PTJ
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

PHYS THER
Vol. 86, No. 4, April 2006, pp. 520-540
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Material
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when Rapid Responses are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lopopolo, R. B
Right arrow Articles by Mangione, K. K
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lopopolo, R. B
Right arrow Articles by Mangione, K. K

Research Reports

Effect of Therapeutic Exercise on Gait Speed in Community-Dwelling Elderly People: A Meta-analysis

Rosalie B Lopopolo, Melissa Greco, Dorianne Sullivan, Rebecca L Craik and Kathleen K Mangione

RB Lopopolo, PT, PhD, MBA, is Professor, Department of Physical Therapy, Arcadia University, Glenside, Pa
M Greco, PT, DPT, is a Staff Therapist at Hazleton General Hospital, Hazleton, Pa. She was completing her DPT studies at Arcadia University at the time of the study
DH Sullivan, PT, DPT, is a Staff Therapist at Nazareth Hospital, Philadelphia Pa. She was completing her DPT studies at Arcadia University at the time of the study
RL Craik, PT, PhD, FAPTA, is Professor and Chair, Department of Physical Therapy, Arcadia University, 450 S Easton Rd, Glenside, PA 19038-3295 (USA)
KK Mangione, PT, PhD, GCS, is Associate Professor, Department of Physical Therapy, Arcadia University

(craik{at}arcadia.edu). Address all correspondence to Dr Craik


Submitted January 19, 2005; Accepted November 9, 2005


   Abstract
 
Background and Purpose. Inconsistent research findings make it unclear whether therapeutic exercise improves gait speed in community-dwelling elderly people. Using meta-analytical procedures, we examined the effect of therapeutic exercise on changing gait speed in community-dwelling older adults and the effect of type, intensity, and dose of therapeutic exercise on gait speed. Method. Studies were retrieved using a comprehensive database search. Two independent reviewers determined study eligibility based on inclusion criteria, rated study quality, and extracted information on study methods, design, intervention, and results. Data were combined to obtain an overall effect size, its 95% confidence interval, and a measure of significance. In addition, analyses to characterize the clinical relevance of the findings were performed. Results. One hundred seventeen studies were evaluated, with 24 studies (n=1,302 subjects) meeting the inclusion criteria for habitual gait speed and 18 studies (n=752 subjects) meeting the inclusion criteria for fast gait speed. Therapeutic exercise—or, more specifically, strength training and combination training (aerobic plus other exercise)—had significant effects (r=.145, P=.017; r=.176, P=.002, respectively) on habitual gait speed. High-intensity (effort expended by subjects) exercise and high-dosage (frequency and duration of exercise sessions) intervention also had a significant effect (r=.184, P=.001; r=.190, P=.001, respectively) on gait speed, whereas there was no effect for moderate- and low-intensity exercise or for low-dosage exercise. No exercise intervention affected fast gait speed in this analysis. Discussion and Conclusion. The results provide support for the belief that therapeutic exercise can improve gait speed in community-dwelling elderly people and that intensity and dosage are important contributing factors. The relatively weak correlation found between therapeutic exercise and gait speed merits further study. [Lopopolo RB, Greco M, Sullivan DH, et al. Effect of therapeutic exercise on gait speed in community-dwelling elderly people: a meta-analysis. Phys Ther. 2006;86:520–540.]

Key Words: Community-dwelling elderly people • Exercise • Gait • Gait speed • Therapeutic exercise


   Introduction
Top
 Abstract
 Introduction
Method
 Results
 Discussion
 Conclusion
 Appendix
 References
 
Acommon complaint of elderly people is that they cannot walk "the way they used to." One aspect of walking that changes with age is gait speed. After the seventh decade of life, habitual gait speed declines at a rate of 12% to 16% per decade and about 20% per decade for fast gait speed.1 Oberg et al2 and Bohannon3 reported that habitual gait speed may decrease by 9% to 11% and fast gait speed may decrease by 8% to 18% between the fourth and eighth decades (Tab. 1). When gait speed slows below 1.0 to 1.2 m/s, elderly people are reported to have more difficulty crossing the street safely before the traffic light changes.4,5


View this table:
[in this window]
[in a new window]

  		Table 1. Gait Speed Reference Values

 
As an outcome measure, gait speed has been shown to be a predictor of functional decline, nursing home placement, and mortality.6–9 Specifically, a decrease in gait speed of 0.1 m/s has been associated with a 10% decrease in the ability to perform instrumental activities of daily living.10 Potter et al11 reported that elderly people with a gait speed of <0.25 m/s were more likely to be dependent in one or more activities of daily living. Additionally, slowed gait speed in the older adult population has been related to an increased risk for falls,12–16 which, in turn, often leads to a loss of independent living and to institutionalization.17

The Guide to Physical Therapist Practice defines therapeutic exercise as "the systematic performance or execution of planned physical movements, postures, or activities intended to enable the patient/client to: (1) remediate or prevent impairments, (2) enhance function, (3) reduce risk, (4) optimize overall health, and (5) enhance fitness and well being."18(p112) Specific interventions listed in the Guide to Physical Therapist Practice as therapeutic exercise include aerobic training, balance/coordination training, body mechanics and postural stabilization, flexibility exercise, gait training, relaxation techniques, and strength training.18

Lower-extremity therapeutic exercise has been reported to improve muscle force-generating capacity and flexibility, which are needed for gait.19 Thus, improving lower-extremity muscle force and flexibility, along with aerobic fitness and upright balance, are expected to result in an improvement in gait speed. Research findings, however, do not consistently support a relationship between therapeutic exercise and gait speed in community-dwelling elderly people.20 The exercise mode (weight training or aerobic training), intensity of exercise (high, moderate, or low), and dosage required to have an effect on gait speed in this population have not been fully explored.21 Dosage was defined as duration of an exercise section multiplied by the frequency of exercise bouts per week.

