Weirich, Grant MS; Bemben, Debra A. PhD; Bemben, Michael G. PhD
Falls are a major public health concern for many individuals, and the risk of falling increases with age.1 Falls occur commonly among older people who have no apparent balance deficits.2 It is estimated that 35% of all community dwellers aged 65 or older fall at least once in a 1-year period,3–5 and estimates even reach as high as 60%.6,7 Often, 10% to 20% of falls require hospital care, and the resultant estimated costs are more than $20 billion per year.8 In fact, almost 13 000 older adults died as the result of a fall in the United States in 2002. By the year 2030, it is estimated that 1 in 5 people living in the United States will be 65 years of age or older; therefore, successful aging and the maintenance of health and independence into old age is becoming increasingly important.9,10
Standing balance is essential for performing the activities of daily living and being able to live independently; however, as people age there is a general deterioration in a number of musculoskeletal and sensory systems that can affect postural control and balance.11 Aging is associated with decreased lower extremity strength and flexibility as well as increased postural sway. In addition, ankle, knee, and hip strength declines by 3% per year beyond the age of 40 years,12 and decreased lower extremity strength has been identified as an independent predictor of poor mobility13 and self-reported disability in community-dwelling older women.14 These physiological changes that occur with increased age could impair the ability to respond quickly and forcefully once a fall begins.12 This is a critical factor for trying to re-establish balance, because 53% of falls among the older adults are due to tripping.12
The range of motion in the lower extremities of older adults has been shown to decline by more than 50% when compared with younger adults,15 and ankle flexibility has accurately predicted walking ability and fall status, especially for individuals between the ages of 50 and 97 years when comparing individuals with history of multiple falls with individuals with no history of falls.16
Strength is related to muscle size; therefore, losses of muscle mass that generally accompany aging result in the loss of strength,17 and studies have reported that age-related increased body fat and low muscle mass are associated with mobility-related disability.10 Both major components of body composition, lean mass and fat mass, have been associated with bone health. Bone mineral density (BMD) is an important physiological determinant of bone strength and being able to maintain adequate BMD is essential because 90% of all hip fractures are the result of a fall and compromised bone health.18 In addition, studies have revealed that femoral neck BMD, quadriceps strength, and body sway were all significant predictors of fractures in women.19
Being able to measure potential factors associated with postural stability and balance, such as lean body mass, lower limb muscle strength, BMD, and lower extremity flexibility, can be beneficial for developing appropriate exercise protocols in an attempt to maintain or improve health and fitness throughout the aging process. Improved physiological function may then help prevent falls and lower the risk of potential injures that can cause functional limitations or even permanent disabilities, especially with advanced age. Therefore, the purpose of this study was to examine the relationships between strength, flexibility, body composition, BMD, and balance in young, middle-aged, and late middle-aged women and to determine whether there are similar physiological predictor variables for balance and postural stability as age increases.
Subjects were actively recruited from the Norman and Oklahoma City communities by way of fliers that were posted in churches, activity(community centers, and school campuses, as well as radio and television announcements on community channels. Eighty-five subjects between the ages of 18 and 64 years eventually volunteered for this study and were categorized into 1 of 3 age groups: young (n = 30, 18-25 years), middle-aged (n = 26, 35-45 years), and late middle-aged (n = 29, 55-64 years). Participants were active and healthy with no medical conditions that would prevent them from participating and each subject read and signed an informed consent form that was approved by the University Institutional Review Board. Participants were required to fill out a health status questionnaire and the Physical Activity Scale for the Elderly (PASE)20 questionnaire in order to document that they were only recreationally active and free from any medical issues.
This cross-sectional study required each subject to be assessed for maximal muscular strength for the lower extremities (knee extension, leg press, and knee flexion), lower extremity flexibility (plantar flexion, dorsiflexion, hip flexion, and sit and reach), hip, leg, lumbar spine, and total BMD and bone mineral content (BMC), total and regional body composition (lean body mass and fat mass), and balance and postural stability (Table 1).
