How does social capital affect individual health among the elderly in rural China?-Mediating effect analysis of physical exercise and positive attitude.

BACKGROUND: The elderly in rural areas comprise over half of the older population in China, and their health problems are a matter of great concern for the Chinese government and society. Among the many factors affecting health, social capital has generated much interest in academic research. Exploring the relationship between social capital and individual health among the elderly in rural China provides ways to improve the health of Chinese people, which has a positive impact on policy.
METHODS: We selected 3719 respondents from the 2016 China Family Panel Studies (CFPS). Structural and cognitive social capital were obtained via exponentiation of variables (mean of zero and a standard deviation of one) and by giving them equal weight. Ordinary least squares (OLS) and two-stage least squares (2SLS) estimators were used to analyze the association between social capital and individual health. We explored the mechanism linking structural and cognitive social capital with individual health through a mediation effect analysis.
RESULTS: After correcting for endogeneity bias, structural social capital had a positive correlation with individual health among the elderly, with a coefficient of 0.062 (95% CI: 0.020-0.104). Cognitive social capital also had a positive correlation with individual health, with a coefficient of 0.097 (95% CI: 0.060-0.135). Physical exercise and positive attitude were two significant mediating variables of the relationship between social capital and individual health in the study group, with mediating effects of 0.018 and 0.054, respectively.
CONCLUSIONS: Cognitive social capital played a stronger role than structural social capital in promoting individual health among the elderly. Physical exercise and positive attitude mediated the relationship between social capital and individual health. Policymakers should not only build basic medical and health care systems but also consistently cultivate and strengthen structural and cognitive social capital among the elderly in rural China.

Introduction

China's population is rapidly aging. By the end of 2018, the number of adults aged 65 years and over had reached 166.58 million, accounting for 11.9% of the total population [1]. With the aging of the population, the risk of disease and disability increases, and the proportion of elderly patients aged 61 and older in 2018 increased 20.5% compared with that in 2017 [2]. Older adults in rural areas constitute the majority of elderly persons in China, and a large number of them live alone at home because their family numbers go out to work in cities with the acceleration of urbanization. These old people do not have much financial income, do not get adequate medical and health services, and often take care of themselves. The main problems rural elderly face are economic poverty, poor physical health, mental loneliness and etc. With the aging of rural areas, the health problem of rural elderly is more prominent than that of urban elderly, which need significant concern of the whole society. Currently, Chinese government has carried out “Healthy China” strategy to meet the challenge of aging. Under such circumstances, knowledge of social determinants of healthy aging are crucial for the development of evidence-based policies and interventions and the sustainable development of Chinese society.

Social capital has been increasingly recognized as significant factor for health [3] and has become an important supplement to the formal medical and health services system. China is a typical Guanxi-based society, and evidence shows Guanxi and social capital have similar connotations and effects [4]. Chinese tend to seeking for social support and maintain social status in the social structure in which they live [5]. In addition, the culture of rural China values trust, mutual assistance and reciprocal exchange, which provide cultural soil for cultivating social capital. Rural residents tend to be more altruistic, honest, and trusting of others, and they reported higher levels of civic cohesion and interpersonal trust than their urban counterparts in China [6]. Some relevant departments do not provide sufficient or formal credit systems for rural elderly, and the elderly often rely more heavily on the development of social capital in daily life. This study would explore the association between social capital and individual health among the elderly in rural area. If social capital does have a promotion on individual health among the elderly in rural China, then providing them with easier access to increasing social capital will make a reference for the implementation of the “Healthy China” strategy.

Definition and classification of social capital

As a formal concept, social capital was first proposed by French sociologist Bourdieu. He defined social capital as “a collection of actual or potential resources related to a lasting network of mutually acquiescent or recognized relationships, which are more or less institutionalized” [7]. Unlike Bourdieu, Coleman defined social capital from the perspective of function. Coleman described social capital as being imbedded in social relationships and as serving as a resource for people to achieve their goals [8]. In his opinion, except for that derived from formal organizations with a position-based structure, all forms of social capital are dependent on the stability of the social structure. The disintegration of social organization and interpersonal relationships means the breakdown of interpersonal interests. In addition, social capital loses the premise for its existence. Putnam defined social capital as “features of social organization such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit” [9]. Putnam actually considered social capital to be a combination of subjective social norms (trust), objective social characteristics (social networks) and outcomes (effectiveness and function). Social capital has powerful explanatory power and influence from multiple perspectives.

