Effects of Limb Exercises on Individuals with Dementia: A Meta-Analysis of Randomized Controlled Trials
With the elderly population growing dramatically, people 65 plus are expected to increase to 20-25% of total population over the next 25 years . It has been well documented that the incidence of dementia and cognitive impairment is increasing with aging . The epidemiological studies reported that older adults who maintain higher levels of physical activities reduced risk for mild cognitive impairment function and dementia. Many research findings have consistently provided the evidence of physical exercises related to cognitive benefits and reduced risk of dementia . Despite the evidence of the benefits of physical exercises or activities for cognitive function or reduce the incidences of dementia, there is a controversy about the effectiveness of types of exercises and activities and systematic practice of physical activities on reducing risk of cognitive function impairment and dementia. Limbs exercises, especially hands exercises and activities, are believed having special contributions to improve cognitive function, however, whether limbs exercises and activities can effectively reduce cognitive function impairment and decrease the risk of dementia, still remaining unclear in the literature. Therefore, the purpose of this study was to use the systematic review and meta-analytic method to analyze the effects of limb exercises and activities on cognition, mood, mobility, and decreasing the risk of cognitive function impairment and dementia among people with Alzheimer’s disease.
Data source and search strategy
This research was conducted following the procedures of the Cochrane Collaboration . Articles published in English language were initially identified by one author from the five major electronic databases from inception until December 27, 2013: PsycINFO, MEDLINE, EMBASE, CINAHL, and SPORTDiscus). Three groups of keywords were used for the search: (i) “Alzheirmer’s disease” OR “dementia” OR “cognitive impairment*”; AND (ii) “exercise” OR “physical activity” OR “upper limb” OR “lower limb” OR “extremity” OR “walking” OR “jogging” OR “hand movement” OR “foot exercise” OR “strength training” OR “stretching”, which were combined with “OR”; “Alzheimer’s disease”, “dementia” and “cognitive impairment”; AND (iii) “randomized controlled trail*” OR “intervention”. Additional articles were manually searched from the relevant reviews and reference lists initially identified above. Inclusion and exclusion criteria Articles that fulfilled all the following inclusion criteria were included: (i) participants were diagnosed with Alzheirmer’s disease; (ii) study design was an RCT; (iii) limb exercises such as walking, hand movement, and stretching training were used as the only or major component of the intervention; (iv) outcomes regarding cognitive, physical, or emotional function were reported. Articles were excluded if (i) participants did not have an Alzheimer’s disease or the study recruited mixed participants without presenting separated outcomes for individuals with dementia; (ii) limb exercises were not used as the (major) intervention or information about the intervention program was lack; (iii) an RCT design was not used; (iv) they did not report cognitive, emotional, or functional mobility outcomes; or (v) they were published in non-English language journals. Two authors independently screened the searched articles by reading titles, abstracts, and full-texts (if necessary) based on these inclusion and exclusion criteria. All the authors agreed with the final list of studies included in this research.
Data extraction and analysis
Two authors independently extracted the following information from the included studies: authors, country, participants’ characteristics, intervention program, and outcome measures, and results (i.e., mean and standard deviation of outcomes). Meta-analysis was conducted through the RevMan 5.2. To assess the heterogeneity of the pooled effect size (ES), Cochran’s Q test and I2 statistic were used (4. Higgins, Thompson, Deeks, & Altman, 2003). The fixed-effect model was selected when a p value was greater than 0.1 and I2 was less than 50%; otherwise, the random effect model was used for meta-analysis. Given the small number of the included studies, further analyses were not conducted to detect the sources of heterogeneity. In the similar vein, we did not evaluate the publication bias within this review.
Risk of bias assessment
To evaluate the methodological quality of the included studies, The Cochrane Collaboration’s tool for assessing risk of bias was used . This assessment tool was agreed to obtain the best available evidence of bias. This tool consisted of seven criteria: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting and other bias. Based on the criteria, a judgment of low bias, unclear bias, or high bias was assigned to each study. Two independent reviewers rated the included studies and a third reviewer was involved in case of disagreement of rating.
Results Study selection
Our search yielded 677 articles. Through screening the information of titles and abstracts of searched studies using the pre-determined inclusion and exclusion criteria, 641 articles were excluded. Subsequently, full-texts of the 36 remaining articles were referred and 31 articles were further excluded. Two studies were identified by hand-searching the relevant review studies and reference lists of the searched studies. Thus, a total of seven articles were included in the current study (see Figure 1).
