Chui, Kevin K. PT, PhD, GCS, OCS; Lusardi, Michelle M. PT, DPT, PhD
Much of geriatric rehabilitation care, across a wide range of medical diagnoses, focuses on restoring the ability to walk and returning the patient to his or her preferred community-living situation. Although there is evidence that walking speed declines with age, there is limited data for spatial and temporal measures of self-selected walking speed (SSWS) and fast walking speed (FWS) for healthy adults older than 75 years. Given that one of the fastest growing population groups are those older than 85 years,1 an understanding of the gait characteristics of persons at very late stages of life would assist health professionals in interpreting walking data from patient populations.
Physical therapists often use walking speed as a baseline measure of function to document improvement during, and outcome following, rehabilitation intervention for patients with stroke,2,3 hip fracture,4,5 Parkinson disease,6 total hip7,8 and total knee9,10 arthroplasty, and for frail older adults hospitalized for a variety of medical conditions.11–13 Slow SSWS is associated with an increased risk of falling,14,15 declines in health and functional status,16–18 likelihood of hospitalization and institutionalization,19–21 and mortality.22 Our ability to use SSWS as an indicator of risk of such outcomes would be enhanced by an understanding of the typical SSWS and associated gait parameters for generally healthy, very old adults living in the community.
REVIEW OF THE LITERATURE
Several studies have reported data for SSWS and FWS using a variety of age groupings, defined populations, and clinic- or laboratory-based methods. Age groupings have varied markedly; studies have (sub)grouped participants by decade23–26 and half decade,27 grouping those aged 65–89 years as aged adults28 or those aged 63 years or older as the eldest group,29 or by middecade to middecade (eg, 65–74 years and 75–82 years) groupings,30 among other conventions. The apparent variability in how participants are grouped makes interpretation and comparing and contrasting of these datasets difficult.
One issue with using groups with large age ranges (eg, 63 years or more or 65–89 years) is that these groups may be, as a function of age, heterogeneous. Furthermore, there may be several homogenous subgroups, perhaps better defined by decade, within a broadly defined group of older adults. Adults aged 70–79 years, for example, may walk differently than adults aged 80–89 years; grouping both decades together may prevent one from finding an important difference.
The varied degree of participant description also makes interpretation of datasets difficult. For examples, inclusion criteria vary from being healthy25 to meeting specific performance, symptom, age, or medical history criteria.26 Few patient characteristics, other than age, height, and weight are typically provided, which further limits the ability to generalize findings. Most studies of gait speed are based on findings from community-dwelling and independently living individuals.
Methods used to measure spatial and temporal walking parameters include stopwatch,23,26,31 instrumented walkway system such as the GAITRite,24,27,28 photocells,25 and kinematic analysis29 (and an unspecified timed method30). Methods that use a stopwatch, although available to most clinicians, can only provide a limited number of walking parameters such as SSWS and FWS. Instrumented walking analysis (eg, GAITRite), although typically not available to most clinicians, can provide SSWS and FWS data and other corresponding walking parameters important to gait analysis. These include cadence, step and stride lengths, percentage of the gait cycle spent in stance and swing, percentage of the gait cycle spent in single- and double-limb support, and the width of the base of support (BOS).
The aim of this study is to provide age- and gender-specific information about average gait characteristics of older adults by age and gender, with emphasis on adults older than 80 years. The purposes of the study are to (1) determine SSWS and FWS by gender and age of participants between 70 and 100 years and (2) examine and compare spatial and temporal walking parameters for SSWS and FWS by gender and age.
Participants were community-dwelling residents of a continuing care retirement community (CCRC) located in a suburban New England town involved in a longitudinal study of function, which was approved by the university institutional review board. Participants were recruited during resident (town hall) meetings and information sessions and by mailings, fliers, and personal invitation by CCRC staff. Participants expressing interest were contacted by the investigators for preliminary health screening. All CCRC residents able to ambulate independently were eligible for participation. Residents were excluded if they had unstable or limiting cardiac disease, respiratory conditions requiring supplemental oxygen, history of neurologic disease with residual impairments, history of fracture within the previous 6 months, severely limiting arthritis, joint instability or back pain, total joint replacement within the previous 6 months, documented dementia or significant depression, cancer-related therapy within the previous 6 months, or acute illness or injury on the day of the functional assessment clinic. Before participation, all participants were provided medical clearance by their primary care physicians, who were affiliated with the CCRC and aware of the study's purpose and physical demands. All eligible participants signed an informed consent document.
