Abstract
Objective:
The relation between mild everyday exercise and cognitive ability in healthy older people was examined using cohort study database.
Methods:
Individually calculated linear regression coefficients in digit cancelation task performances for 11 years age from 65 to 75 years were compared between mild exercise habit holders and non-holders.
Results:
Exercise habit holders showed significantly smaller age-related decline than non-holders, irrespective of task difficulty.
Discussion:
The results suggested that even mild exercise habit for long years possesses benefits on sustaining cognitive function in older people as well as the physical activities such as programmed in a sport gym. It also becomes clear that it is difficult for ordinary older adult to continue exercising habits for many years. Therefore, more substantial ways are required for local health officials to advertise the effectiveness of mild exercise habits and to devise the necessary work to become a habit.
Keywords
Introduction
The increase in medical expenses for the older adult has become a common problem in the developed countries of the so-called aged society. To increase the population that can live as healthy aged dwellers in the community, the physical and cognitive point of view is one of the most important public policies for the public health authorities. Then, they recommend nutritional guidance on diabetes and obesity countermeasures, cognitive training and physical activity, etc. repeatedly. Actually, a positive effect of various activities in everyday life on the anti-age-related cognitive decline has been reported.1–10 However, there is insufficient evidence to support the use of dietary supplements and lifestyle intervention, although physical activity such as aerobic exercise intervention might be promising. 11
The term “physical activity” is an umbrella term that includes various kinds of activities from standing/sitting alone to treadmill running, and bench press in the sports gym. According to American College of Sports Medicine (ACSM), 12 physical activity is classified into four categories, aerobic exercise training (AET) in which the body’s large muscles move in a rhythmic manner for sustained periods, resistance exercise training (RET) that causes the muscles to work or hold against an applied force or weight, activities designed to preserve or extend range of motions (ROM) around a joint, and balance training (BT) activities designed to increase lower body strength and reduce the likelihood of falling. Furthermore, high-intensity interval exercise (HIT) was examined if it is regarded as more effective than traditional moderate-intensity continuous exercise training. 13 Although previous studies mostly concern AET and RET on physical health, cognitive health, and mental health, our concern is mild exercise habit (MEH) such as walking and radio exercise in everyday life for the older adult such as ROM and BT.12,14–19 This is not only because MEH has been recommended by local health authorities as an effective method to maintain physical and cognitive function, but also because MEH habit does not require much financial burden for ordinary community dwellers. Demonstration of a clear link between MEH and cognitive function is an important issue for public health authorities in health-promoting program.
Studies that compare MEH with cognitive performances have been done so far.1,20 Their research compared the presence or absence of exercise habits and the results of cognitive examinations, by reminding the past exercise history when cognitive function was measured. Therefore, even though the group with exercise habits has excellent cognitive performance results, the possibility that includes confounding factor such that the cognitive function is superior at the beginning of the exercise remains.
This study relied on the longitudinal study method, which measured both cognitive performances and exercise history simultaneously at plural points successively and did not depend on retrospective memories. Furthermore, because it was based on the developmental trajectories of cognitive performances, our longitudinal research project possesses a prospective feature that might offer approximate prediction of developmental changes in later life. To assume that age-related changes in physical ability and cognitive ability are considered to be linear functions with positive correlation seems reasonable. It does not imply if this assumption is true after the physical function deterioration became conspicuous or the decline of cognitive function is classified as a dementia. That is, this study is a prospective type study that is crucially different from previous studies.
It was our concern to confirm whether MEH habit possesses benefits similar to AET and RET for age-related cognitive function, especially in a frontal lobe function. There is a large body of studies that support that exercise with high burden of the lung and heart to increase vital capacity and blood flow organized by sports gym trainer or something similar. Actually, evidences on the cognitive benefits of cardiovascular fitness training (CFT) normally in aging adults have been reported in numerous reviews and meta-analyses.15,21,22 Colcombe and Kramer 15 reported that although several studies failed to observe the benefits of CFT on cognitive performances,23,24 there are many studies that showed notable successes in improving cognitive performances with an enhanced CFT. According to their literature search of 167 articles during 1996 to 2001, CFT intervention induces positive benefits for cognition, especially for executive function, although the effects depend on programmatic and methodological factors such as intervention type, length, duration of sessions, and gender difference of participants.
