Data Abstraction and Data Synthesis
Study data were analyzed according to study design, setting, participant and provider characteristics, country of intervention, duration and types of intervention, the way the intervention was conducted, cognitive measure tools used, results, and attrition rate. Extracted data were collated and displayed in tabular summaries. Cross-checking was used to reduce data handling error risk. The narrative synthesis of findings from selected trials was constructed from textual descriptions, complemented with tabular summaries and results.
All 12 trials examined used randomized controlled designs. Most (66.7%) were conducted by a multidisciplinary research team, and most took place in the United States. The first authors' professional areas tended to be in the sports sciences (58.3%) rather than in medicine, physical therapy, or nursing. The number of participants in each trial ranged from 30 to 202. Older adult participants with and without cognitive impairment were 50%, respectively, and only 16.7% of the trials focused on female participants.
Intervention Duration and Types
The wide range of exercise regimens used included walking, treadmill running, extremity stretching exercise, weight-bearing strength training, and swimming. The interventions used in the trials included moderate- and high-intensity resistance exercise programs that were often designed for a healthy older population (Carral & Perez, 2007; Cassilhas et al., 2007) and aerobic and strength-and-flexibility activities (Colcombe et al., 2006, Hoffman et al., 2008; Smiley-Oyen, Lowry, Francois, Kohut, & Ekkekakis, 2008). The duration of interventions ranged from 6 weeks to 12 months (Colcombe et al., 2006; Kwak, Um, Son, & Kim, 2008). The majority lasted 24 weeks. The exercise sessions themselves lasted from anywhere from an unknown amount of time (Hoffman et al., 2008) to 2 hours (Christofoletti et al., 2008). Five trials (41.7%) with exercise sessions of 60 minutes comprised the majority. Only one trial had a session lasting 2 hours. Three times per week was the most common frequency (10 trials, 83.3%).
Cognitive Measure Tools
Cognitive measure tools vary based on trial requirements (Table 1). The 12 trials examined in this study employed a total of 63 different cognitive tests. Cognitive measure tools in the study samples can be summarized into global and domain specific. The Mini-Mental State Examination (MMSE), Alzheimer's Disease Assessment Scale-Cognitive Behavior section, Clinical Dementia Rating, and MRI brain volume were included in the global cognitive measures. Domain-specific cognitive measures included memory, executive functions, verbal fluency, information processing speed, recognition, and reaction time. Only 3 trials used global cognitive measures, with the rest using a combination of global and domain-specific cognitive measures.
Because some trials listed only the items measured and not the tools employed, it was difficult to determine whether a consistent tool had been used to measure the same item from study to study. Therefore, this review did not analyze cognitive measuring tools.
The systematic review included participants who were divided into two groups with normal cognitive function and impaired cognitive function, as described in the following. The cognitive status of trial participants, normal or impaired, was adopted per the author's definition in the trial and not by cognitive tests.
Effectiveness of exercise on older people with normal cognitive function
Of 12 exercise trials, 6 focused on older people with normal cognitive function. Colcombe et al. (2006) randomly assigned 59 healthy older adults (average MMSE score = 29.4) to participate in either a cardiovascular exercise group (EG) or a nonaerobic exercise control group (CG) for three 1-hour exercise training sessions per week lasting for a 6-month period. They measured magnetic resonance imaging and maximal oxygen uptake before and after the 6-month fitness intervention. After the intervention, the aerobically exercising older adults showed a significant increase in maximal oxygen uptake and brain volume in the dorsal anterior cingulate cortex/supplementary motor area, middle frontal gyrus, left superior temporal lobe, and anterior white matter tracts as compared with the older adult controls.
Carral and Perez (2007) selected 62 community-dwelling older women with MMSE scores ≥22 participating in a randomized trial. They assigned participants either to a combined program of aquatic exercise plus high-intensity strength training (Group 1) or callisthenic training (Group 2). Both groups trained 5 days a week for 5 months. After the program, both groups obtained significant cognitive improvement in MMSE scores (Group 1, p = .034; Group 2, p = .021).
Cassilhas et al. (2007) reported that a 24-week (3 days/week) regimen of moderate- and high-intensity resistance exercise program designed for healthy older people produced equally beneficial effects on cognitive function. Sixty-two participants were randomly assigned to control (n = 23), moderate-resistance (n = 19), and high-resistance (n = 20) groups. The results showed the two experimental groups performed significantly better than the CG in forward digit span, Corsi's block-tapping (backward, similarities), Toulous-Pieron test (error), and Rey-Osterrieth (immediate recall).