During data collection, participants completed a battery of tests and measures including vital signs, a general health questionnaire developed for the study, the 36-Item Short Form Health Survey questionnaire,32 Berg Balance Scale,33 Timed Sit to Stand test,34 Timed Up and Go Test,35 Four-Square Step test,36 2-Minute Walk test,37 Short Physical Performance Battery,38 and walking analysis. Tests and measures were completed in random order and at separate testing stations. Participants were allowed adequate rest between stations to minimize the effects of fatigue. All tests and measures were completed in a single 120-minute session. This article focuses on results of the analysis of walking; results of other functional tests have been presented elsewhere.39
Analysis of Walking
Walking data was collected using the GAITRite, an instrumented walkway system. The GAITRite is a pressure-sensitive mat (61 × 366 cm) that is organized in a 48 × 288 grid pattern with sensors placed at 1.27 cm on center. The sampling frequency is 60 Hz. The GAITRite connects to a personal computer via a serial interface cable; data was processed using GAITRite software version 3.8g.
The validity and reliability of temporal and spatial walking parameters from the GAITRite has been examined. High levels of concurrent validity between GAITRite and the following methods have been reported: paper-and-pencil (ICC = 0.96),40 video-based motion analysis (ICC = 0.95–0.99),41 insole pressure (ICC = 0.99),42 and stop watch (ICC = 0.75-97).43 High test-retest reliability of spatial and temporal parameters has also been reported on the same day,40 within 24 hours,44 and over 1-week45 and 2-week46 periods.
During data collection, participants were instructed to stand from an armchair that was 3.5 m from the start of the GAITRite mat to assume the starting position. Starting this distance from the GAITRite mat allowed participants to accelerate as they began to walk and reach a steady SSWS before they stepped onto the GAITRite mat. In addition, participants continued to walk an additional 3.5 m past the end of the GAITRite to minimize possibility of deceleration before stepping off the GAITRite. Participants were allowed to practice up to 3 times if needed.
For SSWS trials, participants were instructed to walk at their “normal speed” over the mat. Between data collection trials, participants rested for 3 minutes in a seated position while their data was processed; this also minimized potential effect of fatigue. After 3 successful SSWS data collection trials, participants rested for 5 minutes before the start of data collection for FWS trials. Instructions for FWS trials and rest periods were the same as the SSWS trials except that participants were instructed to walk as “fast as you feel comfortable going.” Participants wore a gait belt and were guarded from a posterior-lateral position by an assistant during all walking trials. All walking data was collected by one rater (K.K.C) for reliability purposes.
Descriptive statistics were calculated for participant characteristics by gender and age (decade). Independent t tests were used to examine for differences in age, resting vital signs, and anthropometric measurements between men and women. One-way analyses of variance were used to examine for differences in resting vital signs and anthropometric measurements between decades. Separate 2-way factorial ([gender effect: men, women] × [decade effect: 70–79, 80–89, 90–99]) analyses of variance were used to assess differences in the gait speeds and spatial and temporal parameters of gait. Factorial analyses of variance were used to examine the main effects of and the interaction effects between gender and decade. Bonferroni post hoc analyses were used when appropriate. Paired t tests were used to compare SSWS and FWS parameters. All statistical analyses were made using SPSS (v16) (Chicago, Illinois) and statistical significance was set at P ≤ .05 for all analyses.
Three functional assessment clinics were held over an 18-month period with a total participation of 118 individuals aged 72–98 years, of whom 70.3% were women. All participants successfully completed 3 trials of SSWS and FWS without incident. Descriptive statistics (number of cases, mean, standard deviation [SD], minimum, and maximum) for age, height, weight, body mass index, resting heart rate, resting oxygen saturation, and resting systolic and diastolic blood pressures are presented in Table 1 with respect to gender. With respect to age, there was no significant difference (P = .83) between genders; there were no significant differences in resting vital signs (resting heart rate, oxygen saturation, or systolic and diastolic blood pressures) between genders. All of the anthropometric measurements differed significantly between genders with men being taller (P < .001), heavier (P < .001), and having a smaller body mass index (P = .02). None of the vital signs or anthropometric measurements differed significantly between decades (P > .14).