We also regarded that it is necessary to take into consideration individual differences in the years-long level habit, because individual difference as for the age-related cognitive decline becomes prominent after the age of 60 years.25–28
Then, the purpose of this study was to provide the evidence using longitudinal cohort study data that MEH relates to sustaining or preventing prefrontal cortex function decline in older adult using cognitive measure. In our previous study, 29 we employed developmental trajectories of the cognitive performances by individually calculating regression coefficients and found that an age-related decline is particularly pronounced in executive function measures D-CAT3 (target was three digits) rather than in elementary perceptual speed measures D-CAT1 (target was single digit). We compared individual linear regression coefficients of community-dwelling individuals that were calculated individually for the period during 11 years from age 65-75 years. The D-CAT (Digit Cancellation Test) 30 is a cognitive test of paper and pencil type that addressed a prefrontal cortex function assessment as precisely described later. The results showed that regression coefficients for executive function showed a significantly steeper decline than that for an elementary perceptual speed, and a large number of participants showed this tendency. The reason why we limited the sample data from age 65-75 years is based on our finding that cognitive measures (D-CAT, Logical Memory Test, and Verbal Fluency Test) in Yakumo Study database showed that trajectory declines were almost linearly after age 65 years. 25 Needless to say, D-CAT does not measure all facets of executive functions, but it measures mainly selective attention and elementary perceptual speed that consisted of executive functions.31,32
In sum, this study aimed to examine the relation between MEH habit and cognitive decline (especially, in the prefrontal cortex function measured by D-CAT) and prospective beneficial expectation concerning everyday MEH.
The working hypothesis was, first, that the participants who have had years-long MEH habit show no decline or milder cognitive decline slope of regression coefficient than those who have not had MEH habit, and second the discrepancy between the MEH habit group and no-habit group is more robust in executive function measures (D-CAT3) rather than in elementary perceptual speed measures (D-CAT1).
Method
Participants
Participants were identical to those of Hatta et al 29 who voluntarily participated in the Yakumo Study and they could manage everyday life themselves. The Yakumo Study has been conducted jointly between the Department of Preventive Medicine by the Nagoya University Medical School and the Yakumo town since 1981. 33 Investigations were conducted in the fields of epidemiology, internal medicine, orthopedics, urology, ophthalmology, otolaryngology, and neuropsychology. This study was undertaken as a part of the neuropsychology section. In the Yakumo Study, community-dwelling participants age more than 40 joined voluntarily. Yakumo is a town with a population of 17 252 (in 2015) and located in the southern part of Hokkaido Island, facing both the Pacific Ocean and the Japan Sea. Agriculture, fishery, and forestry are the major occupations of Yakumo residents and it can be regarded as the representative rural towns in modern Japan. The age distribution of Yakumo town residents and the population distribution by gender are those that almost agreed with those of the whole country. Participants in the study had been engaged in a variety of jobs, not only in white collar, but also in agriculture, fishery, and forestry (see, eg, http://www.pref.hokkaido.lg.jp).
The analyzed data were selected from the Yakumo Study database by the following criteria. First, the participants’ age was from 65 to 75 years, second they had been administered D-CAT 34 more than four times during the 11-year period (once per year), and finally they had not shown any sign of mild cognitive dementia (participants were omitted if they scored below 23 points on the Mini-Mental State Examination [MMSE]). In total, 104 participants were selected by this criterion from the database, which included 2969 participants (aged 40-91 years) who had completed D-CAT from 2001 to 2013. Table 1 shows the characteristics of the participants.
Characteristics of participants and means of regression coefficients.
Cognitive measure
D-CAT: In D-CAT, the participants are asked to find a given target digit in the randomly arranged digit sheet and mark it with a slash. In D-CAT1 condition, the participants mark the single target digit 6 for 60 s and they mark the three target digits 8, 3, and 7 in D-CAT3 condition as fast and accurately as possible for 60 s. It can be regarded that performance on D-CAT1 represents mainly low-visual search load tasks or elementary perceptual speed tasks, whereas performance on D-CAT3 represents high-visual search load tasks or top-down attentional guidance that requires prefrontal cortex function more, compared with D-CAT1. The reliability and validity of D-CAT as a cognitive test for attention that addresses the distinctive facets of perceptual speed and executive function have been reported elsewhere.34–36 For example, these studies demonstrated that reliability by repetition is very high and the learning effect by repeated testing of D-CAT is negligible and finally that the brain activation of the prefrontal cortex by brain imaging method is higher in D-CAT3 than in D-CAT1.
