Various physical activity programs have been developed for relieving symptoms of dementia [
The physical activities that are recommended for patients with dementia include aerobic exercises, muscle-strengthening exercises, hydrotherapy, exercises involving music [
Nevertheless, the general population may be confused because of the diversity among the options about the effects of physical activity. Therefore, there is a need to suggest the most effective type of exercise and create guidelines for designing physical activity programs.
Some meta-analyses on the effects of exercise for subject with dementia were published. But, they included the articles that targeted not only elderly patients with dementia but also elderly patients with mild cognitive impairment for analyzing [
In this study, a meta-analysis of the physical activity programs for elderly patients with dementia was conducted. Our purpose was to confirm that physical activity program improves the symptoms of dementia and the most effective physical activity to help establish exercise programs.
Three databases, PubMed (158), Science Direct (129), and Wiley online library (450), were used to collect papers on dementia treatment. The databases were published between January 2005 and December 2015. Keywords such as “dementia,” and “physical activity” were used in searching for papers. Outcomes were not included in the searching. Two independent reviewers screened the tile and abstracts of all studies to identify randomized controlled trials or duplicate study.
We considered the following primary outcomes: the cognitive function, physical capacity, ability of activity of daily living, and psychological state. The secondary outcomes were Geriatric Depression Scale, Wechsler Memory Scale, Rivermead Behavioural Memory Test, Symptom Check List Anxiety, cardiopulmonary function, muscle strength, endurance, flexibility, activity of daily living disability score, 6 M walk, Neuropsychiatric Inventory Total Score, Montgomery-Asberg Depression Rating Scare, Mini Mental State Examination, Alzheimer’s Disease Assessment Scale, Functional Reach Test, Time Up and Go Test, sit to stand test, Barthel Index, Instrumental ADL, and Cambridge Neuropsychological Test Automated Battery.
Analyses were conducted through the following process. Based on the agreements among the coauthors of this study, data coding was conducted with items comprising the author’s name, year of publication, publishing type, study model, study participants, genders of the participants, survey tool used, program type, main residence of the participants, and program effects. The coding was conducted by one graduate school student and one meta-analysis specialist. When a discrepancy in coding occurred, the opinions of the coauthors were reflected to address it. Accordingly, the reliability and consistency between the people who were coding were not calculated together.
Analysis of the tables and calculation pertaining to effect sizes and
The result of the homogeneity test for the sampling.
Number of studies |
|
df |
|
|
---|---|---|---|---|
9 | 102.17 | 8 | 0 | 92.17 |
1646 articles were included for searching strategy. From these articles, 1379 were excluded for a detailed analysis and 267 articles were screened. After full-text reading, thirty-three of full-text articles were excluded for the following reasons. First, eight articles included the inadequate control group (there was not control group or the subject of control group is not dementia patient). Second, four articles were consisted of the different intervention duration between groups. Third, eight articles described unclear result that did not provide enough statistical data pertaining to meta-analyses. Forth, two articles unclearly described the number of participants. Fifth, four articles included the unclear participants that mixed the subject with dementia and MCI. Sixth, three articles used quality evaluation. Seventh, four articles described only value of pretest or posttest as results. Finally, nine studies were selected in the second screening of the meta-analyses. Figure
Characteristics of included trials.
Study | Type of intervention |
|
Primary/secondary outcome | Items of intervention | Frequency | Duration (h) |
---|---|---|---|---|---|---|
Cheng et al. [ |
Special physical activity | 12 | Psychology state/GDS | Mahjong, Tai Chi | 3 (60 min) | 24 |
|
||||||
Eggermont et al. [ |
Special physical activity | 30 | Cognitive function/Wechsler Memory Scale revised, RBMT, Eight-Word Test (recognition) | Hand movements | 5 (30 min) | 6 |
Psychology state/GDS and SCL Anxiety | ||||||
|
||||||
Kwak et al. [ |
Combined exercise | 15 | Daily activities | Stretching, thera-band, swiss ball, wall bar, dumb bell | 5 (60 min) | 6 |
Physical capacity/cardiopulmonary function (m), muscle strength (kg W), muscular endurance (time), flexibility (cm), balance (s), agility (s)] | ||||||
Cognitive function/MMSE | ||||||
|
||||||
Rolland et al. [ |
Walking with combined exercise | 56 | Activities of daily living | Collective exercise program (walk, strength, balance, and flexibility training) | 2 (60 min) | 48 |
Physical capacity/(6 min walk speed (m/s), Get up and go (test score) | ||||||
Psychology state (NPI, MADRS) | ||||||
|
||||||
Sung et al. [ |
Special physical activity | 18 | Psychology state (occurrence of agitated behavior) | Music with movement intervention | 2 (30 min) | 4 |
|
||||||
Venturelli et al. [ |
Walking | 11 | Physical capacity/6 min walk (m) | Walked up and down hallway | 4 (30 min) | 24 |
Activities of daily living/Barthel, cognitive function/MMSE | ||||||
|
||||||
Vreugdenhil et al. [ |
Walking with combined exercise | 20 | Cognitive function/MMSE, ADAS-Cog | Community-based home exercise program [brisk walking (30 min), strength, balance training] + usual treatment | 7 (30 min) | 16 |
Physical capacity/functional reach test (cm), Time Up and Go (s), sit to stand (number), waist/hip ratio, body mass index (kg/m2) | ||||||
Abilities of daily living/Barthel index of ADL, Instrumental ADL | ||||||
Psychology state/GDS | ||||||
|
||||||
Yágüez et al. (2011) [ |
Combined exercise | 15 | Cognitive function/CANTAB-Expedio | The Brain Gym training (activate balance, stretching, circular movements) | 1 (90 min) | 6 |
|
||||||
Eggermont et al. [ |
Walking | 51 | Cognitive function/Wechsler Memory Scale, RBMT | Walking at a self-selected speed | 5 (30 min) | 6 |
GDS: Geriatric Depression Scale, RBMT: Rivermead Behavioral Memory Test, SCL Anxiety: Symptom Check List, ADL: activities of daily living disability score, NPI: Neuropsychiatric Inventory, MADRS: Montgomery-Asberg Depression Rating Scale, ADAS-Cog: Alzheimer’s Disease Assessment Scale-Cognitive Subscale, CANTAB: The Cambridge Neuropsychological Test Automated Battery.
