| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
1 Department of Rehabilitation Medicine, Emory University School of Medicine, Atlanta, Georgia.
2 Department of Biostatistics, Rollins School of Public Health, Emory University, Atlanta, Georgia.
3 Geneva University Hospitals, Switzerland.
Address correspondence to Steven L. Wolf, PhD, PT, Department of Rehabilitation Medicine, Emory University School of Medicine, 212 C Rehabilitation Center, 1441 Clifton Rd. NE, Atlanta, GA 30322. E-mail: swolf{at}emory.edu
 |
A |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Methods. This 48-week randomized clinical trial was provided at 10 matched pairs of congregate living facilities in the Atlanta metropolitan area to 291 women and 20 men, who were transitionally frail, 
70 years old, and had fallen at least once within the past year. Pairs of facilities were randomized to either TC exercise (n = 158) or WE (control) interventions (n = 153) over 48 weeks. Physical performance (freely chosen gait speed, reach, chair-rises, 360° turn, picking up an object from the floor, and single limb support) and hemodynamic outcomes (heart rate and blood pressure) were obtained at baseline and after 4, 8, and 12 months.
Results. Mean percent change (baseline to 1 year) for gait speed increased similarly in both cohorts (TC: 9.1% and WE: 8.2%; p =.78). However, time to complete three chair-rises decreased 12.3% for TC and increased 13.7% for WE (p =.006). Baseline to 1 year mean percent change decreased among TC and increased within WE cohorts for: body mass index (–2.3% vs 1.8%; p <.0001), systolic blood pressure (–3.4% vs 1.7%; p =.02), and resting heart rate (–5.9% vs 4.6%; p <.0001).
Conclusions. TC significantly improved chair-rise and cardiovascular performance. Because TC training reduced fall occurrences in this cohort, factors influencing functional and cardiovascular improvements may also favorably impact fall events.
Although reducing falls is important to older adults who are becoming frail, exploring the extent to which TC will affect hemodynamic outcomes and physical performance in this cohort is also important, because mobility and health-related quality of life are also impacted by these outcomes. This article explores the extent and time course over which TC impacts measures of physical performance and cardiovascular function in this cohort.
|
METHODS |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Medical history and physical examinations were obtained at baseline to rule out the possibility of occult cardiovascular disease or other medical or psychological conditions that could adversely influence participation. Participants averaged 5.6 comorbidities, the most prevalent of which included osteoarthritis, visual dysfunction, and hypertension.
Interventions
The TC intervention consisted of 1-hour group exercise conducted twice weekly. Participants were asked to complete and return weekly exercise logs and were taught behavioral skills and strategies relevant to each movement form. Six of the 24 simplified TC forms were used. All TC exercise was standardized by having the two instructors practice with one another until their execution of the movement forms to be taught in each class were identical.
Participants progressed from often being dependent on assistive devices for upright support to performing 2 continuous minutes of TC without support. "Intense" TC was defined as two sessions per week at increasing durations starting at 60 minutes of contact time and progressing to 90 minutes over 48 weeks. In addition, participants were asked to supplement their TC training with home-based exercise, with the goal of exercising 4–5 days per week, 10–30 minutes per day. The actual "work" time, exclusive of warm-up and cool-down, progressed from approximately 10 to 50 minutes.
The WE program was given at participating facilities for an hour each week and consisted of prepared lectures on falls prevention; exercise and balance; diet and nutrition; pharmacological management; legal issues relevant to health; age-related changes in body function; and mental health (stress, depression). The total time for individual attention from each instructor to participants in each group was comparable. All outcome measures were obtained at baseline and after 4, 8, and 12 months of the interventions.
Physical Performance
The participants' physical status was measured by performance-based tests, chosen because of their previously established reliability and validity, simplicity to administer, and clinical utility and included gait speed, functional reach test, time to rise from a chair three times, time to complete a 360° turn, time required to pick up an object from the floor, and duration of single limb support with eyes open and closed. All tests were performed within a single testing session with adequate rest between tests. The average of three measurements was recorded for each task. All testing was undertaken by evaluators who were not involved in either the TC or WE interventions and who were blinded to intervention allocation. Participants were instructed not to disclose the intervention they received to the evaluators.
Dropouts
Dropouts were defined as participants who missed more than 8 consecutive weeks of the intervention. Of 311 participants randomized, 12 individuals randomized to TC and 12 randomized to WE chose not to participate after randomization. One individual with Parkinson's did not contribute data beyond baseline. The remaining 286 participants (92%) provided longitudinal data, but 69 participants did not complete the 48-week intervention. Baseline characteristics of the study completers and noncompleters for the two intervention groups have been previously reported (15).