To examine the working hypotheses, a linear regression coefficient was calculated for D-CAT1 and D-CAT3 individually. It is known that to get a reliable regression coefficient, more than three measuring points are needed to examine cognitive decline in aging, 37 and the mean measuring point in our database was 6.9 points (SD = 1.9). This means that our regression coefficients were highly reliable from a statistical perspective.
Physical activity habit
Participants were given the questionnaire booklet where they were asked to complete question items on their health, medical history, and daily lifestyle including physical activity. Questionnaire items were not completely identical for 11 years; however, the following questionnaire items on exercise habits were identical: “Have you been doing sports such as radio exercise, park golf and walking through the past 1-2 years?” (4 alternatives: (a) did not do, (b) sometimes did, (c) once a week, and (d) twice a week or more).
In this study, the participants were classified into MEH habit holder and no-habit holder groups. The habit holder group participants marked the last two alternatives for successively 5 years from the first year of participation, whereas the no-habit holder group marked the first two alternatives for 5 years. The habit holder group consisted of 6 participants (2 men and 4 women) and the no-habit holder group consisted of 98 participants (38 men and 60 women). At the first measuring point, 20 participants were assigned to the habit holder group; however, only 6 participants could sustain their habit for 5 years continuously.
Results
To examine the working hypotheses, a linear regression coefficient was calculated individually for D-CAT1 and D-CAT3. Individual differences in performances were large as we expected, and using the individual regression coefficient was crucial. As described earlier, trajectories of D-CAT1 and D-CAT3 after age 65 showed a linear decline. To simplify the comparison, the z score transformation was administered using means and standard deviation from the first measurement point. Table 2 shows the mean regression coefficients in D-CAT1 and D-CAT3 for the habit holder and no-habit holder groups. Figure 1 shows the mean regression coefficients of D-CAT1 and D-CAT3 as a function of the two groups.
Mean regression coefficients in D-CAT1 and D-CAT3 for the habit holder and no-habit holder groups.

Mean regression coefficients of D-CAT1 and D-CAT3 in the habit holder and the no-habit holder groups. The vertical axis represents the linear regression coefficient.
To address our concerns, regression coefficients (decline slope) were compared between the habit holder and no-habit holder groups by t test. The results for D-CAT1 showed a significant difference in the non-habit holder group (t = 3.080, df = 97, P = .0001), whereas there was no difference in the habit holder group (t = 0.483, df = 5, P = .6495). The results for D-CAT3 showed a similar finding that the non-habit holder group showed marginally significant difference (t = 1.530, df = 5, P = .060), whereas habit holders showed non-significant difference (t = 0.908, df = 97, P = .183). As can be seen from the figure, the habit holder group showed better performance (non-steep decline slope) in both the D-CAT1 and D-CAT3 conditions. This means that the MEH habit holder could sustain or showed slower decline than the non-habit holders, irrespective of the degree of visual search load.
