Exercise and Brain Health: Food for Thought?

By Nancy J. Selfridge, MD, Associate Professor, Integrated Medical Education, Ross University School of Medicine, Commonwealth of Dominica, West Indies. Dr. Selfridge reports no financial relationships relevant to this field of study.

The number of people in the United States age 65 and older has grown from 35 million in 2000 to 40 million in 2010, a 15% increase. This number is expected to be about 55 million by 2020, a 36% increase for the coming decade.1 Decline in cognitive health, just like many other aspects of health, is associated with aging and exists on a continuum from normal functioning to mild measurable impairment to dementias such as Alzheimer’s disease (AD). The health burden of mild cognitive impairment on individuals and society is hard to estimate, but AD currently afflicts one in eight people age 65 and older and has become the fifth leading cause of death in this age group. If present trends continue, it is estimated that as many as 16 million people will have AD by 2050 and health care costs related to this disease will increase from a current $183 billion to more than $1.1 trillion. The Centers for Disease Control and Prevention (CDC) reported in its most recent 2010 HealthStyles survey of 4183 U. S. adults that 70% of respondents expressed concern about memory loss and 20% feared becoming cognitively impaired.2 Thus, identification of modifiable risk factors in the development of cognitive decline and dementia is important to stem the tide of this evolving problem.


Exercise and physical activity are known to positively affect many of the common consequences of aging, including loss of muscle mass and strength, diminished bone density, impaired balance, reduced cardiorespiratory endurance, and increased incidence of chronic debilitating illnesses such as cardiovascular disease and diabetes mellitus.3 A substantial body of both animal and human research suggests that exercise also appears to have a salutary effect on aspects of cognitive health, the components of which include: language, thought, memory, executive function, judgment, attention, perception, remembered skills, and the ability to live a purposeful life.2,5,6 However, variability in research design and rigor has prevented the CDC and other agencies from developing clinical guidelines or recommendations for preserving cognitive functioning with exercise.2

Mechanism of Action

Several mutually compatible hypotheses exist for the beneficial mechanisms of physical activity and exercise on cognitive function in health and in dementia. One of the hallmarks of AD is amyloid β plaque deposition in the brain. Laboratory studies have shown reduced amyloid β plaque formation in mice provided exercise interventions compared to sedentary mice. In humans, amyloid β brain load, plasma concentration, and serum levels have been shown to be lower in individuals with higher exercise and activity levels. These effects may be mediated by the ability of exercise to raise neurotransmitter levels (see below), increase testosterone levels, and increase growth factors such as brain-derived neurotrophic factor (BDNF) and insulin-like growth factor 1 (IGF-1). Increase in cerebral blood flow associated with exercise also may play a role.4

Apart from an effect on amyloid β deposition, the increase in BDNF and IGF-1 may reduce brain atrophy (normal hippocampal volume loss is 1-2% per year in older adults) and may even increase brain mass and volume. BNDF is known to be associated with neurogenesis and increased survival of neurons, and IGF-1 mediates both exercise induced angiogenesis and neurogenesis. In fact, higher levels of fitness in older adults have been associated with increased mass in the frontal and hippocampal areas of the brain. Exercise causes significant increases in several neurotransmitters including serotonin, norepinephrine, acetylcholine, dopamine, and epinephrine, all of which are known to decline with aging.4,5

Telomeres are nucleotide sequences on the ends of chromosomes that protect their integrity. Telomeres shorten with each successive cell division and eventually lose their protective effect. Thus, progressive telomere shortening leading to cell and tissue growth arrest, damage, and senescence is one of the main theoretical mechanisms of physical and mental decline due to aging.6 Telomerase is a ribonucleoprotein complex that preserves telomere length in proliferating cells. Telomerase activity appears to be upregulated in exercising mice and humans.7

Sex hormones may have a neuroprotective effect, and the increases in testosterone levels seen with exercise may play a role in preserving and improving cognitive function. Exercise may help reduce serum cortisol levels, one of the few hormones that increases with age and which may have a role in declining hypothalamic function with aging.7

Insulin resistance and type 2 diabetes have been implicated as having roles in AD because of the way that they alter amyloid β processing. Exercise has a profound effect in improving insulin sensitivity and some of its beneficial effects on cognition may be mediated by this mechanism.4

The APOE ε4 allele has been associated with the strongest risk of late onset AD. Carriers of this allele are more susceptible to amyloid deposition if they are sedentary, but in exercising individuals the allele does not appear to increase brain amyloid load.4

