Happiness, depression, physical activity and cognition among the middle and old-aged population in China: a conditional process analysis

Introduction

Cognition is the most complex function of the brain. Cognitive decline is a significant concern among middle and old-aged individuals, particularly in China, a country with rapidly aging population. Age-related cognitive changes are cumulative irreversible changes occurring in a long-term process (Li, Huxhold & Schmiedek, 2004). According to studies on cognitive aging, with increasing age, individuals generally have cognitive decline, which affects many cognitive domains, including memory attention, reasoning, and executive function (Park et al., 1996; Salthouse, 2010; Schaie, Willis & Caskie, 2004). The decline in cognitive ability is related to impaired daily functioning among the elderly; however, severity varies among different individuals. Therefore, knowledge of factors that prevent cognitive impairment in old age is necessary and should be elucidated.

Studies have shown that besides age, which is an irreversible factor, many factors are related to changes in human cognitive function. Individuals are often concurrently exposed to both risk and protective factors, which may interact with each other during the life course of individuals, and the net effect of these interactions determines the overall risk of cognitive impairment. Educational level (Lövdén et al., 2020), sleep (Dzierzewski, Dautovich & Ravyts, 2018), physical activity (Gheysen et al., 2018), smoking (Amini, Sahli & Ganai, 2021), depression (Shimada et al., 2014), malnutrition (Corish & Bardon, 2019), and happiness (Tan et al., 2019) may all be linked to cognitive decline. Among these factors, the effect of happiness on cognitive function deserves added attention because the potential mechanisms of action are likely distinct from those of other interventions in improving cognition.

Happiness, a variable of subjective well-being, is defined as “the overall appreciation of one’s life-as-a-whole” (Danner, Snowdon & Friesen, 2001). Happiness confers several protective effects on health. Happy individuals tend to live longer and be healthier and more resilient (Danner, Snowdon & Friesen, 2001; Fredrickson, 2003; Veenhoven, 2008). According to previous studies, individuals with low subjective well-being are more likely to experience cognitive decline and dementia (Boyle et al., 2010; Sutin, Stephan & Terracciano, 2018). Despite the known associations between happiness and cognitive function, the potential mechanisms underlying these associations are currently poorly understood. A better understanding of how happiness is associated with cognition is crucial, given the beneficial effects of happiness on health and the deleterious effect of cognitive impairment on life.

Older individuals with higher levels of happiness are more likely to be physically active (Kim, Strecher & Ryff, 2014; Steptoe, 2019) and live healthier and longer (Veenhoven, 2008). Many longitudinal studies have shown that physical activity reduces cognitive decline and the risk of dementia (Blondell, Hammersley-Mather & Veerman, 2014; Carvalho et al., 2014; Fratiglioni, Paillard-Borg & Winblad, 2004). Happiness is a positive psychological state, whereas depression is a negative psychological state. According to several reports, individuals with depression tend to have more severe cognitive impairment (Kim et al., 2016; Wong et al., 2015). A cohort study with over three years of follow-up in Spain revealed that depression significantly increased the risk of mild cognitive impairment among older adults (Lara et al., 2017).

Since happiness, depression, and physical activity may all be related to cognition, and happiness is related to depression and physical activity, we hypothesized that happiness may have an indirect effect on cognition through depression and physical activity aside its direct effect on cognition. In other words, depression and physical activity may play mediatory roles between happiness and cognition. To the best of our knowledge, no study has explored the possible association between happiness and cognition with depression and physical activity as mediatory factors. Therefore, this retrospective cross-sectional study was conducted to assess the direct effect of happiness on cognition and assess the mediatory effects of depression and physical activity on the relationship between happiness and cognition.

Material and Methods

Data source and study population

The survey data were obtained from the China Family Tracking Survey (CFPS) (http://www.isss.pku.edu.cn/cfps/), which was organized and conducted by the Institute of Social Sciences of Peking University. The CFPS sample was drawn from 25 provinces/municipalities/autonomous regions in China, and they account for 95% of the total Chinese population. Therefore, CPFS can be regarded as a nationally representative sample (Zhang et al., 2021). All the sub-sampling frames of CFPS were obtained in three stages: the primary sampling unit consisted of administrative districts/counties; the second stage sampling unit consisted of administrative villages/neighborhood communities; and the third stage sampling unit consisted of households. All the selected households participated in the survey. The baseline survey was conducted in 2010, and follow-up surveys were conducted biennially. This study was reviewed and approved by the Biomedical Ethics Committee of Peking University (Approval No.: IRB00001052-14010).

In this study, the CFPS 2018 survey data were selected. Of the 33,097 observed participants, 18,753 participants were excluded. Among the excluded group, 17,971 individuals were younger than 45 years, and 782 individuals had missing data for required variables. The final number of participants was 14,344 individuals; these ones had available data on cognitive test and covariates included in this study (Fig. 1).

