Loss of muscle mass and muscle weakness are closely associated with functional decline, loss of independence, and mortality in older adults (27), with women typically living longer than men with more disability and functional limitations (11). Preferential atrophy of the faster type II muscle fibers and reduced tendon stiffness, along with slowing of muscle twitch properties and a decrease in maximal motor unit discharge frequency (24,26), may contribute to more substantial reductions in power compared with reductions in strength or endurance with aging (9). Power is defined as the amount of work performed in a given period and, as such, is the product of torque and velocity (1). Cross-sectional studies have demonstrated that functional capabilities such as the ability to get up and down from a chair, climb stairs, and walk quickly are more closely associated with power than strength (2,31) and that limitations in power may be more related to the etiology of falls (28).
Prolongation of response initiation times and reduced rapid torque generation capabilities (9,14) put older adults at risk in situations that demand fast movement responses. In some situations (e.g., to avoid a fall or to brake a vehicle), being able to generate torque quickly may take equal or greater importance to being able to generate large magnitudes of torque (14). Cross-sectional data have demonstrated that older adults at high risk for falling demonstrate slower response times and slower movement times than those at low risk for falling (29), with women exhibiting slower maximum velocities compared with men when recovering balance (35). Slower reaction/movement times have also been shown to be associated with a greater risk of motor vehicle crashes (15).
Measures of ankle strength and power have been shown to be significantly related to chair rise time, stair climb time, gait velocity, balance, and the occurrence of falls in older adults (3,14,28,31,34,36). In an isokinetic study, Whipple et al. (34) found that nursing home residents with a history of falls demonstrated lower strength and power measures about the knee and ankle compared with nonfallers. Specifically, ankle dorsiflexion (DF) power was the most limited in the fallers. Similarly, Skelton et al. (28) also demonstrated reductions in DF strength and in knee and ankle power measures in self-reported fallers. Concentric exercise performed with the intention to move quickly (power training) has been shown to result in improvements in strength, endurance, power, balance, and walking speed in older adults (7,17,19). Whereas most resistance training studies have used large, facility-based resistance training machines, strength gains have been reported in older adults using elastic bands that are much less expensive and more appropriate for home programs (10). The majority of power training studies to date have focused on the larger lower extremity muscles around the hip and knee, with very little emphasis on the distal leg muscles. In addition, the effectiveness of ankle plantarflexion (PF) and DF training on improving movement time has not been previously examined.
The objective of this study was to determine the effects of ankle power training on movement time in a group of mobility-impaired older women. Specifically, we were interested in determining whether movement time changed within each training group. Secondarily, we were interested in measuring changes in PF and DF strength and power in response to training. We hypothesized that movement time would become faster and PF strength and power would improve in the weight-trained group. As well, we anticipated that the smaller DF muscles would achieve a training effect with lower loads; therefore, DF strength and power measures would improve in both the weight and band training groups.
Participants in this controlled trial were randomly assigned in blocks, by a person external to the study, to one of three groups (two resistance training groups and one exercise control group). This meant that, once three subjects had agreed to participate in the study, each was randomly assigned to one of the three groups. This was done to keep the size of the intervention groups similar and to control for the effects of season (the study spanned spring, summer, and fall, and activity levels could potentially differ on the basis of weather conditions, etc.). Before randomization and after 12 wk of training, testing was conducted by an examiner blinded to the study objectives, number of groups, and types of interventions.
The study was advertised in newspapers, newsletters, posters, and television and radio shows. Women, 70 yr or older, with mobility limitation (self-reported inability to walk 1 mile at a moderate pace) who were willing to be randomized and able to attend for 12 wk, were eligible to participate. Exclusion criteria included unstable acute or chronic disease, participation in an exercise program more than once per week in the past 6 months, and neurological or musculoskeletal impairment that would interfere with the ability to participate. Potential participants who responded to advertisements completed a telephone interview that included the Physical Activity Readiness Questionnaire. A physician's signed Physical Activity Readiness Medical Examination form was required when potential contraindications to exercise were identified (e.g., history of cardiac disease, uncontrolled hypertension, hernia, detached retina).
One hundred and ninety-two women responded to study advertisements and underwent telephone screening. Of these initial responders, most were deemed ineligible because they were too young or too active. Seventy-five eligible participants were sent further information, after which 13 participants reported they were no longer interested, leaving 62 participants (32% of initial respondents) who underwent initial testing (Fig. 1). One woman was further excluded because she did not meet all study criteria, leaving 61 participants to be randomized to the three study groups. Fifty of these women completed 12 wk of training. At the initial evaluation session, participants provided their written informed consent. The Education/Nursing Research Ethics Board of the University of Manitoba granted ethical approval for this study.
