Geriatric Nursing 35 (2014) 9e19

Contents lists available at ScienceDirect

Geriatric Nursing journal homepage: www.gnjournal.com

Feature Article

Effects of Tai Chi on cognitive function in community-dwelling older adults: A review Sally M. Miller, MS, RN *, Ruth E. Taylor-Piliae, PhD, RN, FAHA College of Nursing, The University of Arizona, 1305 N. Martin, P.O. Box 210203, Tucson, AZ 85721-0203, USA

a r t i c l e i n f o

a b s t r a c t

Article history: Received 14 July 2013 Received in revised form 17 October 2013 Accepted 21 October 2013 Available online 16 November 2013

As the population of the United States ages, activities to maintain or improve cognitive function will become increasingly important to preserve functional ability, independence and health-related quality of life. This article is a review of recent research on Tai Chi and cognitive function in community-dwelling older adults. Of the 12 studies reviewed, 10 reported improvement in measures of executive function, language, learning, and/or memory. Several design features make comparisons across studies challenging. As a moderate-intensity, low-impact form of exercise, Tai Chi is appropriate for older adults and seems to offer positive cognitive benefits. Recommendations for future research are provided. Ó 2014 Mosby, Inc. All rights reserved.

Keywords: Tai Chi Cognitive function Older adults Mind-body exercise

Introduction

Exercise and cognitive function

The population of the United States is aging, and by the year 2030 the number of Americans older than 65 will almost double.1 Although older adults exhibit wide variability in the aging process, age-related changes in cognitive function are common, including declines in executive function, information processing speed, and attention.2 Changes in cognitive function can be viewed on a continuum, with age-related cognitive changes on one end of the continuum, mild cognitive impairment as a mid-stage, and dementia as the most severe end-point on the continuum of cognitive decline. Age-related cognitive changes cause minimal to no interference in instrumental activities of daily living (IADL) and independence,3 while mild cognitive impairment causes minor but noticeable changes and abnormal findings in one or more cognitive domains.4 The final stage of the cognitive decline continuum is dementia, characterized by a broad range of severe cognitive deficits, behavioral changes and functional decline, leading to dependence.5 Cognitive impairment places older adults at risk for progression to dementia,6 and strategies are needed to maintain or enhance cognitive function, and prevent cognitive decline. One promising intervention to enhance cognitive function in older adults is through exercise.

The health benefits of exercise in older adults are well-documented.7,8 Exercise is important for physical and cognitive function, and is a strategy to maintain or increase independence in the growing population of older adults in the United States. In spite of this, adults aged 65 and older are the least active group of Americans,9,10 and can benefit from regular physical exercise. Exercise is classified by the energy cost to the individual using metabolic equivalents of task (METs).11 For example, low-intensity activities such as cooking or dusting have a MET value 130) were excluded. Data from studies meeting the inclusion criteria were extracted including: country of study, study design, Tai Chi style and exercise dose, comparison group, cognitive domains measured, and findings. Results Study designs and populations A total of 12 articles were included in this review examining the effects of Tai Chi on cognitive function in community-dwelling older adults (Fig. 1). Study designs included randomized controlled trials (RCT) (n ¼ 6),17,31e35 quasi-experimental (n ¼ 3),36e38 and cross-sectional (n ¼ 3)39e41 studies (Table 1). The cross-sectional studies compared Tai Chi to other forms of exercise such as walking, stretching/toning, and/or no exercise. The studies reviewed were conducted in Asia, North America, and South America and included community-dwelling adults ranging in age from 55 to 90, with a majority of female participants (average ¼ 71%). Nine of the 12 studies listed exclusion criteria of participants with unstable conditions, substance abuse, diagnoses for which physical activity would be contraindicated, or active

Excluded Full-text articles excluded (n=35) for the following reasons: 20 = not cognitive measures 6 = review articles 4 = dementia 2 = non community-dwelling 3 = other (abstracts, interim reports, not Tai Chi)

Fig. 1. Study selection flow diagram showing reasons for exclusion.

Table 1 Summary of studies measuring the effect of Tai Chi on cognitive function. First author, year, country

Tai Chi style/duration

Cognitive domain/measures

Findings

n ¼ 15 Community-dwelling adults Age range 62e85 (72.2  7.7) Female: 67% Cognitive function/dementia exclusion criteria: MMSE < 24

Yang style (24 form) 1.5 h 2  wk  12 wks Control (health-related topics): 1hr 2  wk

Lam, 2012, Hong Kong

n ¼ 389 Adults at risk of cognitive decline >65 years old Female: 76% TC, n ¼ 171 Stretch/tone, n ¼ 218 Cognitive function/dementia inclusion criteria CDR ¼ 0.5 Baseline MMSE: TC ¼ 25.1, Control ¼ 24.4

Lavretsky, 2011, United States

n ¼ 73 Adults with a major depressive disorder >60 years old Female: 62% TC, n ¼ 33 Health education, n ¼ 35 Cognitive function/dementia exclusion: MMSE < 26

