Tai Chi and perceived health status in older adults who are transitionally frail: a randomized controlled trial.

About 30% of older adults living in the community fall each ear. (1) Of those who fall, approximately 5% will sustain fractures, and an additional 5% to 11% will sustain other serious injuries. (2) Among older adults who sustain hip fractures, 90% of these fractures occur following a fall. (3) Whether experiencing injurious or noninjurious falls, many older adults will never return to preinjury function. The resulting functional limitations also are associated with depression, fear of falling, and declines in health status and related quality of life.

Some studies (4-6) have demonstrated the effectiveness of tai chi (TC), a type of Chinese exercise derived from a form of martial arts, in maintaining physical functioning and reducing falls among older adults. The slow, rhythmic movements that characterize TC also provide a holistic approach to exercise that combines mind and body experiences. This meditative aspect of TC could result in general improvements in health status and quality of life, in addition to improvements in physical functioning.

Several studies have demonstrated improvements in quality of life among older adults with osteoarthritis, (7) patients with chronic heart failure, (8) people who have had breast cancer, (9) and patients with fibromyalgia (10) following TC practice. Although a few studies on older adults who are at risk for falls have examined the effect of TC on physical health, (6,11) the effect of TC on the multiple dimensions of health status has not been well studied. In this study, a single-blind, randomized clinical trial examining the benefits of TC, we hypothesized that a group of older women who are transitionally frail would report greater improvements in perceived health status following an intense 48-week TC program than would women taking part in a wellness education (WE) program of similar duration.

Method

Subjects and Procedure

The study methods have been described in detail elsewhere. 12 Briefly, study subjects were recruited from 20 independent senior living facilities in the greater Atlanta, Ga, area between December 1997 and August 1999. Participants were at least 70 years of age, reported one or more falls during the last year, were ambulatory (with or without an assistive device), and were classified as transitioning to frailty on the basis of criteria developed by Speechley and Tinetti. (13)

Speechley and Tinetti defined adults as vigorous, frail, or transitionally frail on the basis of 10 attributes: age, gait and balance, walking activity for exercise, other physical activity for exercise, presence or absence of depression, use of sedatives, near-vision status, upper- and lower-extremity strength (force-generating capacity), and lower-extremity disability. Adults who are vigorous are defined as those who have at least 3 vigorous and no more than 2 frail attributes. Adults who are frail are defined as those who have at least 4 frail attributes and no more than 1 vigorous attribute. Adults who are transitionally frail are those who do not meet the criteria for the frail or vigorous group. Subjects were excluded if they had a severe or trustable medical condition, had significant cognitive impairment (Mini-Mental State Examination score of <24), or had a condition in which physical activity was contraindicated.

Of the 354 people screened, 311 (291 women and 20 men) met eligibility criteria and were enrolled in the study. Further details regarding enrollment and reasons for exclusion were previously reported# Because of sex differences regarding self-perceived health, (14-16) we excluded the 20 men who were part of the original sample (Figure). Of the 291 women who were enrolled, 148 were randomly assigned to participate in the TC intervention and 143 were randomly assigned to participate in the WE intervention. Ten of the women who were randomly assigned to participate in the TC intervention and 11 of the women who were randomly assigned to participate in the WE intervention withdrew from the study before the start of the intervention. Of the 21 women who withdrew, 5 did not want to participate, 5 had poor health status, 3 experienced catastrophic events, 2 died, and 6 did not receive physician approval to participate. One other woman was later excluded because of previously nondiagnosed Parkinson disease. This process resulted in a total of 137 women in the TC group and 132 women in the WE group. During the course of the intervention, 34 participants in the TC group and 30 participants in the WE group discontinued their interventions. The reasons for dropping out are given in the Figure. The study analysis was based on the 269 participants who either completed or participated in the TC or WE intervention.

Eligible subjects who agreed to participate were interviewed and underwent functional assessments at their facilities prior to randomization and intervention. The initial interview was used to characterize the subject's health status and to obtain information regarding demographic and behavioral characteristics, including fear of falling. The functional assessments consisted of performance measures designed to assess muscle strength, range of motion, gait, and balance. Following baseline assessments, participants were randomly assigned to either the TC group or the WE group. Participants were reassessed at 1 year, after completion of the TC and WE programs. A total of 103 participants in the TC group and 102 participants in the WE group completed 1-year follow-up assessments. Study investigators who administered assessments and interviews were unaware of participant intervention groups. Written informed consent was obtained from all subjects before enrollment.

Characteristics of the Study Population

Table 1 summarizes the baseline frailty, demographic, and selected health and functional characteristics of the study participants. The frequency of frailty characteristics ranged from 16.1% to 99.2%. Near-vision impairment (97.8% in the TC group and 99.2% in the WE group) and impaired gait or balance (92.7% in the TC group and 94.7% in the WE group) were the most common frailty characteristics; upper-extremity impairment (16.1% in the TC group and 18.9% in the WE group) was the least common. More than three quarters of the study participants were white and had completed high school, and more than one half were at least 80 years of age. Baseline Sickness Impact Profile (SIP) scores were in the moderate disability range for the TC and WE groups, although baseline SIP scores were significantly lower for the TC group, indicating better perceived health status, for subjects in the TC group.

