The Effectiveness of Tai Chi on Improving Balance in Older Adults: An Evidence-based Review

Transcription

1 The Effectiveness of Tai Chi on Improving Balance in Older Adults: An Evidence-based Review Sachiko Komagata, PT 1, Roberta Newton, PT, PhD 2 1 PhD Candidate, Department of Physical Therapy, Temple University 2 Professor, Department of Physical Therapy,Temple University ABSTRACT Purpose: This paper provides a systematic critical review of the effectiveness of Tai Chi for balance improvement and fall reduction, and determines the effectiveness of the Quality Index (QI) as a critical review system. Methods: Eleven of 30 articles on Tai Chi met the criteria for systematic review. Both the QI and Methodological Rigor (MR) scores were used to evaluate the methodological quality. Results: Our review revealed moderate evidence of the effectiveness of Tai Chi on balance improvement. D-indices, calculated to determine the relative effect of the intervention on balance measurements, showed moderate negative correlations with the QI (r = -0.52) and minimal negative correlations with the MR score (r = ). Conclusion: Tai Chi is effective for balance improvement but not effective for reduction of falls. This review is unique in demonstrating the feasibility of using the methodological QI designed for both randomized and nonrandomized studies. Key Words: Tai Chi, balance, fall, review, EBM INTRODUCTION Tai Chi, a philosophy that originated in China several centuries ago, has gained the status of a cultural-historical treasure. 1,2 Tai Chi Chuan is the martial art aspect of Tai Chi philosophy. Despite the various interpretations of the history of Tai Chi, 1-3 current consensus indicates that the 3 main styles of Tai Chi Chuan are Chen, Yang, and Wu. 1,2 Yang s style is the most commonly found in the United States, followed by the Chen style. Since a discussion of Tai Chi as a philosophy and the principles of Tai Chi Chuan are beyond the scope of this review, the term Tai Chi will be used throughout this review to denote any derivatives of the main styles of Tai Chi Chuan. The objectives of Tai Chi include harmony of the mind, promotion of health, and attainment of rejuvenation and longevity. 1 Tai Chi movements can be characterized as slow, controlled, and continuous weight shifting with various animal-like postures. 4 However, when Tai Chi is performed with the intention to fight, the fluid soft forms easily turn into selfdefense or fighting movements. There are 3 advantages of Tai Chi as an exercise program for older adults. 1,5-8 Firstly, Tai Chi can be performed by anyone, regardless of age or gender, as long as the individual is able to stand. Secondly, there is no Please address correspondence to: Sachiko Komagata, Department of Physical Therapy, Temple University 3307 N. Broad St., Philadelphia, PA 10140, Ph: , Fax: (sachiko@komagata.net). need for special equipment or large space. Finally, the length of each session is flexible according to the individual s needs and tolerance. The paucity of qualified instructors in community-based or rehabilitation facilities presents a restriction to its true practice, because the practice of Tai Chi takes years to develop. 2,9 Therefore,Tai Chi is performed in a variety of forms in fitness centers, health care practices, and research studies. 2,9 Studies of the health benefits of Tai Chi have been accumulating slowly over the last decade. Studies have focused on the cardiorespiratory system, 10,11 muscular strength, 7,12,13 flexibility, 7,13 pain, 14 mood, 14 and balance. 5,6,12,13,15-19 Various health care practitioners, such as physicians, nurses, and occupational and physical therapists, have published reports indicating the benefits of incorporating Tai Chi into their practice, particularly for older adults. It is critical for physical therapists working with older adults to be able to evaluate the effectiveness of Tai Chi in geriatric physical therapy practice. Although nonrandomized studies or observational studies can yield valuable evidence, 20 critical reviews in medical science often emphasize randomized controlled studies. 21,22 This emphasis is based in part on the concept of strength of the evidence introduced by Sackett and colleagues. 23 Few metaanalyses assess the methodological quality of randomized studies. 24 Therefore, the overestimation of the effectiveness of the intervention may occur among low quality randomized controlled studies within the meta-analysis. 24 Chan and Bartlett published a meta-analysis of current studies to determine the effectiveness of Tai Chi as a therapeutic intervention to improve balance and postural control. 