Simplified TC (24 form) TC master plus occupational therapists Induction: Weekly  4e6 weeks plus practice Maintenance: video with either TC or control exercises Both: No less than 30 min/day No less than 3 days/wk Refresher lessons every month for both Chih style (with Qi Gong) 2h 1/wk  10 wks

Other SOT Response time to auditory or visual stimulus while stationary (single-task) and while walking (dual-task) Executive function DR DSB TMT VSB Language CVFT Global cognition ADAS-Cog CDR sum of boxes MMSE

Nguyen, 2012, Vietnam

n ¼ 96 Community-dwelling adults Female: 100% TC, n ¼ 48 Mean age ¼ 69.2  5.3 Control, n ¼ 48 Mean age ¼ 68.7  4.9 Exclusion criteria: MMSE  25

TC style-not specified (24-form) 60 min 2  wk  6 mo

Executive function TMT-A and B

Taylor-Piliae, 2010, United States

n ¼ 132 Community-dwelling adults Age range 60e84 Female: 70% TC, n ¼ 37 Mean age ¼ 70.6  5.9 Western exercise (WE), n ¼ 39 (endurance, resistance/strength, flexibility) Mean age ¼ 68.5  5.0 Control, n ¼ 56 (healthy aging topics) Mean age ¼ 68.2  6.2 Exclusion criteria: severe cognitive impairment (method not specified)

Yang style-24 postures Adoption phase: 60 min 2  wk  6 mo (class) 3  wk  6 mo (home) Maintenance phase: 45 min 1  wk  6 mo (class) 3  wk  6 mo (home) Daily practice encouraged

Executive function DSF DSB Language AN

Other No significant effect of time, group or interaction for either equilibrium or verbal reaction scores (dual task) Increased dual task cost for TC group; improvement in control group under single and dual tasks Decreased obstacle avoidance under dual tasks; no improvement in TC group NS improvement in control group Executive function NS improvement in DSB, VSB Completers only: improved DR, (group difference at 1 year ¼ 0.17, p ¼ 0.05) VSB greater after 5th month in TC group, NS after 1 yr Language NS difference Global cognition NS improvement in ADAS-Cog 21% lower CDR sum of boxes in TC group than control (beta ¼ 0.79, 95% CI 0.63e0.99, p ¼ 0.04) MMSE e no change Executive function NS difference Learning/memory CVLT long delayed recall, cued showed improvement in escitalopram with TC than escitalopram with health education (group  time interaction: F[1,65] ¼ 5.29; p < .05) Global cognition NS difference Executive function Significant difference between TC and control in TMT-A (F[1,71] ¼ 78.37, p < 0.001) and TMT-B (F [1,71] ¼ 175.90, p < 0.001) Significant interactions in TMT-A for time (eta squared ¼ 0.746, p ¼ 0.000) and time  group (eta squared ¼ 0.708, p ¼ 0.000); and TMT-B between time (eta squared ¼ 0.860, p ¼ 0.000) and time  group (eta squared ¼ 0.854, p ¼ 0.000) Executive function Adoption phase: TC had greater improvement in DSB than WE or control (F[3129] ¼ 7.75, p < 0.001) Maintenance phase: TC maintained improvement in DSB compared with WE (t[2,74] ¼ 2.52, p ¼ 0.014) Statistically significant within-group improvement in DSB in TC (t[2,74] ¼ 3.26, p ¼ 0.003) Language Maintenance phase: within-group improvement in TC (t [2,74] ¼ 3.27, p ¼ 0.003) and WE (t[2,74] ¼ 2.81, p ¼ 0.008); NS between groups

Executive function ST TMT-A and B Learning/memory CVLT Global cognition MMSE

S.M. Miller, R.E. Taylor-Piliae / Geriatric Nursing 35 (2014) 9e19

Sample

Randomized clinical trials Hall, 2009, United States

(continued) 11

First author, year, country Wang, 2010, Japan

Quasi-experimental studies Kasai, 2010, Brazil

Reid-Arndt, 2012, United States

Cross-sectional studies Chan, 2005, Hong Kong

Sample

Tai Chi style/duration

Cognitive domain/measures

Findings

n ¼ 34 Adults with cerebrovascular disease Female: 74% TC, n ¼ 17 Mean age ¼ 76.5  9.7 Control, n ¼ 17 Mean age ¼ 77.6  12.3 Exclusion: MMSE < 20

Yang style 50 min 1/wk  12 wk Control: Non-resistance (20 min) plus resistance (60 min) Total: 80 min 1  wk  12 wk

Other ERP e P300

Other NS difference

n ¼ 26 Adults with subjective memory decline, no dementia diagnosis Mean age ¼ 74 Female: 100% TC, n ¼ 13 Control, n ¼ 13 Cognitive function/dementia exclusion: MMSE corrected for educational level (cut-off score not specified) n ¼ 20 Community-dwelling adults Age range 68e84 (Mean age ¼ 76.5) Female: 95% Screened for dementia and mild cognitive impairment (method not specified) n ¼ 23 Female cancer survivors Mean age ¼ 62.3  10.8 Female: 100% Screening for cognitive impairment or dementia not specified