Interventions

The experimental intervention, TC, consists of slow, rhythmic movements that emphasize trunk rotation, weight shifting, coordination, and gradual narrowing of the lower-extremity stance. Six simplified forms of exercise procedures that best characterized these movements were used. (17) Participants were de-conditioned and were expected to progress slowly. Thus, the goal for the participants was to perform 2 minutes of unassisted TC exercise. The facilities at which the TC intervention was used were divided between 2 instructors. All instruction was standardized by having the 2 instructors practice with one another until their execution of the movement forms to be taught in each class was identical. The TC intervention consisted of 2 sessions per week at increasing durations starting at 60 minutes of contact time and progressing to 90 minutes over the course of 48 weeks. Sessions included warm-up and cool-down periods. Actual TC practice time, excluding warm-up and cool-down periods, progressed from approximately 10 minutes to 50 minutes. However, participants were allowed to proceed at their own pace and could sit or rest as needed. Therefore, the intensity of TC practice time varied somewhat by participant and the capability of each participant on the basis of the physiological effort that she was required to exert. Individual differences could not be monitored.

The control intervention consisted of a WE program. The WE classes were held for 1 hour per week and consisted of instruction on falls prevention; exercise and balance; diet and nutrition; pharmacological management; legal issues relevant to health; changes in body function; and mental health issues, such as stress, depression, and life changes. Interactive materials were provided, but there was no formal instruction in exercise. The total amounts of time for individual attention given to participants in each group by the instructors were comparable.

Variables and Measures

Perceived health status. The SIP is a reliable and valid measure of perceived health status, (18-20) has been used widely across patients with a variety of diseases and injuries, (20) takes 20 to 30 minutes to administer, (21-29) and has been applied extensively to older adults. (23,30-32) Specifically, the reliability and validity of SIP data among older adults have been well demonstrated across a variety of older populations, including nursing home residents who are frail, (33,34) older adults with chronic illnesses, (35) and older veterans. (36) The internal consistency of the overall SIP is .96, with individual category scores ranging from .63 to .90. (37) In addition, the clinical validity of SIP data has been well established through comparisons with clinical data for a variety of situations and diagnoses, including total hip replacement, hyperthyroidism, stroke, angina pectoris, and rheumatoid arthritis. (38) Finally, multiple studies have demonstrated that the SIP is responsive to changes over time. (37) The SIP consists of 136 statements divided into 12 categories: sleep and rest, eating, work, home management, recreation and pastimes, ambulation, mobility, body care and movement, social interaction, alertness behavior, emotional behavior, and communication (Tab. 2). Four of the categories--social interaction, alertness behavior, emotional behavior, and communication--combine to form a psychosocial health dimension. Ambulation, mobility, and body care and movement combine to form a physical health dimension. Participants are asked whether the statements describe themselves today and whether the statements are related to their health. Categories and dimensions can be scored independently or combined to produce an overall summary score. Scores are derived by use of a weighted algorithm. Overall SIP scores and categorical scores range from 0 to 100, with higher scores indicating poorer health status. Scores of less than 4 are found in the general population and indicate no disability, scores of 4 to 9.9 indicate minor disability, scores of 10 to 19.9 indicate moderate disability, and scores of [greater than or equal to] 20 indicate severe disability. (39-41) Differences of 2 or 3 points indicate meaningful differences in function. (42,43)

Self-rated health (SRH). Participants also were asked to rate their health as excellent, very good, good, fair, or poor at baseline and 1-year interviews. Self-rated health as a single indicator has been shown to be associated with mortality (44,45) and with future health and disability. (46,47)

Data Analysis

Baseline characteristics were summarized by use of means and standard deviations for continuous data and frequency distributions for categorical data. Because the distribution of the SIP was significantly skewed, the data were grouped into clinically meaningful categories as described in previous studies. (39-41) Repeated measures ordinal logistic regression (proportional odds) was used to model the various domains within the SIP as the dependent variables for the evaluation of intervention x

time interaction effects. The repeated measures were analyzed by use of generalized estimating equations. A significant interaction effect would indicate a larger change in function in one of the groups between baseline and follow-up. Significance was set at P= .05. Participants who did not complete the intervention were analyzed in the groups to which they initially were assigned by use of an intention-to-treat analysis. We also compared this analysis with a complete-case analysis that included only participants who completed the intervention.

Results

Attendance and Dropouts

Attendance did not differ between the TC participants and the WE participants. The mean attendance for the TC participants was 86% (SD=10.4), and the mean attendance for the WE participants was 82% (SD= 10.4). Dropouts were defined as participants who missed more than 8 consecutive weeks of the intervention; dropping out usually was attributable to a decline in health. Beyond this interval, learning TC movement forms or WE materials already covered by the classes would have impeded the progress of those classes.