Statistical Analysis
The primary analyses of the data were performed according to participants' original treatment assignment for all 311 participants randomized. Baseline measurements of TC and WE participants were compared using a permutation test in which the unit of randomization was the independent living facility pair.
Repeated-measures analysis included participant level characteristics and congregate living facilities clustering. Participants were clustered within living facilities with living facility pairs as a random effect and the participant as the experimental unit. Repeated-measures analyses, using mixed linear models, were performed for six physical performance, three hemodynamic, and three anthropomorphic measurements and percent change from baseline with a means model using SAS Proc Mixed (version 8; SAS Institute, Cary, NC). An unstructured variance–covariance form in repeated measurements was assumed for each outcome. Variance components included the cluster and the between- and within-participant components (16). The intra-class correlation coefficient (ICC) ranged from 0.005 (functional reach) to 0.12 (gait speed). Because the data for chair-rises were right-skewed, natural log transformations were performed prior to analysis and results were reported based on back transformation of the log values to the usual arithmetic scale. Sensitivity analyses to evaluate the implications of missing observations included "completers" analysis—that is, an analysis including only participants who completed follow-up, (for chair-rise data) last observation carried forward, and baseline observation carried forward. These analyses showed similar results to intent-to-treat repeated-measures analyses which suggest that the missing data were noninformative.
Statistical tests were two-sided. A Bonferroni adjustment (p <.0125) was used to compare TC versus WE at each of the four time points. Statistical tests were not adjusted for multiple outcomes.
|
RESULTS |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
|
|
|
Mean total body weight for TC participants decreased by 1.49 kg (p <.0001) and by only.29 kg (p =.39) for WE individuals. The mean percent change from baseline to 1 year was –2.2% and –0.40% in the TC and WE cohorts, respectively (p =.01, Figure 1C).
Hemodynamic Changes
Systolic blood pressure (SBP) also changed in significantly different directions during the interventions (p =.005, Table 2). Mean SBP declined over the intervention for the TC cohort (p =.0002) but not for the WE cohort (p =.09). Mean SBP was significantly lower in the TC cohort than in the WE group at 1 year (p <.0001, Table 2). The mean percent change from baseline to 1 year was –3.4% in the TC cohort and 1.7% in the WE cohort (p =.017, Figure 1D).
Diastolic blood pressure (DBP) in the study cohorts was consistently different (p =.0009), with TC participants having lower mean DBP than WE participants at every follow-up time point. Figure 1E indicates a significant decline in the percent change from baseline to 4 months in both cohorts (p <.0001).
Resting heart rate changed in significantly different directions for the two study cohorts (p <.0001, Table 2). Mean heart rate was significantly lower in the TC cohort than in the WE cohort at 4 months (p <.0001). The mean difference at 4 months was 6 bpm (95% CI, 4 bpm–9 bpm). The mean percent change from baseline to 4 months in resting heart rate was –6.5% for the TC cohort and 4.9% for the WE cohort (p <.0001, Figure 1F), and remained stable thereafter.
Physical Changes
Mean gait speed increased over time in both cohorts (Table 2, p <.0001). Mean gait speed at 4 months was 1.08 m/s in the TC cohort and 0.99 m/s in the WE cohort (mean difference = 0.09 m/s; 95% CI, 0.02–0.17; p =.017), and persisted through 8 months. However, when expressed as a mean percent change over time, the increases from baseline to 4 months were 9.4% for the TC cohort and 6.0% for the WE cohort (p =.20, Figure 1G) and 9.1% and 8.2%, respectively, at the 12-month evaluation (p =.78).
The mean time necessary for completion of the three chair-stands was similar at baseline (Table 1) and at each follow-up visit for the two cohorts (Figure 1H, Table 2). Expressed as a mean percent change from baseline, the three chair-stands time decreased in the TC cohort (p =.005, Figure 1F) and was substantially different after 8 months. At month 12, the mean percent change from baseline was –12.3% in the TC cohort and 13.7% in the WC cohort (p =.006). Thirty TC and 29 WE participants could not complete the chair-stands at baseline. At a subsequent evaluation after the baseline session, eight of the TC and five of the WE individuals within this subgroup were able to do so. Despite the missing data, sensitivity analyses consistently found that TC improved chair-rise performance.
Functional reach was consistently different between cohorts (p =.002, Table 2) with TC participants having a significantly greater functional reach than WE participants, but the percent increase at 12 months was similar for the two cohorts (Figure 1I).