Following completion of the questionnaires and signing of the informed consent forms, subjects were required to attend 2 days of testing, separated by 1 day of rest. During the first visit, subjects were first assessed for lower extremity flexibility. This testing took approximately 15 minutes to complete and required minimal exertion. Subjects were then familiarized with the 3 exercise machines (Cybex [Cybex International, Inc, Medway, Massachusetts]) that would be used to assess muscular strength. Once the subjects were comfortable with the equipment and a short warm-up session was completed, strength was assessed for each exercise in random order. Each separate test required about 10 minutes to complete, and subjects had a 5-minute rest period between tests. On the second day of testing, subjects first had their bone densities and body composition assessed by dual-energy x-ray absorptiometry. Subjects were required to lie still for about 15 minutes for this test. The last assessment was to obtain measures of balance and postural stability. This testing required about 45 minutes to complete.
Muscular strength was assessed using Cybex isotonic weight-training equipment, and flexibility was measured by the sit-and-reach test and a Leighton Flexometer (Leighton Flexometer, Inc, Spokane, Washington). Bone mineral density, BMC, and measures of body composition were obtained by dual- energy x-ray absorptiometry, Prodigy enCORE software version 10.50.086 (DXA [GE Lunar Prodigy, Madison, Wisconsin]), and balance and postural stability were assessed by using the NeuroCom (NeuroCom International, Inc, Clackamas, Oregon).
Isotonic muscular strength/1 repetition maximum assessment
1 repetition maximum (1RM) testing required an initial warm-up session of 5 minutes of cycling on a stationary bicycle at 1 kp resistance. The standardized 1RM testing procedure has been shown to be both a reliable and valid measure of maximal strength.21 Test retest reliability values in our laboratory for 1RM measures are between r = 0.95 and r = 0.99. Subjects were positioned at 1 of the 3 testing stations (seated knee extension, seated knee flexion, or a semireclined 2 leg press) in random order and were instructed on proper body position and lifting technique. They then warmed up with a very light weight (2-3 plates or 11.5-17 kg), completing 10 repetitions. Following a 1-minute rest, the load was then subjectively increased based on the perceived exertion of the subjects and then the subjects performed the repetition. If the lift was successful, the weight was increased again, depending on the perceived effort of the previous attempt, and another attempt was made following a 1-minute rest period. Maximal strength was achieved in 5 attempts for each of the lower body exercises. There was a 5-minute rest period between each exercise.
The sit-and-reach test, a well-established measure of lower back and lower leg flexibility, was used to measure lower back and hamstring flexibility.22 Subjects were seated on the floor with legs extended in front of the body and the feet (with shoes removed) placed with soles flush against the flexibility box. Subjects then leaned forward as far as possible, pushing the measuring gauge away from their bodies with tips of fingers. The subjects performed 3 separate trials, and the intraclass correlation coefficient for the trails was r = 0.98 (P = .001).
Flexibility was also measured with a Leighton Flexometer. This goniometer has a 360-degree dial and a weighted indicator that operate freely and independently, allowing measurement of range of motion.23 For hip flexion, the flexometer was strapped to the right midthigh with the dial on the outside of the thigh while the subjects were in a supine position. The subjects raised their right leg upward in a straight leg raise with an extended knee. The subjects' left leg was stabilized by the tester on the table while the test was being performed. Three trials were obtained. For assessment of ankle dorsiflexion and plantar flexion, subjects sat on a table in a long sitting position, with the ankles and feet off the edge of the table. The flexometer was placed on the right foot while the foot was in a neutral position, with the sole of the foot at a 90% angle to the table. The subjects actively moved the foot into as much dorsiflexion as possible and then pointed the toes away from the body for measurement of plantar flexion. Once again, 3 trials were obtained. The intraclass correlation for the trails was r = 0.68 (P = .007).