Perceived Health Status and SRH

The SIP scores at baseline and at 1 year after the intervention and the group x time interactions (intervention effects), determined by use of an intention-to-treat analysis, are summarized in Table 3. Participants in the TC group reported significantly lower SIP scores in the physical dimension (P=.016) and ambulation category (P=.013). In addition, participants in the TC group reported borderline significant scores in the physical dimension category of body care and movement (P=.051). These scores indicated improved physical functioning (group x time interaction), most notably in ambulation, following 48 weeks of TC practice relative to the findings for the WE group.

No significant intervention effects were observed in the psychosocial dimension or other categorical scores. Because most of the participants were retired, there was little variation in the work subcategory, and although it was used to calculate total SIP scores, it was omitted from categorical analyses. In addition, there were no statistically significant differences within intervention groups over time or between groups at baseline and at times after intervention, except for between-group differences in the physical dimension indicating better physical functioning among the TC participants than among the WE participants (P=.005). The data were reanalyzed after participants who did not complete the TC or WE intervention were excluded. As determined with a complete-case analysis, TC participants had lower scores in the following SIP dimensions and categories: physical dimension (P=.010), ambulation category (P=.013), and body care and movement category (P=.018).

The majority of participants in both the TC and the WE groups rated their health as good or better, with at least 40% rating their health as very good or excellent and less than 30% rating their health as fair or poor at baseline. Self-rated health was not significantly different between the TC participants and the WE participants at baseline. In addition, SRH did not change significantly from baseline to 1 year for either the TC group or the WE group.

Discussion

In this study, we examined the effect of TC exercise on perceived health status and SRH in a group of women who were transitionally frail and [greater than or equal to] 70 years of age. The pre-intervention total SIP score (X [+ or -] SD) of 14.1 [+ or -] 10.0, indicating moderate disability, lies between that of older adults who are healthy (mean SIP score=3.4) (48) and that of nursing home patients who are very frail (mean SIP score=43.5). (49) The major finding of this study was that adults who are transitionally frail who participated in a 48-week TC intervention reported significant improvements in perceived physical health, specifically in the category of ambulation, and borderline significance in the category of body care and movement. Significant changes in perceived health status were not observed for psychosocial health or the independent categories of home management, sleep and rest, and eating. Although we did not expect changes in eating behavior, we had hypothesized that TC would affect other areas of health status.

In previous studies in which the effect of TC on perceived health status was examined, conflicting findings were reported. (38,50,51) These discrepancies may be attributable to differences in the intensity of the intervention, the study population, sample size, the study methodology, or the measure used to assess perceived health status. In only 1 of 3 studies in which the Medical Outcomes Study Short-Form questionnaire (38,50,51) was used to assess health status was a significant improvement in perceived health status reported. That study (50) targeted older adults with balance deficits; in the other 2 studies, (38,51) older adults who were healthy were recruited. In her review of TC interventions, Wu (52) reported that changes in health status may be limited in studies that target older adults who are healthy, in whom baseline health status scores are already high.

The findings of Li and colleagues, (6,11,53) who conducted a randomized trial of TC among physically inactive older adults, lend support to this idea. The initial findings of Li and colleagues supported the conclusion that TC improved perceived physical functioning; however, in a reanalysis stratifying their sample into high and low scores on the basis of baseline physical functioning, Li and colleagues found significant improvements among participants with low baseline physical functioning scores following 6 months of a TC intervention, whereas participants with high baseline physical functioning scores did not show any change in physical functioning, relative to the findings for control participants performing their usual daily activities. 53 Our own findings of improved physical functioning among TC participants in this study and negative findings for older adults who were healthy recruited in the Frailty and Injuries: Cooperative Studies of Intervention Techniques (FICSIT) TC trial (38) also support the conclusion that older adults with lower levels of functioning may be more likely to report improvements in perceived physical health than older adults who are healthy.

Within the physical function dimension, ambulation significantly contributed to improvements in physical functioning, the body care and movement category was borderline significant, whereas no significant differences were reported with mobility. Items in both the ambulation category and the body care and movement category focus on activities that require balance. For example, the ambulation category includes items such as walking up or down hills, walking up and down stairs, and getting around only by using a walker, crutches, cane, walls, or furniture. Body care and movement activities primarily include activities of daily living, many of which require balance to perform well. Such items include standing up, kneeling, stooping, bending down, getting into or out of cars or bathtubs, and maintaining balance. Several studies (52,54,55) have shown improvements in balance following TC. In contrast, mobility items describe an individual's ability to move around within the community versus staying at home. Although balance certainly plays a role in an individual's ability to do so, illnesses and various chronic health conditions also may affect this area, consequently reducing the effect of TC in the area of mobility. For example, subjects may stay home most of the time because they feel too sick to leave home, irrespective of walking and balance capabilities.