Given their resting vital signs, body mass index; independent living arrangement; medical clearance necessary to participate; and successful completion of all of the functional testing, including all of the walking trials, we contend that this was a group of older adults who are generally healthy with no evidence of frailty or functional limitation.
Self-Selected Walking Speed
There was a significant main effect for decade; SSWS data showed a significant age-related decline (F = 12.35, P < .001). Post hoc analyses showed significant SSWS differences between all 3 decades (P ≤ .001) with a mean (SD) SSWS of 138.6 (23.7) cm/s for those aged 70–79 years, 113.5 (10.1) cm/s for those aged 80–89 years, and 86.5 (17.3) cm/s for those aged 90–99 years. There was also a significant main effect for gender; SSWS data showed a significant gender-related difference with a mean (SD) SSWS 129.7 (14.8) cm/s for men participants and 105.3 (10.8) cm/s for women participants (F = 12.47, P < .001). For SSWS, there was no significant interaction between decade and gender (F = .083, P = .92). Descriptive statistics (number of cases, mean, SD, standard error of the measurement [SEM], confidence interval, and minimal detectable difference [MMD]) for SSWS are presented in Table 2 by gender, decade, and the entire sample.
Fast Walking Speed
There was a significant main effect for decade; FWS data showed a significant age-related decline (F = 10.30, P < .001). Post hoc analyses showed significant FWS differences between all 3 decades (P ≤ .03) with a mean (SD) velocity of 179.0 (32.6) cm/s for those aged 70–79 years, 154.2 (14.0) cm/s for those aged 80–89 years, and 116.4 (25.6) cm/s for those aged 90–99 years. There was also a significant main effect for gender; FWS data showed a significant gender-related difference with a mean (SD) 176.6 (20.8) cm/s for men and 140.5 (14.3) for women (F = 19.49, P < .001). For FWS, there was no significant interaction between decade and gender (F = 0.680, P = .51). Descriptive statistics for FWS are presented in Table 3 by gender, decade, and the entire sample.
Normalized Self-Selected Walking Speed and Fast Walking Speed
Because men were also significantly taller than women, it is unclear whether the significant gender differences in SSWS and FWS were confounded by gender-related difference in height. Self-selected walking speed and FWS were therefore normalized (adjusted) by leg length (a function of total height), and supplemental analyses were performed. Normalized SSWS (cm/s per leg length in centimeters) data also showed a significant difference between genders: 1.40 (0.29) cm/s per leg length in centimeters for males and 1.23 (.36) cm/s per leg length in centimeters for women (F = 6.58, P = .01). Therefore, when normalizing for leg length, the SSWS of men was still significantly faster than women. Normalized FWS data also showed a significant difference between genders: 1.91(0.40) cm/s per leg length in centimeters for men and 1.64 (0.49) cm/s per leg length in centimeters for women (F = 8.47, P = .004). Therefore, when normalizing for leg length, the FWS of men remained significantly faster than women.
Self-Selected Walking Speed Parameters
Descriptive statistics for SSWS parameters are presented in Table 4 by age, gender, and the entire sample. When examining parameters from SSWS, there was a significant age effect for cadence (P < .001), step length (P < .001), stride length (P < .001), percentage of cycle spent in stance (P < .003), percentage of cycle spent in swing (P = .003), percentage of cycle in single-limb support (P = .003), and percentage of cycle in double-limb support (P = .002). Post hoc analyses showed significant differences in cadence (P ≤ .023), step length (P ≤ .006), and stride length (P ≤ .006) between all age groups. In fact, cadence, step length, and stride length significantly decreased with increasing age (decade). Those aged 90–99 years were significantly different than those aged 70–79 years and 80–89 years for percentage of the gait cycle spent in stance (P ≤ .013) and swing (P ≤ .013) and for percentage of the gait cycle spent in single- (P ≤ .013) and double- (P ≤ .008) limb support. There were no differences in these SSWS parameters between those aged 70–79 years and 80–89 years.
Table 4-a. Character...Image Tools
Men had significantly longer step length (P < .001), longer stride length (P < .001), and wider BOS (P = .012) when compared with women. Between genders, there were no significant differences in cadence (P = .695), percentage of the gait cycle spent in stance (P = .080) and swing (P = .081), and percentage of the gait cycle spent in single- (P = .081) and double- (P = .072) limb support.