The purpose of this meta-analysis was to conduct a systematic review of the published literature from 1995 to 2003 on the effect of therapeutic exercise on gait speed in community-dwelling elderly people. Our hypotheses were the following:

Hypothesis 1: Therapeutic exercise will have a significant positive effect on habitual walking speed in community-dwelling elderly people.

Hypothesis 2: Therapeutic exercise will have a significant positive effect on fast walking speed in community-dwelling elderly people.

Hypothesis 3: High-intensity exercise (strength training programs performed at 60%–80% of the 1-repetition maximum [RM] level or combination training programs performed at 70%–85% of heart rate reserve [HRR] or at 80% of age-predicted heart rate maximum [APHRM]) and high-dosage exercise (180 minutes of exercise per week; 60 minutes of treatment, 3 times a week) will have a significantly greater effect on walking speed than exercise performed at low intensity or low dosage.


   Method
Top
 Abstract
 Introduction
 Method
Results
 Discussion
 Conclusion
 Appendix
 References
 
We searched the literature published in English between 1995 and 2003 (see Fig. 1 for a flow chart of the search and selection process.) The starting date was chosen because it corresponded to a time when the literature on aging was beginning to emphasize the effect of exercise on physical performance. The initial search uncovered 25,000 articles that might be applicable to the meta-analysis. Inclusion criteria for the studies required subjects to be 60 years of age or older, community dwelling, and ambulatory with or without an assistive device. In addition, the studies had to have clearly defined exercise interventions and either sufficient data or test statistics to obtain an effect size of their findings. Exclusion criteria eliminated review articles and other meta-analyses or studies in which the participants were selected based on medical conditions (eg, stroke, Parkinson disease, osteoarthritis, fractures, peripheral vascular disease).


Figure 1
View larger version (15K):
[in this window]
[in a new window]

  		Figure 1. Flow chart of process used to search for and select studies of the effect of therapeutic exercise on habitual and fast gait speeds.

 
Using these criteria, 117 abstracts appeared relevant for inclusion in the study. Two independent reviewers read the full text of each of these 117 articles and determined that only 33 articles met the inclusion criteria for the meta-analysis.22–54 Of the 84 excluded articles, 76% had no gait speed measures, 20% had not used community-dwelling elderly people as subjects, and 11% had a poorly described exercise intervention or no exercise intervention. Using a systematic review process and criteria established by content experts, the 2 reviewers then independently abstracted the data from the 33 studies. The abstracted data included: subject demographics, sample size, study design, interventions, and specific data on the outcome measures, data analyses, and study results. Any disagreement between the reviewers was discussed until agreement was reached. In addition, a third, independent reviewer was given a random sample of 5 articles and used the same procedure to check the accuracy of the data abstraction. Agreement was 100% among the 3 reviewers for data abstraction in this random sample. When the reviewers had questions about data or results in an article, the authors were asked for clarification or to provide the data required for our analysis. Once all of the appropriate data were extracted from the articles, a computerized database of all data was created to facilitate computation and data analysis (see Tabs. 2 and 3 for the descriptions of data used for the analysis of habitual and fast gait speeds, respectively.)


View this table:
[in this window]
[in a new window]

  		Table 2. Studies Included in the Analysis of Habitual Gait Speed

 

View this table:
[in this window]
[in a new window]

  		Table 3. Studies Included in the Analysis of Fast Gait Speed

 
Participant Demographics

The participants in the 33 included studies were community-dwelling, ambulatory elderly people over 59 years of age with minimal reported limitations in their ability to walk either with or without an assistive device. Only 2 studies reported participants’ use of an assistive device. Comorbidities for the participants were reported in 20 of the studies. Twenty-two studies included both men and women, and 11 studies excluded men. The age of the participants ranged from 60 to 89 years. The demographics of the study participants are shown in Table 4.


View this table:
[in this window]
[in a new window]

  		Table 4. Demographics of Participants in Each Study

 
Study Variables

Gait speed
Of the 33 articles, 17 reported only habitual gait speed, 9 reported only fast gait speed, and 7 reported both habitual and fast gait speeds. Thus, 24 studies (Tab. 2) were included in the analysis of habitual gait speed, and 18 studies (Tab. 3) were included in the analysis of fast gait speed. One article39 reported 3 independent samples, which were included as 3 studies.

Gait speed was measured in a variety of ways across studies (eg, in meters per second, centimeters per second, feet per second, or the distance participants could walk in a predetermined time). Regardless of the measure used, gait speed was converted to meters per second to ensure consistency across studies in calculations used for the meta-analysis. Habitual gait speed was defined as the comfortable walking speed (eg, self-selected speed, preferred speed, normal pace). Fast gait speed was defined in several different ways. In 14 of the 18 studies, subjects were asked to walk at as fast as possible (ie, at maximal gait speed). In the other 4 studies, the instructions were to walk faster than normal (eg, self-selected fast speed or rapid, but safe, speed). If the study did not specify the instructions for obtaining a gait speed, we assumed that participants were asked to walk at habitual gait speed, which was more typically reported. When the reported gait speed was not specified as habitual or fast in the study, gait speed values were compared with age-referenced values to verify speed assignment.2,3 When a study reported data for both habitual and fast gait speeds, we included the appropriate data from the study in our analyses. Reference values and previous research provided support to analyze the 2 gait speeds separately.55

Therapeutic exercise intervention
Most of the therapeutic exercise sessions occurred in a facility-based setting; however, the intervention occurred entirely at home in 5 studies and in both settings in 2 studies. All except 2 studies32,46 reported that the exercise intervention lasted for 8 to 12 weeks, although the duration of the programs ranged from 2 to 52 weeks. The frequency of exercise intervention ranged from 3 times per week in 19 of the studies, 2 times per week in 6 studies, and 3 times a week for the first week followed by 2 times or 1 time per week thereafter in 4 studies.