Given that TC has meditative components to complement its exercise aspects and because improved physical functioning could affect other areas of health, we hypothesized that TC would result in improved psychosocial health. Previous studies support our hypothesis. Improvements in psychosocial health following TC have been demonstrated among participants with a variety of diagnoses and chronic conditions, including osteoarthritis, (7) breast cancer, (9) and chronic heart failure. (8) Two randomized controlled studies in which the psychological effects of TC on sedentary older adults were examined revealed improvements in life satisfaction and general well-being among TC participants relative to control participants. (56,57) In another randomized controlled study, Kutner et a1 (38) examined the effects of TC on general health status among community-dwelling older adults. Although improvements in general health, mental health, and social functioning were not demonstrated, as measured with the Medical Outcomes Study Short-Form questionnaire, exit interviews revealed that TC participants were more likely than control participants to report benefits from participation and to report that the intervention (TC) had a noticeable effect on their fives. Therefore, our failure to demonstrate improvements in psychosocial health and other aspects of health status, with the exception of physical health, was surprising.

Several factors could account for our findings. First, our study participants were more frail than those in previous studies, and comorbid conditions may have exerted a greater influence on psychosocial health. Second, the SIP may not be sensitive enough to identify modest changes in psychosocial health among older adults. Third, differences in study methodology may account for conflicting results. Further research is needed with psychosocial measures that are sensitive in older adults to better understand whether TC has any direct or indirect effect on psychosocial health or whether the trends that we observed are spurious.

Several factors could account for why our results failed to demonstrate an effect of TC or WE on SRH. First, three quarters of all participants rated their health at baseline as good or better, a finding that supports the positive perceptions of SRH among older adults in previous studies. (58-60) Older adults tend to view their health in context with their peers and may perceive declining functional status as part of the normal aging process. Thus, high baseline scores may have limited our ability to detect changes in SRH. Second, individuals' perceptions of SRH not only may be a reflection of their current health status but also may reflect a more lasting serf-concept based on preexisting beliefs regarding their own health. (61) Therefore, an improvement in physical health may have been only one of several factors that could have influenced participants' assessments of SRH at 1 year. Finally, Bailis and colleagues (61) found that a change in SRH is more likely to be influenced by health status changes that are most important to an individual. To demonstrate changes in SRH, we may have needed to identify a priori those individuals who had a specific goal of improving their physical functioning.

We used an intention-to-treat analysis because it minimizes bias (62) and may provide a more realistic estimate of clinical relevance. Excluding dropouts and participants who do not adhere to a study protocol weakens the unbiased estimate that is gained from randomization. Participants who drop out tend to fare worse that those who complete an intervention, even after controlling for known predictive factors. (62) Thus, the remaining participants would be expected to have a better outcome than those who dropped out. Such a scenario may account for the greater significance in physical health and in body care and movement that we observed with the complete-case analysis. Thus, the use of a complete-case analysis instead of an intention-to-treat analysis may result in an outcome suggesting greater success than is warranted. Finally, an intention-to-treat analysis gives the clinician a more realistic representation of the likely success of an intervention. In reality, not every individual will fully adhere to or complete a treatment regimen. Thus, an intention-to-treat model takes into account issues of nonadherence.

The interpretation of data from this study has some limitations. First, on the basis of the definition developed by Speechley and Tinetti, (13) we classified older adults who met neither the frailty criterion nor the vigorous criterion as transitioning to frailty. This definition included older adults with a broad range of health characteristics and functional abilities. Although mean SIP scores ranked in the moderate level of disability, individual scores varied widely, ranging from no disability to severe disability, suggesting that our definition of transitional frailty may have been imprecise. Recently, Fried and colleagues (63) presented more simplified definitions of frailty and vigor that include a transitional group that they define as "intermediate" frailty. Although this classification system has appeal because of its simplicity, further research is needed to demonstrate its sensitivity and specificity for classifying older adults, especially those who are transitioning to frailty.

Second, participants were recruited from congregate living facilities for a clinical intervention trial and may not be representative of older adults who are transitionally frail and live in the community or those who would not volunteer for an exercise intervention. Third, although the instructor gradually increased work time from 10 minutes to 50 minutes over the 48-week intervention, participants were allowed to rest as needed. Although participants generally increased their work time, with an instructor ratio of 1:15, the instructor was unable to monitor individual changes. Fourth, older adults with cognitive deficits (Mini-Mental State Examination score of <24) were excluded because participants needed to have sufficient cognitive skills to be able to answer interview questions. Fifth, few African-American women were enrolled in the study, despite attempts to enroll greater numbers. Low enrollment may have been primarily attributable to the composition of the congregate living facilities, which was primarily female and white, with the exception of 4 primarily minority facilities. Only women were included in our analysis; therefore, the findings cannot be extended to older men.

Finally, although our analyses of intervention effects took into account baseline differences in SIP scores, the higher baseline SIP scores among the TC participants could have created some unintended biases. Because TC participants had better perceived health, they may have had more optimistic views about their health and may have been more likely to perceive their health as improved after an intervention. On the other hand, previous research (53) suggested a greater benefit among TC participants with lower baseline physical functioning scores, which could have reduced the treatment effects for TC participants relative to WE participants.