Table 4-b. Character...Image Tools
There were no significant interaction effects between decade and gender for any of the SSWS parameters.
Fast Walking Speed–Parameters
Descriptive statistics for FWS parameters are presented in Table 5 by age, gender, and the entire sample. When examining parameters from FWS, there were significant gender and age effects for step length and stride length. During fast walking, men had longer step length (P < .001) and stride length (P < .001) than women. Post hoc analyses showed difference between all age groups for step length (P ≤ .024) and stride length (P ≤ .021). Both step and stride length decreased as age increased. When walking fast, men had a wider BOS than women (P = .03). When walking fast, there were no gender or age effects for cadence, percentage of the gait cycle spent in swing, in stance, or in single- and double-limb support. There were no significant interaction effects between decade and gender for any of the FWS parameters.
Table 5-a. Character...Image Tools
Comparison of Self-Selected Walking Speed and Fast Walking Speed–Parameters
Table 5-b. Character...Image Tools
For the entire group, there was a significant difference between the 2 speeds of walking (mean differenceFWS–SSWS) of 38.68 (20.83) cm/s, P < .001. Several gait parameters appear to collectively contribute to the difference between FWS and SSWS: an increase in cadence of 24.42 (11.99) steps/min, P < .001; in step length of 6.24 (4.56) cm, P < .001; and in stride length of 12.37 (9.15) cm, P < .001. In addition, the mean percentage of gait cycle spent in swing increased by 2.00 (1.46), P < .001 and the mean percentage of gait cycle spent in single-limb support increased by 4.02 (2.94), P < .001. There was a corresponding significant mean decrease in percentage of the gait cycle spent in stance of −2.00% (1.46%), P < .001; and in percentage of the gait cycle spent in double-limb support of −4.08% (2.77%), P < .001. There was no significant change in the BOS when comparing SSWS and FWS (P = .91).
The SSWS of men and women aged 70–79 years in this study was based on a relatively small number when compared with previous studies.23–27 In fact, in our study, those aged 70–79 years represented only 16.2% of our total sample. This percentage is, however, consistent with the relative number of residents aged 70–79 years in the CCRC examined. Therefore, the remainder of this discussion will focus on the gait parameters of those aged 80 years and older, who represent the predominant group in this study.
Of the studies that reported their findings by decade23–26 or half decade,27 only Lusardi et al24 and Steffan et al26 reported SSWS for older adults aged 80–89 years. The SSWS data of 80–89 year old men and women from this study are similar to findings reported by Steffan et al26 and markedly faster than the SSWS reported by Lusardi et al.24 Although all of the participants in our study and those in the studies of Lusardi et al24 and Steffan et al26 were community dwelling, Steffan et al26 excluded those who used an assistive device, while Lusardi et al24 included those who used an assistive device. In fact, 29.4% of those aged 80–89 years in the study by Lusardi et al24 used an assistive device, and their SSWS was considerably slower than the participants who did not use a device. Although our study did not exclude those who use an assistive device for ambulation, none of our participants used an assistive device during gait analysis. We, therefore, acknowledge that selection bias may have occurred.
Only Lusardi et al24 examined the SSWS of those older than 90 years (ie, aged 90–101 years). Our study included a sample that was defined by decade (ie, 90–99 years). The men and women participants aged 90–99 years in our study walked markedly faster than those in the study of Lusardi et al,24 of which 58.8% used an assistive device. Not surprisingly, those who used an assistive device in the Lusardi et al24 study walked markedly slower than those who did not. There is very little published information on gait parameters for those aged 90 years and older. This is likely due to the lower percentage of older adults reaching this age, and the inclusion criteria that researchers typically use to define health.
In our study, we provided the SEM47 and minimal detectable difference47 for each spatial and temporal gait parameter for the aggregate and by age and gender. The SEM is a measure of response stability and is related to measurement error. The smaller the SEM (ie, less measurement error) the more reliable the measurement is. The SEM is used to calculate the minimal detectable difference, which is the smallest amount of change that exceeds the measurement error and is considered a true change or difference. For example, on the basis of the data in Table 2, a woman aged 80–89 years would have to increase (or decrease) her SSWS by 8.64 cm/s or more for the change to be considered a true change. Changes in her SSWS less than 8.64 cm/s would not be considered a true change; it would be considered a function of measurement error.