Yang style 60 min 2  wk  6 mos Brochure instructed to practice at home

Executive function DSF DSB Learning/memory RBMT Other SMC Performance of TC forms tested at 3 mos

Executive function At 3 mos TC group improved in DSF (p ¼ 0.031) but no improvement in DSB (p ¼ 0.164) Learning/memory Improvement in RBMT (p ¼ 0.007) Positive correlation (p ¼ 0.008) between learning TC test and RBMT (r not supplied) Other Significant decrease in SMC in TC group (p ¼ 0.023)

Sun style 50 min. 3  wk  10 wks

Executive function TMT-A and B DSym CDT

Modified Yang Style 1h 2/wk  10 wk

Executive function DS DSym TMT-A and B ST WMS-LM I Language COWAT Learning/memory RAVLT WMS-LM II Other MASQ

Executive function Significant improvement in time to complete TMT-B (pre-intervention mean ¼ 106.30  33.48; postintervention mean ¼ 89.00  32.78; p ¼ 0.029) Significant difference in CDT: on a seven-point scale (pre-intervention mean ¼ 6.20  0.77; postintervention mean ¼ 6.50  0.69; p ¼ 0.029) Executive function Significant changes in ST (F[1,22] ¼ 6.46; p ¼ 0.019); TMT-A (F[1,22] ¼ 12.29; p ¼ 0.002); and TMT-B (F [1,22] ¼ 18.49; p < .001) Language Significant change (F[1,22] ¼ 12.00; p ¼ 0.002) in COWAT Learning/memory Significant improvement in immediate memory: RAVLT 1 (F[1,22] ¼ 5.74; p ¼ 0.026), RAVLT 1e5 (F[1,22] ¼ 5.03; p ¼ 0.035); and WMS-LM I (F[1,22] ¼ 26.32; p ¼ 1; CDR 0 mean MMSEC ¼ 27.26  1.94; CDR 0.5 mean MMSE-C ¼ 24.48  2.76

TC style-not specified (included yoga practitioners) Average duration of regular exercise (stretching, aerobic or Minde Body) ¼ 5.57  7.66 yrs Average time spent per session ¼ 33.5  32.0 min

Executive function DR DSB DSF VSB Language CVFT Global cognition ADAS-Cog (Chinese) CDR MMSE-C (Chinese version)

Man, 2010, Hong Kong

n ¼ 135 Community-dwelling adults Age range: 55e90 Female: 53% TC, n ¼ 42 Exercise, n ¼ 49 Control, n ¼ 44 Cognitive function/dementia exclusion criteria: MMSEC ¼ 24

Ng style (108 forms) Practiced TC 45 min 3  wk; 3 yrs or more

Executive function CTT 1 and 2 Learning/memory HKLLT RBMT-CV

Abbreviations: ADAS-Cog (Chinese) ¼ Alzheimer’s Disease Assessment Scale-Cognitive Subscale (Chinese); AMIC ¼ Abbreviated Memory Inventory for the Chinese; AN ¼ Animal Naming; BNT-C ¼ Chinese version Boston Naming Test; CDR ¼ Clinical Dementia Rating; CDT ¼ Clock Drawing Test; COWAT ¼ Controlled Oral Word Association Test; CTT ¼ Color Trails Test; CVFT ¼ Category Verbal Fluid Test; CVLT ¼ California Verbal Learning Test; DR ¼ Delayed Recall; DS ¼ Digit Span; DSB ¼ Digit Span Backward; DSF ¼ Digit Span Forward; DSym ¼ Digit Symbol; ERP ¼ Event Related Potentials; HKLLT ¼ Hong Kong List Learning Test; HKLLT-1 ¼ verbal memory test, delayed recall; MASQ ¼ Multiple Abilities Self-Report Questionnaire; MMSE ¼ Mini-Mental State Examination; MMSE-C ¼ Mini-Mental State Examination-Cantonese; NS ¼ Not Significant; RAVLT ¼ Rey Auditory Verbal Learning Test; RBMT ¼ Rivermead Behavioral Memory Test; RBMT-CV ¼ Rivermead Behavioral Memory Test-Chinese Version; RCT ¼ Randomized Controlled Trial; SD ¼ Standard deviation; SMC ¼ Subjective Memory Complaint; SOT ¼ Sensory Organization Test; ST ¼ Stroop Test; TC ¼ Tai Chi; TMT ¼ Trail Making Test; VSB ¼ Visual Span Backward; WMS-LM ¼ Wechsler Memory Scale-Logical Memory.

S.M. Miller, R.E. Taylor-Piliae / Geriatric Nursing 35 (2014) 9e19

Lam, 2009, Hong Kong

Cardiovascular and MindeBody group recalled significantly more words than No Exercise group (F [3,136] ¼ 6.39; p