The type of therapeutic exercise varied across studies and included various forms of strength, aerobic, flexibility, balance, gait, or relaxation training, or Tai Chi. Because multiple forms of exercise were used, the therapeutic exercise interventions were categorized into 2 groups for statistical analysis. The first group used only strength training and consisted of 13 studies (8 for habitual gait speed and 5 for fast gait speed). For the second group, we found that pure forms of aerobic training were not used. Instead, we found that multiple forms of therapeutic exercise, including aerobic training, stretching or flexibility exercise, balance and relaxation training, and Tai Chi were used. Therefore, we placed the studies using aerobic exercise plus other interventions into a group we called "combination training." Combination training was used in 29 of the studies, 16 studies examined changes in habitual gait speed and 13 studies examined changed in fast gait speed (Tabs. 2 and 3). We did not include warm-up or cool-down as a component of the exercise intervention because they were included in almost every study. The Appendix contains descriptions of the type of exercise used in the habitual and fast gait speed studies.

Exercise intensity and dosage
The effect of exercise intensity was examined in the 24 habitual gait speed studies. Participants in 10 studies performed a high-intensity exercise program, participants in 5 studies performed a moderate-intensity exercise program, and those in 9 studies performed a low-intensity exercise program (Tab. 2). Exercise intensity was categorized based on primary research and guidelines56–58 from the available information and data in the articles. Research results support the use of 60% to 80% of the 1-RM level to produce a strengthening effect in untrained individuals.54 For strength training programs, the exercise programs were classified as high intensity if exercises were performed at 60% to 80% of the 1-RM level. The exercise programs were classified as moderate-intensity strength training when resistance was progressed, but not in a specified manner, and when unsupervised exercise sessions were included. The exercise programs were classified as low-intensity strength training when insufficient information was provided to place them into either of the other groups. For the combination training programs, high-intensity exercise was denoted when participants worked at 70% to 85% of HRR or 80% of APHRM.57 For moderate intensity, 50% to 69% of HRR or 60% to 70% of APHRM was used. Finally, <50% of HRR or <60% of APHRM was considered a low-intensity exercise program.57

As with exercise intensity, the effect of exercise dosage (duration of the exercise sessions and frequency of exercise bouts per week) was examined in the 24 studies for habitual gait speed. Research findings56 and American College of Sports Medicine (ACSM) guidelines on strength training58 recommend that training be done for a minimum of 2 days a week, 4 sets per muscle group, with a warm-up and a cool-down. For aerobic exercise, the literature and ACSM guidelines support the use of 20 to 60 minutes of exercise per session, for 3 to 5 days per week, to produce an effect for endurance training.58,59 To determine exercise dosage, the total minutes of therapeutic exercise per week for each study was calculated from the duration of each exercise session and frequency per week of exercise bouts. The number of repetitions per muscle group was not available for all studies; therefore, it could not be included in the calculations. We arbitrarily decided that 180 minutes of exercise per week (60 minutes of treatment, 3 times a week) would be considered a high-dosage exercise intervention and anything less than 180 minutes of exercise per week would be considered low-dosage exercise intervention. Based on this decision, high-dosage exercise was performed in 14 of the 24 studies measuring habitual gait speed, and low-dosage exercise was performed in the remaining 10 studies. Table 2 provides identification of dosage by study.

Other Data Coding Decisions

Direction of effect
Studies were coded as positive if they reported statistically significant improvements in gait speed for the exercise group compared with the control group. All other studies were coded as negative (Tabs. 2 and 3 show the direction of effect for each study measuring habitual and fast gait speeds, respectively). Twenty-two studies that analyzed habitual gait speed (92%) reported a significant improvement in gait speed and 14 studies that analyzed fast gait speed (78%) reported a significant improvement in gait speed.

Weighting the articles
To account for differences in study quality, we used a modified version of Sackett and colleagues’60 levels of evidence to weight the studies included in the meta-analysis. If a study had a true control group and reported means, mean changes, and standard deviations or reported a focused F-statistic or t-test statistic, it received a weighting of 4, the highest score. If a study used a quasi-control group and reported means, mean changes, and standard deviations or reported a focused F-statistic or t-test statistic, it received a weighting of 3. If the study had no control group but reported means, mean changes, and standard deviations or reported a focused F-statistic or t-test statistic, it received a weighting of 2. Finally, if the study only reported a P value, it received a weighting of 1 regardless of the study design used. One study47 had such a small sample size (n=3) that it also was weighted with a value of 1. For this meta-analysis, 17 studies measuring habitual gait speed were determined to be high quality and weighted with a value of 3 or 4 (Tab. 2). For fast gait speed, 12 studies had a weighting of 3 or 4 (Tab. 3).

Blocking
When there was a significant effect of therapeutic exercise intervention on gait speed, further analyses were done to identify possible underlying factors (ie, exercise intensity and dosage) that may have contributed to this finding.61 As previously described, exercise intervention was blocked into combination or strength training; high, moderate, or low exercise intensity, and finally, high or low exercise dosage.