Of the 5 studies that reported their findings by decade or half decade, FWS was reported in 2 studies24,26 for those aged 80–89 years, and in one study24 for those older than 90 years. Men and women in our study aged 80–89 years had a FWS similar to their counterparts in the study by Steffan et al26 and faster than those in the study of Lusardi et al,24 which included participants who used assistive devices. Similar differences between the FWS of 90- to 99-year-old participants in our study were found with the study of Lusardi et al.24
Of the studies that used an instrumented walkway system such as the GAITRite,24,27,28 only two27,28 provided spatial and temporal gait parameters other than SSWS or FWS. For our study, we provided cadence, step length, stride length, stance percentage, swing percentage, single- and double-limb support—percentage, and BOS data from SSWS and FWS, disaggregated by decade and gender. When compared with other studies, additional spatial and temporal parameters were reported in our study to assist practitioners with gait analysis of SSWS and FWS.
Callisaya et al27 examined a younger population of older adults and provided cadence, step length, step width, and double-limb support–percentage data from SSWS. When disaggregated by half decade and gender, their data from participants aged 80–86 years are similar to our data from participants aged 80–89 years. Similar to the findings of our study, Callisaya et al27 also found significant differences in step length and step width (also referred to as BOS) between genders and no difference in double-limb support–percentage between genders. The majority of the participants in the study of Callisaya et al27 were younger than 80 years (81.1%), whereas the majority of the participants in our study were aged 80 years or older (83.8%). This difference may explain why Callisaya et al27 found a significant gender effect for cadence, while our study did not.
Laufer28 provides cadence, stride length, swing phase percentage, and double-limb support–percentage data from SSWS, dichotomized by age (young [20-31 years] and aged [65-89 years]) and gender. With a wide range of ages included in the aged population (aged 65–89 years), comparison of spatial and temporal data between the study of Laufer28 and our study is difficult. Given the heterogeneity of some of the spatial and temporal gait parameters by decade found in our study and the studies of others, we do not advise grouping older adults by large age brackets that span more than one decade.
Our study is one of the few studies that statically examined the main effects for or an interaction effect between gender and decade for SSWS and FWS and their corresponding gait parameters in older adults. Callisaya et al27 reported significant interactions between age (half decade) and gender for SSWS, cadence, and double-limb support. In contrast, we did not find any significant interactions. One possible explanation for these differences is that Callisaya et al27 examined adults who were predominantly younger than 80 years, whereas we examined adults who were mostly aged 80 years or older. Together, these findings suggest that the aging process may affect gait differently in men and women younger than 80 years but have similar effects in men and women aged 80 years or older.
There were several limitations to our study. One limitation was the relative proportion of females. Older women (70.3%) were slightly overrepresented in this sample when compared with national (61.6%) and state (62.9%) census estimates for 2008 (http://factfinder.census.gov; accessed September 18, 2010). Future work should include stratification strategies so that when the data is aggregated by decade alone, the mean values will better represent the population. Although the sample size for men aged 90–99 years is small, this is a group whose gait characteristics have not previously been reported in the literature. Clinicians, therefore, should be cautious when interpreting the results for men aged 90–99 years as other than an estimate for gait speed and the other gait parameters presented. The use of self-selected participants from only one CCRC also limits the generalizability of our findings. Generalizability would be improved if future studies included multiple sites where healthy older adults dwell and strive for larger percentages of participation of residents.
This study reports gait data for SSWS, FWS, and select walking parameters from healthy adults aged 72–98 years. These gait data are derived from a large and well-defined group of healthy older adults and is further examined by gender and age. Differences in walking performance found between genders and across age groups are consistent with prior studies. The results for participants in the 80–89 and 90–99 subgroups add valuable information about walking not previously reported in the literature. Access to spatial and temporal parameters of gait for community-living older adults, sorted by both decade and gender, can assist health professionals in interpreting walking data from patient populations.
1. Boscoe FP. Subdividing the age group of 85 years and older to improve US disease reporting. Am J Public Health. 2008; 98:1167–1170.
2. Duncan PW, Jorgensen HS, Wade DT. Outcome measures in acute stroke trials: a systematic review and some recommendations to improve practice. Stroke. 2000; 31:1429–1438.