Results

Mediatory role of depression and physical activity

The mediatory effect of depression and physical activity were tested by path analysis. After eliminating the influence of control variables, the direct effect of happiness was statistically significant. The estimate value was 0.0287, which accounted for 57.86% of the total effect of happiness on cognition. The indirect effect of mediators was statistically significant, and the estimate value was 0.0209, which accounted for 42.14% of the total effect. For two mediatory variables, the mediatory effect of depression (path of happiness →depression →cognition) accounted for 38.31% of the total effect. The mediatory effect of physical activity (path of happiness →exercise times →cognition) accounted for 3.02%, and the joint mediatory effect of depression and physical activity (path of happiness →depression →exercise times →cognition) accounted for 0.81%. These results indicated that depression and physical activity had mediatory effects, indicating that happiness had an indirect effect on cognition through depression and physical activity. Therefore, Hypotheses 2, 3, and 4 were all accepted. Mediation analysis results were shown in Tables 4 and 5, and Fig. 2.

Table 4:

Mediatory effect of depression and physical activity in the association between happiness and cognition (n = 14, 344).

Effect	Estimate	BootSE	BootLLCI	BootULCI	Relative effect value (%)
Total effect (happiness)	0.0496	0.0063	0.0372	0.0620	–
Direct effect (happiness)	0.0287	0.0066	0.0158	0.0415	57.86
Indirect effect (depression)	0.0190	0.0019	0.0154	0.0228	38.31
Indirect effect (exercise times)	0.0015	0.0014	0.0017	0.0023	3.02
Indirect effect (depression and exercise times)	0.0014	0.0011	0.0012	0.0016	0.81
Total indirect effect of mediators	0.0209	0.0019	0.0172	0.0248	42.14

DOI: 10.7717/peerj.13673/table-4

Notes:

BootSE, BootLLCI, BootULCI refer to the standard error, lower limit, and upper limit of 95% confidence interval of indirect effects estimated by the bias corrected percentile Bootstrap method, respectively.

Table 5:

Mediatory model of depression and physical activity between happiness and cognition (n = 14, 344).

Variables	Model 1				Model 2				Model 3
	B	se	t	p	β	se	t	p	β	se	t	p
Constant	0.119	0.055	2.165	0.03	−0.676	0.210	−3.213	0.01	−0.461	0.041	−11.225	<0.001
Happiness	−0.290	0.008	−34.324	<0.001	0.161	0.034	4.783	<0.001	0.029	0.007	4.375	<0.001
Depression	–	–	–	–	−0.153	0.032	−4.805	<0.001	−0.066	0.006	−10.541	<0.001
Exercise times	–	–	–	–	–	–	–	–	0.009	0.002	5.627	<0.001
Residence	−0.177	0.016	−11.182	<0.001	0.462	0.061	7.630	<0.001	0.148	0.012	12.473	<0.001
Age	0.005	0.001	6.256	<0.001	0.051	0.003	16.394	<0.001	−0.014	0.001	−22.250	<0.001
Sex	−0.278	0.019	−14.761	<0.001	−0.089	0.073	−1.223	0.22	0.231	0.014	16.295	<0.001
Smoking	0.028	0.020	1.409	0.16	−0.202	0.076	−2.648	0.01	−0.038	0.015	−2.579	0.01
Napping	0.007	0.004	1.767	0.08	−0.116	0.015	−7.820	<0.001	−0.021	0.003	−7.316	<0.001
Reading	−0.052	0.024	−2.233	0.03	0.769	0.090	8.573	<0.001	0.318	0.018	18.103	<0.001
Education	−0.097	0.007	−13.080	<0.001	0.371	0.028	13.037	<0.001	0.482	0.006	86.247	<0.001
Satisfaction	−0.128	0.009	−14.573	<0.001	0.086	0.034	2.526	0.01	−0.043	0.007	−6.523	<0.001
Income level	0.001	0.009	0.123	0.90	0.069	0.034	2.022	0.04	−0.045	0.007	−6.795	<0.001
Social status	−0.043	0.009	−4.809	<0.001	0.062	0.034	1.840	0.07	−0.027	0.007	−4.145	<0.001
R2	0.197				0.071				0.553
F	318.677				91.612				1366.076

DOI: 10.7717/peerj.13673/table-5

Notes:

Model 1 takes depression as the dependent variable; Model 2 takes exercise times as the dependent variable; Model 3 takes cognition as the dependent variable; In all three models, residence, age, smoking, napping, reading, education, satisfaction, income level and social status related associated with cognition were included into the equation as covariables.

Discussion

The main objective of this study was to explore the association between happiness, depression, physical activity and cognition of the middle and old-aged Chinese individuals. Using CFPS data from 2018, the findings supported our hypothesis that study participants who were happier had higher scores on cognitive tests after controlling for relevant covariates. Moreover, depression and physical activity played mediatory roles between happiness and cognition. Direct effects accounted for 57.86% of the total effects of happiness on cognition, and indirect effects accounted for the remaining 42.14% through depression and physical activity.

Mental factors are widely acknowledged to affect physical functioning. Moreover, well-established evidence suggest that mental distress, such as depression, has negative effects on physical health, whereas positive mental state, such as happiness, has beneficial effects on physical health (Zautra, 2003). Happiness has substantial effects on psychological and physical health. Currently, happiness is known to foster physical health, but the mechanism underlying this association remains unknown.