Discussion
A pile of evidence that physical activity and lifestyle relate to cognitive function in older adult people has been accumulated since the seminal work of Spirduso 8 and Spirduso and Clifford. 9 Until that time, most studies emphasized the possible benefits of CFT in lifestyle against diabetes and prevention of heart disease. Since Spirduso’s studies, the association between physical activity or CFT (mostly categorized as AET and RET by ACSM) and cognitive function was taken up. Nowadays, it can be regarded that the cognitive benefits of the CFT across the adult lifespan have been well documented.12,16,18,38,39
Actually, it has been reported in numerous reviews and meta-analyses that evidence on the cognitive benefits of CFT in aging adults is reliable.15,21,22 There are many studies that showed notable successes in improving cognitive performances with an enhanced CFT. According to the literature search, most studies concluded that CFT intervention induces positive benefits especially for executive function not only in healthy older adults2,3,5 but also in older adults with mild cognitive impairment (MCI).40,41
Although it may be slightly overstated, there are several limitations for their conclusion. Firstly, most of the quoted work in the review was those of examination between the relation of CFT and the cognitive function that relied on rather short-term (at most a few months) intervention. They did not rely on examinations of CFT and cognition for a long time period. As the participants were older adults, it is difficult to examine complexly intertwined effects such as age-related physical function and mental function on the relations, unless it is examined on a yearly basis intervention. Secondly, CFTs in the review were mostly those employing high burden of the lung and heart organized by the sports gym trainer or something similar. For this reason, systematically planned exercise might be effective for increasing vital capacity and blood flow. However, except for the high-functioning older adult in a privileged environment, physical activities that are recommended by public health authorities to ordinal community dwellers are walking, radio exercises, stretching, park golf, etc. that do not need a financial burden, although the load on the internal organs of activity is relatively small compared with that of sports gym training programs. Thirdly, many researchers studied the relationship between CFT and cognitive functions; however, the causal relationship is not necessarily transparent. Finally, the participants are high-functioning owners of educated older adult who are interested in maintaining health and originally have higher cognitive functions, and they might not be necessarily typical representative of older adult community dwellers.
The purpose of this study was to provide further evidence from the Yakumo Study longitudinal data that MEH habit holders in older age could provide neurocognitive benefits or prevent steep cognitive decline associated with aging as well as CFT habit holders. The participants of the Yakumo Study were typical local residents of Japan and participants have engaged in clerical work, agriculture, fishery, and forestry (some of them are still active workers). There is no sports gym of private company management in Yakumo town, and physical activities of community dwellers in their daily lives are mainly radio exercises, walks, jogging, and park golf recommended by local health authorities. The year of education of our participants was shorter than that of western societies. Education history changes depending on the times and culture, and our participants’ educational history of the older adult is representative of the typical Japanese older adult. Therefore, this study examined the influence of continuing exercise with a minor load on the heart and pulmonary function on cognitive function, especially on prefrontal cortex–related attention rather than CFT. Although Karimooy et al 42 reported positive effects of MEH on MMSE scores for older adult participants who had exercised habit for 2-10 years, their findings relate to cognitive function in general, and it is not possible to mention the relationship between elements constituting cognitive function and brain parts.
As for the first working hypothesis, the present findings revealed that the regression coefficient of MEH habit holders is smaller than that of the non-habit holders in both D-CAT 1 and D-CAT 3. This predicts that age-related attentional function deterioration in the habit holders is slow compared with non-holders and working hypothesis was supported. However, there was no significant regression coefficient decline between D-CAT1 and D-CAT3, and the second working hypothesis was not supported. That is, continuous MEH habit in older adult people relates to sustaining good attention function irrespective of cognitive load difference. In other words, it is expected prospectively that MEH habit relates to sustaining the prefrontal cortex function.
The present finding revealed another important issue. That is, it is an extremely difficult matter for almost ordinary older adult to continue MEH habit for 5 years. As observed from the findings, only 6 out of 104 could sustain MEH habit for 5 years successively and 94.2% older adult people gave up exercise habit. We realize that the sample size of habit holders is unexpectedly small to invite reliable statistical conclusion; however, the present examination offers an important point of how difficult it is to sustain a habit for a long period for ordinary community dwellers. As they are voluntarily participating in the annual health examination, it can be regarded that the possible reasons why they failed to continue exercise habit are not due to problems of physical function necessary for mild exercise, but what was not able to continue could be psychological problem such as an increase in feeling of annoying, bothering that may be partially due to a decrease in physical and mental vitality due to aging.