3. Lord SE, Rochester L. Measurement of community ambulation after stroke: current status and future developments. Stroke. 2005; 36:1457–1461.
4. Mangione KK, Craik RL, Lopopol R, Tomlinson JD, Brenneman SK. Predictors of gait speed in patients after hip fracture. Physiother Can. 2008; 60:10–18.
5. Palombaro KM, Craik RL, Mangione KK, Tomlinson JD. Determining meaningful changes in gait speed after hip fracture. Phys Ther. 2006; 86:809–816.
6. Canning CG, Ada L, Woodhouse E. Multiple-task walking training in people with mild to moderate Parkinson's disease: a pilot study. Clin Rehabil. 2008; 22:226–233.
7. Jan MH, Hung JY, Lin JC, Wang SF, Liu TK, Tang PF. Effects of a home program on strength, walking speed, and function after total hip replacement. Arch Phys Med Rehabil. 2004; 85:1943–1951.
8. van den Akker-Scheek I, Stevens M, Bulstra SK, Groothoff JW, van Horn JR, Zijlstra W. Recovery of gait after short-stay total hip arthroplasty. Arch Phys Med Rehabil. 2007; 88:361–367.
9. Yoshida Y, Mizner RL, Ramsey DK, Snyder-Mackler L. Examining outcomes from total knee arthroplasty and the relationship between quadriceps strength and knee function over time. Clin Biomech. 2008; 23:320–328.
10. Giaquinto S, Ciotola E, Margutti F. Gait in the early days after total knee and hip arthroplasty: a comparison. Disabil Rehabil. 2007; 29:731–736.
11. Sullivan DH, Wall PT, Bariola JR, Bopp MM, Frost YM. Progressive resistance muscle strength training of hospitalized frail elderly. Am J Phys Med Rehabil. 2001; 80:503–509.
12. Purser JL, Weinberger M, Cohen HJ, et al. Walking speed predicts health status and hospital costs for frail elderly male veterans. J Rehabil Res Dev. 2005; 42:535–546.
13. van Iersel MB, Munneke M, Esselink RA, Olde Rikkert MG. Gait velocity and the Timed Up and Go test were sensitive to changes in mobility in frail elderly patients. J Clin Epidemiol. 2008; 61:186–191.
14. Verghese J, Holtzer R, Lipton RB, Wang C. Quantitative gait markers and incident fall risk in older adults. J Gerontol A Biol Sci Med Sci. 2009; 64:896–901.
15. Runge M, Hunter G. Determinants of musculoskeletal frailty and the risk of falls in old age. J Musculoskelet Neuronal Interact. 2006; 6:167–173.
16. Studenski S, Perera S, Wallace D, et al. Physical performance measures in the clinical setting. J Am Geriatr Soc. 2003; 51:314–322.
17. Rothman MD, Leo-Summers L, Gill TM. Prognostic significance of potential frailty criteria. J Am Geriatr Soc. 2008; 56:2211–2116.
18. Brach JS, VanSwearingen JM, Newman AB, Kriska AM. Identifying early decline of physical function in community-dwelling older women; performance based and self-report measures. Phys Ther. 2002; 82:320–328.
19. Montero-Odasso M, Schapira M, Soriano ER, et al. Gait velocity as a single predictor of adverse events in healthy seniors aged 75 years and older. J Gerontol A Biol Sci Med Sci. 2005; 60:1304–1309.
20. Rabadi MH, Blau A. Admission ambulation velocity predicts length of stay and discharge disposition following stroke in an acute rehabilitation hospital. Neruorehabil Neural Repair. 2005; 19:20–26.
21. Giuliani CA, Gruber-Baldini AL, Park NS, et al. Physical performance characteristics of assisted living residents and risk for adverse health outcomes. Gerontologist. 2008; 48:203–212.
22. Hardy SE, Perera S, Rumani YF, Chandler JM, Studenski SA. Improvement in usual gait speed predicts better survival in older adults. J Am Geriatr Soc. 2007; 55:1727–1734.
23. Bohannon RW. Comfortable and maximum walking speed of adults 20–79 years; reference values and determinants. Age Ageing. 1997; 26:15–19.
24. Lusardi MM, Pellechia G, Schulman M. Functional performance in community living older adults. J Geriatr Phys Ther. 2003; 26:14–22.