Flow diagram of studies included.
To ensure the validity of the meta-analysis results, the publication bias was verified. No publication bias was confirmed (Table
The results of publication bias verification results.
Studies trimmed | Point estimate | 95% CI |
|
||
---|---|---|---|---|---|
Lower limit | Upper limit | ||||
Observed values | 0.71 | 0.42 | 0.99 | 102.17 | |
Adjusted values | 0 | 0.71 | 0.42 | 0.99 | 102.17 |
The improvement in the dementia symptom of physical capacity was 1.05 (high effect size, 95% CI: 0.03 to 0.73), activities of daily living were 0.73 (slightly high effect size, 95% CI: 0.23 to 1.23), cognitive function was 0.46 (medium effect size, 95% CI: 0.26 to 0.66), and psychological state was 0.39 (lower than the medium effect size, 95% CI: 0.01 to 0.77) (Table
Effect size according to improvement of the dementia symptom.
Group | Number of studies | Point estimate | 95% CI |
|
---|---|---|---|---|
Physical capacity | 14 | 1.05 | 0.03–0.73 | 0 |
Ability of activity of daily living | 5 | 0.73 | 0.23–1.23 | 0.004 |
Cognitive function | 29 | 0.46 | 0.26–0.66 | 0 |
Psychological state | 8 | 0.39 | 0.01–0.77 | 0.045 |
According to type of physical activity, the effect size of combined exercise program was 1.17 (high effect size, 95% CI: 0.86 to 1.47). The walking program was 0.46 (medium effect size, 95% CI: 0.18 to 0.74). Special physical activity program was 0.44 (medium effect size, 95% CI: 0.16 to 0.72). Walking program with other activities was 0.41 (medium effect size, 95% CI: 0.16 to 0.67) (Table
Effect size according to the type of physical activity program.
Category | Number of studies | Point estimate | 95% CI |
|
---|---|---|---|---|
Combined | 15 | 1.17 | 0.86–1.47 | 0 |
Walking | 13 | 0.46 | 0.18–0.74 | 0.001 |
Special physical activity | 13 | 0.44 | 0.16–0.72 | 0.002 |
Walking with other activities | 15 | 0.41 | 0.16–0.67 | 0.002 |
In this study, various physical activity programs for elderly patients with dementia were analyzed using the papers published since 2005 to find out the programs that are effective for relieving dementia symptoms and to suggest the most effective program.
The physical activity programs were very effective in improving the physical capability and ADL of patients with dementia but showed a small effect size in cognitive function or psychological states.
With aging, the frequency of occurrence of depression is increasing in the women with low level of physical activity [
In this study, the assessment tools of depression were Geriatric Depression Scale and Montgomery-Asberg Depression Rating Scale that lower score indicates normal state. We found that physical activity induces lower score of depression. This result can interpret positive effect in depression even though the effect size was small.
In Heyn et al.’s meta-analysis study [
In terms of the effect size by the type of physical activity, the combined exercise method showed the largest size while the walking-alone method showed the smallest size among all of the methods. These results are similar to those of Scherder’s meta-analysis [
Many studies recommend walking exercises because they are effective for cognition [
The World Health Organization (WHO) [
Brain-derived neurotropic factor (BDNF) increases with aerobic and strengthening exercises. BDNF is known to help in supporting the survival of existing neurons and encouraging the growth of new neurons [
Boyle et al. [
Marosi et al. [
Lista and Sorrentino [
Additionally, physical activity can affect the apolipoprotein E (
According to earlier studies, aerobic exercise and strength exercise can affect the cognitive decline that is related to dementia, even though the mechanism remains unclear. Therefore, combined exercise, which consists of aerobic exercise and strength exercise, might be beneficial for elderly people with dementia. The outcomes of this study can serve as a scientific groundwork for the claim that integrated exercises, combining various exercises, are the most appropriate exercises among the various types of known exercise. Therefore, combined exercise may be required for elderly patients with dementia.
We selected the article that targeted subject with dementia and the assessment tools of the cognition that the reliability and validity were proven were used in the paper. We decided the assessment of tool in the targeting paper was for subject with dementia and it was not associated with normal older people. Therefore, we did not consider the difference of cognition tool among the papers.
In this study, the effects of exercise programs were reviewed; however, the appropriate intensity of the exercises was not analyzed. The number of analyzed papers was most likely too small because the papers that included subjects without a diagnosis of dementia were excluded, many papers described the intervention program effects qualitatively, and many papers were nonrandomized controlled (RCT) studies or did not have control groups.
In this study, the effects of physical activities by type on patients with dementia were objectively meta-analyzed to suggest a treatment program guideline for such patients. The papers with control groups and those that conducted before-after measurements were selected to determine the effect sizes of physical activity programs. The physical activity for patients with dementia had an effect on the improvement of physical capacity. Among the physical activities for patients with dementia, combined exercise, consisting of multiple exercises, showed the largest effect size. Therefore, physical activity programs that include various activities are recommended when treatment for elderly patients with dementia is determined. However, taking into consideration the diversity of the papers that were reviewed in this study, this outcome must be carefully applied to clinical cases.
All authors declare that there is no conflict of interests regarding the publication of this article.