5 Their analysis, based on Sackett and colleagues guidelines for critical appraisal, 25 revealed that Tai Chi appeared to influence balance abilities in response to internal (self-induced) perturbations but not external perturbations. This article represented the first critical review of Tai Chi s positive effect on balance that indicated the inverse relationship between studies Methodological Rigor (MR) score and their effect size. However, the validity of their MR score sheet has not been established. 5 In a search for a valid tool among 25 checklists available to judge the methodological rigor of clinical studies, none was applicable to assessing nonrandomized studies, such as cohort and case control studies. 26 Downs and Black created and tested a checklist that measures the Quality Index (QI), applicable for both randomized studies and nonrandomized studies. 27 Subscales such as confounding factors and bias were added to assess the study s strengths and weaknesses in terms of the quality of reporting and internal validity. An external validity subscale also was added to the checklist to determine the study s generalizability to clinical practice. It is beneficial to use Downs and Black s QI to assess the methodological quality of Tai Chi studies because the majority of them are nonrandomized studies. Journal of Geriatric Physical Therapy Vol. 26;2:03 9

2 The purpose of this critical review was to examine the evidence-based support for Tai Chi as an effective adjunct to traditional physical therapy interventions to improve balance and reduce falls in older adults. The examination involved 2 methods of critical review: the use of the QI 27 and the MR scores. sure of balance or falls. Summary data included research questions and/or variables, study design, and the main findings of 11 studies selected for review. Of 40 articles identified in the searches, 11 met our criteria for review. Each study s methodological quality was assessed using the QI 27 and MR instruments. 5 METHODS The search strategy was based on the method described by Sackett and colleagues (Figure 1). 23,28,29 We searched MED- LINE, EMBASE, the Social Sciences Citation Index, and PsycINFO, using keywords Tai Chi$, fall$, and therapy$ (where $ indicates truncation of the word). Additional searches were performed by combining the keywords. We also used the Cochrane Database of Systematic Reviews. Studies were screened by reviewing the title and abstract and by applying the following inclusion criteria: (1) the article contained original research data (reviews and summaries were excluded), (2) Tai Chi was used as the therapeutic intervention for balance training, and (3) the article included at least 1 outcome mea- Figure 1. Flow Chart Illustrating the Process of Literature-Searching. Reprinted with permission from Elsevier. 29(p14) Quality Index Qaulity index is a 27-question instrument, developed and tested by Downs and Black, that can be used to review critically both randomized and nonrandomized studies. 27 The questions are subdivided into 5 categories: reporting (10 items), external validity (3 items), internal validity-confounding (selection bias, 6 items), internal validity (bias, 7 items), and power (1 item). Reporting assesses the objectives of the study, main outcomes, characteristics of the subjects, interventions, distribution of the confounding factors in each group, main findings, estimates of random variability, adverse events, follow-up, and actual probability values (eg, rather than < 0.05). External validity assesses how well the subjects, staff, facility, and treatment received represent the population and intervention in question. Internal validity-confounding assesses whether the subjects in different groups originated from the same population, if subject recruitment occurred in the same time frame and was randomized to intervention, if the study was blinded for both subjects and experimenter, adequate adjustment for confounding factors, and follow-up of dropouts. Internal validity-bias assesses the reliability and validity of the main outcome measures, appropriateness of the statistical tools, blinding subjects and examiners to the intervention, and describing any unplanned data analysis. Power addresses statistical power, the ability of a statistical test to find a significant difference that really does exist. 30(p689) Using a standard power calculation, sample size is calculated for 6 different levels of power (70, 75, 80, 85, 90, 95) based on the statistical significance level (p<0.05) and what constitutes a clinically significant difference in the effect. Then each study is rated by applying the smallest intervention sample size to the 6 levels of sample sizes. The scores of each category indicate whether the key items are clearly described in the study. All items, except for power and 1 item in the reporting subscale, are scored on a 0, 1 basis. Zero indicates no or unable to determine and 1 indicates yes or able to determine. Power was scored on a 0 to 5 scale and 1 item in the reporting subscale scored on a 0 to 2 scale. The maximum number of points attainable is 32 and is termed QI in order to quantify the overall methodological quality of each study. Psychometric properties of the QI have been assessed. 27 High internal consistency has been demonstrated for both randomized and nonran- 10 Journal of Geriatric Physical Therapy Vol. 26;2:03