25. Oberg T, Karsznia A, Oberg K. Basic gait parameters: reference data for normal subjects, 10–79 years of age. J Rehabil Res Dev. 1993; 30:210–223.
26. Steffan TM, Hacker TA, Mollinger L. Age- and gender related test performance in community dwelling elderly people: Six Minute Walk Test, Berg Balance Scale, Timed Up and Go Test, and gait speeds. Phys Ther. 2002; 82:128–137.
27. Callisaya ML, Blizzard L, Schmidt MD, McGinely JL, Srikanth VK. Sex modifies the relationship between age and gait: A population-based study of older adults. J Gerontol A Biol Sci Med Sci. 2008; 63A:165–170.
28. Laufer Y. Age- and gender-related changes in the temporal-spatial characteristics of forwards and backwards gaits. Physiother Res Int. 2003; 8:131–142.
29. Hinman JE, Cunningham DA, Rechnitzer PA, Paterson DH. Age-related changes in speed of walking. Med Sci Sports Exerc. 1988; 20:161–166.
30. Kaneko M, Morimoto Y, Kimura M, Fuchimoto K, Fuchimoto T. A kinematic analysis of walking and physical fitness testing in elderly women. Can J Sport Sci. 1991; 16:223–228.
31. Bohannon RW, Andrews AW, Thomas MW. Walking speed: reference values and correlates for older adults. J Ortho Sports Phys Ther. 1996; 24:86–90.
32. Ware JE. SF-36 Health Survey: Manual and Interpretation Guide. Boston, MA: Quality Metric Inc; 2003.
33. Riddle DL, Stratford PW. Interpreting validity indexes for diagnostic tests: an illustration using the Berg Balance Test. Phys Ther. 1999; 79:939–948.
34. Bohannon RW. Sit-to-stand test for measuring performance of lower extremity muscles. Percept Mot Skills. 1995; 80:163–166.
35. Podsiadlo D, Richardson S. The timed “Up & go”: a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 1991; 67:387–389.
36. Dite W, Temple VA. A clinical test of stepping and change of direction to identify multiple falling older adults. Arch Phys Med Rehabil. 2002; 83:1566–1571.
37. Brooks D, Davis AM, Naglie G. The Feasibility of Six-Minute and Two-Minute Walk Tests in in-patient geriatric rehabilitation. Can J Aging. 2007; 26:159–162.
38. 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 Med Sci. 1994; 49:M85–M94.
39. Lusardi MM, Chui KK, King MB, Carlin C. How do community living adults >80 typically perform on functional measures used in rehabilitation? Paper presented at: the 2008 Annual Meeting of the Gerontological Society of America. National Harbor, MD. November 21, 2008.
40. McDonough AL, Batavia M, Chen FC, Kwon S, Ziai J. The validity and reliability of the GAITRite system's measurements: a preliminary evaluation. Arch Phys Med Rehabil. 2001; 82:419–425.
41. Cutlip RG, Mancinelli C, Huber F, DiPasquale J. Evaluation of an instrumented walkway for measurement of the kinematic parameters of gait. Gait Posture. 2000; 12:134–138.
42. Bilney B, Morris M, Webster K. Concurrent related validity of the GAITRite®
walkway system for quantification of the spatial and temporal parameters of gait. Gait Posture. 2003; 17:68–74.
43. Youdas J, Hollman J, Aalbers M, Ahrenholz H, Aten R, Cremers J. Agreement between the GAITRite walkway system and a stopwatch—footfall count method for measurement of temporal and spatial gait parameters. Arch Phys Med Rehabil. 2006; 87:648–1652.
44. Batey P, Rome K, Finn P, Hanchard N. Assessing reliability of measurement of gait velocity. Physiotherapy. 2003; 89:313–317.
45. van Uden CJT, Besser MP. Test-retest reliability of temporal and spatial gait characteristics measured with an instrumented walkway system (GAITRite®
). BMC Musculoskeletal Disorder. 2004; 5:13.
46. Menz HB, Latt MD, Tiedemann A, Kwan MMS, Lord SR. Reliability of the GAITRite®
walkway system for the quantification of temporo-spatial parameters of gait in young and older people. Gait Posture. 2004; 20:20–25.
47. Portney LG, Watkins MP. Foundations of Clinical Research: Application to Practice. Upper Saddle River, NJ: Prentice Hall; 2009.
older adults; walking speed