Additionally, our study sought to investigate whether happiness can predict cognitive decline in older adults. We found that happiness had a significant positive total effect on cognition. Consistent with our findings, those from studies in the United States and Singapore have shown that individuals with high levels of happiness had a reduced risk of mild cognitive impairment and dementia (Boyle et al., 2010; Zaninotto, Wardle & Steptoe, 2016). Contrarily, a 12-year follow-up study in Maastricht showed no association between positive emotions and cognitive function among the elderly (Berk et al., 2017). Therefore, more evidence is needed to draw firm conclusions. The situation was further complicated by indications that cognitive decline in older adults led to lower happiness levels in the elderly (Wilson et al., 2013), implying that the relationship between happiness and cognition was bidirectional.

Regarding the total effect of happiness on cognition, the indirect effect of depression and physical activity on cognition cannot be overemphasized. Evidence suggest that the associations between happiness and depression were complex and may coexist (Larsen et al., 2017). Consistent with several previous studies (Byers & Yaffe, 2011; Miskowiak et al., 2012; Xu et al., 2015), this study showed that depression was one of the important risk factors for cognitive decline. Meanwhile, happiness led to reduced risk of depression, and lower depression scores were associated with higher scores of cognition; 38.31% of the total effect between happiness and cognition was mediated by depression. A meta-analysis of longitudinal studies supports our findings that later-life depression increased the risk of cognitive impairment (Diniz et al., 2013). Several biologic mechanisms have been observed to be related to the association between depression and mild cognitive impairment. They include presence of high amyloid plaque number, hypothalamic-pituitary-adrenal axis dysregulation leading to increased glucocorticoid production (Butters et al., 2008), and chronic inflammation (Hermida et al., 2012). All of these mechanisms may contribute to frontostriatal abnormalities (Hermida et al., 2012) or hippocampal atrophy (Rapp et al., 2006), resulting in cognitive decline.

In this study, happy individuals had more physical activity levels. Similar to some retrospective studies, this study found physical activity to be one of the protective factors in reducing the risk of cognitive decline (Ruuskanen & Ruoppila, 1995; Stephens, 1988). Additionally, 3.02% of the total effect between happiness and cognition was mediated by physical activity, implying that happiness was associated with physical activity, and more physical activity in later life confers a protective effect on cognition. Similarly, a systematic review and meta-analysis of longitudinal studies (Blondell, Hammersley-Mather & Veerman, 2014) revealed an association between higher levels of physical activity and reduced risk of cognitive decline and dementia. Although the quality of the data is limited, most trial results support the conclusion that moderate levels of physical activity in later life benefit cognitive function in older adults. The cognitive benefits of physical activity include not only maintaining optimal cognitive function but also enhancing existing cognitive function and delaying the onset and progression of cognitive impairment (Barha et al., 2017; Daskalopoulou et al., 2018; Almeida et al., 2014). Furthermore, several studies have reported a dose–response relationship between physical activity and cognitive function, providing additional evidence of the beneficial effect of physical activity on cognitive function (Gillum & Obisesan, 2010; Kasai et al., 2010; McAuley et al., 2011). These studies have suggested a strong relationship between increased physical activity and better cognitive function among older adults. Likewise, several hypotheses exist to explain the beneficial effects of physical activity on cognition. For example, physical activity can promote amyloid clearance and increase cognitive reserve (Fratiglioni, Paillard-Borg & Winblad, 2004), brain volume (Rovio et al., 2010), and brain-derived neurotrophic factor levels (Komulainen et al., 2008). Another trial found that aerobic physical activity can have a direct effect on the body’s hippocampal volume, which was beneficial to cognitive function (Erickson et al., 2011).

Therefore, ensuring that middle and old-aged individuals in China are happy should not be of utmost importance to society. Strategies to ensure happiness in this population with favorable effects on their mental and physical conditions need to be recognized and developed. Although happiness has a genetic underpinning (Okbay et al., 2016), multiple factors can affect it. Tkach and Lyubomirsky (Tkach & Lyubomirsky, 2006) summarized eight strategies for increasing happiness; they include mental control, party attendance, affiliation, goal pursuit, religion, inactive hobbies, active leisure, and intentional attempt. These strategies suggest that it is possible to work toward happiness at any age. Effective interventions towards ensuring happiness among middle and old-aged individuals cannot only improve their subjective well-being but also improve cognitive function. Such improvements carry great potential for reducing public health burdens related to cognitive aging and dementia and are necessary interventions needed in a country with an aging population.

Conclusion

This cross-sectional study in a representative population sample suggested that happiness was positively correlated with cognitive function. Mediation analysis further found significant association of depression and physical activity on the relationship between happiness and cognition. This finding provided a better understanding of how happiness has an effect on cognition and provided clues for future research concerned with links between happiness and cognitive function. Effective interventions for happiness of middle and old-aged individuals in China carries great potential for reducing public health burdens related to cognitive aging.