Smiley-Oyen et al. (2008) designed an RCT consisting of two interventions and pre-post tests. Fifty-seven older adults participated in either thrice weekly aerobics (CARDIO group, n = 28) or strength-and-flexibility (FLEX-TONE group, n = 29) exercise training for 10 months. Cognitive function at 10 months for the two EGs both surpassed levels at 4-5 months and at 1 month. Particularly in the CARDIO group, Stroop test values such as word-color conflict (p = .001) and word-color conflict error (p = .002) showed significant improvement.
Marmeleira et al. (2009) randomly assigned 32 healthy older individuals (MMSE scores ≥28) to either an EG (n = 16) or a CG (n = 16). Participants in the EG in the supervised exercise program faced largely cognitive demands, including activities such as dual-tasking, working with peripheral vision, planning and decision making, dependence on working memory, targeting speed processing, and response inhibition. Each session lasted approximately 60 minutes, 3 days a week for 12 weeks. Significant positive cognitive effects from participation in the EG were identified at the 12-week follow-up.
In conclusion, the results of the previously mentioned studies indicated a positive effect of exercise on cognitive function in healthy older people. On the contrary, Hoffman et al. (2008) concluded little evidence to support the benefits of an aerobic exercise intervention on neurocognitive performance in patients with major depressive disorder but without a bipolar disorder or psychotic depression. A total of 202 sedentary advanced older individuals with major depressive disorder with no cognitive impairment were randomly assigned to one of four groups, namely, (1) supervised exercise, (2) home-based exercise, (3) sertraline (an antidepressant of the selective serotonin reuptake inhibitor), and (4) placebo pill. After the 16-week treatment, the performance of participants who exercised was no better than that of participants who received placebo across all neuropsychological tests (executive function, verbal memory, verbal fluency/working memory).
Effectiveness of exercise on older people with impaired cognitive function
Six exercise trials were designed for older people who had impaired cognitive function. Scherder et al. (2005) compared the effectiveness of two types of exercise with varying intensities: walking and hand/face exercises. The study randomly assigned the 43 older individuals with mild cognition impairment (mean short form MMSE score = 9.61) into three groups, namely, walking, hand/face exercises, and control (n = 15, 13, 15, respectively). All participants received individual treatment for 30 minutes a day, three times a week, for 6 weeks. Performance on Category Naming and Trail-making A + B was significantly better in both EGs after 6 weeks of treatment, but no significant difference among the three groups at the 12-week follow-up was observed. These findings suggested that the treatment effect could not be maintained during nontreatment periods. However, only 2 of 12 trials continued to track the long-term effects after exercise intervention, and both focused on the cognitively impaired older people.
Kwak et al. (2008) conducted a trial with a prospective, two-group design, where 30 female patients with senile dementia who participated in the study were assigned into two groups: an EG (n = 15) and a CG (n = 15). The EG completed an exercise program, such as stretching, upper extremity exercise, lower extremity exercise, and walking, lasting 30-60 minutes per day, two to three times per week, for 12 months. The exercise program resulted in significant improvements in MMSE scores (pre: 14.53, post: 19.07; p < .01), as compared with those in the CG.
In another study by Lautenschlager et al. (2008), a significant difference in the Alzheimer's Disease Assessment Scale-Cognitive Behavior section, words list delay recall, and Clinical Dementia Rating sum of boxes between the experimental group and the CG was observed after a walking exercise program trial in older adults at risk for Alzheimer for 50-minute sessions, three times a week, over a 24-week span.
However, opposite results have also been observed. Christofoletti et al. (2008) recruited 54 institutionalized older people with mixed dementia (mean MMSE score = 15.3) and allocated them into three groups, namely, Group 1, an interdisciplinary program composed of physiotherapy, occupational therapy, and physical education (n = 17); Group 2, physiotherapy program (n = 17); and Group 3, control (n = 20). The two experimental groups received physiotherapy of varying intensities for 6 months. Results showed that global cognition did not improve through treatment, but attenuation was observed in the decline of global cognition on two specific cognitive domains (clock drawing, verbal fluency test) in Group 3.