Data Analysis

The data for habitual and fast gait speeds were analyzed separately. Because meta-analysis requires only a test statistic, some of the reported studies did not provide actual gait speed data, which accounts for some of the empty cells in Tables 2 and 3. Several studies included sufficient description of their methods to support inclusion of the study in the meta-analysis, although they did not include specific data on gait speeds of the included groups. An attempt was made to contact the authors for specific data, but no author responded. For these studies, the weighting was lowered to reflect the omission of data, although the test statistic and sample size reported were sufficient to calculate an effect size for the study. Thus, we were able to include them in the meta-analysis.61 A summary of all relevant statistical data pertaining to each set of analyses is presented in Table 5.


View this table:
[in this window]
[in a new window]

  		Table 5. Results of Meta-analytical Hypotheses Testing for Habitual Gait Speed and Fast Gait Speeda

 
Because effect size, measured as Pearson correlation coefficient (r), is the variable of interest in this meta-analysis and is typically not reported in research articles, we had to calculate the effect size for each study from available data. In addition, standardized r values (Fisher Z [Zr]) and 95% confidence intervals (CIs) around the effect size (r) were calculated for each analysis.62,63

Before the data could be combined, the homogeneity of the effect sizes (r values) for included studies was ascertained.63 As a general test for homogeneity, the distributions of the effect sizes of the 24 habitual gait speed studies and 18 fast gait speed studies were examined visually using a stem-and-leaf display and associated descriptive statistics. As a more definitive test for homogeneity, chi-square analyses were performed on the 2 samples of effect sizes (Tab. 5).61,63,64 Nonsignificant findings (P≥.10) for the chi-square analyses indicated that the effect sizes were sufficiently homogeneous to permit combination.

Once the homogeneity of the effect sizes for the studies was established, the effect sizes were combined using a random-effects model to determine the overall effect (rcombined and Zrcombined) of therapeutic exercise on gait speed. In addition, the level of significance (P≤.05) of this effect size was determined.61 If the effect of therapeutic exercise was significant, the analyses of the blocked studies were undertaken in a similar fashion. Additionally, to better demonstrate the clinical relevance of the findings, a forest plot for each analysis was produced and a binomial effect size display (BESD) and odds ratio for the combined effect sizes were calculated for each analysis.62 The BESD provides a clearer interpretation of the clinical meaningfulness of the size of the r values than the traditionally reported small, medium, and large effect sizes.63

Meta-analyses that combine study statistics are often criticized for publication bias (ie, producing an effect size that is overinflated).65 This criticism is based on the belief that studies that show an absence of statistical significance may be under-reported in the literature. To identify the presence of publication bias in the studies included in the meta-analyses of habitual and fast gait speeds, a funnel plot (scatter plot) of sample size versus effect size was examined.65 Additionally, a fail-safe number, the number of unpublished studies finding an absence of statistical significance that would need to be added to the meta-analysis to drive the P value above the .05 level, was calculated61 (Tab. 5). The fail-safe number was obtained using the file-drawer method.63,65


   Results
Top
 Abstract
 Introduction
 Method
 Results
Discussion
 Conclusion
 Appendix
 References
 
Effect of Therapeutic Exercise on Habitual Gait Speed

A total of 1,302 subjects were included in the 24 studies measuring habitual gait speed. The visual display and the statistical comparison of effect sizes of the 24 studies revealed no significant heterogeneity ({chi}2=23.70, df=23, P=.42); thus, the effect sizes of all 24 studies were combined. In addition, the funnel plot displayed a characteristic funnel shape, suggesting that publication bias was not a major problem for these studies.65 The combined effect size was rcombined=.165 (P=.000, 95% CI=0.112–0.218), which was significant at the P≤.05 level (Tab. 5 and Fig. 2). This effect size is considered to be small by the Cohen convention.66 These results support hypothesis 1; that is, therapeutic exercise has a significant, positive effect on habitual walking speed. In terms of the clinical relevance of these findings, there was a 57% success rate for exercise training to improve habitual gait speed in the study population.61 Stated differently, exercise training was 1.83 times more likely to improve habitual gait speed when compared with the control. Furthermore, exercise training resulted in an overall gait speed change of 0.01 m/s in the 24 studies. Finally, the fail-safe analysis indicated that it would take an additional 109 studies with no significant findings for the P value to no longer be significant (P≤.05 level). These findings support hypothesis 1.


Figure 2
View larger version (18K):
[in this window]
[in a new window]

  		Figure 2. Forest plot of 24 studies measuring habitual gait speed.

 
The factors that may have produced these results (ie, the blocking of studies by type of therapeutic exercise [strength training and combination training], intensity of exercise [high, moderate, and low], and exercise dosage [high and low]) were examined next. As indicated in Table 5, when the effect sizes of the studies in each block were combined, the combined effect size of the studies that used moderate- and low-intensity exercise (14 studies, Ncombined=953) and low-dosage exercise (10 studies, Ncombined=703) had Z-scores that were not significant (P≥.05). Thus, further analysis was not indicated.61,62 The combined effect sizes for the remaining blocks of studies were significant and are presented below (Tab. 5).

Exercise Mode

Combining the 8 strength training studies (n=280) resulted in an overall effect size of rcombined=.145 (P=.017, 95% CI=0.028–0.258), which was significant at P≤.05 (Tab. 5 and Fig. 3). Although the combined effect size was small,66 the success rate for strength training to improve habitual gait speed was 56%. Strength training was 1.61 more likely to improve gait speed as compared with the control. The gait speed change from strength training was 0.02 m/s. The fail-safe number was only 8 studies.