Body composition and bone mineral density and content
Dual-energy x-ray absorptiometry was used to measure the BMD (g/cm2) and BMC (g) of the total body as well as all body composition parameters. Quality assurance testing using a scanning calibration block and a phantom spine block of known densities was performed each day to ensure that the DXA was operating properly. For the total body scan, subjects were placed in a supine position, with arms close to their sides and legs fastened together by Velcro straps (Velcro USA Inc, Manchester, New Hampshire).
Balance and postural stability
Various measurement tools have been used to assess balance including the Tinetti Assessment, the Berg Balance Scale, the Timed Up and Go test, and the Dynamic Gait Index. One problem with these tests is that results may be influenced by the consistency and expertise of the rater.24 The ability to use the NeuroCom and the tests that have been developed for this equipment provides a more objective, quantitative result that eliminates problems related to rater subjectivity and increases the accuracy and reliability of the findings.25 To ensure consistency, the standardized instructions for each test were read to the subjects. Subjects also viewed the instructions on a computer screen. All tests began by asking the subjects to place their feet in the designated testing position. All outcome measures were presented as percentages, ratios, or graphs and pictures of movement patterns.
The balance and stability tests were divided into 2 categories: (1) impairment level tests, which utilize visual, vestibular, somatosensory, automatic, and voluntary motor skills that aid in balance and mobility during changing conditions; and (2) functional tests, which examine the ability to safely perform everyday activities. The tests were presented in the same order across all subjects by a trained laboratory technician. Four different tests were conducted; the intraclass correlation coefficients for the different trials ranged from r = 0.50 to r = 0.77 (P = .01 to P = .003).
Impairment level tests
The Modified Clinical Test of Sensory Interaction on Balance (MCTSIB) measured functional balance by quantifying postural sway velocity (SV) while changing the subject's sensory condition, such as eyes closed on a foam surface. The less postural sway after being instructed to “hold still” showed the subjects' ability to remain stable. Three 10-second trials were recorded. The SV was the ratio of the distance traveled by the center of gravity to the time of trial. A low score indicated less sway or movement, reflecting greater postural stability.
Unilateral stance (US) condition measures balance by quantifying postural SV while the subject stood on 1 foot, with eyes opens (EO) or closed (EC), instructed to “hold still” for as long as possible. Low postural sway indicated better stability. Three 10-second trials were recorded for each of the following conditions: EO and EC left foot; EO and EC right foot. The average SV for both EO and EC conditions was calculated by combining the values of left foot and right foot trials.
Tandem Walk (TW) measured postural control as the subject walked the length of the platform as quickly as possible using a heel-to-toe method. End sway velocity (SV, °/s) or the amount of body movement at the end of the task was calculated. The mean of 3 trials was calculated and used in subsequent analyses.
An activity named Step Quick and Turn (SQT) was used to assess dynamic balance by quantifying turn performance after taking 2 steps forward and then quickly pivoting 180°. The measured parameters were turn-time (s) and turn-SV (°/s). The mean of 3 trials was calculated and used in subsequent analyses.
Physical activity assessment
Physical activity was assessed with the PASE questionnaire. This measure included questions about leisure time activities and household activities.20 Test retest reliability assessed over a 3- to 7-week interval has been reported to be 0.75. The validity of the PASE questionnaire has been established by comparison to strength measures, heart rate, and perceived health status scores in various populations.26 Subjects completed the questionnaires independently, asking clarifying questions about items if they had any concerns or uncertainties. Scoring indicated the frequency of each activity using a 4-point response scale (never, seldom, sometimes, or often). Point values were calculated by number of hours per day that were accrued in 1 week for each specific activity performed by the subject. Scores could vary greatly (0-360) across participants, because both type and duration of activity differed among participants.
Data were analyzed using SPSS for windows, version 12.0 (SPSS, Inc, Chicago, Illinois). The data were reported as mean (standard error). One-way repeated measures analysis of variance (ANOVA) was used to evaluate tests with multiple trials. If there were no significant differences between the trials' values, the values were averaged and used in further analyses. A 1-way ANOVA was used to compare the outcome measures across age groups. Bonferroni post hoc tests were used when significant group effects were noted. Pearson correlation coefficients were used to evaluate the relationships between predictor variables and balance measures. Stepwise linear regression was used to predict balance and stability from predictor variables (age, strength, flexibility, and muscle mass) for each age group. Statistical significance was set at a P value of .05 or less.
