Background
Alzheimer’s disease and related dementias top the world’s most prevalent and costly medical conditions [
1] and prevention and treatment of these disorders has been deemed a public health priority by the World Health Organization [
2]. It has become widely accepted that unhealthy lifestyle behaviours contribute to an increased prevalence of cognitive impairment and dementia worldwide [
3] and that engaging in physical activity is an effective strategy for preserving brain health in older adults [
4‐
8]. Broadly defined, physical activity is any activity that involves bodily movements and the use of skeletal muscles. This includes everyday activity that is not structured, such as occupational or household activity, as well as recreational physical activity, such as exercise, which is structured and intended to improve fitness [
9]. The actions by which physical activity affect the brain are suspected to be both indirect, such as improving health conditions, as well as more direct, including increasing brain neurotrophins [
6,
10], improving cerebrovascular function [
11,
12], and enhancing brain plasticity [
13,
14]. Understanding how different forms of physical activity (i.e., recreational activity vs. household activity) contribute to brain health is crucial for developing strategies to reduce the risk of cognitive decline and dementia in older adults.
Much of the research in this regard has focused on recreational physical activities, in particular the effects of higher intensity exercise. Moderate to vigorous intensity exercise and higher aerobic fitness have been positively associated with whole brain volume, gray matter volume, and improved white matter integrity [
5,
6,
15‐
19]. Moreover, exercise interventions have demonstrated effectiveness at increasing whole brain volume and gray matter volume in older adults [
20‐
22]. Similar findings are also evident with low intensity forms of exercise independent of moderate to vigorous intensity activity [
23‐
26]. A comprehensive review of the literature by Erickson et al. (2014) concluded that physical activity and fitness levels were routinely associated with larger hippocampal and prefrontal cortex volumes, and that exercise interventions were able to evoke positive changes in these regions specifically [
6]. Cognitive function in older adults is also positively associated with cardiovascular fitness and in some studies has been shown to improve following exercise interventions [
4,
27] with greatest benefits being apparent in executive function and attentional control. Not all studies show this effect, however. For example, a recent review of physical activity interventions concluded that evidence for cognitive improvement following interventions was insufficient, largely due to the heterogeneous cognitive measures used across studies, and underpowered sample sizes [
28].
We know that not all brain benefits are derived from recreational physical activity exclusively. Total physical activity is associated with reduced cognitive decline and dementia risk [
29‐
31], decreased brain atrophy [
32], as well as increases in brain volume [
33‐
36]. While ample evidence exists that recreational physical activity contributes to these associations, less is known about how other everyday physical activities such as household activity contribute to brain health. Household activities may be particularly interesting to study, because these types of activities (e.g., cleaning chores, meal preparation) are inherent aspects of many people’s daily life, providing some degree of physical activity and decreasing sedentary behaviour naturally. Studies examining the role of household physical activity on brain health are limited, however. Engagement in household physical activity has been shown to be negatively associated with frailty [
37] and leisure time activity incorporating household chores has been found to be associated with decreased dementia risk [
38,
39]. The purpose of this study was to assess how household physical activity specifically correlates with brain volume and cognition among a sample of cognitively unimpaired older adults. We hypothesized that household physical activity would be positively associated with both brain volume and cognition.
Results
Table
1 describes the demographic, neuroimaging, and cognitive characteristics of the study sample.