Stevens and Killeen (2006) also found results showing exercise as ineffective. They examined the cognitive effect of exercise on nursing home residents with mild to moderate dementia. Seventy-five residents were randomly assigned to three groups. Group 1 (n = 30) received no intervention; Group 2 (n = 21) received a social visit equivalent in duration and frequency to the exercise program in group 3; and Group 3 (n = 24) undertook a 30-minute group exercise program three times per week for 12 weeks. Findings from the clock drawing test indicated that exercise may slow the dementia-related decline of cognitive symptoms, although results lacked statistical significance (p = .524).
Researchers van Uffelen, Chinapaw, van Mechelen, and Hopman-Rock (2008) designed an RCT in which 152 participants with mild cognitive impairment were assigned to two interventions. These included (1) either a 60-minute, twice-weekly, group-based program of moderate-intensity walking (n = 77) or a low-intensity placebo activity program (n = 75) for 1 year and (2) daily vitamin B complex pills (n = 78) or placebo pills (n = 74) for 1 year. Findings revealed that the walking program and/or vitamin B supplementation did not improve cognition in either the 6-month or 1-year follow-up. However, in the walking program, each percentage increase in attendance by women enhanced their attention performance (Stroop Color Word Test-Attention, p = .04). Men with at least 75% attendance showed memory benefits (Auditory Verbal Leaning Test, p = .04).
Although the 12 trials followed disparate exercise regimens, frequency, intensity, and duration were similar. Trials showed a positive effect for exercise on cognition when the exercise regimen lasted for 6 weeks and occurred at least three times per week for 60 minutes. However, there were issues about research design and quality, participants, tools, and results that should be further clarified and discussed.
In terms of research design and quality, although all trials were randomized, prospective, two-or-more group pre-post test designs, only three trials used large samples sizes (Hoffman et al., 2008; Lautenschlager et al., 2008; van Uffelen et al., 2008). The trials included in this systematic review were relatively small, with trial participants unevenly distributed between healthy and cognitively impaired older individuals. The score for reporting quality was centralized at 11 points (medium quality). However, only two (16.7%) studies (Lautenschlager et al., 2008; van Uffelen et al., 2008) reported explicitly on concealment of allocation, appropriate blinding, and the use of intention-to-treat for handing attrition in the analysis. Thus, application of the findings of this study should relatively conservative due to the ambiguous methodological description of the trials examined.
Although 2 of the 12 studies (Lautenschlager et al., 2008; Scherder et al., 2005), both focusing on older people with impaired cognition, followed up on the long-term effects of exercise intervention, their findings were completely opposite from one another. In examining the results of two trials, we found that Lautenschlager et al. (2008) analyzed only participants who achieved the equivalent of 70,000 steps or more per week after the exercise program, thus correlating high compliance and continuous exercise with continued positive effects on cognitive function.
In terms of subject issues, participants covered in this study were of two cognitive function types, namely, normal and impaired. The cognitive function presented by the MMSE score was not consistent in terms of "healthy" or "impaired" status. The average MMSE score for older people with impaired cognition (van Uffelen et al., 2008; MMSE score = 27-28) was greater than in those with normal cognitive function (Carral & Perez, 2007; MMSE score = 22.07-24.39).
Four trials, however, did not present MMSE scores (Cassilhas et al., 2007; Hoffman et al., 2008; Smiley-Oyen et al., 2008; Stevens & Killeen, 2006). In addition, MMSE score was profoundly influenced by years of education and age (Anthony, Leresche, Niaz, VonKorff, & Folstein, 1982; Tombaugh & McIntyre, 1992). However, most studies examined did not record years of education in their demographic data or the adjusted MMSE cutoff score by years of education (Carral & Perez, 2007; Cassilhas et al., 2007; Kwak et al., 2008; Stevens & Killeen, 2006; van Uffelen et al., 2008). Therefore, any inferences that the positive effect on cognition was found more frequently in the trials designed for healthy older people should be carefully considered.
Subsequently, discussing research tools, most studies adopted multiple domain-specific cognitive measures. Marmeleira et al. (2009) applied 17 cognitive tests, and Smiley-Oyen et al. (2008) spent up to 90 minutes for cognitive tests. Because of the need for a special computer or instruments for measuring, testing often took place in a laboratory rather than clinical setting. Such a complicating, time-consuming, and difficult implementation of cognitive tests in a scientific setting may obstruct observations of cognition in terms of the intended effects as well as disallow comparison of cognitive effects on various trials. Angevaren et al. (2008) recommended the development and use of a smaller battery of cognitive tests to render research on cognition transparent and heighten result reproducibility.