Subjects did not differ by height or body weight by age group (P = .23 and P = .13, respectively). There were significant group differences (P < .001) for each of the 3 strength measures (Table 2). The youngest subjects (18-25 years) were stronger (P < .001) than the subjects in the late middle-age group (55-64 years) for leg press, knee extension, and knee flexion. Younger subjects were also stronger (P < .001) than the middle-aged subjects (35-45 years) for knee extension. Middle-aged subjects were stronger (P < .001) than the late middle-aged subjects for both knee extension and flexion. There were no differences (P = .06) among groups in physical activity as determined by the PASE questionnaire (Table 2).
The only measure of flexibility that differed by age was hip flexion; the youngest subjects were significantly (P = .02) more flexible than only the middle-aged subjects (Table 3).
There were many differences across groups on measures of body composition (Table 4) and bone health (Table 5). In general, the youngest subjects had lower percentage of fat and fat mass in all regions of the body compared with the middle-aged and oldest subjects (P < .001). Middle-aged subjects had lower percentage of fat compared with the oldest group for both arms and legs (P < .001 for both arms and legs). The youngest group had higher lean body mass than the late middle-aged group as measured at arms (P < .001) and legs (P < .001).
There were a number of differences between the groups for both BMC and BMD (Table 5). Both the young and middle-aged groups had higher BMC for the arms compared with the older group (P = .04). The youngest group had higher total BMC values compared with the oldest group (P = .01). The youngest group had higher BMD than both the 2 older groups for the legs (P < .001 for both groups). The young and middle-aged groups demonstrated higher BMD at the pelvis than did the late-middle-aged group (P = .002 for both groups). The youngest group had a higher BMD at the lumbar spine compared with the oldest group (P = .03). Both young and middle-aged groups had higher total body BMD than the oldest group (P = .02).
Table 6 presents the balance and physical activity data for each age group. As expected, measures of balance were better for the youngest subjects and lowest for the oldest group, although there were only 2 tests (US EO and SQT SV) that resulted in a significant group effect (P < .001 and P = .02, respectively). Both young and middle-aged groups had better balance than the oldest group during US EO (P < .001). The middle-aged group demonstrated better performance on SQT than the oldest group (P = .02).
Pearson correlation coefficients were computed to determine which variables of interest had the strongest relationships to balance and postural stability for each separate age group, as well as for the entire sample. Because greater SV indicates less effective postural control, we anticipated negative correlations would occur with the predictor variables (strength, flexibility, body composition, and bone health). Independent of age group, most measures of balance were only weak or moderately related to predictor variables. These relationships were computed to prepare for the stepwise regression analyses in an attempt to predict balance and postural stability while avoiding issues of multicolinearity by using only 1 measure of a given predictor with the highest correlation to balance.
Prediction equations were developed for each measure of balance for each age group. Subsequently, data were combined into a single regression analysis using age as one of the potential predictors of balance. For the youngest age group, the predictor variables were useful to predict only TW SV (r = 0.63, r2 = 0.40, standard errors of estimates [SEE] = 1.02). The regression equation was as follows:
Equation (Uncited)Image Tools
For the middle-aged group, the physiological measures were able to predict 2 measures of balance: US EC SV (r = 0.50, r2 = 0.25, SEE = 0.30) and SQT Sway (r = 0.52, r2 = 0.27, SEE = 3.93). The regression equations were as follows:
Equation (Uncited)Image Tools
Each measure of the balance (MCTSIB SV EC, US EO SV, US EC SV, TW SV, and SQT End Sway) could be predicted by relationships to the physiological parameters of interest for the oldest age group (Table 7).