Regarding the neurological mechanism for sustaining the frontal lobe function with a slight but long-term exercise habit, the findings of Colcombe et al 43 are helpful. They reported experiments where high-functioning community-dwelling older adult people were asked to participate in CFT such as 1-mile walking and treadmill exercise, and conducted an attention-related cognitive test and the functional magnetic resonance imaging (fMRI) measurement in Study 1. In Study 2, community-dwelling older adult people participated in CFT for 6 months longitudinally and the same cognitive and fMRI measurements were conducted before and after CFT intervention. They reported that CFT results in the increased functioning of attentional networks during a cognitive task. The fMRI analysis showed that CFT participants showed greater task-related activity in regions of the prefrontal and parietal cortices that are involved in the spatial selection and inhibitory functioning compared with the control participants. Additionally, in both studies there exist group-wise differences in the activation of the anterior cingulate cortex, which is thought to monitor for conflict in the attentional system and signal the need for adaptation in the attentional network. These data suggest that increased CFT can affect improvements in the plasticity of the aging human brain and may serve to reduce both biological and cognitive senescence in humans. According to their proposal, one possibility was proposed that CFT increases the number of synapses in the frontal and parietal gray matter that allow greater systematic recruitment of these areas under high-cognitive load attention task even though the duration is as short as 6 months. They also proposed another possibility that the increases in CFT lead to increases in the blood supply in these regions, which in turn provides the metabolic resources necessary to coherently respond during a cognitive task performance. Similar proposal has been reported elsewhere.1,44,45
Many recent studies seem to support the first possibility, ie, CFT increases neural networks in not only white matter, but also gray matter. 46 For example, Basso et al 47 reported an experiment where they compared acute CFT effects of two types of cognitive performances: one related to executive function mediated by the prefrontal cortex such as Stroop test and the other related to hippocampus-related brain function such as episodic memory test and control groups. The CFT group participants showed improved performance in prefrontal cortex–related tests, but not in hippocampus-related tests compared with the control participants, although the participants were not older adult people but younger adults. Best et al 48 reported an intervention experiment. In their study, 65- to 75-year-old participants were classified into three groups; the RT2 group participants were given sports gym exercise (RET) twice per week and once per week at RT1. The BT group participants were given stretch and relaxation twice per week. Intervention duration was 52 weeks and then the effects of intervention were measured after 1 year and again after 2 years. The results of measurements in both years for RT1 and RT2 showed better performance in executive function–related cognitive tests and a smaller decrease in the size of white matter that suggests the durability of the protective effects in white matter compared with BT, ie, possible neuroplasticity due to exercise lasting even after 2 years. Reiter et al 49 reported a mild exercise intervention study where physically inactive older adults (14 with MCI and 16 healthy controls; their age ranged from 61 to 88 years) underwent magnetic resonance imaging (MRI) before and after participating in a 12-week moderate-intensity walking intervention. The results showed a CFT effect on cortical thickness change in the bilateral insula, precentral gyri, precuneus, posterior cingulate, and inferior and superior frontal cortices. These suggest that even MEH is a beneficial intervention to counteract cortical atrophy and may provide protection against a future cognitive decline in both healthy elders and those with MCI.
These studies suggest that MEH can affect on cognitive sustaining against age-related decline in older people. In addition to these studies based on experimental design, the present longitude data examination demonstrated that not only exercise like a CFT but also even a mild exercise, such as long-term walking habit prevent the declining speed of the frontal lobe attention function in older adult.
It also becomes clear that it is extremely difficult for the ordinary older adult people to continue exercising habits for many years, even if they have been informed by the local health authority repeatedly. Therefore, more substantial new ways are required for local health officials to advertise the effectiveness of MEHs and to devise the necessary work to become a habit.
One of the primary limitations of this study is that the participants’ concrete exercise characteristics are not clear. Recent studies pointed out the relation between gait and cognitive function including the underlying neural mechanisms in older adults.50–54 Further examinations are necessary to identify what kind of physical activities offers large benefits on the anti-age-related decline of executive function.
However, our evidence that there is a link between mild physical exercise and cognitive decline seems to provide support for public authorities to promote ideas for public effort.
Footnotes
Acknowledgements
We wish to thank the participants and staff of the health examination program for residents of Yakumo Town, Hokkaido, Japan.
Funding:
This study was supported by the Grant-in Aid for Scientific Promotion, Ministry of Education, Science, and Culture Japan under Grant No. 23330219 to the first author.
Declaration of Conflicting Interests:
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Authors’ Note
Kazumi Fujiwara is now at Toho Universiy, Tokyo, Japan.
Author Contributions
TH: Research planning and context writing; KK: Data collection and statistical analysis; Al: Data collection and database revision; TH: Data collection and statistical analysis; KF: Data collection; El: Data collection; YH: Research Planning.
Ethical Approval
Ethical approval was obtained from the Ethical Committee of Nagoya University Medical School for Yakumo Study (2011 # 643) and written informed consent for participation and data publication was obtained from each participant.