3 domized studies (Kuder-Richardson 20 = 0.89). 27 Good testretest reliability (r = 0.88), interrater reliability (r = 0.75), and criterion validity (r = 0.90) have been shown. Downs and Black indicated that raters required an average 20 to 25 minutes to assess each paper. 27 Methodological Rigor The second tool for assessing the methodological quality was MR, an 18-item check list created by Chan and Barlett. 5 Thirteen of the items in the checklist had a 3 point score and the remaining 5 items had a 2 point score for a maximum of 49 points. Of the 11 articles reviewed, 6 had been critically analyzed by Chan and Barlett. 5 The relationship between the MR scores assigned by Chan and those assigned by the author (SK) was determined by calculating the Pearson Product correlation coefficient. Then the same coefficient was calculated for correlation analysis of 5 additional articles reviewed by the author (SK) using both QI and MR. Analysis of outcome involved the effect size d-index. 5 The effect size is defined by Cohen as the degree to which the phenomenon is present in the population on the degree to which the null hypothesis is false. 31(p325) An effect size of 0.2 is considered small, 0.5 is moderate, and greater than 0.8 is large. 32 In meta-analysis, the d-index is used as an estimate of the effect size. The d-index, therefore, is a relative value that allows the investigator to compare the effects of the intervention between the groups or within the group over time in a standard unit (scale-free measure). The d-index was calculated for the outcome measures of interest. The formula is d = (me - mc)/σ where: me = means of the outcome measure in the experimental group (or post-test in the same group); mc = means of the outcome measure in the control group (or pretest in the same group); σ = The mean standard deviation between the two groups (or pretest and post-test in one group). 5(p6) Categorization of Outcome Measures To examine the effectiveness of Tai Chi on balance, we used the same categorization of balance as described by Chan and Barlett. 5 These outcome measures are categorized into 3 groups: static conditions (eg, velocity of sway in AP in static condition, lateral body stability), internal perturbations (eg, single leg stance, eyes closed, reach testing), and external perturbations (eg, dispersion toes, up/eyes open, loss of balance during sensory organization testing). These categories are intended to describe no movement of the center of mass; voluntary movement of the center of mass, or involuntarily movement of the center of mass. A one-way ANOVA was used to determine the differences in the d-index among the above 3 categories. RESULTS Our search resulted in 40 journal articles published between 1981 and No studies were found in the Cochrane Database of Systematic Reviews. Limiting the search to randomized studies and/or qualitative studies yielded only 1 article. 16 Out of 30 potential articles 5-19,33-46 to review,11 (Table 1) 6,9,13-19,41,42 met our inclusion criteria and were fully evaluated for their methodological quality using the QI and MR checklists (Table 2). 27 The QI scores of the 11 articles ranged from 8 to 22 out of a maximum of 32 points. A total of 495 subjects participated in the 11 studies, with 250 participating in Tai Chi intervention. The number of subjects per study ranged from 11 to 110 subjects with an age range of 20 to 92 years. With the exception of 2 studies that used younger adults (ranging in age from 20 to 45 years), 13,17 all of the studies used subjects who were healthy community-dwelling elderly ranging from 65 to 92 years old. To minimize the effects of gender, Hong et al 15 studied male subjects, while Judge et al 6 exclusively studied women. The remaining articles included both men and women. Tai Chi was used as one of the main interventions for balance or postural control in most of the studies. 