As can be seen from the generally low r2 values and the relatively large SEE from the regression equations, only a few of the regression equations provided meaningful predictions of balance that were able to account for a fairly large portion of the variance. These included TW SV for the youngest (R2 = 0.40) and oldest groups (R2 = 0.56) and SQT end sway for the oldest group (R2 = 0.47). When all subjects were pooled together in an attempt to find a common set of possible predictors for balance and postural stability, 4 more equations were generated, but the relatively small R2 (0.11-0.24) provided little predictive power; however, they did suggest that age was an important factor.
It is interesting to note that the number of outcome variables that had a significant relationship to measures of balance and stability increased as age increased in the current study. This is somewhat surprising, because our oldest subjects were only 55 to 64 years of age and were fitter than others of the same age and older samples described in other studies.
In previous work, the travel distance of the center of pressure obtained from a balance platform, which is similar to our measures of SV, is related to the probability of a fall.27 Other reported predictors of fall include the inability to rise from the floor, stair climbing, 1-leg stance, and maximal walking speed.28,29
Performance-based measures of physical capacity, especially measurements of balance and gait impairments, have been reported as strong predictors for the risk of falling for older individuals.28–30 Studies examining healthy adults often report moderately strong positive correlations between strength, power, and functional ability, such as being able to rise from a chair.31,32 Previous studies have also reported that subjects with a history of falls compared with control subjects had severe impairments in both ankle dorsiflexion and plantar flexion strength and flexibility.32 It appears that low muscle strength is strongly associated with poor mobility, independent of age and gender. Lauretani et al13 reported that individuals with deficits in isometric leg extension peak torque had greater difficulties with chair rising, gait speed, stair ascent, and stair descent. In addition, hip extensor strength has been used to predict individuals who fall versus individuals who do not16 and declines in general physical activity levels.33
The strength values for each of the 3 muscle groups assessed decreased with increasing age, as expected, but not at the rate of 3% per year beyond age 40 as often reported in the literature.12 The actual mean strength values for each of the 3 muscle groups and for each age group were also higher than the norms generally reported in the literature. This may be a limitation of the subject selection process because all subjects were volunteers and potentially more interested and probably more fit than their sedentary counterparts. Average PASE values for the oldest subjects in this study are reported to be 112, whereas our group of 55- to 64-year-olds averaged a score of 160.20
Range of motion in the lumbar spine region has been shown to decrease by 25% to 50% between the ages of 20 and 80 years.15 Similarly, ankle dorsiflexion decreases with increased age and has been shown to be an accurate predictor of fall status, especially between the ages of 50 and 97 years when comparing individuals with a history of multiple falls with individuals with no history of falls.16 Ankle plantar flexion is related to the support phase and push-off phase during walking and is often used as a potential indicator of gait speed.34 Ankle flexibility is crucial for walking, especially as the surface may change, because approximately 30% of rotation occurs at the ankle joint during level walking.34
In a recent study, Nemmers and Miller35 reported that balance confidence and physical activity in a group of 60- to 96-year-olds was significantly positively correlated, but the relationship was weak (r = 0.18). They also reported that age, physical activity levels, and balance confidence were all significant predictors of balance stability and accounted for about 43% of the variance associated with balance stability.
The overall effectiveness of the outcome variables in the current study to predict balance and postural stability was reasonable (R2 ranging from 0.24 to 0.56). In general, sit and reach flexibility, plantar flexion flexibility, and dorsiflexion flexibility were all involved in various prediction equations for each of the 3 separate age groups. Measures of strength for the knee extensors and leg press were both involved in prediction equations for the middle-aged and oldest subjects. With respect to the different variables used to assess body composition and bone health, most measures of body fat (leg fat, percentage of leg fat, trunk fat, and percentage of trunk fat) were included in significant regression equations while only total body lean mass was included in 1 regression equation for the oldest subjects. Finally, total BMD and trunk bone mineral content were the only 2 measures of bone health that were included in any of the significant prediction equations across all age groups.