Conclusion

Perceived health status measures, such as the SIP, provide a profile of patient-reported dysfunction that encompasses not only physical functioning but also psychosocial health and various aspects of daily function. The SIP is easy to administer and can be an important adjunct to performance measures by providing additional information that is not clinically assessed. Although performance and serf-perceived health are highly correlated, the correlation is by no means perfect. (30) Perceived health and performance may be influenced differently by external factors, such as depression and cognitive status. Thus, an assessment of perceived health can provide a more complete representation of functionality for both the researcher and the clinician. Finally, the SIP can provide useful clinical information not only about the quality of performance, such as "I walk more slowly," but also about the level of mobility, such as "I am getting around only within 1 building."

In summary, the results of this study suggest that an intense program of TC targeting older women who are transitionally frail can have a positive effect on their sell-perceived physical health, primarily in the areas of ambulation and sell-care. Ambulation, sell-care, and movement activities are important functions for maintaining independent lifestyles. More community-based studies are needed to determine whether these improvements are sustained and result in older adults adopting more active lifestyles that have a broader impact on their overall health status and quality of life.

This study was approved by the Emory University Human Investigation Committee.

This research was supported by National Institutes of Health grant AG14767 from the National Institute on Aging and coupons for redeemable products from Kroger Corporation and CVS Pharmacies for each participant upon completion of participation.

The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the funding agency.

This article was received December 1, 2005, and was accepted January 9, 2007.

DOI: 10.2522/ptj.20050378

References

(1) Gregg EW, Pereira MA, Caspersen CJ. Physical activity, falls, and fractures among older adults: a review of the epidemiologic evidence. J Am Geriatr Soc. 2000;48: 883-893.

(2) Kannus P, Parkkari J, Koskinen S, et al. Fall-induced injuries and deaths among older adults. JAMA. 1999;281:1895-1899.

(3) Cumming RG, Nevitt MC, Cummings SR. Epidemiology of hip fractures. Epidemiol Rev. 1997;19:244-257.

(4) Wolf SL, Sattin RW, Kutner M, et al. Intense tai chi exercise training and fall occurrences in older, transitionally frail adults: a randomized, controlled trial. J Am Geriatr Soc. 2003;51:1693-1701.

(5) Li F, Harmer P, Fisher KJ, et al. Tai Chi and fall reductions in older adults: a randomized controlled trial. J Gerontol A Biol Sci Med Sci. 2005;60:187-194.

(6) Li F, Harmer P, McAuley E, et al. An evaluation of the effects of Tai Chi exercise on physical function among older persons: a randomized controlled trial. Ann Behav Med. 2001;23:139-146.

(7) Hartman CA, Manos TM, Winter C, et al. Effects of T'ai Chi training on function and quality of life indicators in older adults with osteoarthritis. J Am Geriatr Soc. 2000;48:1553-1559.

(8) Yeh GY, Wood MJ, Lorell BH, et al. Effects of tai chi mind-body movement therapy on functional status and exercise capacity in patients with chronic heart failure: a randomized controlled trial. Am J Med. 2004;117:541-548.

(9) Mustian KM, Katula JA, Gill DL, et al. Tai Chi Chuan, health-related quality of life and self-esteem: a randomized trial with breast cancer survivors. Support Care Cancer. 2004;12:871-876.

(10) Taggart HM, Arslanian CL, Bae S, et al. Effects of T'ai Chi exercise on fibromyalgia symptoms and health-related quality of life. Ortbop Nurs. 2003;22:353-360.

(11) Li F, Harmer P, McAuley E, et al. Tai Chi, self-efficacy, and physical function in the elderly. Prey Sci. 2001;2:229-239.

(12) Wolf SL, Sattin RW, O'Grady M, et al. A study design to investigate the effect of intense Tai Chi in reducing falls among older adults transitioning to frailty. Control Clin Trials. 2001;22:689-704.

(13) Speechley M, Tinetti M. Falls and injuries in frail and vigorous community elderly persons. J Am Geriatr Soc. 1991;39: 46-52.

(14) Wood RH, Gardner RE, Ferachi KA, et al. Physical function and quality of life in older adults: sex differences. South Med J. 2005;98:504-512.

(15) Diehr P, Patrick DL. Probabilities of transition among health states for older adults. Qual Life Res. 2001;10:431-442.

(16) McCullough ME, Laurenceau JP. Gender and the natural history of self-rated health: a 59-year longitudinal study. Health Psychol. 2004;23:651-655.

(17) Wolf SL, Coogler C, Xu T. Exploring the basis for Tai Chi Chuan as a therapeutic exercise approach. Arch Phys Med Rebabil. 1997;78:886-892.

(18) Bergner M, Bobbitt RA, Carter WB, et al. The Sickness Impact Profile: development and final revision of a health status measure. Med Care. 1981; 19:787-805.

(19) Patrick DL, Deyo RA. Generic and disease-specific measures in assessing health status and quality of life. Med Care. 1989;27:S217-S232.