Table 1
Participant characteristics
Demographics |
Age (years) | 71 ± 4 (65–83) |
Sex (count) | 41 Females (62%) |
Education (years) | 16 ± 3 (10–23) |
Framingham Risk Score | .17 ± .10 (.03–.37) |
Household Physical Activity - Phone-FITT Score | 10.67 ± 4.28 (1–20) |
Recreational Physical Activity - Phone-FITT Score | 9.39 ± 5.40 (0–21) |
Neuroimaging |
Intracranial Volume (cm3) | 1487 ± 171 (1090 – 1862) |
Whole Brain Volume (cm3) | 1066 ± 111 (799–1311) |
Gray Matter Volume (cm3) | 569 ± 53 (429–672) |
White Matter Volume (cm3) | 434 ± 58 (323–584) |
Cognition |
Memory Domain |
CVLT - Learning | 49.76 ± 9.58 (29–69) |
CVLT - Delayed Free Recall | 10.98 ± 3.16 (0–16) |
Visual Paired Associates - Learning | 12.58 ± 3.59 (5–18) |
Visual Paired Associates – Delayed Free Recall | 5.53 ± 0.93 (2–6) |
Verbal Paired Associates - Learning | 17.92 ± 3.14 (10–24) |
Verbal Paired Associates – Delayed Free Recall | 7.09 ± 1.06 (5–8) |
Faces – Immediate Recall | 35.53 ± 3.90 (28–45) |
Attention / Working Memory |
Digit Span Forwards | 10.97 ± 2.47 (5–16) |
Digit Span Backwards | 7.68 ± 2.52 (3–14) |
Letter – Number Sequencing | 10.62 ± 2.68 (4–17) |
Arithmetic | 12.67 ± 3.23 (6–19) |
Mental Control | 25.11 ± 4.95 (15–38) |
Executive Function |
WCST errors | 22.12 ± 12.75 (7–52) |
Trail Making Test – Version B | 82.82 ± 29.35 (36–173) |
Phonemic Fluency (FAS) | 43.55 ± 12.40 (23–74) |
Processing Speed |
Digit Symbol Coding | 61.11 ± 12.95 (28–91) |
Trail Making test – Version A | 36.48 ± 11.71 (19–87) |
Bivariate correlations between variables are displayed in Table
2. Age was negatively associated with gray matter volume (
r = −.40,
p < .001). Years of education was positively associated with attention/working memory (
r = .33,
p = .006) and executive function (
r = .39,
p = .001). FRS was negatively associated with whole brain (
r = −.28,
p = .026), white matter volume (
r = −.28,
p = .024), memory (
r = −.39,
p = .001), and executive function (
r = −.30,
p = .015). There were no significant sex differences in physical activity or brain volume measurements; however, memory performance was better in women (
p < .001).
Table 2
Bivariate correlations
Age | 1 | | | | | | | | | | | |
Education | 0.01 | 1 | | | | | | | | | | |
FRS | 0.18 | −0.11 | 1 | | | | | | | | | |
Household Physical Activity | −0.18 | − 0.23 | −0.11 | 1 | | | | | | | | |
Recreational Physical Activity | 0.04 | 0.01 | −0.15 | 0.19 | 1 | | | | | | | |
Whole Brain Volume | −0.24 | −0.15 | −0.28a | 0.36b | 0.19 | 1 | | | | | | |
Gray Matter Volume | −0.40c | −0.07 | − 0.22 | 0.36b | 0.16 | 0.71c | 1 | | | | | |
White Matter Volume | −0.07 | −0.17 | −0.28a | 0.27a | 0.15 | 0.80c | 0.23 | 1 | | | | |
Memory | −0.06 | 0.12 | −0.39b | 0.10 | 0.13 | 0.36b | 0.19 | 0.26a | 1 | | | |
Working Memory | −0.04 | 0.33b | −0.03 | 0.00 | 0.18 | 0.17 | 0.15 | 0.08 | 0.21 | 1 | | |
Executive Function | −0.20 | 0.39b | −0.30a | − 0.14 | 0.19 | 0.35b | 0.35b | 0.23 | 0.35b | 0.57c | 1 | |
Processing Speed | −0.13 | 0.24 | −0.05 | 0.08 | 0.14 | 0.31a | 0.25a | 0.17 | 0.35b | 0.44c | 0.61c | 1 |
The MANOVA revealed that household physical activity (
F (3, 57) = 3.07,
p = .035), but not recreational physical activity (
F (3, 57) = 0.32,
p = .812) was significantly associated with brain volume measurements. Age (
p = .043), sex (
p = .003), and FRS (
p = .025) were significant covariates in this analysis. Neither household (
F (4, 56) = .1.52,
p = .210) nor recreational physical activity (
F (4, 56) = .74,
p = .571) were associated with overall cognitive performance, but women had better cognitive performance (
p < .001). Table
3 displays the results from the post hoc multiple regression analysis. Household physical activity was positively associated with gray matter volume (
p = .015) and age was the only significant covariate (
p = .008).