Several limitations associated with this study must be addressed. First, the age range for the subjects was 18 to 64 years and was considerably younger than in most studies assessing balance and stability. Readers should note that the intent of this study was not to assess individuals who currently had balance issues but rather to try and determine which measures of strength, flexibility, body composition, and bone health might be used to help predict an individual who might be more susceptible to balance problems.
Participants in this study were all volunteers and, as such, more than likely represent a healthier and more active population than participants who are normally considered to be at risk for balance and stability problems. Even so, a number of parameters were still found to be reasonable predictors of balance and stability. Finally, the use of the PASE questionnaire to assess activity levels might have limited the ability to detect differences in physical activity levels between the 3 age groups because the questionnaire was designed for an older population.
Future research should focus on individuals older than 65 years of age and should also provide some longitudinal follow-up (perhaps 1-2 years) to see if any individuals who might have been predicted to have balance issues actually had any falls.
Similar to many previous studies, most relationships between measures of strength, flexibility, physical activity, body composition, and bone health and measures of balance and postural stability were only weak to moderate and quite variable, even though they may have been statistically significant. However, in general, as age increased in the current study, more variables of strength, flexibility, and body composition were more strongly related to measures of balance. In addition, there were more significant correlations between possible predictors of balance when balance was assessed in a dynamic fashion (TW and SQT) as compared with static measures of postural stability (MCTSIB and US). This may be an obvious finding, but it is a very important point because most falls occur during some form of activity rather than during times of stationary positioning. Finally, it appeared as though there was not a similar set of independent physiological variables that was able to predict balance and postural stability for the 3 different age groups used in this study.
1. Hong Y, Li JH, Robinson PD. Balance control, flexibility, and cardiorespiratory fitness among older Tai Chi practitioners. Br J Sports Med. 2000; 34:29–34.
2. Kannus P, Sieuanen H, Palvanen J, Parkkari J. Prevention of falls and consequent injuries in elderly people. Lancet. 2005; 366:1885–1893.
3. Condron JE, Hill KD. Reliability and validity of a dual-task force platform assessment of balance performance: effect of age, balance impairment, and cognitive task. J Am Geriatr Soc. 2002; 50:157–162.
4. Hornbrook MC, Steven VJ, Wingfield DJ, Hollis JF, Greenlick MR, Org MG. Preventing falls among community-dwelling older persons: results from a randomized trial. Gerontologist- 1994; 34:16–23.
5. Hausdorff JM, Rios DA, Edelber HK. Gait variability and fall risk in community-living older adults: a 1-year prospective study. Arch Phys Med Rehabil. 2001; 82:105–106.
6. Lord SR, Sherrington C, Menz HB. Falls in Older People. Risk Factors and Strategies for Prevention. New York, NY: Cambridge University Press; 2001.
7. Bergland A. Validity of an index of self-reported walking for balance and falls in elderly women. Adv Physiother. 2002; 4:65–73.
8. Englander F, Hodson TJ, Terregrossa RA. Economic dimensions of slip and fall injuries. J Forensic Sci. 1996; 41:733–746.
9. Stevens JA, Corso PS, Finkelstein EA, Miller TR. The cost of fatal and nonfatal falls among older adults. Inj Prev. 2006; 12:290–295.
10. Davison KK, Ford ES, Cogswell ME, Dietz WH. Percentage of body fat and body mass index are associated with mobility limitations in people aged 70 and older from NHANES III. J Am Geriatr Soc. 2002; 50:1802–1809.
11. Isles R, Low Choy NL, Steer M, Nitz JC. Normal values of balance tests in women aged 20-80. J Am Geriatr Soc. 2004; 52:1367–1372.
12. Pavol MJ, Owings TM, Foley KT, Grabiner MD. Influence of lower extremity strength of healthy older adults on the outcome of an induced trip. J Am Geriatr Soc. 2002; 50:256–262.