(20) Deyo RA, Inui TS, Leininger JD, et al. Measuring functional outcomes in chronic disease: a comparison of traditional scales and a self-administered health status questionnaire in patients with rheumatoid arthritis. Med Care. 1983;21:180-192.

(21) Tak E, Staats P, Van Hespen A, et al. The effects of an exercise program for older adults with osteoarthritis of the hip. J Rheumatol. 2005;32:1106-1113.

(22) Simko LC, McGinnis KA. What is the perceived quality of life of adults with congenital heart disease and does it differ by anomaly? J Cardiovasc Nuts. 2005;20: 206-214.

(23) Clark NM, Janz NK, Dodge JA, et al. Changes in functional health status of older women with heart disease: evaluation of a program based on self-regulation. J Gerontol B Psychol Sci Soc Sci. 2000; 55:S117-S126.

(24) Redeker NS, Wykpisz E. Effects of age on activity patterns after coronary artery bypass surgery. Heart Lung. 1999;28:5-14.

(25) MacKenzie EJ, Bosse MJ, Castillo RC, et al. Functional outcomes following trauma-related lower-extremity amputation. J Bone Joint Surg Am. 2004;86:1636-1645.

(26) Mock C, MacKenzie E, Jurkovich G, et al. Determinants of disability after lower extremity fracture. J Trauma. 2000;49:1002-1011.

(27) Lipsett PA, Swoboda SM, Dickerson J, et al. Survival and functional outcome after prolonged intensive care unit stay. Ann Surg. 2000;231:262-268.

(28) Grady KL, Meyer PM, Dressier D, et al. Longitudinal change in quality of life and impact on survival after left ventricular assist device implantation. Ann Thorac Surg. 2004;77:1321-1327.

(29) Holzner B, Kemmler G, Kopp M, et al. Preoperative expectations and postoperative quality of life in liver transplant survivors. Arch Phys Med Rebabil. 2001;82:73-79.

(30) Cress ME, Schechtman KB, Mulrow CD, et al. Relationship between physical performance and self-perceived physical function. J Am Geriatr Soc. 1995;43:93-101.

(31) Cress ME, Buchner DM, Questad KA, et al. Exercise: effects on physical functional performance in independent older adults. J Gerontol A Biol Sci Med Sci. 1999; 54:M242-M248.

(32) Breeze E, Jones DA, Wilkinson P, et al. Area deprivation, social class, and quality of life among people aged 75 years and over in Britain. Int J Epidemiol. 2005;34: 276-283.

(33) Rothman ML, Hedrick S, Inui T. The Sickness Impact Profile as a measure of the health status of noncognitively impaired nursing home residents. Med Care. 1989;27:S157-S167.

(34) Gerety MB, Cornell JE, Mulrow CD, et al. The Sickness Impact Profile for nursing homes (SIP-NH. J Gerontol. 1994;49: M2-M8.

(35) Goldsmith G, Brodwick M. Assessing the functional status of older patients with chronic illness. Fam Meal. 1989;21:38-41.

(36) Weinberger M, Samsa GP, Hanlon JT, et al. An evaluation of a brief health status measure in elderly veterans. J Am Geriatr Soc. 1991;39:691-694.

(37) Damiano AM. Sickness Impact Profile: User's Manual and Interpretation Guide. Baltimore, Md: The Johns Hopkins University; 1996.

(38) Kutner NG, Barnhart H, Wolf SL, et al. Self-report benefits of Tai Chi practice by older adults. J Gerontol B Psychol Sci Soc Sci. 1997;52:P242-P246.

(39) Butcher JL, MacKenzie EJ, Cushing B, et al. Long-term outcomes after lower extremity trauma. J Trauma. 1996;41:4-9.

(40) Jurkovich G, Mock C, MacKenzie E, et al. The Sickness Impact Profile as a tool to evaluate functional outcome in trauma patients. J Trauma. 1995;39:625-631.

(41) McCarthy ML, McAndrew MP, MacKenzie EJ, et al. Correlation between the measures of impairment, according to the modified system of the American Medical Association, and function. J Bone Joint Surg Am. 1998;80:1034-1042.

(42) MacKenzie EJ, Bosse MJ, Pollak AN, et al. Long-term persistence of disability following severe lower-limb trauma: results of a seven-year follow-up. J Bone Joint Surg Am. 2005;87:1801-1809.

(43) Deyo RA, Diehl AK, Rosenthal M. How many days of bed rest for acute low back pain? A randomized clinical trial. N Engl J Med. 1986;315:1064-1070.

(44) Mossey JM, Shapiro E. Self-rated health: a predictor of mortality among the elderly. Am J Public Health. 1982;72:800-808.

(45) Idler EL, Benyamini Y. Self-rated health and mortality: a review of twenty-seven community studies. J Health Soc Behav. 1997;38:21-37.

(46) Wilcox VL, Kasl SV, Idler EL. Serf-rated health and physical disability in elderly survivors of a major medical event. J Gerontol B Psychol Sci Soc Sci. 1996;51: S96-S104.