9,13,14,17-19,41 An exception is the study of Wolfson et al 16 in which Tai Chi was used as a maintenance program after the main balance training program was completed. Hong et al 15 and Tse and Bailey 42 compared the balance abilities of Tai Chi practitioners and nonpractitioners. The intensity, duration, and frequency of the Tai Chi training varied from 8 to 16 weeks, 45 to 60 minutes per session, and once a week to 3 times per week. Most of the Tai Chi training was conducted in a group setting. The instructor s qualifications were not indicated in all studies under review. Table 3 lists MR and QI assigned by the author (SK) and MR reported by Chan and Barlett. All 3 are positively correlated to each other. The effect size d-index was calculated for 33 out of the potential 35 outcome measures (Table 4). Ross et al 14 did not report the means or standard deviations, thus the d- index was not determined. The overall mean d-index was 1.99 (SD = 2.8) (Table 5). The mean d-index of measures corresponding to the static conditions, internal perturbation, and external perturbations was 0.52 (SD = 0.8), 2.48 (SD = 3.53), and 0.34 (SD = 0.39), respectively (Table 5). No statistically significant difference in the d-index occurred among the 3 outcome categories (F (2, 32) = 0.60, p>0.05). The relationship between the d-index and the QI score has moderate negative correlation at r = , and the relationship between the d- index and the MR score was r = DISCUSSION The movement pattern of Tai Chi is unique among other therapeutic interventions for people with balance impairments. The movement is slow but continuous, so the individual learns how to move most effectively within the postures and forms. Tai Chi is posture or form-oriented so that the person learns to use his or her visual or kinesthetic frame of reference throughout the movement. Therefore, the learner s behavior emerges from a self-organizing function of various subsystems. 47 It is also hypothesized that the Tai Chi movement patterns facilitate specific breathing patterns, which in turn trigger the autonomic nervous system for self-adjustment, ie, homeostatis. 48 Unlike some therapeutic interventions that have little kinesthetic demands, Tai Chi is postulated to improve kinesthetic awareness. By performing slow, continuous, and fluid patterns, the individual becomes aware of his or her own postural limitations. Such kinesthetic Journal of Geriatric Physical Therapy Vol. 26;2:03 11

4 Table 1. Summary of Studies Tai Chi s Effect on Balance. The articles are listed in reverse chronological order Author(s) and Year Question and /or variables Study design/sample size Findings (Indicate if Clinically significant) of publication and profile Hong et al, 2000 To determine the effects of TCC on total body rotation Cross sectional study Long term and regular practice of TCC may flexibility, heart rate responses, and single leg stance decrease the rate of age-related declines in with eyes closed. Age/gender matched control study balance, flexibility, and cardiovascular N (TCC) = 28 N (Sedentary) = 30 capacity. Hain et al, 1999 To determine if TCC significantly improves balance. One-group pretest-post-test TCC training improves balance. Moving platform posturography Romberg testing Reach testing Dizziness Handicap Inventory Modified MOS-SF N = 22 (divided into 3 age groups) Ross et al, 1999 To evaluate the therapeutic effects of a short-term Pretest-post-test quasi-experimental Positive changes in mood, pain perception, Tai Chi exercise program for the elderly. design flexibility, balance, and sway were demonstrated. Test items are flexibility, balance, sway, pain, and mood. N = 17 (4 drop out) Shih, 1997 To examine the effect of TCC on the average velocity of One-group pretest-post-test Sway in static condition before and after sway. the TCC training was not statistically sig- N = 11 nificant. The reduction in dynamic sway condition after the TCC was statistically significantly reduced as compared to pretest. Jacobson et al, 1997 To assess the effect of TCC training on lateral Pretest-post-test control group design Statistically significant differences between stability, kinesthetic sense, and strength of groups were found in lateral body stability, voluntary knee extension. N (TCC) = 12 N (CONT) = 12 kinesthetic sense at 60 o, and strength of the dominant knee extension. Wolf et al, 1997 To examine whether 2 exercise programs (TCC and BT) Multigroup pretest-post-test BT group showed significant improvement would affect the ability to minimize postural sway of control group design in postural stability while TCC and control relatively inactive older adults. did not. N (TCC) = 24 N (BT) = 24 Dispersion EO, Dispersion EC, COB-X/EO, COB-X/EC, N (CONT) = 24 TCC delayed onset of first or multiple