13. Lauretani F, Russo CR, Bandenelli S, et al. Age-associated changes in skeletal muscles and their effect on mobility: an operational diagnosis of sarcopenia. J Appl Physiol. 2003; 95:1851–1860.
14. Suzuki T, Bean JF, Fielding RA. Muscle power of the ankle predicts functional performance in community-dwelling older women. J Am Geriatr Soc. 2001; 49:1161–1167.
15. Daley MJ, Spinks WL. Exercise mobility and aging. Sports Med. 2000; 29:1–12.
16. Daubney ME, Culham EG. Lower-extremity force and balance performance in adults aged 65 years and older. Physical Ther. 1999; 79:1177–1185.
17. Newman AB, Kupelian V, Visser M, et al. Sarcopenia: alternative definitions and associations with lower extremity function. J Am Geriatr Soc. 2003; 51:1602–1609.
18. Skelton DA, Beyer N. Exercise and injury prevention in older people. Scan J- Med Sci Sports. 2003; 13:77–85.
19. Nguyen T, Sambrook P, Jones G, Lord S, Freund J, Eisman J. Prediction of osteoporotic fractures by postural instability and bone density. BMJ. 1993; 30:1111–1115.
20. Washburn RA, Smith KW, Jette AM, Janney CA. The Physical Activity Scale for the Elderly (PASE): development and evaluation. J Clin Epidemiol. 1993; 46:153–162.
21. Hoeger WWK, Hopkins DR, Barette SL, Hale DF. Relationship between repetitions and selected percentages of one repetition maximum: a comparison between untrained and trained males and females. J Appl Sport Sci Res. 1990; 4:47–54.
22. Johnson B. Practical Flexibility Measurement with the Flexomeasure. Portland, TX: Brown and Littleman; 1977.
23. Verducci F. Measurement Concepts in Physical Education. St. Louis, MO: Mosby; 1980.
24. Berg K. Measuring balance in the elderly: preliminary development of an instrument. Physiother Can. 1989; 41:304–311.
25. Fjeldstad C, Pardo G, Frederiksen C, Bemben D, Bemben M. Assessment of postural balance in multiple sclerosis. Int J MS Care. 2009; 11:1–5.
26. Hagiwara A, Ito N, Sawai K, Kazuma K. Validity and reliability of the Physical Activity Scale for the Elderly (PASE) in Japanese elderly people. Geriatr Gerontol Int. 2008; 8:143–151.
27. Pajala S, Era P, Koskenvuo M, Kaprio J, Tormakangas T, Rantanen T. Force platform balance measures as predictors of indoor and outdoor falls in community-dwelling women aged 63-76 years. J Gerontol. 2008; 63:171–178.
28. Berglund A, Jarnlo G-B, Laake K. Predictors of falls in the elderly by location. Aging Clin Exp Res. 2003; 15:43–50.
29. Berglund A, Wyller TB. Risk factors for serious fall related injury in elderly women living at home. Inj Prev. 2004; 10:308–313.
30. Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among elderly persons living in the community. N Engl J Med. 1988; 319:1701–1707.
31. Skelton DA, Young A, Greig CA, Malbut KE. Effects of resistance training on strength, power, and selected functional abilities of women aged 75 and older. J Am Geriatr Soc. 1995; 43:1081–1087.
32. Whipple RH, Wolfson LI, Amerman PM. The relationship of knee and ankle weakness to falls in nursing home residents: an isokinetic study. J Am Geriatr Soc. 1987; 35:13–20.
33. Visser M, Kritchevsky SB, Goodpaster BH, et al. Leg muscle mass and composition in relation to lower extremity performance in men and women aged 70 to 79: the health, aging and body composition study. J Am Geriatr Soc. 2002; 50:897–904.
34. Vandervoort AA. Ankle mobility and postural stability. Physiother Theory- Pract. 1999; 15:91–103.
35. Nemmers T, Miller J. Factors influencing balance in healthy community-dwelling women age 60 and older. J Geriatr Physical Ther. 2008; 31:93–100.
aging; falls; muscular strength; stability assessment