Table 3
Associations between physical activity and individual brain volume measurements
Household Physical Activity | 2.82 (1.17) | 2.41 | .019 | 2.17 (0.86) | 2.51 | .015 | 1.82 (0.99) | 1.85 | .070 |
Recreational Physical Activity | 0.82 (0.92) | 0.90 | .371 | 0.46 (0.66) | 0.69 | .495 | 0.39 (0.77) | 0.50 | .616 |
Exploratory regression analyses revealed that household physical activity was significantly associated with hippocampal (
p = .015) and frontal lobe (
p = .010) volumes (Table
4). To investigate the lack of relationship between recreational physical activity and brain volume, an additional analysis was conducted excluding light intensity activities (e.g., light walking, stretching/balance exercises, gardening, golf) from our recreational measure. As with the previous analyses, recreational physical activity was not associated with gray (
p = .598) or white (
p = .149) matter volumes.
Table 4
Associations between physical activity, hippocampal volume, and frontal lobe volume
Household Physical Activity | 2.17 (0.86) | 2.51 | .015 | 7.15 (2.67) | 2.68 | .010 |
Recreational Physical Activity | 0.46 (0.66) | 0.69 | .495 | 1.76 (2.09) | 0.84 | .403 |
Discussion
This study investigated the associations between household physical activity, brain volume, and cognition in a sample of cognitively unimpaired older adults. These results supported our hypothesis that household physical activity is positively associated with brain volume, however no significant associations with cognition were observed. The overarching association with brain volume was driven by gray matter volume, but not white matter volume. We believe that this was reflected in the trend towards a significant association with total brain volume, however this association failed to reach statistical significance after Bonferroni correction. The hippocampus and frontal lobe have been identified as brain areas that are particularly sensitive to exercise [
6] and our work supports this notion in the context of household physical activity. Surprisingly, no associations were observed between recreational physical activity and brain volume or cognition.
This is the first study to identify an association between household physical activity and gray matter volume, and contributes to the growing body of research helping to guide physical activity recommendations for older adults. Our work uniquely points to an association with previously unexplored components of total physical activity. While there is ample evidence for the benefits of recreational physical activity, we now show how a specific type of activity that is an inherent aspect of many people’s daily life relates to brain health. Highlighting the brain benefits associated with household chores (e.g., cooking, cleaning, home maintenance) may motivate older adults to be more active by providing a more sustainable, low risk form of physical activity.
It is commonly hypothesized that increases in brain volume result from improved fitness and enhanced blood flow following exercise [
11]. Total physical activity is associated with enhanced cardiorespiratory fitness and decreased incidence of cardiovascular disease and mortality [
57]. It has been suggested that the vascular effects from non-structured everyday activities may be similar to those resulting from low intensity aerobic exercise [
58]. For example, Sanchez-Lopez et al. [
59] observed positive associations between non-structured physical activity and brain activity, and posited that results may have been due to vascular mechanisms based on hematological differences between study subjects. Another potential mechanism linking household physical activity to brain volume is enhanced neuroplasticity resulting from the planning and organization required for completing household chores. Interventions focusing on goal management and multitask training have shown to be effective at improving brain function in older adults [
60‐
63]. Housework includes a wide range of tasks and may share many of the same features as cognitive training interventions [
64]. On the contrary, it may be that individual's brain volume influences their level of household physical activity engagement. For example, someone experiencing greater than normal atrophy may be less likely to engage in household tasks. Lastly, an intriguing consideration is that the increased household physical activity scores in our sample may reflect lower levels of sedentary behavior. Approximately 67% of older adults report sitting for more than 8.5 h per day [
65] and sedentary time is associated with adverse health outcomes despite regular exercise engagement [
66,
67]. Prolonged sitting increases venous pooling and coagulation factors which in turn disrupt blood flow [
68] and can lead to spikes in blood glucose and plasma triglycerides that are detrimental to the vasculature [
69]. Sedentary behavior is associated with brain atrophy [
29,
70] and research suggests that replacing sedentary time with light activity may promote optimal brain health [
71‐
73]. We did not collect measures of sedentary behavior in this study but is an important measure to consider in future trials exploring the link between household physical activity and brain volume.