falls. COB-Y/EO, COB-Y/EC, CO Schaller 1996 To determine the effects of TCC on balance, flexibility, Quasi-experimental pretest-post-test Significant difference between the mood, health status, and blood pressure in a sample design 2 groups on balance. of community-dwelling elders. N (TCC) = 24 N (CONT) = 22 Single limb stance test Modified Sit and Reach test Profile of Mood status SF-36 SBP/DBP by sphygmomanometer Wolfson et al, 1996 To determine the effect of 3 months balance Randomized multigroup pretest design BT meaningfully improved balance and/or weight training followed by 6 months of measures. ST increased ISOK. There was Tai Chi training on balance and strength. no interaction between BT and ST. N (BT) = 19 N (ST) = 18 Significant balance gain was maintained LOB during sensory organization testing, SST, after 6 months of TCC. voluntary limits of stability (FBOS), summed N (BST) = 16 N (CONT) = 16 isokinetic torque of 8 lower extremity movements (ISOK) and usual gait velocity (GVU). Province et al, 1995 To determine if short-term exercise reduces falls and Meta-analysis of the FICSIT Trials Treatments for elderly adults including fall-related injuries in the elderly. exercise, reduce the risk of falls. 100<N<1323 per study Judge et al, 1993 To test whether vigorous exercise would improve Two-group pretest-post-test design SST improved with vigorous exercise the single-stance balance of healthy older women training. Double stance did not change and lower their risk of falls and fall-associated injuries. N (COMBO) = 12 N (FLEX) = 9 with training. Sway difference between the groups was not statistically significant. Tse & Bailey, 1992 To determine the potential value of TCC for postural Nonequivalent post-test-only control The TCC subjects performed significantly control in elderly. group design better than the non-tcc subjects on 3 out of 5 balance tests (1, 2, & 5). Males per- 5 balance tests: N (TCC) = 9 N (CONT) = 9 formed significantly better than females 1. SLS-EO (R) on 3 out of 5 balance tests (1, 2, & 5). 2. SLS-EO (L) 3. SLS-EC (R) 4. SLS-EC (L) 5. Heel-to-toe walk-eo Abbreviations TCC = Tai Chi, BT = balance training, ST = strengthening training, BST = balance and strengthening training, SST = single stance time, MOS SF = medical outcome study short form, LOB = loss of balance, SLS = single leg stance, EO = eyes open, EC = eyes closed, (R) = right, (L) = left, CONT = control, COMBO = combination of resistive, walking, flexibility, and simple tai chi, FLEX = flexibility training, COB-X = center of balance in side-by-side direction, COB-Y = center of balance in anterior-posterior direction 12 Journal of Geriatric Physical Therapy Vol. 26;2:03

5 Table 2. Scores of Quality Index and Subcategories Scale Quality Subcategories (items/max score) Index Reporting External Confounding Bias Power (27/32) (10/11) Validity (3/3) (6/6) (7/7) (1/5) Hong et al, Hain et al, Ross et al, Shih, Jacobson, et al Wolf et al, Schaller, Wolfson et al, Province et al, Judge et al, Tse & Bailey, Range of Scores Table 3. Methodological Rigor Scores (MR) and Quality Index (QI) of 11 Articles Reviewed by Chan and Komagata Chan & Barlett Komagata Komagata using MR* using MR* using QI* Tse & Bailey, Wolfson et al, Schaller, Jacobson et al, Shih, Wolf et al, Hong et al, 2000 N/A Hain et al, 1999 N/A Ross et al, 1999 N/A Province et al, 1995 N/A Judge et al, 1993 N/A * MR = Methodological Rigor score, QI = Quality Index awareness could decrease the incidence of falls. 18,43 This review suggests that that activities incorporated into Tai Chi appear to be capable of favorably influencing balance. This influence does not seem to be specific to internal perturbation category as reported by Chan and Barlett. 5 The majority of the articles reviewed selected a frequency of Tai Chi as once a week for approximately 1 hour each training session. Such a length of training may fail to show improvements in postural control. Many postures in Tai Chi Journal of Geriatric Physical Therapy Vol. 26;2:03 13 are unintuitive at first. For example, moving the right arm forward while the right foot is in front of the left can be an awkward movement to learn unless the person is familiar with boxing or fencing. If one has difficulty with overcoming the initial awkwardness of Tai Chi postures, the learner may feel discouraged to continue. Peer and instructor encouragement is one method to facilitate individuals to learn the forms. In addition, most of the studies reviewed here did not clearly state their confounding factors, such as gender, age, and lifestyle. Group Tai Chi sessions, rather than individual sessions, were used in most studies without addressing the potential