Participants completed the Physical Activity Scale for the Elderly (PASE) (33) as well as a health/demographic questionnaire. Resting blood pressure, HR, body mass, and active range of motion for the ankle were measured using standard procedures. The Short Physical Performance Battery (SPPB), a test of standing balance, timed walking, and repetitive chair stands, was also performed (8).
Foot reaction and movement time was assessed with the Lafayette Timer, Model 63017 (Lafayette Instruments Co., Lafayette, IN). Participants were instructed to move their right foot as quickly as possible from the right switch ("the gas") to the left switch ("the brake") when a red light was presented. Participants were cued to upcoming trials with a white light. A standardized 1-, 2-, or 3-s delay was provided after the white light to discourage anticipation of the red light. Ten familiarization trials were followed by 10 test trials. Average reaction times and movement times (of the 10 test trials) were used in analyses. Reaction time was the time between onset of the red light and onset of movement from the "gas" switch, and movement time was the time between onset of movement from the "gas" switch to depression of the "brake" switch. The reliability of this test was assessed in a separate test-retest study at a 1-wk interval in our laboratory (n = 30 women, mean age = 73.3 ± 4.7 yr). The intraclass correlation coefficient was 0.90 for movement time, and SEM was 18.3 ms.
Ankle Strength and Power Measures
DF and PF torque, position, and velocity were measured using a Biodex System 3 Pro dynamometer (Biodex Medical Systems, Inc., Shirley, NY). Calibration of the dynamometer was verified each day before testing. A 4-min warm-up was provided on a treadmill before participants were positioned on the Biodex with the right lateral malleolus aligned with the axis of rotation of the dynamometer, the right knee flexed 45°-55°, and the trunk reclined 5° from vertical. Participants were instructed to keep their arms folded across their chest during test trials, and belts were used for stabilization. The end limits of range of motion were set at 10° DF and 30° PF for all tests.
To obtain a measure representative of passive resistive torque about the ankle, the difference between end-range DF and PF torque measures was determined from passive trials (five repetitions, 5°·s−1). There were no changes in passive resistive torque after training (P = 0.59).
Maximal effort isokinetic concentric ankle DF and PF tests were performed at 30°·s−1 and 90°·s−1. DF tests preceded PF tests, and testing at 30°·s−1 preceded that at 90°·s−1. Participants were given three to five submaximal practice trials to familiarize themselves with the concentric movement before five test trials were conducted. A 2-min rest period was provided between test velocities.
Biodex data were collected at a frequency of 100 Hz and exported for analyses in SigmaStat (Version 3.10; Systat Software, Inc., San Jose, CA). Power (W) was calculated as the product of torque (N·m) and velocity (rad·s−1). In our laboratory, the test-retest reliability of DF and PF peak torque and peak power measures in older women yielded intraclass correlation coefficients in the range of 0.86 to 0.97. SEM results were as follows: peak torque, DF 30°·s−1 = 1.2 N·m, PF 30°·s−1 = 6.6 N·m; peak power, DF 90°·s−1 = 0.8 W, PF 90°·s−1 = 8.1 W.
Torque, velocity, and power data were analyzed for the concentric DF and PF tests. At 30°·s−1, peak torque and peak power frequently occurred at the same joint angle (and were therefore redundant), so only peak torques are reported for the slower speed of testing. At 90°·s−1, peak torque frequently occurred before a constant velocity was achieved, whereas peak power usually occurred at or very near 90°·s−1. For this reason, peak power (and not peak torque) values are reported for the faster speed of testing.
Participants were randomly assigned to one of three groups who met for 45-min training sessions, twice per week for 12 wk. All exercise sessions were supervised by kinesiology graduates and/or a licensed physiotherapist. Sessions began and finished with a standardized warm-up and cool-down routine that consisted of a total of 25-30 min of general lower body exercises performed in sitting for all groups (e.g., marching, knee extensions, ankle circles, toe tapping).
Weight training group (weights).