Figure 1: Diagram of The Study Selection Process
Characteristics of participants
The characteristics of studies included are shown in Table 1. Most of the studies were published after 2005 (1991-2011). These studies were conducted in United States (n = 3), Netherlands (n = 2), South Korea (n = 1), and France (n = 1). A total of 334 participants (experimental group = 175, control group = 159) were recruited with the mean total sample size for each study was 47.7 (30-97). Participants’ mean age ranged from 71.8 to 88.2 years.
Six studies used walking or combined walking with other components (e.g., upper extremity exercise) as the intervention. One study employed hand movement  in its intervention program. The duration of intervention program ranged from six weeks to 12 months. Participants were generally required to involve in two to five training sessions each week and each training session lasted for about 30 to 60 min.
Five studies assessed participants’ cognitive function using the Mini-Mental State Examination , theWechsler Memory Scale-Revised , the Rapid Evaluation of Cognitive Functions Test , and the Riverhead Behavioral Memory Test .Two studies determined participants’ functional mobility using the 6-Mintue Walk  and the Acute Care Index of Function . Two studies evaluated participants’ mood through the Geriatric Depression Scale , the Anxiety from the Symptom Check List , the Dementia Mood Assessment Scale , and the Alzheimer’s Mood Scale .
|First author (year)||Country||Age (years)||Participants (n)||Study duration||Frequency||Experimental
|United States||Exp: 71.8
|12 weeks||30 min/session, 3 times/week||Walking & conversation||Conversation||Cognitive function|
|South Korea||Exp: 79.67
|12 months||30-60 min/session, 2-3 times/week||Walking, stretching & upper/lower extremity exercises||No treatment||Cognitive function &
|Williams (2008)||United States||87.9||Exp: 19
|16 weeks||30 min/session, 5 times/week||Walking||Conversation||Mood|
|Eggermonta (2009)||Netherlands||85.4||Exp: 51
|6 weeks||30 min/session, 5 times/week||Walking||Social visits||Cognitive function|
|Eggermonta (2009)||Netherlands||Exp: 85.0
|6 weeks||30 min/session, 5 times/week||Hand movement||Read aloud||Cognitive function & Mood|
|15 weeks||60 min/session, 3 times/week||Walking, equilibrium, stamina & intellectual activities||Intellectual activities||Cognitive function|
|Roach (2011)||United States||Con: 87.31
|16 weeks||30 min/session, 5times/week||Walking||Conversation||Physical function|
Note. Exp = Experimental Group; Con = Control Group; NA = Not Available.
Table 1: Characteristics of Studies Included
Figure 2 and Table 2 presents the results of methodological assessment of studies included. Most studies (n = 5) adequately generated the allocation sequence. Only three studies showed concealed allocation. All seven studies either had bias of binding of participants or presented insufficient details for our judgment. However, four studies blinded the assessors. Three studies failed to adequately address the incomplete outcome data. In general, most studies (n = 4) were free of suggestion of selective outcome reporting and three studies reported insufficient information for us to determine.
Figure 2: Methodological Assessment for Risk of Bias Graph: Review Authors’ Judgments About Each Risk of Bias Item Presented as Percentages Across All Included Studies.
|Random sequence generation||Allocation concealment||Binding of participants||Blinding of outcome assessment||Incomplete outcome data||Selective reporting||Other bias|
Table 2:Results of Assessment of Risk of Bias
Three studies assessed participants’ overall cognitive function [16,17]. The intervention improved participants’ overall cognitive function, standardized mean difference (SMD) = 0.70, 95%CI (0.36, 1.10), p < .01, I2 = 44%, Q (2) = 3.56, p = .17 (see Figure 3).
Figure 3: Meta-Analytic Result of Overall Cognitive Function
Two studies evaluated participants’ executive function [18,19]. The results indicated that the experiment show no effect on improving participants’ executive function, SMD = 0.04, 95%CI (-0.38, 0.47), p = .85, I2 = 44%, Q(1) = 1.78, p = .18 (see Figure 4).
Figure 4: Meta-Analytic Result of Executive Function
Two studies determined participants’ learning memory and short-term memory ability [18,19]. The meta-analytic results suggested that limb exercises cannot facilitate participants’ learning memory, SMD = 0.03, 95%CI (-0.50, 0.55), p = .92, I2 = 63%, Q(1) = 2.69, p = .10 as well as their short-term memory ability, SMD = 0.03, 95%CI (-0.28, 0.34), p = .86, I2 = 0%, Q(1) = 0.24, p = .63 (see Figures 5 and 6).