effect of the group process or motivational factors on the outcome. Internal validity will improve by addressing these potential confounding factors. In addition, studies designed to address the participants compliance status would improve the study s internal validity as well as the effectiveness of the program. To minimize selection bias, future studies should use randomized assignments when possible. Only 4 studies clearly documented the subject s compliance with the intervention. The qualifications of Tai Chi instructors also play a major role in the effectiveness of Tai Chi in balance training. The instructors may affect the participants levels of motivation, compliance, and accuracy in performing the forms and postures. Observational studies investigating the quality of instruction and movement may provide the opportunity to discover a deeper understanding of Tai Chi s hidden effects on an individual s movement and wellness. Since the effect of Tai Chi on balance has been tested in elderly subjects who have no history of neurological or musculoskeletal diseases, it may be beneficial to use this type of intervention to a specific disease population, such as stroke, Parkinson disease, or peripheral neuropathy. Physical therapists are skilled in analyzing people s movement patterns based on motor control and biomechanical theories. Therefore, they have the potential to determine through systematic study what specific characteristics of Tai Chi improve balance as compared with traditional physical therapy balance training techniques. The use of a sham-tai Chi paradigm can be used to examine a placebo effect of Tai Chi. Furthermore, due to the nationwide managed care trends, neuromusculoskeletal rehabilitation has been evolving to use more innovative formats. This usually entails fewer visits, a shorter length of treatment sessions, or the use of group sessions with more emphasis on education and home exercise programs. Tai Chi and other movement therapies can supplement a rehabilitation program and can be provided by health

6 Table 4A. Effect Size Index (d), Quality Index (QI) Scores, and Methodological Rigor Score (MR) for Static Balance Measures Author(s) and Publication Static Balance Measure d QI MR Year Shih, 1997 Velocity of sway in AP in static condition Jacobson et al, 1997 Lateral body stability (stability platform) Wolf et al, 1997 Dispersion, eyes open Dispersion, eyes closed COB-X/EO COB-X/EC COB-Y/EO COB-Y/EC Province MA et al, 1995 Average falls/subject Abbreviations: COB-X/EO = center of balance in side-by-side direction/ eyes open, COB-X/EC = center of balance in side-by-side direction/eyes closed, COB-Y/EO = center of balance in anterior-posterior direction/eyes open, COB-Y/EC = center of balance in anterior-posterior direction/eyes closed Table 4B. Effect Size Index (d), Quality Index (QI) Scores, and Methodological Rigor Score (MR) for Internal Perturbation Balance Measures Author(s) and Internal Perturbation Balance d QI MR Publication Year Measure Hong et al, 2000 Single leg stance, eyes closed ( R ) Single leg stance, eyes closed ( L ) Hain et al, 1999 Moving platform posturography Tandem Romberg testing Reach testing Ross et al, 1999 Single leg stance NA Sway during tandem walking NA Shih, 1997 Velocity of sway in AP in dynamic condition Schaller, 1996 Single limb stance test Wolfson et al, 1996 Single stance time (SST) Voluntary limits of stability (FBOS) Judge et al, 1993 Single leg stance Double stance EO Double stance EC Tse & Bailey, 1992 SLS-EO (R) SLS-EO (L) SLS-EC (R) SLS-EC (L) Heel-to-toe walk-eo Abbreviations: SLS-EO = single leg stance-eyes open, SLS-EC = single leg stance-eyes closed Table 4C. Effect size index (d), Quality Index (QI) Scores and Methodological Rigor Score (MR) for External Perturbation Balance Measures Author(s) and External Perturbation Balance d QI MR Publication Year Measure MR Wolf, et al, 1997 Dispersion toes, up/eyes open Dispersion toes, up/eyes closed COB-X/toes up/eo COB-X/toes up/ec COB-Y/toes up/eo COB-Y/toes up/ec Wolfson, et al, 1996 Loss of balance (LOB) during sensory organization testing Abbreviations: COB-X/toes up/eo =center of balance in side-by-side direction/toes up/eyes open, COB-X/toes up/ec =center of balance in sideby-side direction/toes up/eyes closed, COB-Y/toes up/eo =center of balance in anterior-posterior direction/toes up/eyes open, COB-Y/toes up/ec =center of balance in anterior-posterior direction/toes up/eyes closed Note: With the directional hypothesis, the sign of d corresponds to the predicted direction. 