Clinical Research

Summary of Epidemiological Studies. Several epidemiological studies have shown that exercise improves or helps maintain cognitive function. Most of these studies have assessed physical activity levels using self-reporting questionnaires but some have measured physical fitness or have made objective measures of physical activity with the use of an accelerometer device to measure body movement. Brown et al provide a systematic review of these.4 In one cross-sectional study of 1927 healthy adults ages 45-70 years who self-reported their physical activity, Angevaren et al found higher intensity physical activity was associated with better processing speed (P < 0.01), memory (P < 0.05), mental flexibility (P < 0.05), and overall cognitive functioning (P < 0.01). In another cross-sectional study, Barnes et al noted higher global cognitive function in 349 healthy subjects 55 years of age and older who had higher levels of measured cardiorespiratory fitness measured by peak oxygen consumption, exercise duration, and oxygen uptake efficiency using a standard treadmill exercise test. Six longitudinal studies in older adults (n = 36,472) reported a statistically significant positive impact on cognitive measures or a reduction in cognitive decline with higher levels of physical activity or an increased risk of cognitive decline in subjects with lower levels of physical activity. P values, when reported for these studies, ranged from 0.02 to 0.001. Two prospective cohort studies by Wilson et al reported no association between self-reported physical activity and incident AD, but there are possible explanations for these negative findings. Both of these studies used smaller sample sizes than other epidemiological studies demonstrating an effect, both had follow-up periods of < 5 years and one of the studies used subjects from a limited and non-representative demographic.4

Sofi et al conducted a meta-analysis of 15 prospective cohort studies that included 33,816 initially healthy older individuals of which 3210 developed cognitive decline. Exercise conferred a significant protective effect on cognitive function. Follow-up periods ranged from 1 to 12 years. The highest levels of exercise, measured in various ways in all of these studies (see conclusions, below) provided the greatest protective effects (hazard ratio 0.62; 95% confidence interval [CI] 0.54-0.70; P < 0.00001); however, even low-to-moderate levels of exercise were beneficial compared to a sedentary lifestyle (hazard ratio 0.65; CI 0.57-0.75; P < 0.00001).8

Hamer et al reviewed data from 16 studies assessing the impact of physical activity on neurodegenerative disease risk (n = 163,797). In their analysis, the relative risk of dementia in the highest physical activity groups compared to the lowest activity or control groups was determined to be 0.72 (CI 0.60-0.86; P < 0.001) and the relative risk of AD was determined to be 0.55 (CI 0.36-0.84; P = 0.006). Again, exercise levels in these studies were measured and reported in highly variable ways.9

In a recent observational study of 104 early-stage AD patients, Winchester et al noted that sedentary patients experienced a significant decline in mini-mental status exam (MMSE) scores, while active patients had an attenuation in global cognitive decline. Those patients who walked for more than 2 hours per week demonstrated a significant improvement in MMSE scores over 1 year.10

In another observational study, Kattenstroth et al reported that subjects who maintained a regular schedule of dancing into old age had better cognitive, motor, and perceptual abilities compared to education-, gender-, and age-matched controls having no history of dancing or sports participation activities.11

Summary of Intervention Studies. Brown et al summarized seven intervention trials assessing the impact of physical activity or exercise on cognitive function in healthy older adults. The interventions in these studies varied widely in type (strength, balance, stretching, aerobic, or combination), intensity, and duration of exercise. All provided supervised exercise, sometimes in group training and sometimes in a home-based program. All studies included men and women subjects; follow-up periods ranged from 6 months to 18 months, and several different measures of cognitive function were used as outcome measures. All but one of these studies demonstrated statistically significant or clinically significant improvement in cognitive performance in subjects after the exercise intervention.4

In a summary of eight studies of exercise intervention in subjects with cognitive decline, van Uffelen et al noted a statistically significant (P < 0.05) beneficial effect of exercise on cognitive decline in two-thirds of these studies. Again, the exercise interventions varied widely in type and volume as did the outcome measures. Attendance/adherence and drop-out rates in intervention and control groups were not reported or not included in intention-to-treat analysis of data in a number of these studies.12

In results from the Austrian Stroke Prevention Study, Sen et al assessed MRIs in 725 elderly community-dwelling subjects for brain parenchymal fraction (a measure of brain atrophy) and volume of white matter lesions (a measure of ischemic cerebral damage), and compared these measures to individuals’ fitness represented by VO2max. VO2max was inversely associated with white matter lesion volume in men (P = 0.02). There was no relationship between fitness level and brain parenchymal fraction in this study.13

In a randomized controlled trial assessing the impact of chronic endurance exercise training (supervised exercise, 3 hours per week for 23 months) in community-dwelling older adults (n = 120) in Italy, Muscari et al reported that MMSE scores decreased significantly in the non-exercising control group (mean difference -1.21, CI -1.83 to -0.60, P = 0.0002) though not significantly in the intervention group (-0.21, CI -0.79 to 0.37, P = 0.47). Odds ratio for the exercising adults to have stable cognitive status at the end of 1 year compared to the control group was 2.74 (CI 1.16-6.48) after adjustment for possible confounders.14


Though epidemiological evidence appears to support regular exercise and maintenance of physical fitness for reducing risk of cognitive decline due to aging, interventional studies are not yet numerous nor robust enough to support the development of clinical guidelines for exercise as a lifestyle strategy to preserve cognition by the CDC. Studies to date have had significant methodological flaws.