Figure 5: Meta-Analytic Result of Learning Memory
Figure 6: Meta-Analytic Result of Short-Term Memory Ability
Two studies measured participants’ mood [18,19]. The meta-analytic result suggested limb exercises cannot improve participants’ mood, SMD = -0.14, 95%CI (-0.55, 0.27), p = .51, I2 = 0%, Q(1) = 0.31, p = .58 (see Figure 7).
Figure 7: Meta-Analytic Result of Mood
Functional mobility outcome
Two studies assessed participants’ functional mobility [21,22]. Our analysis showed that the intervention had no effect on promoting participants’ functional mobility, SMD = 0.80, 95%CI (-0.76, 2.35), p = .32, I2 = 90%, Q(1) = 9.79, p < .01 (see Figure 8).
The present meta-analytic study firstly investigated the benefits of limb exercises on cognitive performance, mood, and functional mobility among people with Alzheimer’s disease. Only seven articles were included in this study and six of which were published in the western countries, implying the need to conduct more studies within the field. According to the results of methodological assessment, most studies provided insufficient details making the judgments difficulty. Future investigations need to account the methodological quality especially for blinding participants/assessors. The rest of this section will revolve around discussing the three outcomes (i.e., cognition, mood, and mobility).
Firstly, our meta-analytic result showed that limb exercises improved participants’ overall cognitive function with a moderate to high ES (.70). This finding is similar to the early study , which found that exercise interventions facilitated overall cognitive function among people with dementia. Our study, however, indicated that limb exercises failed to change participants’ executive function, learning memory, and short-term memory. One possibility is that the duration of the intervention programs of the included studies were too short (i.e., six weeks) to find a positive effect. A limb exercise program with longer duration may lead to positive changes of executive function, learning memory, and short-term memory.
Secondly, based on analyzing two included studies [18,19], we found that participants’ mood was not improved by limb exercise programs (i.e., walking). This might be because the control group of these two included studies used either social visits or casual conversations as a comparison. Social interaction has been found to improve old adults’ mood .
Lastly, the effectiveness of limb exercises on participants’ mobility was meta-analytically investigated based on the two included studies [21,22]. The result showed no positive effect of limb exercises on mobility but with a large effect size (0.80). If we look into each of the included studies, it seems that the duration of intervention resulted in a difference. The study by Kwak et al. (2008) found a significant and large effect (SMD = 1.63) with a 12-month intervention period, whereas Raoch et al. (2011) used a 4-month intervention program and found no change. Another possibility is the difference of the intervention program used between the two studies. The one with mixed components (i.e., walking combined with stretching and upper/lower extremity exercises) showed more positive effect on improving participants’ mobility.
Similar to most of the previous meta-analytic studies, this study was also limited to include only those articles published in English, which may exclude relevant studies in other languages. Bearing this limitation in mind, this study showed that there is a trend that limb exercises can enhance overall cognitive function among people with dementia. Given a small number of studies included, it is still unclear whether limb exercises can bring positive changes of mood and mobility for this population. We believe that the current review study shall shed light on the future research efforts within the field.
There is a trend that limb exercises can positively improve overall cognitive function in individuals with Alzheimer’s disease. There is lack of evidence regarding the effectiveness of limb exercises on executive function, learning memory, short-term memory, mood, and mobility for this population. More well-designed studies are needed to examine the efficacy of limb exercises on these interested outcomes.
- Alexopoulos GS, Abrams RC, Young RC, Shamoian CA. Cornell Scalefor Depression in Dementia. Biological Psychiatry. (1988), 23(3): 271-284.
- Arrindell WA, Ettema JH (1986). SCL-90. Manual to a multi-dimensional psychopathology indicator [In Dutch: Handleiding bij een multidimensionele sychopathologie-indicator]. Lisse: Swets & Zeitlinger.
- Barber SE, Clegg AP, Young JB. Is there a role for physical exercise in preventing decline in people with mild cognitive impairment? Age Ageing. (2012), 41(1): 5-8.
- Colcombe S, Kramer AF. Fitness effects on the cognitive function of old adults: A meta-analytic study. Psychol. Sci. (2003), 14(2): 125-130.
- Chiu HFK, Lee HC, Chung WS, Kwong PK. Reliability and validity of the Cantonese version of Mini-Mental State Examination – a preliminary study. Hong Kong Journal of Psychiatry. (1994), 4(2): 25-28.
- Eggermonta LHP, Knolb DL, Hol EM. Hand motor activity, cognition, mood, and the rest–activity rhythm in dementia: A clustered RCT. Behavioural Brain Research. (2009), 196(2): 271-278.