40 Table 5. Mean and Standard Deviation of D-Index Among the 3 Outcome Measure Categories Overall Static Internal External Conditions Perturbations Perturbations Mean Standard Deviation care professionals and practitioners other than physical therapists. A trend in current physical therapy practice is to offer post rehabilitation maintenance programs. Tai Chi may be an effective group therapy activity for people who undergo formal outpatient physical therapy programs for balance and gait training. When this review article was nearly completed, the authors found Wu s review article on Tai Chi s effectiveness for improving balance and preventing falls in the older population. 49 Thus our recommendations for future studies overlap significantly with the study by Wu. CONCLUSION The majority of the studies reviewed here support the effectiveness of Tai Chi to improve balance. However, quality of studies requires further improvement in treating confounding factors, addressing compliance, and randomizing the assignments. Systematic review using QI and d-index is warranted due to the existence of nonrandomized studies. Although the 3 balance outcome measure categories do not show a statistically significant difference in their effect size d- index, the internal balance perturbations category appears most suited to demonstrate Tai Chi s effect on the participants balance. Demystifying the oriental martial art of Tai Chi through scientific studies and incorporating it in physical therapy practice, particularly aiming for balance improvement and fall prevention, can be justified. 14 Journal of Geriatric Physical Therapy Vol. 26;2:03

7 ACKNOWLEDGEMENTS We would like to acknowledge the support and valuable feedback provided by Susan L. Michlovitz, PT, PhD, CHT and the editors and reviewers for the Journal of Geriatric Physical Therapy. REFERENCES 1. Jou TH. The Tao of Tai-Chi Chuan: Way to Rejuvenation. 3rd ed. Warwick, NY: Tai Chi Foundation; Guang ZW. Kantan Taikyokuken:Guangbo Taijiquan (Simpler Tai Chi Chen:Guangbo Taijiquan). Tokyo, Japan: Baseball Magazine Co; Koh TC. Tai Chi Chuan. Am J Chin Med. 1981;9(1): Plummer JP. Acupuncture and tai chi chuan (Chinese shadow boxing):body/mind therapies affecting homeostasis. In: Lau Y, ed. The Scientific Basis of Traditional Chinese Medicine. Hong Kong: University of Hong Kong; Chan WW, Bartlett DJ. Effectiveness of Tai Chi as a therapeutic exercise in improving balance and postural control. Phys Occup Ther Geriatr. 2000;17(3): Judge JO, Lindsey C, Underwood M,Winsemius D. Balance improvements in older women: Effects of exercise training. Phys Ther. 1993;73: Lan C, Lai JS, Chen SY,Wong MK.Tai Chi Chuan to improve muscular strength and endurance in elderly individuals: A pilot study. Arch Phys Med Rehabil. 2000;81: Ross MC, Presswalla JL. The therapeutic effects of Tai Chi for the elderly. J Gerontol Nurs. 1998; 24: Schaller KJ. Tai Chi Chih: An exercise option for older adults. J Gerontol Nurs. 1996;22(10): Lan C, Lai J, Wong M, Yu M. Cardiorespiratory function, flexibility, and body composition among geriatric Tai Chi Chuan practitioners. Arch Phys Med Rehabil. 1996;77: Lan C, Chen SY, Lai JS, Wong MK. The effect of Tai Chi on cardiorespiratory function in patients with coronary artery bypass surgery. Med Sci Sports Exerc. 1999;31: Wolfson L,Whipple R, Judge JO, Amerman P, Derby C, King M. Training balance and strength in the elderly to improve function. J Am Geriatr Soc. 1993;41: Jacobson BH, Ho-Cheng C, Cashel C, Guerrero L.The effect of Tai Chi Chuan training on balance, kinesthetic sense, and strength. Percept Mot Skills. 1997;84: Ross MC, Bohannon AS, Davis DC, Gurchiek L. The effects of a short-term exercise program on movement, pain, and mood in the elderly. J Holist Nurs. 1999;17: Hong Y, Li JX, Skelton D. Balance control, flexibility, and cardiorespiratory fitness among older Tai Chi practitioners. Br J Sports Med. 2000;34: 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: Shih J. Basic Beijing twenty-four forms of T ai Chi exercise and average velocity of sway. Percept Mot Skills. 1997;84: Wolf SL, Barnhart HX, Ellison GL, Coogler CE. The effect of Tai Chi Quan and computerized balance training on postural stability in older subjects. Phys Ther. 1997;77: Hain T, Fuller L, Weil L, Kotasias J. Effects of T ai Chi on balance. Arch Otolaryngol Head Neck Surg. 1999;125: Trudeau DL. The value of observational studies in neurotherapy. J Neurother. 2001;4(3): Salerno SM, Browning R, Jackson JL. The effect of antidepressant treatment on chronic low back pain: A metaanalysis. Arch Intern Med. 2002;162: McAlister FA, Lawson FME, Teo KK, Armstrong PW. A systematic review of randomized trials of disease management program in heart failure. Am J Med. 2001;110: Sackett DL, Straus SE, Richardson WS, Rosenberg W, Haynes RB. Evidence-based Medicine: How to Practice and Teach EBM.2 nd ed. London: Churchill Livingstone; Moher D, Pham B, Jones A, et al. Does quality of reports of randomized trials affect estimates of intervention efficacy reported in meta-analysis? Lancet. 