(47) Idler EL, Kasl SV. Self-ratings of health: do they also predict change in functional ability? J Gerontol B Psychol Sci Soc Sci. 1995;50:S344-S353.

(48) Andresen EM, Patrick DL, Carter WB, et al. Comparing the performance of health status measures for healthy older adults. J Am Geriatr Soc. 1995;43:1030-1034.

(49) Mulrow CD, Gerety MB, Cornell JE, et al. The relationship between disease and function and perceived health in very frail elders. J Am Geriatr Soc. 1994;42: 374-380.

(50) Hain TC, Fuller L, Weil L, Kotsias J. Effects of T'ai Chi on balance. Arch Otolaryngol Head Neck Surg. 1999;125:1191-1195.

(51) Schaller KJ. Tai Chi Chih: an exercise option for older adults. J Gerontol Nurs. 1996;22:12-17.

(52) Wu G. Evaluation of the effectiveness of Tai Chi for improving balance and preventing falls in the older population: a review. J Am Geriatr Soc. 2002;50:746-754.

(53) Li F, Fisher KJ, Harmer P, et al. Delineating the impact of Tai Chi training on physical function among the elderly. Am J Prev Med. 2002;23:92-97.

(54) Wolf SL, Barnhart HX, Ellison GL, Coogler CE. The effect of Tai Chi Quan and computerized balance training on postural stability in older subjects. Atlanta FICSIT Group. Frailty and Injuries: Cooperative Studies on Intervention Techniques. Phys Ther. 1997;77:371-381.

(55) Husted C, Pham L, Hekking A, et al. Improving quality of life for people with chronic conditions: the example of t'ai chi and multiple sclerosis. Altern Ther Health Med. 1999;5:70-74.

(56) Brown DR, Wang Y, Ward A, et al. Chronic psychological effects of exercise and exercise plus cognitive strategies. Med Sci Sports Exerc. 1995;27:765-775.

(57) Li F, Duncan TE, Duncan SC, et al. Enhancing the psychological well-being of elderly individuals through Tai Chi exercise: a latent growth curve analysis. Structural Equation Modeling. 2001;8: 53-83.

(58) Stadnyk K, Calder J, Rockwood K. Testing the measurement properties of the Short Form-36 Health Survey in a frail elderly population. J Clin Epidemiol. 1998;51: 827-835.

(59) Mitrushina MN, Satz P. Correlates of self-rated health in the elderly. Aging (Milano). 1991;3:73-77.

(60) Dening TR, Chi LY, Brayne C, et al. Changes in self-rated health, disability and contact with services in a very elderly cohort: a 6-year follow-up study. Age Ageing. 1998;27:23-33.

(61) Bailis DS, Segall A, Chipperfield JG. Two views of self-rated general health status. Soc Sci Med. 2003;56:203-217.

(62) Montori VM, Guyatt GH. Commentary: intention-to-treat principle. Can Med Assoc J. 2001;165:1339-1341.

(63) Fried LP, Tangen CM, Walston J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci. 2001;56:M146-M156.

Al Greenspan, PT, MPH, DrPH, is Senior Scientist, National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 4770 Buford Hwy NE, Mailstop K-63, Atlanta, GA 30341 (USA). Address all correspondence to Dr Greenspan at: AGreenspan@ cdc.gov.

SL Wolf, PT, PhD, FAPTA, is Professor, Department of Rehabilitation Medicine, Emory University School of Medicine, Atlanta, Ga.

ME Kelley, PhD, is Research Assistant Professor, Department of Biostatistics, Rollins School of Public Health, Emory University, Atlanta, Ga.

M O'Grady, MD, is currently in private practice. Dr O'Grady was Assistant Professor, Department of Rehabilitation Medicine, Emory University School of Medicine, at the time of the study.

[Greenspan AI, Wolf SL, Kelley ME, O'Grady M. Tai chi and perceived health status in older adults who are transitionally frail: a randomized controlled trial. Phys Ther. 2007;87:525-535.]

Dr Greenspan, Dr Wolf, and Dr O'Grady provided concept/idea/project design and data collection. Dr Greenspan and Dr Wolf provided writing. Dr Greenspan, Dr Wolf, and Dr Kelley provided data analysis. Dr Greenspan, Dr Wolf, and Dr O'Grady provided fund procurement. Dr Wolf and Dr O'Grady provided subjects. Dr Wolf provided facilities/equipment and institutional liaisons. Dr O'Grady provided consultation (including review of manuscript before submission). The authors acknowledge Nana Freret, RN, MS, for her tireless effort 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; Mary Jowers, RN, for her assistance with evaluations and fall surveillance; Tingsen Xu, PhD, tai chi Grand Master, for devising and implementing the tai chi program; and Gregory Bailey for provision of tai chi exercise classes at half of the facilities.