- Eggermonta LHP, Swaab DF, Hol EM, Scherder EJ. Walking the line: A randomised trial on the effects of a short term walking programme on cognition in dementia.Journal of Neurology, Neurosurgery & Psychiatry. (2009), 80(7): 802-804.
- Etnier JL, Nowell PM, Landers DM, Sibley BA. A meta-regression to examine the relationship between aerobic fitness and cognitive performance. Brain Res. Rev. (2006),52(1): 119-130.
- Farina N, Rusted J, Abet N. The effect of exercise interventions on cognitive outcome in Alzheimer’s disease: A systematic review.International Psychogeriatrics. (2014), 26(1): 9-18.
- Folstein MF, Folstein SE, Mchugh PR. Mini-Mental State-practical method for grading cognitive state of patients for clinician. Journal of Psychiatric Research. (1975), 12(3): 189-198.
- Forbes D, Forbes S, Morgan DG, Markel-Reid M, Wood J, et al. Physical activity programs for persons with dementia (Review). Cochrane Database Systematic Review, 16, CD006489. (2008).
- Friedman RM. The effect of planned walking on communication in Alzheimer’s disease. Journal of the American Geriatrics Society. (1991), 39(7): 650-654.
- Gates N, Fiatarone Singh MA, Sachdev PV, Valenzuela M. The effect of exercise training on cognitive function in older adults with mild cognitive impairment: A meta-analysis of randomized controlled trials. American Journal of Geriatric Psychiatry. (2013), 21(11): 1086-1097.
- Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of Interventions, version 5.1.0. The Cochrane Collaboration. (2011).
- Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BritishMedical Journal. (2003), 327(7414): 557-560.
- Lautenschlager NT, Cox KL, Flicker L, Foster JK, van Bockxmeer FM. Effects of physical activity on cognitive function in older adults at risk for Alzheimer disease. JAMA. (2008), 300(9): 1027-1037.
- Kemoun G, Thibaud M, Roumagne N, Carette P, Albinet C. (2010). Effects of a physical training programmer on cognitive function and walking efficiency in elderly persons with dementia Dementia and Geriatric Cognitive Disorders. (2010), 29: 109-114.
- Kok RM, Heeren ThJ, van Hemert AM. De geriatric depression scale. Tijdschr Psychiatr. (1993), 35: 416-421.
- Kwak YS, Um SY, Son TG, Kim DJ. Effect of regular exercise on senile dementia patients. International Journal of Sports Medicine. (2008), 29(6): 471-474.
- Muangpaisan W, Intalapaporn S, Assantachai P. Digit span and verbal fluency tests in patients with mild cognitive impairment and normal subjects in Thai-community. Journal of Medical Association of Thailand. (2010), 93(2): 224-230.
- Roach KE, Tappen RM, Kirk-Sanchez N, Williams CL, Loewenstein D. A randomized controlled trial of an activity specific exercise program for individuals with Alzheimer disease in long-term care settings. Journal of Geriatric Physical Therapy. (2011), 34(2): 50-56.
- 22. Roach KE, Van Dillen LR. Development of an acute care index of functionfor the neurological patient.Physical Therapy. (1988), 68(7): 1102-1108.
- Sunderland T, Alterman I, Yount D, Hill J, Tarior P, et al. A new scale for the assessment of depressed mood in demented patients. American Journal of Psychiatry. (1988), 145(8): 955-959.
- Tappen RM, Roach KE, Buchner D, Barry C, Edelstein J. Reliability of physical performance measures in nursing home residents with Alzheimer’s disease. Journals of Gerontology Series A: Biological Sciences and Medical Sciences. (1997), 52: M52-M55.
- Tappen RM, Williams C. Alzheimer’s Mood Scale: Validity and reliability. Abstract presented at the Gerontological Society of America 52nd annual scientific meeting. San Francisco, CA. (1999).
- Tsai HH, Tsai YF. Changes in depressive symptoms, social support, and loneliness over 1 year after a minimum 3-month videoconference program for older nursing home residents. Journal of Medical Internet Research. (2011), 13(14): 373-384.
- Williams CL, Tappen RM. Exercise training for depressed older adults with Alzheimer’s disease. Aging & Mental Health. (2008), 12(1): 72-80.
- Wilson B, Cockburn J, Baddeley A. The Rivermead Behavioural Memory Test. Titchfield: Thames Valley Test Company. (1987).
- Corrada M, Brookmeyer R, Paganini-Hill A. Dementia incidence continues to increase with age in the oldest old. Ann Neurol. (2010), 67(1): 114-121.
- Mather M. Fact sheet: Aging in the United States. (2016).