Figure 3
View larger version (20K):
[in this window]
[in a new window]

  		Figure 3. Forest plot of 8 studies using strength training to affect habitual gait speed.

 
The overall effect size of the 16 combination training studies (n=1,022) was rcombined=.176 (P=.002, 95% CI=0.116–0.235) and significant at the P≤.05 level. This again is considered to be a small effect size,66 but the success rate was 58% and the odds ratio was 1.95, indicating that combination training would be more likely to improve habitual gait speed than the control (Tab. 5 and Fig. 4). The gait speed change from combination training was 0.01 m/s. For this group of studies, the fail-safe number was 40 studies. These findings support hypothesis 1.


Figure 4
View larger version (20K):
[in this window]
[in a new window]

  		Figure 4. Forest plot of 16 studies using combination training to affect habitual gait speed.

 
Exercise Intensity

Combined data from the 10 studies (n=349) that used a high-intensity exercise program resulted in a combined effect size of rcombined=.184 (P=.001, 95% CI=0.082–0.285), which is significant at the P≤.05 level and supports hypothesis 3. However, the gait speed change was 0.02 m/s, which is considered small.66 The success rate was 59% and the odds ratio was 2.12, showing that high intensity exercise training would be more likely to improve habitual gait speed compared with the control. The fail-safe number was 32 studies (Tab. 5 and Fig. 5).


Figure 5
View larger version (18K):
[in this window]
[in a new window]

  		Figur 5. Forest plot of 10 studies using high-intensity exercise to affect habitual gait speed.

 
Exercise Dosage

Fourteen studies (n=599) included in the high-dosage exercise training group yielded an overall small effect size66 of rcombined=.190 (P=.001, 95% CI=0.112–0.266) and, therefore, also supported hypothesis 3. As with previous analyses, the use of a high-dosage exercise program would be 58% more successful or 1.89 times more likely to improve habitual gait speed as compared with a control (Tab. 5 and Fig. 6). The 14 studies yielded an overall gait speed change of 0.02 m/s. The fail-safe number was 42 studies.


Figure 6
View larger version (20K):
[in this window]
[in a new window]

  		Figure 6. Forest plot of 14 studies using high-dose exercise to affect habitual gait speed.

 
Effect of Therapeutic Exercise on Fast Gait Speed

The 18 studies (n=752) included in the analysis for fast gait speed were found to be homogeneous ({chi}2=23.28, df=17, P=.14); therefore, the data were combined across studies. Again the funnel plot indicated that publication bias was not a major problem for these studies. The combination of effect sizes from these studies yielded an overall and statistically insignificant effect size of rcombined=.055 (P=.173, 95% CI=–0.017–0.126) (Tab. 5 and Fig. 7). Hypothesis 2 was not supported, and no further analysis was performed on the fast gait speed studies.61


Figure 7
View larger version (33K):
[in this window]
[in a new window]

  		Figure 7. Forest plot of 18 studies of the effect of exercise on fast gait speed.

 

   Discussion
Top
 Abstract
 Introduction
 Method
 Results
 Discussion
Conclusion
 Appendix
 References
 
The results of this meta-analysis show that therapeutic exercise, in general, improves habitual gait speed in community-dwelling elderly people, although the relationship between the 2 factors was not large. A positive effect of therapeutic exercise is consistent with Keysor and Jette’s19 review, which reported that exercise improved strength, flexibility, or endurance in 14 of 21 studies of elderly people. As the fail-safe number was 109 studies, we feel confident about this finding because it is unlikely that there are 109 additional unpublished studies with nonsignificant findings. An overall change in gait speed between 0.01 to 0.02 m/s, however, is not clinically meaningful in the opinion of the authors of this review. Furthermore, it is not possible to determine, at this time, whether this finding was due to the variety of interventions, different gait speed measures used across studies, or some other factor that is not apparent from the review.

Blocking the studies further clarified the initial findings. Although both strength training and combination training improved gait speed, the effect was small for both. However, with a fail-safe number of only 8 studies for strength training, this finding has to be considered tentative. A fail-safe number of 40 studies for combination training, however, suggests more confidence in these results. Furthermore, combination training produced a slightly larger correlation between exercise and habitual gait speed than did strength training. Perhaps because of the specificity of training principle, gait speed would be expected to improve when walking is part of the training. Specificity may explain why our results differ from those of Buchner et al,28 who found no difference in gait speed between combination training and strength training. In their study, cycling, not walking, was the aerobic component for combination training.

Further analysis of the habitual gait speed data revealed that only high-intensity and high-dosage exercise training produced a significant effect on habitual gait speed. Our findings add support to prior research findings suggesting that a high-intensity exercise program is required to produce improvements in strength, power, and functional abilities, including gait, for community-dwelling older adults.67 The meta-analysis findings suggest a dosage-response relationship between therapeutic exercise and improved gait speed. However, the effect size was again small.

As Buchner et al28 suggested, if strength and endurance capacities are adequate prior to exercise, further increases in capacity are not expected to alter basic tasks such as usual walking speed. The small effect of exercise on gait speed of elderly people in our meta-analysis supports this theory. If the population that we examined was sufficiently fit prior to intervention, the literature supports the assumption that habitual gait speed will not be significantly slowed. The exercise intensity or dosage, therefore, may have been insufficient to produce a substantial change in gait speed, which would be one factor to account for the small effect size obtained in the meta-analysis. However, if the participants were very frail elderly people, with slowed gait speed and multiple chronic conditions prior to intervention, the intensity or dosage of exercise offered in the investigations may have been a sufficient training stimulus to invoke change in muscle force-production and gait speed. The former possibility (ie, that the combined samples studied were not frail) is supported by the inclusion criteria for subjects examined across studies. For the most part, subjects lived independently in the community.