The first three training sessions for participants in the weights group were designed to promote familiarization with the Hammer Strength Tibia Dorsi Flexion and Super Horizontal Calf weight training machines (Life Fitness, Schiller Park, IL). After these sessions, each participant's one-repetition maximum (1RM) load was estimated from a submaximal test (8-10RM test). Thereafter, participants completed the standardized warm-up routine and then performed three sets of 8-10 repetitions of concentric ankle DF and PF against 80% of a 1RM load. DF was performed unilaterally, and PF was performed bilaterally with 2 min of rest between sets. Loads were progressed every 2-3 wk to present a consistent challenge while still maintaining the speed of contraction as assessed visually.
Elastic resistance training group (bands).
The bands group performed three sets of eight repetitions of concentric ankle DF and PF movements against Thera-Band (The Hygenic Corporation, Akron, OH) elastic resistance bands after warming up. Movements were performed unilaterally with 2 min of rest between sets. Participants were progressed to more difficult resistance bands every 2-3 wk.
Participants in both the weights and bands groups were encouraged to perform the concentric portions of the training movements "as fast as possible," whereas the eccentric phases of the movements were completed in a slow and controlled manner for 2-3 s (17).
The placebo control group participants performed static neck stretches and shoulder range of motion exercises and received education about proper head and neck posture between their warm-up and cool-down exercises.
On the basis of preliminary movement time data (the primary outcome variable), sample size calculations were conducted (minimally detectable difference = 10 ms, expected SD of residuals = 15 ms, three groups with α = 0.05, and desired power = 0.8). It was determined that 45 participants were required for the study.
Analyses were conducted using SigmaStat (Version 3.10; Systat Software, Inc.) and SPSS (SPSS 15.0; SPSS, Inc., Chicago, IL) software packages. Descriptive statistics were calculated as mean ± SD for normally distributed variables and as medians and ranges for those that were not normally distributed. Mean differences between variables at baseline were assessed using one-way ANOVA tests (P < 0.05). When normality failed, a one-way Kruskal-Wallis ANOVA on ranks was performed. A series of two-way ANOVA tests with repeated measures were used to investigate between (group) and within (time) factors, with movement time, peak torque, and peak power values as dependent variables (P < 0.05). Preplanned within-group comparisons were conducted to investigate changes in movement time using the Bonferroni adjustment. Because three within-group comparisons were planned, the significance level was set at 0.017 (0.05/3) for these analyses, so that the overall probability of making a type I error was still 0.05 (25).
Eleven of the 61 women who were randomized to the three study groups did not complete 12 wk of training. Also, one woman completed all tests except the movement time test because she could not see the indicator lights because of poor vision. Reasons for not completing training included unexpected move from the province (n = 1), joining a fitness facility (n = 1), sustaining injuries in a fall unrelated to the study (n = 1), increasing time commitments elsewhere (n = 4), and illness related to persistent upper respiratory tract infection (n = 1). In addition, three women discontinued training because they felt it was exacerbating preexisting musculoskeletal conditions. One woman in the weights group reported worsening knee pain in response to the initial testing procedures, and two participants in the control group complained of worsening hip/back pain related to exercise. Of the 50 participants who trained, 36 attended 90%-100% of scheduled sessions, 12 attended 75%-89% of sessions, and 2 attended 63%-67% of sessions. There were no differences in median attendance among the groups (P = 0.88).
Participant characteristics are listed in Table 1. Participants ranged in age from 70 to 88 yr. On average, participants had three chronic medical conditions and had performance scores consistent with moderate mobility limitations. There were no significant differences (P ≥ 0.19) between the training groups at baseline in their descriptive characteristics (Table 1).
Movement time and reaction time.
There were no significant differences in reaction time (P = 0.15) or movement time (P = 0.80) among groups at baseline. Results of the two-way repeated-measures ANOVA are presented in Table 2. Movement time demonstrated a significant main effect for time. A priori within-group comparisons demonstrated a significant decrease in movement time in the bands group (P = 0.003, significance level set at P < 0.017 with Bonferroni adjustment), whereas changes in the weights group (P = 0.03) and control group (P = 0.38) were not significant. Reaction time demonstrated no main effects for group or time. In percent change, movement time decreased 12% in the bands group, 8% in the weights group, and 2% in the control group.
The effect size was calculated using the means and SD from the pre- and postintervention measures (25). The effect sizes for change in movement time were 0.60 in the bands group, 0.28 in the weights group, and 0.17 in the control group (Table 2). According to Cohen, effect sizes of 0.80 represent large changes, effect sizes of 0.50 represent moderate changes, and effect sizes of 0.20 represent small changes (6).