Most of the longitudinal or cross-sectional epidemiological studies have used self-report questionnaires to measure exercise and activity levels, a notoriously unreliable way to gauge physical activity or fitness levels. Further, the questionnaires have not employed consistent definitions of low, moderate, and high levels of physical activity and exercise. These two shortcomings create a significant problem when trying to determine what levels and types of exercise are most effective for preventing cognitive decline. Some of the studies have used an accelerometer device or measures of physical fitness that sidestep this dilemma to a degree. But even measures of physical fitness, such as the VO2max or oxygen consumption, require maximal treadmill or cycle ergometer exercise tests wherein subjects are required to run or cycle until they reach their maximal power; the test is terminated when the subject reports exhaustion or the supervising physician orders it for medical reasons. Thus, even these measures may be strongly influenced by subjective sensations and motivation. Many diverse tools have been used in these studies to measure cognitive functioning and different outcomes chosen, as well. Some have looked at level of cognitive function, some at cognitive decline, and others have used a diagnosis of AD as an outcome. The clinical significance of outcomes is important and needs to be considered and addressed in all studies, but often is not. An increase in a point or two on the MMSE score, for example, may take a person out of the range associated with early dementia (less than 21) or mild cognitive decline (21 to 24). Further, since not all cognitive decline results in functional impairment and not all cognitive decline in aging results in dementia, results and conclusions are difficult to compare and interpret clinically.

The interventional studies to date have looked at the effect of exercise both on healthy elderly and in persons already afflicted with dementia. These studies have employed very diverse exercise interventions ranging from “individualized programs,” to purely aerobic programs in prescriptive weekly doses, to combinations of aerobic, strength, balance, and flexibility training. Volumes of exercise interventions have varied as have durations of the interventions. As previously mentioned, some of the interventional studies have failed to report adherence and dropout rates and some have failed to perform intention-to-treat analysis on data. In some instances, the control group was sedentary, and in others the control group also exercised or took part in some other non-exercise group activity, helping to control for group effect on outcomes. Again, cognitive function measures were very diverse and comparisons and clinically relevant conclusions are difficult to discern.


Exercise for middle aged and elderly people has proven benefit for prevention and attenuation of many chronic diseases strongly supported by research evidence. Larger and better designed interventional studies addressing the precise types and volumes of exercise needed to prevent cognitive decline or dementia are necessary before clinical guidelines can be made for exercise as a lifestyle intervention for these problems, though. Nonetheless, because of the low cost and low risk of increasing physical activity, physicians may counsel their aging patients that among other health benefits accumulating 150 minutes per week of moderate exercise according to present clinical guidelines may help promote brain health and prevent decline in cognitive function while we await the research supporting definitive guidelines for this specific benefit.


1. Department of Health & Human Services. Administration on Aging. Aging Statistics. Available at: Accessed Jan. 8, 2013.

2. Centers for Disease Control and Prevention. The CDC Healthy Brain Initiative: Progress 2006-2011; Atlanta, GA: CDC; 2011.

3. Agency for Healthcare Research and Quality and the Centers for Disease Control. Physical Activity and Older Americans: Benefits and Strategies. June 2002. Available at: Accessed Jan. 8, 2013.

4. Brown BM, et al. Multiple effects of physical activity on molecular and cognitive signs of brain aging: Can exercise slow neurodegeneration and delay Alzheimer’s disease? Mol Psychiatry 2012 [Epub ahead of print].

5. Lista I, Sorrentino G. Biological mechanisms of physical activity in preventing cognitive decline. Cell Mol Neurobiol 2010;30:493-503.

6. Blackburn EH. Telomere states and cell fates. Nature 2000;408:53-56.

7. Kaliman P, et al. Neurophysiological and epigenetic effects of physical exercise on the aging process. Ageing Res Rev 2011;10:475-486.

8. Sofi F, et al. Physical activity and risk of cognitive decline: A meta-analysis of prospective studies. J Intern Med 2011;269:107-117.

9. Hamer M, Chida Y. Physical activity and risk of neurodegenerative disease: A systematic review of prospective evidence. Psychol Med 2009;39:3-11.

10. Winchester J, et al. Walking stabilizes cognitive functioning in Alzheimer’s disease (AD) across one year. Arch Gerontol Geriatr 2013;56:96-103.

11. Kattenstroth JC, et al. Superior sensory, motor, and cognitive performance in elderly individuals with multi-year dancing activities. Front Aging Neurosci 2010;2.pii:31.

12. van Uffelen JG, et al. The effects of exercise on cognition in older adults with and without cognitive decline: A systematic review. Clin J Sport Med 2008;18:486-500.

13. Sen A, et al. Association of cardiorespiratory fitness and morphological brain changes in the elderly: Results of the Austrian Stroke Prevention Study. Neurodegener Dis 2012;10:135-137.

14. Muscari A, et al. Chronic endurance exercise training prevents aging-related cognitive decline in healthy older adults: A randomized controlled trial. Int J Geriatr Psychiatry 2010;25:1055-1064.