Finally, 8 of the 12 trials (66.7%) demonstrated the positive effect of exercise on cognition when the exercise regimen lasted for 6 weeks and occurred at least three times per week for 30 minutes. This finding is similar to the one reported in the meta-analysis by Colcombe and Kramer (2003), whereby they concluded that 30-minute aerobic exercise enhanced the cognitive capacity of adults aged 65-70 years.
From analysis of the five studies that focused on healthy older people and the three that focused on older people who had impaired cognition at baseline that showed positive effects, it appeared that simple, one-component exercise was better for older people with cognitive impairment (2/8, 25%). Healthy older people demonstrated positive effects with multicomponent exercise (5/8, 62.5%; Table 2). This indicated that exercise programs should not be too complicated for cognitively impaired older people and be diversified for healthy older people.
Strengths of the current review are its selection of RCTs with medium to high quality and its focus on older people. Positive findings added support to the beneficial effects of exercise on cognitive function in older adults. This review continued the work of two previous systematic reviews. Because of the short review time, the limited number of research articles may have biased the results. Also, as only articles published in English were selected, it should not be ruled out that studies in which exercise demonstrated statistically significant effects on cognitive function have been published in other languages. Finally, all included trials differed in several dimensions, including participant characteristics, inclusion/exclusion criteria, exercise content, and outcome evaluations, which may limit the generalizability of exercise as a therapeutic modality for improving cognitive function in older people.
Implications for Nursing Research
Findings suggest the potential of exercise to improve older people's cognitive function. However, additional larger research trials with more rigorous methodology are required to evaluate the sustained effectiveness of exercise on cognitive function and to determine whether multicomponent exercise is more effective in improving cognitive function, particularly for healthy older people. The use of the MMSE as a cognitive screening tool must include the MMSE score, and years of education must be part of demographic data. In addition, an adjusted MMSE with a cutoff score by years of education should be included. Consequently, one could compare the cognitive function of participants at baseline and infer whether the positive effect on cognition is found more frequently in the studies designed for healthy older people.
Finally, the attrition rate in the examined studies ranged from 0% to 24.1%. This factor is crucial to assessing the effectiveness of exercise on cognitive function. Personal preference, type of exercise, and motivation in older people should also be considered. Qualitative research should be used to interview those older people unable to sustain exercise in order to ascertain the reasons for such. On the basis of qualitative results, researchers would then be better equipped to design individualized exercise programs and introduce incentives for older people to enhance exercise participation and decrease dropout rates.
This review highlighted the efficacy of exercise intervention in enhancing cognitive function in older people. However, generally small sample sizes; variations in frequency, intensity, and duration of exercise; lack of evaluation of sustained effects; various outcome measures of cognition; and generally high dropout rates were prevalent weaknesses and limitations in these studies. There continues to be an urgent need in this research area for good-quality research with appropriately larger sample sizes. It would be beneficial to develop an appropriate and sustainable exercise regimen to put into practice, monitor short- and long-term effects on cognitive function, and derive and validate simple indices for cognitive assessment. Such should be examined in future studies.
*References marked with an asterisk indicate studies included in the meta-analysis.
Angevaren, M., Aufdemkampe, G., Verhaar, H., Alemen, A., & Vanhees, L. (2008). Physical activity and enhanced fitness to improve cognitive function
in order people without know cognitive impairment. Cochrane Database of Systematic Reviews
, 1, 1-104.
Anthony, J., Leresche, L., Niaz, U., VonKorff, M., & Folstein, M. (1982). Limits of the "Mini-Mental State" as a screening test for dementia and delirium among hospital patients. Psychology Medicine
Bennett, J. A. (2005). The Consolidated Standards of Reporting Trials (CONSORT). Guidelines for reporting randomized trials. Nursing Research
*Carral, C. J. M., & Perez, C. A. (2007). Effects of high-intensity combined training on women over 65. Gerontology
*Cassilhas, R. C., Vuiana, V. A. R., Grassmann, V., Santos, R. T., Santos, R. F., Tufik, S., et al. (2007). The impact of resistance exercise
on the cognitive function
of the elderly. Medicine & Science in Sports & Exercise
*Christofoletti, G., Oliani, M. M., Gobbi, S., Stella, F., Teresa, L., Gobbi, B., et al. (2008). A controlled clinical trial on the effects of motor intervention on balance and cognition in institutionalized elderly patients with dementia. Clinical Rehabilitation
*Colcombe, S. J., Erickson, K. I., Scalf, P. E., Kim, J. S., Prakash, R., McAuley, E., et al. (2006). Aerobic exercise
training increases brain volume in aging humans. Journals of Gerontology. Series A: Biological Sciences & Medical Sciences
, 61A(11), 1166-1170.