Strength and power.
No significant differences in DF or PF strength or power were noted among groups at baseline (DF peak torque (P = 0.84), PF peak torque (P = 0.98), DF peak power (P = 0.89), and PF peak power (P = 0.84)). Main effects for time were significant for all strength and power variables, but group × time interactions were not significant (Table 3). Percent change and effect size results were similar among groups for DF strength and power; however, changes in PF strength and power were greater in the weights group.
The primary objective of this study was to determine the effects of power training on movement time in mobility-impaired older women. In this study, ankle training with elastic bands resulted in significant improvements in movement time as assessed with the braking task. This is an important finding because slower reaction/movement times have been shown to be associated with increased risk of falling (13), reduced sit-to-stand performance (12), slower speeds ascending and descending stairs (32), and greater risk of motor vehicle crashes (15).
Although many brake response studies have not differentiated between reaction time and movement time in evaluating speed of response, a recent study by Marques et al. (16) demonstrated that patients at 10 and 30 d after total knee arthroplasty had longer movement times (and similar reaction times) compared with their preoperative results. Thirty days after surgery, brake movement times for women were 20 ms slower, and movement times for men were 25 ms slower than preoperative values. In this study, training with elastic bands resulted in an average improvement in movement time of 24 ms, which, although small, exceeded the measurement error for the test and was approximately equal to the decrement in movement time that occurred after surgery in knee arthroplasty patients. This suggests that concentric training with elastic bands performed with the intention to move quickly may play an important role in the rehabilitation of older adults to improve movement time response. Falls and motor vehicle crashes represent the leading injury causes of death in the older population (4,30). Safety may be improved if older adults are able to move and respond more quickly to balance threats and/or dangerous driving situations. Because training with elastic bands is relatively inexpensive and more practical than most other modes of resistance training, these results may have important public health implications.
Although participants in both resistance training groups demonstrated improvements in movement time in this study, only the change in the bands group reached statistical significance. Changes in movement time in the weights group verged on being significant (P = 0.03) but did not meet the modified level for multiple comparisons (P < 0.017 with Bonferroni adjustment). This statistical finding was supported by the effect size data (i.e., effect sizes for change in movement time were small in the weights group (0.28) and moderate in the bands group (0.60)). This was contrary to our original hypothesis that stated that participants in the weights group would demonstrate faster movement times after the exercise intervention. It was not possible to measure the speeds attained during training in this study; however, participants who performed the DF and PF "as fast as possible" against elastic bands conceivably trained at higher velocities and against lower loads compared with those using 80% of 1RM loads with the Hammer Strength equipment. It seems likely that training at higher velocities resulted in greater improvement in the brake movement time task in the bands group and that the higher loads experienced by the weights group may have limited the velocities achieved in training and reduced the carryover effect to the braking task.
Orr et al. (19) found that older adults who performed rapid concentric (and slower eccentric) lower extremity resistance exercises at lower loads (20% of 1RM) demonstrated greater improvements in balance (measured using a computerized force platform) compared with those who trained at higher loads (50% and 80% of 1RM). Changes in peak power capabilities were similar in all three training groups in their study. In addition, attainment of lower peak velocities at 20% and 40% of 1RM during baseline testing independently predicted improvements in balance with training. Increased average velocity at posttest for 40% of 1RM demonstrated a trend (P = 0.065) toward significantly predicting balance improvement. These results seem to suggest that for some aspects of balance, movement velocity may be more important than torque or power generation. In situations where the load is relatively small, being able to move quickly may be the most important factor to determine balance.
Similarly, velocities attained during training may have had the greatest influence on movement time in our study. However, factors responsible for improved movement time in this cohort were not directly determined. Analyses did demonstrate that changes in movement time were not correlated with changes in DF or PF strength or power as measured on the dynamometer. Because the brake response task involved coordinated activation of both DF and PF and may have also included a small degree of hip and/or knee movement, it is not surprising that changes in movement time were not closely related to changes in single-joint, velocity-constrained strength and power tests on the dynamometer.
Orr et al. (19) have suggested that training with fast concentric contractions against low loads may result in enhanced neural function (reduced response latency, increased recruitment of postural muscles, and improved interpretation of sensory information) and improved force control (decreased cocontraction). Moreover, in the upper extremity, rapid concentric contractions against relatively light loads (45% of 1RM) have been associated with greater average muscle activity throughout the concentric action and a shorter deceleration phase at the end of the movement (18). These types of neuromuscular adaptations may have occurred in response to the concentric training with elastic bands and contributed to improvements in movement time, which required coordinated activation of the DF muscles, quickly followed by deceleration of DF and activation of PF.