Among other physical measures, the percent change from baseline to 12 months was similar for the two cohorts (p =.95 for 360° turn; p =.23 for pick up an object, data not shown), with no differences between groups seen across time for the single limb support data.
|
DISCUSSION |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
After 4 months, TC participants demonstrated a reduced BMI with a loss in total body weight of almost 1.5 kg, whereas no decrease was observed in the WE participants. A decrease in height was seen among the WE group but remained unchanged in TC participants suggesting that the training contributed to participants' standing tall and losing weight. This notion seems plausible because TC has been shown to improve static and dynamic postural control in older persons with less (10,17–19) or more (20,21) experience in practicing TC.
The significant decreases in SBP and resting heart rate among our participants are comparable to those from Hong and colleagues (22), who reported that the average resting heart rates for older healthy TC practitioners with at least 10 years of experience and inactive age- and body-type matched controls were 69.01 ± 9.6 and 76.88 ± 9.32, respectively. These remarkably similar findings suggest that transitionally frail older persons retain the ability to adapt to TC training as well as do experienced robust individuals.
Our TC training protocol was not uniformly effective at improving physical function and mobility. Both cohorts showed comparable changes in gait speed and functional reach over time. In contrast to our observations, Mak and Ng (23) explored the impact of TC on adults (mean age 62 years vs 80.8 years in the present study) who had been practicing for years, and reported that they had faster gait speed (1.11 m/s vs 1.01 m/s), greater functional reach, and improved postural control as measured by sway parameters during single-leg stance than did nonpractitioners. Our participants practiced different forms of TC and had no previous experience with this exercise. Thus the intensity of our TC training may not have been sufficient to elicit significant improvements in gait speed or postural control.
TC training did, however, produce a significant mean percent reduction from baseline in chair-stand time by the eighth month of training but not in other functional measures. However, given that 30 TC individuals could not chair-stand before the intervention, improved chair-rise performance among all TC participants might not apply to the frailest older adults. Among our participants, chair-stands without the use of arm support place a concurrent demand on lower extremity strength and phasic changes in balance that might not be manifest in the other performance measures.
Our participants required more time or intense training to effect behavioral, hemodynamic, functional, or fall-related improvement. This 4-month latency may be related to physical (improved strength and balance) and psychological (a sense of mastery and perceived physical competence) factors known to be associated with latent improvements in physical performance (24–26). Interestingly, 4 months of TC training were required before these participants demonstrated a significant reduction (40%) in fall rates (12) and 8 months were required before significant improvements in balance confidence were achieved (15).
The primary study limitation is missing outcome data. Although changes observed over time in the TC group might be due to differential dropout of more frail participants, the intention-to-treat analyses and sensitivity analyses were consistent and suggested that TC improved chair-rise performance and cardiovascular outcomes. Despite the consistency of study results, the findings may not generalize to the entire population of transitional older people or to those who could not chair-stand at baseline.
Summary
Forty-eight weeks of bi-weekly TC exercise among transitionally frail older adults results in significant, beneficial changes in several physiological parameters that are manifest after at least 4 months of training. Future research examining the effects of TC training designed to slow the progression of frailty should better specify the inter-relationships between intensity, duration, and type of TC training and the magnitude of comorbidities confronting participants. Such studies might entertain using a contemporary, operational definition of "intermediate frail" derived from a recent, exploratory study by Fried and colleagues (27) whose participants, like many of our own, were at risk for disability, hospitalization, nursing home placement, and/or adverse health outcomes, such as falls.
|
A |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
We are indebted to Nana Freret, RN, MS, for her efforts at coordinating this project; Susan Murphy for her assistance in project coordination; Lois Ricci, RN, EdD, and Carol Holbert, PhD, RN, for provision of the wellness program; Tingsen Xu, PhD, and Tai Chi Master for devising and implementing the Tai Chi program; and Gregory Bailey for provision of Tai Chi exercise classes at half the facilities.
|
F |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Received May 11, 2005
Accepted August 28, 2005
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. Ferrucci and E. M. Simonsick A Little Exercise J. Gerontol. A Biol. Sci. Med. Sci., November 1, 2006; 61(11): 1154 - 1156. [Full Text] [PDF]
|
|
|
|
|
|
The LIFE Study Investigators* [*See Appendix for L Effects of a Physical Activity Intervention on Measures of Physical Performance: Results of the Lifestyle Interventions and Independence for Elders Pilot (LIFE-P) Study J. Gerontol. A Biol. Sci. Med. Sci., November 1, 2006; 61(11): 1157 - 1165. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