Colcombe, S., & Kramer, A. F. (2003). Fitness effects on the cognitive function
of older adults: A meta-analytic study. Psychological Science
Cotman, C. W., & Berchtold, N. C. (2002). Exercise
: A behavioral intervention to enhance brain health and plasticity. Trends in Neurosciences
Department of Statistics, Ministry of the Interior. (2011, January 8). MOI statistical bulletin: Analysis of population structure at the end of 2010.
Retrieved February 22, 2011, from http://www.moi.gov.tw/files/news_file/week10002.doc
. (Original work published in Chinese)
*Hoffman, B. M., Blumenthal, J. A., Babyak, M. A., Smith, P. J., Rogers, S. D., Doraiswamy, P. M., et al. (2008). Exercise
fails to improve neurocognition in depressed middle-aged and older adults. Medicine & Science in Sports & Exercise
*Kwak, Y., Um, S., Son, T., & Kim, D. (2008). Effect of regular exercise
on senile dementia patients. International Journal of Sports Medicine
*Lautenschlager, N. T., Cox, K. L., Flicker, L., Foster, J. K., van Bockxmeer, F. M., Xiao, J., et al. (2008). Effect of physical activity on cognitive function
in older adults at risk for Alzheimer disease: A randomized trial. Journal of the American Medical Association
*Marmeleira, J. F., Godinho, M. B., & Fernandes, O. M. (2009). The effects of an exercise
program on several abilities associated with driving performance in older adults. Accident
, Analysis and Prevention
Netz, Y., Wu, M. J., Becker, B. J., & Tenenbaum, G. (2005). Physical activity and psychological well-being in advanced age: A meta-analysis of intervention studies. Psychology and Aging
Pang, F. C., Chow, T. W., Cumming, J. L., Leung, V. P., Chiu, H. F., & Lam, L. C. (2002). Effect of neuropsychiatric symptoms of Alzheimer's disease on Chinese and American caregivers. International Journal of Geriatric Psychiatry
Pereira, A. C., Huddleston, D. E., Brickman, A. M., Sosunov, A. A., Hen, R., McKhann, G. M., et al. (2007). An in vivo correlate of exercise
-induced neurogenesis in the adult dentate gyrus. Proceedings of the National Academy of Sciences of the United States of America
Radak, Z., Kaneko, T., Tahara, S., Nakamoto, H., Pucsok, J., Sasvari, M., et al. (2001). Regular exercise
improves cognitive function
and decrease oxidative damage in rat brain. Neurochemistry International
Richard, M., & Sacker, A. (2003). Lifetime antecedents of cognitive reserve. Journal of Clinical and Experimental Neuropsychology
*Scherder, E. J., Van Paasschen, J., Deijen, J. B., Van Der Knokke, S., Orlebeke, J. F., Burgers, I., et al. (2005). Physical activity and executive functions in the elderly with mild cognitive impairment. Aging and Mental Health
*Smiley-Oyen, A. L., Lowry, K. A., Francois, S. J., Kohut, M. L., & Ekkekakis, P. (2008). Exercise
, fitness, and neurocognitive function in older adults: The "selective improvement" and "cardiovascular fitness" hypotheses. Annals of Behavioral Medicine: A Publication of the Society of Behavioral Medicine
*Stevens, J., & Killeen, M. (2006). A randomised controlled trial testing the impact of exercise
on cognitive symptoms and disability of residents with dementia. Contemporary Nurse: A Journal for the Australian Nursing Profession
Tombaugh, T. N., & McIntyre, N. J. (1992). The Mini-Mental State Examination: A comprehensive review. Journal of the American Geriatrics Society
*van Uffelen, J. G. Z., Chinapaw, M. J., van Mechelen, W., & Hopman-Rock, M. (2008). Walking or vitamin B for cognition in older adults with mild cognitive impairment? A randomised controlled trial. British Journal of Sports Medicine
Yu, F., Kolanowski, A. M., Strumpf, N. E., & Eslinger, P. J. (2006). Improving cognition and function through exercise
intervention in Alzheimer's disease. Journal of Nursing Scholarship
Keywords:© 2011 Lippincott Williams & Wilkins, Inc.
exercise; cognitive function; older people