We proposed that our main results may be explained by the cognitive involvement of household chores, however physical activity was not associated with cognition in our sample. It is possible that more sensitive measures of planning and multitasking are required to confirm this association. Not all research supports the link between physical activity and cognition. A recent systematic review by Brasure et al. [
28] concluded that evidence of a link between physical activity and dementia prevention is insufficient. The authors acknowledged limitations of the existing studies, including heterogeneous protocols and cognitive measures, as well as underpowered sample sizes. We may not have had sufficient power to observe the relationship between physical activity and cognition that has been reported in the literature [
27‐
31]. While associations between physical activity and cognition were not apparent in our sample, brain volumetric measurements are strong predictors of longitudinal cognitive change [
74,
75].
The lack of association between recreational physical activity and indicators of brain health in our sample was unexpected, but is perhaps explained by the removal of intensity scores from the analysis. We chose not to analyze intensity scores as per the recommendations of the Phone-FITT authors over concerns about the reliability of self-reported exercise intensity. Furthermore, although Phone-FITT recreational physical activity scores incorporate moderate to vigorous intensity physical activities, several low intensity activities such as gardening, golf, tai chi, stretching, and balance exercises are included in the questionnaire. The inclusion of these activities may have attenuated associations between recreational physical activity, brain volume, and cognition. Exploratory analyses were conducted on our dataset by removing light intensity activities (light walking, stretching, golf, and gardening) however these analyses did not lead to significant changes to our results. The removal of intensity scores may also help to explain the lack of associations with white matter volume. White matter abnormalities are a hallmark sign of cerebral small vessel disease, a condition that is attributable to poor cardiovascular health [
3]. It is likely that white matter volume would show greater associations with participation in higher intensity exercise aimed at improving cardiovascular health and fitness. When light intensity activities were excluded form recreational physical activity scores, a closer association was observed between recreational physical activity and white matter volume, however, it still failed to reach significance.
As the first study to explore the relationships between household physical activity and brain volume, we first investigated more general areas of the brain, but conducted exploratory analysis to reveal regional associations. Consistent with the literature, associations were observed between hippocampal and frontal lobe volume. Further work with larger samples would allow for more precise characterizations of areas of the brain associated with household physical activity. Furthermore, while we proposed several mechanisms that may explain our results, we were unable to test these notions statistically. We did not collect measures of sedentary behaviour or quantify the cognitive involvement relating to household physical activity. Although we did calculate cardiovascular burden (FRS), larger studies with higher power are needed to test how cardiovascular health mediates the relationship between household physical activity and brain volume. It would be a worthwhile objective for future studies to investigate these potential mechanisms as mediators.
Like all self-report measures, the Phone-FITT is prone to social desirability and recall bias. Using the previous month as the recall period may avoid some of the difficulty with recalling behaviour over longer time periods but it is still likely that individuals feel the need to overestimate healthy behaviors. The Phone-FITT also provides an arbitrary score that cannot be translated to common physical activity parameters and objective measures such as accelerometers are preferred when assessing physical activity levels. Furthermore, the cross sectional design of this study does not allow for the determination of causal relationships. Household physical activity may influence brain volume or brain volume may influence engagement in household physical activity. Future research examining the direction of this relationship is needed. For example, longitudinal studies assessing engagement in household physical activity and brain volumetric changes over time or interventions studies focused on increasing engagement in household physical activity. Aside from the mentioned limitations, strengths of this study include objective measures of cardiovascular risk, the use of advanced neuroimaging techniques, and the assessment of household and recreational physical activity separately.
In conclusion, household physical activity was positively associated with brain volume, specifically total gray matter volume in our sample. This is the first study to highlight associations with previously unexplored components of total physical activity and contributes to the growing body of literature guiding physical activity recommendations for older adults.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.