The results of this meta-analysis did not support a relationship between therapeutic exercise and changes in fast gait speed in community-dwelling elderly people. This finding is in agreement with Daley and Spinks’13 review of literature on exercise, mobility, and aging, which concluded that exercise influences only slow walking speed in elderly people. It is not possible to determine why there was not a relationship between exercise and fast gait speed. It is plausible that impairments limiting fast gait speed did not change because the intensity or duration of the exercise was inadequate. However, the lack of a relationship may have been due to the variety of fast gait speed instructions used across studies. Future prospective studies could begin to address these issues.

Study Limitations

We calculated exercise dosage from the available data as minutes per week and did not account for total weeks of exercise intervention. Fox et al57 and Chandler and Hadley68 indicated that in order to produce true strength training effects, a minimum of 6 to 8 weeks is required to surpass motor learning and the neural components of strengthening. The majority of our studies used between 8 and 12 weeks of exercise intervention; however, because there was a large variation in program duration, we were unable to incorporate this variability into our calculations of dosage. Thus, our measure of dosage, especially at the low level, may have affected our results. Furthermore, the findings from our study regarding exercise intensity may be limited by the method used to categorize this variable. Research evidence as to what is high- versus low-intensity exercise is only tentative and specific thresholds to distinguish clearly among intensity categories have not been articulated. Thus, only guidelines and expert opinion are available to establish intensity categories. Finally, our inability to distinguish among the various forms of exercise included in the combination training group may have prevented us from capturing subtle differences based on program type. Future research should carefully focus on the specifics of the exercise program to determine which of them actually influence gait speed.

This research provides a framework for future study. Through future research, perhaps we can provide more precise interventions that can have a significant effect on improving the mobility of the elderly population and in doing so can improve their function and quality of life.


   Conclusion
Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusion
Appendix
 References
 
Therapeutic exercise appears to improve gait speed in community-dwelling elderly people, although the effect is rather small. Furthermore, this meta-analysis suggests that exercise type, intensity, and dosage may be important in producing a change in the habitual walking speed of community-dwelling elderly people.


   Appendix
Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusion
 Appendix
References
 


Figure 1
Figure 1
Figure 1
Figure 1
View larger version (437K):
[in this window]
[in a new window]

  		Appendix. Description of Exercises in Studies Included in the Analysis of Habitual and Fast Gait Speedsa

 


   Footnotes
 
Dr Lopopolo and Dr Craik provided concept/idea/research design and project management. All authors provided writing and data analysis. Dr Greco and Dr Sullivan provided data collection. Dr Craik provided facilities/equipment.

This study was approved by the Arcadia University Committee on the Protection of Research Subjects. The rights of human subjects were protected.


   References
Top
 Abstract
 Introduction
 Method
 Results
 Discussion
 Conclusion
 Appendix
 References
 