1998;352(9128): Sackett DL, Hayes RB, Guyatt GH, Tugwell P. Clinical Epidemiology: A Basic Science for Clinical Medicine.2 nd ed. Boston, Mass: Little Brown and Company; Moher D, Jadad AR, Nichol G, Penman M, Tugwell P, Walsh S. Assessing the quality of randomized controlled trials: An annotated bibliography of scales and checklists. Control Clin Trials. 1995;16: Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomized and non-randomized studies of health care interventions. J Epidemiol Community Health. 1998;52: Evidence-Based Working Group. Evidence-based medicine: A new approach to teaching the practice of medicine. JAMA. 1992;268: Ridsdale L. Evidence-based Practice in Primary Care. London, UK: Churchill Livingstone; Portney LG, Watkins MP. Foundations of Clinical Research: Applications to Practice. Stamford, Conn: Appleton & Lange; Bickman L, Rog DS. Handbook of Applied Social Research Methods. Thousand Oaks, Calif: Sage Publication; Kazis LE, Anderson JJ, Meenan RF. Effect sizes for interpreting changes in health status. Med Care. 1989;27 (Suppl 3):S178-S Lan C, Lai JS, Chen SY,Wong MK. 12-month Tai Chi training in the elderly: Its effect on health fitness. Med Sci Sports Exerc. 1998;30: Kessenich CR.Tai Chi as a method of fall prevention in the elderly. Ortho Nurs. 1998;17: Farrell SJ, Ross AD, Sehgal KV. Eastern movement therapies. Phys Med Rehabil Clin N Am. 1999;10(617): Cerrato PL. Tai chi: A martial art turns therapeutic. RN. 1999;62: Journal of Geriatric Physical Therapy Vol. 26;2:03 15

8 37. Cassileth BR. Complementary therapies: Overview and state of the art. Cancer Nurs. 1999;22: Chen KM, Snyder M. A research-based use of Tai Chi/ movement therapy as a nursing intervention. J Holist Nurs. 1999;17: Luskin FM, Newell KA, Griffith M, et al. A review of mind/body therapies in the treatment of musculoskeletal disorders with implications for the elderly. Altern Ther Health Med. 2000;6: Zwick D, Rochelle A, Choksi A, Domowicz J. Evaluation and treatment of balance in the elderly: A review of the efficacy of the Berg Balance Test and Tai Chi Quan. Neurorehabilitation. 2000;15: Province MA, Hadley EC, Hornbrook MC, et al. The effects of exercise on falls in elderly patients. JAMA. 1995;273: Tse SK, Bailey DM.T ai Chi and postural control in the well elderly. Am J Occup Ther. 1992;46: Kutner NG, Barnhart H, Wolf SL, McNeely E, Xu T. Selfreport benefits of Tai Chi practice by older adults. J Gerontol Psychol Sci. 1997;52B: Yan JH. Tai Chi practice reduces movement force variability for seniors. J Gerontol Med Sci. 1999;54A:M Wolf SL, Coogler C, Xu T. Exploring the basis for Tai Chi Chuan as a therapeutic exercise approach. Arch Phys Med Rehabil. 1997;78: Kirsteins AE, Dietz F, Hwang S. Evaluating the safety and potential use of a weight-bearing exercise, Tai-Chi Chuan, for rheumatoid arthritis patients. Am J Phys Med Rehabil. 1991;70: Heriza C. Motor Development: Traditional and Contemporary Theories. In: Lister MJ, ed. Contemporary Management of Motor Control Problems: Proceedings of the II STEP Conference. Alexandria, Va: Foundations for Physical Therapy; 1991: You S, Hashi I. Kenko Taikyokuken Kihan Kyoutei:Yo Meiji Taikyokuken Jitsugi Tekisto I (Step-by-step Textbook on Tai Chi for Health:Meiji Yo s Tai Chi Practice Textbook I). Tokyo, Japan: Baseball Magazine Co.; 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: Merrill BA. A global look at compliance in health/safely and rehabilitation. J Orthop Sports Phys Ther. 1994;19: Friedrich M, Gittler G, Halberstadt Y, Cermak T, Heiller I. Combined exercise and motivation program: Effect on the compliance and level of disability of patients with chronic low back pain: A randomized controlled trial. Arch Phys Med Rehabil. 1998;79: Grahn B, Stigmar K, Ekdahl C. Motivation for change in patients with prolonged musculoskeletal disorders: A qualitative two-year follow-up study. Physiother Res Int. 1999;4: Scott AH.Wellness works: Community service health promotion groups led by occupational therapy students. Am J Occup Ther. 1999;53: COMING THIS FALL Balance, Mobility & Falls in Neurologic and Geriatric Patients Anne Shumway-Cook, PhD, PT Leslie Allison, MS, PT September 27 & 28, 2003 Oakland, CA SAVE the DATES and mark your calendar NOW! Details about this Neurology Section sponsored course will be posted at as they become available. Contact Janice Ford, janiceford@apta.org, , ext Journal of Geriatric Physical Therapy Vol. 26;2:03