The PF and DF peak torques measured in this study were, as expected, lower than those previously reported for similarly aged, more active women (23). Contrary to our hypothesis, changes in strength and power were similar among the groups. The absolute and percentage change values for all groups for DF strength and power increased similarly. All participants (including controls) attended up to 24 exercise sessions away from home, which may have resulted in a form of "functional" training (increased overall activity level) for these mobility-impaired individuals. In addition, motor skill training has been shown to result in increased excitability of the motor cortex, especially when the training requires coordination of visual input and motor performance (20,21). Many of the warm-up exercises performed by all participants required a high degree of concentration and coordination (e.g., marching and toe tapping exercises with alternating legs). Regular participation in the warm-up protocol may have resulted in greater excitability of the motor cortex, and given that there are more pronounced corticospinal projections to the tibialis anterior motoneurons than to any other lower extremity muscle group in humans (5,22), this may partially explain why DF strength and power improved in all participants.
For PF strength and power, absolute and percentage increases in the weights group were two to four times greater than changes seen in the control group, but these differences among groups did not reach statistical significance. Similarly, effect sizes for changes in PF strength and power were greater in the weights group (0.35 and 0.38, respectively) compared with the control group (0.15 and 0.07, respectively). Because movement time was the primary variable of interest in this study, sample size calculations were conducted on the basis of pilot movement time data. A much larger sample size would be required to detect significant changes in ankle strength or power.
Strengths of this study include the fact that participants were randomized, group allocation was concealed, a placebo control group was included, and strength and power were tested by a blinded assessor using equipment that differed from that used in the intervention. It is recommended that further investigations with larger sample sizes be conducted to determine changes in strength and power with similar training protocols. Future research may also examine the actual velocities attained during training using bands and weight machines. For example, further research is needed to determine whether weight training performed with lower loads (and therefore at higher velocities) is equal to or better than band training.
In conclusion, our results demonstrated that concentric ankle movements performed as quickly as possible against elastic bands improved foot movement time, which may have important implications in circumstances when rapid generation of ankle torque is required (e.g., to avoid a fall or to prevent a vehicle crash). Training with bands represents a low-cost, practical form of exercise that could be considered as an addition to programs designed to address mobility limitations in older adults.
The authors thank the participants and acknowledge Nancy Maskus and Heather Klassen for their assistance with the training groups in this study.
This study was supported by funding from the Natural Sciences and Engineering Research Council of Canada; the Canadian Institutes of Health Research; and CanDRIVE, a New Emerging Team funded by the Institute of Aging of the Canadian Institutes of Health Research.
The results of the present study do not constitute endorsement by American College of Sports Medicine.
1. American College of Sports Medicine. Position Stand. Progression models in resistance training
for healthy adults. Med Sci Sports Exerc
2. Bean JF, Kiely DK, Herman S, et al. The relationship between leg power and physical performance in mobility-limited older people. J Am Geriatr Soc
3. Bendall MJ, Bassey EJ, Pearson MB. Factors affecting walking speed of elderly people. Age Ageing
4. Binder S. Injuries among older adults: the challenge of optimizing safety and minimizing unintended consequences. Inj Prev
. 2002;8(4 suppl):IV2-4.
5. Brouwer B, Ashby P. Corticospinal projections to upper and lower limb spinal motoneurons in man. Electroencephalogr Clin Neurophysiol
6. Cohen J. Statistical Power Analysis for the Behavioral Sciences
. 2nd ed. Hillsdale (NJ): Lawrence Erlbaum Associates; 1988. 567 p.