  1. Judge JO, Ounpuu S, Davis RB III. Effects of age on the biomechanics and physiology of gait. Gait and Balance Disorders. 1996;12:659–678.
  2. 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.[ISI][Medline]
  3. Bohannon RW. Comfortable and maximum walking speed of adults aged 20–79 years: reference values and determinants. Age Ageing. 1997;26:15–19.[Medline]
  4. Langlois JA, Keyl PM, Guralnik JM, et al. Characteristics of older pedestrians who have difficulty crossing the street. Am J Public Health. 1997;87:393–397.[Abstract/Free Full Text]
  5. Hoxie RE, Rubenstein LZ. Are older pedestrians allowed enough time to cross intersections safely? J Am Geriatric Soc. 1994;42:241–244.[ISI][Medline]
  6. Spirduso WW, Cronin DL. Exercise dose-response effects on quality of life and independent living in older adults. Med Sci Sports Exerc. 2001;33:S598–S608.[ISI][Medline]
  7. Guralnik JM, Ferrucci L, Simonsick EM, et al. Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability. N Engl J Med. 1995;332:556–561.[Abstract/Free Full Text]
  8. Gill TM, Williams CS, Tinetti ME. Assessing risk for the onset of functional dependence among older adults: the role of physical performance. J Am Geriatr Soc. 1995;43:603–609.[ISI][Medline]
  9. Brach JS, Van Swearingen JM, Newman AB, Kriska AM. Identifying early decline in physical function in community-dwelling older women: performance-based and self-report measures. Phys Ther. 2002;82:320–328.[Abstract/Free Full Text]
  10. Judge JO, Schechtman K, Cress E. The relationship between physical performance measures and independence in instrumental activities of daily living. J Am Geriatr Soc. 1996;44:1332–1341.[ISI][Medline]
  11. Potter JM, Evans AL, Duncan G. Gait speed and activities of daily living function in geriatric patients. Arch Phys Med Rehabil. 1995;76:997–999.[ISI][Medline]
  12. Rubenstein LZ, Powers CM, MacLean CH. Quality indicators for the management and prevention of falls and mobility problems in vulnerable elders. Ann Intern Med. 2001;135:686–693.[Free Full Text]
  13. Daley MJ, Spinks WL. Exercise, mobility and aging. Sports Med. 2000;29:1–12.[ISI][Medline]
  14. Sadeghi H, Prince F, Zabjek KF, et al. Knee flexors/extensors in gait of elderly and young able-bodied men (II). Knee. 2002;9:55–63.[ISI][Medline]
  15. Prince FF, Corriveau H, Hebert R, Winter DA. Gait in the elderly. Gait Posture. 1997;5:128–135.[ISI]
  16. Bootsma-van der Wie A, Gussekloo J, de Craen AJM, et al. Common chronic diseases and general impairments as determinants of walking disability in the oldest-old population. J Am Geriatr Soc. 2002;50:1405–1410.[ISI][Medline]
  17. Rogers ME, Rogers NL, Takeshima N, Islam MM. Methods to assess and improve the physical parameters associated with fall risk in older adults. Prev Med. 2003;36:255–264.[ISI][Medline]
  18. Guide to Physical Therapist Practice. 2nd ed. Phys Ther. 2001;81:9–744.[ISI][Medline]
  19. Keysor JJ, Jette AM. Have we oversold the benefit of late-life exercise? J Gerontol A Biol Sci Med Sci. 2001;56:M412–M423.[Abstract/Free Full Text]
  20. Jette AM, Lachman M, Giorgetti MM, et al. Exercise—it’s never too late: the strong-for-life program. Am J Public Health. 1999;89:66–72.[Abstract/Free Full Text]
  21. Fiatarone MA. Fitness and function at the end of life. J Am Geriatr Soc. 1995;43:1439–1440.[ISI][Medline]
  22. Barnett A, Smith B, Lord SR, et al. Community-based group exercise improves balance and reduces falls in at-risk older people: a randomized controlled trial. Age Ageing. 2003;32:407–414.[Abstract/Free Full Text]
  23. Barrett C, Smerdely P. A comparison of community-based resistance exercise and flexibility exercise for seniors. Aust J Physiother. 2002;48:215–219.[ISI][Medline]
  24. Bean J, Herman S, Keily DK, et al. Weighted stair climbing in mobility-limited older people: a pilot study. J Am Geriatr Soc. 2002;50:663–670.[ISI][Medline]
  25. Berg WP, Lapp BA. The effect of a practical resistance training intervention on mobility in independent, community-dwelling older adults. J Aging Phys Act. 1998;6:18–35.[ISI]
  26. Brandon LJ, Boyette LW, Gaasch DA, Lloyd A. Effects of lower extremity strength training on functional mobility in older adults. J Aging Phys Act. 2000;8:214–227.[ISI]
  27. Brown M, Sinacore DR, Ehsani AA, et al. Low-intensity exercise as a modifier of physical frailty in older adults. Arch Phys Med Rehabil. 2000;81:960–985.[ISI][Medline]
  28. Buchner DM, Cress ME, de Lateur BJ, et al. The effect of strength and endurance training on gait, balance, fall risk, and health services use in community-living older adults. J Gerontol A Biol Sci Med Sci. 1997;52:M218–M224.[Abstract]
  29. Cress ME, Buchner DM, Questad KA, et al. Exercise: effects on physical performance in independent older adults. J Gerontol A Biol Sci Med Sci. 1999;54:M242–M248.[Abstract]
  30. Earles DR, Judge JO, Gunnarsson OT. Velocity training induces power-specific adaptations in highly functioning older adults. Arch Phys Med Rehabil. 2001;82:872–878.[ISI][Medline]
  31. Greendale GA, Salem GJ, Young JT, et al. A randomized trail of weighted vest use in ambulatory older adults: strength, performance, and quality of life outcomes. J Am Geriatr Soc. 2000;48:305–311.[ISI][Medline]
  32. Hickey T, Wolf FM, Robins LS, et al. Physical activity training for functional mobility in older persons. J Appl Gerontol. 1995;14:357–371.[Abstract/Free Full Text]
  33. Hunter GR, Treuth MS, Weinsier RL, et al. The effects of strength conditioning on older women’s ability to perform daily tasks. J Am Geriatr Soc. 1995;43:756–760.[ISI][Medline]
  34. Kerschan K, Alacamlioglu Y, Kollmitzer J, et al. Functional impact of unvarying exercise program in women after menopause. Am J Phys Med Rehabil. 1998;77:326–332.[ISI][Medline]
  35. King MB, Whipple RH, Gruman CA, et al. The performance enhancement project: improving physical performance in older persons. Arch Phys Med Rehabil. 2002;83:1060–1069.[ISI][Medline]
  36. Krebs DE, Jette AM, Assmann SF. Moderate exercise improves gait stability in disabled elders. Arch Phys Med Rehabil. 1998;79:1489–1495.[ISI][Medline]
