med Sci 1994;2:M72-84. 8. Potter JM, Evans AL, Duncan G. Gait speed and activities of daily living function in geriatric patients. Arch Phys Med Rehabil 1995;76:997-9. 9. Giorgetti MM, Harris BA, Levenson C, Heislin D, Jette AJ. Reliability of clinical balance measures in the elderly [abstract]. In: Proceedings of the International Congress of the World Confederation for Physical Therapy; June 1995; Washington, DC. Alexandria (VA): American Physical Therapy Association; 1995. Abstract PL-RR-0643-F. 10. Weiner DK, Bongiomi DR, Studenski SA, Duncan PW, Kochersberger GG. Does functional reach improve with rehabilitation? Arch Phys Med Rehab 1993;74:796-9. 11. Nashner LM, McCollum G. The organization of human postural movements: a formal basis and experimental synthesis. Behav Brain Sci 1985;8:135-71. 12. Horak FB, Nashner LM. Central programming of postural movements: adaptation to altered support-surface configurations. J Neurophysiol 1986;15:1369-81. 13. Horak FB. Clinical measurement of postural control in adults. Phys Ther 1987;67:1881-5. 14. Shumway-Cook A. Critical analysis of measurement in balance: a clinical approach. In: Proceedings of the 16th Annual Eugene Michels Researchers’ Forum, American Physical Therapy Association, Combined Sections Meeting; Feb 1996; Atlanta, GA. Alexandria (VA): American Physical Therapy Association; 1996. p. 5-8. 15. Black FO. Shunert CL. Horak PB. Nashner LM. Abnormal postural ccntrol’associatkd with peripheral vestibular disorders. Progr Brain Res 1988;76:263-74. 16. Diener HC, Horak FB, Nashner LM. Influence of stimulus parameters on human postural responses. J Neurophysiol 1988;59: 1888-905.

FUNCTIONAL

REACH,

17. Horak FB, Nashner LM, Deiner HC. Postural strategies associated with somatosensory and vestibular loss. Exp Brain Res 1990;82: 167-77. 18. Herdman SJ, Sandusky AL, Hain TC, Zee DS, Tusa R. Characteristics of postural stability in patients with aminoglycoside toxicity. J Vestib Res 1994;4:7 l-80. 19. Shumway-Cook A, Horak FB. Rehabilitation strategies for patients with vestibular deficits. Diagn Neurotol 1990;8:441-57. 20. Shupert CL, Horak FB, Black FO. Hip sway associated with vestibulopathy. J Vestib Res 1994;4:231-44. 21. Schenkman M. Interrelationship of neurological and mechanical factors in balance control. In: Duncan PW, editor. Balance. Proceedings of the APTA Forum; 13-l 5 June 1989; Nashville, TN. Alexandria (VA): American Physical Therapy Association; 1990. p. 29-41. 22. Kirkpatrick R, Tucker C, Ramirez J, Parker SW, Gill KM, Riley PO, et al. Center of gravity control in normal and vestibulopathic gait. In: Woollacott M, Horak FB, editors. Posture and gait: control mechanisms, Vol. 1, XI International Symposium of-the Society for Postural and Gait Research. Portland (OR):, Universitv of Oregon Books; 1992. p. 260-3. 23 Riley PO, Mann RW, Hodge WA. Modeling of the biomechanics of posture and balance. J Biomech 1990;23:503-6. 24. Krebs DE, Gill-Body KM, Riley PO, Parker SW. Double blind, placebo-controlled trial of rehabilitation for bilateral vestibular vestibular rehabilitation: preliminary report. Otolaryngol Head Neck Surg 1993;109:735-41, 25. Benda BJ, Riley PO, Krebs DE. Biomechanical relationship between center of gravity and center of pressure during standing. IEEE Trans Rehaba Eng-1994;2:3-9. 26. Antonsson EK, Mann RW. Automatic 6-DOF kinematic trajectory acquisition analysis. ASMEJ Sys Meas Control 1989;111:3 1-9.

269

Wernick-Robinson

27. Mann GC, Whitney SL, Refer-n MS, Borello-France DF, Furman JM. Functional reach and single leg stance in patients with peripheral vestibular disorders. J Vestib Res 1996;6:343-53. 28. Mulch G, Petermann W. Influence of age on results of vestibular function tests. Ann Otol Rhino1 Laryngol 1979;88(2 Pt 2 Suppl 56):1-17. 29. Wall C 3d, Black FO, Hunt AE. Effects of age, sex and stimulus parameters upon vesitibulo-ocular responses to sinusoidal rotation. Acta Otolarvngol (Stockh) 1984:98:270-8. 30. Peterka RJ: Black FO, Schoenhoff MB. Age related changes inhuman vestibulo-ocular refelexes: sinusoidal rotation and caloric tests. J Vestib Res 1990;1:49-59. 3 I. Baloh RW, Jacobson KM, Socotch TM. The effect of aging on visual-vestibuloocular responses. Exp Brain Res 1993;95:509-16. 32. Paige GD. Sensescence of human visual-vestibular interactions: smooth pursuit, optokinetic, and vestibular control of eye movements with aging. Exp Brain Res 1994;98:255-72. 33. Horak FB, Henry SM, Shumway-Cook A. Postural perturbations: new insights for treatment of balance disorders. Phys Ther 1997;77:517-33. 34. Murray MP, Seireg AA, Sepic SB. Normal postural stability and steadiness: quantitative assessment. J Bone Joint Surg 1975;57A: 510-15. 35. Blaszczyk JW, Lowe DL. Hansen PD. Ranges of postural instability and their changes in the elderly. Gait Posture 1994;2: 10-7. 36. Cho CY, Kamen G. Detecting balance deficits in frequent fallers using clinical and quantitative evaluation tools. J Am Geriatr Sot 1998:46:426-30. Supplier a. SELSPOT 11camera; Selcom, Partille, Sweden.

Arch

Phys

Med

Rehabil

Vol 80, March

1999