7. de Vos NJ, Singh NA, Ross DA, Stavrinos TM, Orr R, Fiatarone Singh MA. Optimal load for increasing muscle power during explosive resistance training
in older adults. J Gerontol A Biol Sci Med Sci
8. Guralnik JM, Simonsick EM, Ferrucci L, et al. A Short Physical Performance Battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol
9. Hakkinen K, Hakkinen A. Muscle cross-sectional area, force production and relaxation characteristics in women at different ages. Eur J Appl Physiol Occup Physiol
10. Jette AM, Lachman M, Giorgetti MM, et al. Exercise-it's never too late: the Strong-for-Life program. Am J Public Health
11. Leveille SG, Penninx BW, Melzer D, Izmirlian G, Guralnik JM. Sex differences in the prevalence of mobility disability in old age: the dynamics of incidence, recovery, and mortality. J Gerontol B Psychol Sci Soc Sci
12. Lord SR. Sit-to-stand performance depends on sensation, speed, balance, and psychological status in addition to strength in older people. J Gerontol A Biol Sci Med Sci
13. Lord SR, Menz HB, Tiedemann A. A physiological profile approach to falls risk assessment and prevention. Phys Ther
14. Mackey DC, Robinovitch SN. Mechanisms underlying age-related differences in ability to recover balance with the ankle strategy. Gait Posture
15. Margolis KL, Kerani RP, McGovern P, Songer T, Cauley JA, Ensrud KE. Risk factors for motor vehicle crashes in older women. J Gerontol A Biol Sci Med Sci
16. Marques CJ, Cabri J, Barreiros J, Carita AI, Friesecke C, Loehr JF. The effects of task complexity on brake response time
before and after primary right total knee arthroplasty. Arch Phys Med Rehabil
17. Marsh AP, Miller ME, Rejeski WJ, Hutton SL, Kritchevsky SB. Lower extremity muscle function following strength or power training in older adults. J Aging Phys Act
18. Newton RU, Kraemer WJ, Hakkinen K, Humphries BJ, Murphy AJ. Kinematics, kinetics, and muscle activation during explosive upper body movements. J Appl Biomech
19. Orr R, de Vos NJ, Singh NA, Ross DA, Stavrinos TM, Fiatarone-Singh MA. Power training improves balance in healthy older adults. J Gerontol A Biol Sci Med Sci
20. Perez MA, Lundbye-Jensen J, Nielsen JB. Changes in corticospinal drive to spinal motoneurones following visuo-motor skill learning in humans. J Physiol
. 2006;573(Pt 3):843-55.
21. Perez MA, Lungholt BK, Nyborg K, Nielsen JB. Motor skill training induces changes in the excitability of the leg cortical area in healthy humans. Exp Brain Res
22. Petersen NT, Pyndt HS, Nielsen JB. Investigating human motor control by transcranial magnetic stimulation. Exp Brain Res
23. Porter MM, Vandervoort AA. Standing strength training of the ankle plantar and dorsiflexors in older women, using concentric
and eccentric contractions. Eur J Appl Physiol Occup Physiol
24. Porter MM, Vandervoort AA, Lexell J. Aging
of human muscle: structure, function and adaptability. Scand J Med Sci Sports
25. Portney LG, Watkins MP. Foundations of Clinical Research: Applications to Practice
. 3rd ed. Upper Saddle River (NJ): Pearson Education, Inc; 2009. 892 p.
26. Roos MR, Rice CL, Vandervoort AA. Age-related changes in motor unit function. Muscle Nerve
27. Roubenoff R. Sarcopenia: effects on body composition and function. J Gerontol A Biol Sci Med Sci
28. Skelton DA, Kennedy J, Rutherford OM. Explosive power and asymmetry in leg muscle function in frequent fallers and non-fallers aged over 65. Age Ageing
29. St George RJ, Fitzpatrick RC, Rogers MW, Lord SR. Choice stepping response and transfer times: effects of age, fall risk, and secondary tasks. J Gerontol A Biol Sci Med Sci
30. Staats DO. Preventing injury in older adults. Geriatrics
31. Suzuki T, Bean JF, Fielding RA. Muscle power of the ankle flexors predicts functional performance in community-dwelling older women. J Am Geriatr Soc
32. Tiedemann AC, Sherrington C, Lord SR. Physical and psychological factors associated with stair negotiation performance in older people. J Gerontol A Biol Sci Med Sci
33. Washburn RA, Smith KW, Jette AM, Janney CA. The Physical Activity Scale for the Elderly (PASE): development and evaluation. J Clin Epidemiol
34. 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
35. Wojcik LA, Thelen DG, Schultz AB, Ashton-Miller JA, Alexander NB. Age and gender differences in single-step recovery from a forward fall. J Gerontol A Biol Sci Med Sci
36. Wolfson L, Judge J, Whipple R, King M. Strength is a major factor in balance, gait, and the occurrence of falls. J Gerontol A Biol Sci Med Sci
. 1995;50 Spec No:64-7.