  37. McCool JF, Schneider JK. Home-based leg strengthening for older adults initiated through private practice. Prev Med. 1999;28:105–110.[ISI][Medline]
  38. Peel C, Utsey C, MacGregor J. Exercise training for older adults with limitations in physical function. J Aging Phys Act. 1999;7:62–75.[ISI]
  39. Puggaard L. Effects of training on functional performance in 65, 75, and 85 year-old women: experiences deriving from community based studies in Odense, Denmark. Scand J Med Sci Sports. 2003;13:70–76.[ISI][Medline]
  40. Sayers SP, Bean J, Cuoco A, et al. Changes in function and disability after resistance training: does velocity matter? A pilot study. Am J Phys Med Rehabil. 2003;82:605–613.[ISI][Medline]
  41. Schlicht J, Camaione DN, Owen SV. Effect of intense strength training on standing balance, walking speed, and sit-to-stand performance in older adults. J Gerontol A Biol Sci Med Sci. 2001;56:M281–M286.[Abstract/Free Full Text]
  42. Sharpe PA, Jackson KL, White C, et al. Effects of a one-year physical activity intervention for older adults at congregate nutrition sites. Gerontologist. 1997;37:208–215.[Abstract]
  43. Sipila S, Multanen J, Kallinen M, et al. Effects of strength and endurance training on isometric muscle strength and walking speed in elderly women. Acta Physiol Scand. 1996;156:457–464.[ISI][Medline]
  44. 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.[ISI][Medline]
  45. Skelton DA, McLaughlin AW. Training functional ability in old age. Physiotherapy. 1996;82:159–167.
  46. Steadman J, Donaldson N, Kaira L. A randomized controlled trial of an enhanced balance training program to improve mobility and reduce falls in elderly patients. J Am Geriatr Soc. 2003;51:847–852.[ISI][Medline]
  47. Stylianides GA. The effects of resistance training programs for elderly women on balance, gait, and strength [abstract]. Dissertations and Abstracts. 1998.
  48. Topp R, Mikesky A, Dayhoff NE, Holt W. Effect of resistance training on strength, postural control, and gait velocity among older adults. Clin Nurs Res. 1996;5:407–427.[Abstract/Free Full Text]
  49. Verfaillie DF, Nichols JF, Turkel E, Hovell MF. Effects of resistance, balance, and gait training on reduction of risk factors leading to falls in elders. J Aging Phys Act. 1997;5:213–228.[ISI]
  50. Von Heideken Wagert P, Littrand H, Johansson A, et al. Jumping exercises with and without raloxifene treatment in healthy elderly women. J Bone Miner Metab. 2002;20:376–382.[ISI][Medline]
  51. Westhoff MH, Stemmerik L, Boshuizen HC. Effects of a low-intensity strength-training program on knee-extensor strength and functional ability of frail older people. J Aging Phys Act. 2000;8:325–342.[ISI]
  52. Williams K, Gill DL, Butki B, Kim BJ. A home-based intervention to improve balance, gait and self-confidence in older women. Activities Adapt & Aging. 2000;24:57–71.
  53. Wolfson L, Whipple R, Derby C, et al. Balance and strength training in older adults: intervention gains and tai chi maintenance. J Am Geriatr Soc. 1996;44:498–506.[ISI][Medline]
  54. Worm CH, Vad E, Puggaard L, et al. Effects of a multicomponent exercise program on functional ability in community-dwelling, frail older adults. J Aging Phys Act. 2001;9:414–424.[ISI]
  55. Bohannon RW, Andrews AW, Thomas MW. Walking speed: reference values and correlates for older adults. J Orthop Sports Phys Ther. 1996;24:86–90.[ISI][Medline]
  56. Rhea MR, Alvar BA, Burkett LN, Ball SD. A meta-analysis to determine the dose response for strength development. Med Sci Sports Exerc. 2003;35:456–464.[ISI][Medline]
  57. Fox E, Bartels R, Billings C, et al. Frequency and duration of interval training programs on changes in aerobic power. J Appl Physiol. 1975;38:481–484.[Abstract/Free Full Text]
  58. American College of Sports Medicine. ACSM’s Guidelines for Exercise Testing and Prescription. 6th ed. Philadelphia, Pa: Lippincott Williams & Wilkins; 2000.
  59. Faria I. Cardiovascular response to exercise as influenced by training of various intensities. Res Q. 1970;41:44–50.[ISI][Medline]
  60. Sackett DL, Strauss SE, Richardson WS, et al. Evidence-Based Medicine: How to Practice and Teach EBM. 2nd ed. Edinburgh, United Kingdom: Churchill Livingstone; 2000.
  61. Rosenthal R. Meta-Analytic Procedures for Social Research. Rev ed. Newbury Park, Calif: Sage Publications; 1991.
  62. Rosnow RL, Rosenthal R. Computing contrasts, effect sizes and counternulls on other people’s published data: general procedures for research consumers. Psychol Methods. 1996;1:331–340.[ISI]
  63. Rosenthal R. Writing meta-analytic reviews. Psychol Bull. 1995;118:183–192.[ISI]
  64. Rosnow RL, Rosenthal R. Effect sizes for experimenting psychologists. Can J Exp Psychol. 2003;57:221–237.[ISI][Medline]
  65. Begg C. The Handbook of Research Synthesis. New York, NY: Russell Sage Foundation; 1994.
  66. Portney LG, Watkins MP. Foundations of Clinical Research: Application to Practice. 2nd ed. Engelwood Cliffs, NJ: Prentice Hall Health; 2000.
  67. Province ME, Hadley EC, Hornbrook MC, et al. The effects of exercise on falls in elderly patients: a preplanned meta-analysis of the FICSIT Trials. Frailty and Injuries: Cooperative Studies of Intervention Techniques. JAMA. 1995;273:1341–1347.[Abstract]
  68. Chandler JM, Hadley ES. Exercise to improve physiologic and functional performance in old age. Gait and Balance Disorders. 1996;12:761–784.



This article has been cited by other articles:


Home page
 ptjournalHome page
K. S Rudolph, L. C Schmitt, and M. D Lewek
Age-Related Changes in Strength, Joint Laxity, and Walking Patterns: Are They Related to Knee Osteoarthritis?
Physical Therapy, November 1, 2007; 87(11): 1422 - 1432.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Material
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when Rapid Responses are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lopopolo, R. B
Right arrow Articles by Mangione, K. K
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lopopolo, R. B
Right arrow Articles by Mangione, K. K

HOME