Table 1.
Baseline Characteristics of Study Participants (Women Only)

Characteristic               Tai Chi              Wellness         P
                              Group              Education
                                                   Group

Age, y (%)
  70-79                        46.7                 41.7          .411
  [greater than or             53.3                 58.3
    equal to] 80
Race (%)
  White                        78.8                 80.3          .422
  African American             18.2                 18.9
  Other                         2.9                  0.8
Education (%)
  Did not complete             21.2                 22.7          .757
    high school
  High school                  78.8                 77.3
    graduate
Marital status (%)
  Married                       6.6                  4.5          .573
  Widowed                      75.2                 76.5
  Divorced or                  10.9                 14.4
    separated
  Single, never                 7.3                  4.5
    married
Frailty
  characteristics (%)
  Age of [greater              54.0                 58.3          .475
    than or equal to]
    80 y
  Impaired gait or             92.7                 94.7          .501
    balance
  No walking for               67.2                 73.5          .256
    exercise
  No other physical            62.0                 61.4          .909
    exercise
  Depression (treated          33.6                 40.2          .264
    or untreated)
  Use of sedatives             23.4                 25.8          .647
  Near vision                  97.8                 99.2          .332
  Impaired upper-              16.1                 18.9          .534
    extremity
    strength
  Impaired lower-              24.1                 27.3          .550
    extremity
    strength
  Lower-extremity              37.2                 40.2          .622
    disability
Sickness Impact
  Profile score
  ([bar.X] [+ or -]
  SD)
  Total                 12.2 [+ or -] 8.9    16.1 [+ or -] 10.7   .001
  Physical dimension    13.1 [+ or -] 9.1    17.0 [+ or -] 10.6   .001
  Psychosocial          10.2 [+ or -] 10.4   13.8 [+ or -] 14.2   .018
    dimension

Table 2.
Sickness Impact Profile: Dimensions, Categories, and Sample Items

Dimension      Category             Selected Items

Physical       Body care and        I stand up only with someone's
                 movement             help
                                    I get dressed only with someone's
                                      help

               Mobility             I stay home most of the time
                                    I am not now using public
                                      transportation

               Ambulation           I get around only by using a
                                      walker, crutches, cane, walls,
                                      or furniture
                                    I walk shorter distances or stop
                                      to rest often

Psychosocial   Emotional behavior   I talk about the future in a
                                      hopeless way
                                    I laugh or cry suddenly

               Social interaction   I am going out less to visit
                                      people
                                    I stay alone much of the time

               Alertness behavior   I do not keep my attention on any
                                      activity for long
                                    I am confused and start several
                                      actions at a time

               Communication        I am having trouble writing or
                                      typing
                                    I do not speak clearly when I am
                                      understress

Independent    Sleep and rest       I sit around half asleep
  categories                        I sleep less at night, for
                                      example, wake uptoo early, do
                                      not fall asleep for a long time,
                                      or awaken frequently

               Home management      I am not doing heavy work around
                                      the house
                                    I am not doing any of the shopping
                                      that I would usually do

               Recreation and       I am going out for entertainment
                 pastimes             less often
                                    I do my hobbies and recreation for
                                      shorter periods of time

               Eating               I feed myself with help from
                                      someone else
                                    I am drinking less fluid

               Work                 I am working shorter hours
                                    I am doing only light work

Table 3.
Perceived Physical Health (Sickness Impact Profile Scores) at Baseline
and at 1 Year by Intervention Type and Level of Disability (a)

Dimension or Category      Score for:

                           Tai Chi Group

                            Baseline    1 y
                            (n = 137)   (n = 103)

Physical dimension
  No or minor disability   40.9        56.3
  Moderate disability      42.3        28.2
  Severe disability        16.8        15.5
Body care and movement
  No or minor disability   57.6        71.9
  Moderate disability      29.2        17.5
  Severe disability        13.1        10.7
Mobility
  No or minor disability   63.5        67.0
  Moderate disability      16.8        18.4
  Severe disability        19.7        14.6
Ambulation
  No or minor disability   21.2        27.2
  Moderate disability      17.5        26.2
  Severe disability        61.3        46.6

Dimension or Category      Score for:

                           Wellness Education
                           Group                   Group x
                                                   Time
                           Baseline    7 y         Interaction (b)
                           (n = 132)   (n = 102)

Physical dimension                                 .016
  No or minor disability   29.5        27.4
  Moderate disability      37.1        35.3
  Severe disability        33.3        37.3
Body care and movement                             .051
  No or minor disability   40.9        48.1
  Moderate disability      34.1        25.5
  Severe disability        25.0        26.5
Mobility                                           .224
  No or minor disability   49.2        49.1
  Moderate disability      22.0        20.6
  Severe disability        28.8        30.4
Ambulation                                         .013
  No or minor disability   13.6        17.7
  Moderate disability      19.7         8.8
  Severe disability        66.7        73.5

(a) No or minor disability = 0-9.9, moderate disability = 10-19.9,
severe disability = [greater than or equal to] 20.

(b) Group x time interaction indicates a treatment effect of tai chi
relative to wellness education between baseline and 1 year.