Regarding the classification of social capital, it has been argued that the nature of social capital is too diverse and too comprehensive to be operationalized [10]. It is necessary to deconstruct the concept of social capital into its major elements. In academic studies, it is common practice to divide social capital into structural and cognitive components. Cognitive social capital mainly includes an individual moral norms, values, attitudes and trust [11]. Structural social capital is mainly an individual social networks, his or her social engagement and other social structural factors [12]. The value of the cognitive component is generally regarded as a novel contribution of social capital theory [13]. Although the two components are related, they are distinguishable and should be separated from each other [14]. The cognitive component refers to an individual subjective perceptions of accessible social resources, whereas the structural component refers to actual social activities in formal or informal networks (civic engagement and social participation) that can be objectively measured [15]. Structural social capital is thoroughly embodied in the relationships within voluntary associations, consistent with Coleman's definition. Cognitive social capital is mainly embodied in social norms and values, especially social credit, as in Putnam's definition. In this study, we selected structural and cognitive aspects as dimensions of social capital.

Structural, cognitive social capital and health

The importance of social capital for health has been increasingly recognized in the public health literature [16]. Scholars have found that social capital plays a significant role in self-rated health, healthy behavior and chronic illness. Regarding cognitive social capital, trust and neighborhood relationships are positively associated with self-rated health [17,18]. Structural social capital, such as social support and social networks, is associated with self-related health and better functioning [19,20], and a well-structured social network is negatively rated to individual mortality [21]. Among the elderly, social capital significantly promotes their self-rated health, activities of daily living, and chronic disease and mental health status, where its most important role is to promote activities of daily living [22]. Cognitive social capital can significantly alleviate depressive symptoms, relieve loneliness and increase life expectancy among older adults [23]. Structural social capital in the form of social networks is more likely to be associated with better mental health than with better self-related health among the elderly [24]. Social capital has been found to be associated with physical and mental health in Chinese population [25,26]. Participation in social activities promotes health among the elderly in China [27], and bonding trust and bonding networks are positively associated with older adults’ physical and mental health in urban China [28]. It has already been found that structural and cognitive social capital are strongly correlated with individual health. We seek to further explore the relationship between structural and cognitive social capital and individual health among the elderly in rural areas.

The mechanism linking social capital and individual health

Identifying the mechanisms linking social capital and health is challenging in part due to the reliance on cross-sectional data [29,30]. Several scholars have explored the mechanism linking social capital and health. Generally, social capital plays a positive role in individual health through two means: healthy behavior and mental state. Chen H and Meng T proposed that the mechanism linking social capital and health mainly involves social support, psychological perception and material resource access [31]. Mohnen et al. found that social capital is mainly correlated with health via physical exercise and social interaction [32]. Kawachi et al. systematically summarized three possible mechanisms by which social capital affects health status: social capital affects the health of community members through the rapid dissemination of health information, increasing the frequency of exercise and other healthy behaviors; cohesive community members are more likely to create social organizations to ensure individuals’ access to community health-related services, such as institutions and entertainment facilities; and social capital can improve personal health by providing emotional support and reinforcing positive social and psychological processes [33]. Scholars have focused on two key issues when analyzing the impact of social capital on health: the acquisition of resources and the adjustment of mentality. Specifically, norms of conduct from structural social capital can encourage the elderly to imitate healthy behavior and exchange health information, which is conducive to improving human physiological function and promoting the consumption of material. Cognitive social capital is often seen as emotional support. It enhances residents’ awareness of disease and prevents its spread through social networks [34]. The elderly in rural areas have relationships of strong trust and dependence on the individuals around them, allowing them to participate in emotional communication and exchange of information through the social network. This is highly beneficial in helping this population manage negative emotion and adjust their physical and mental state. Thus, we mainly used healthy behavior (e.g., physical exercise) and emotional state (e.g., positive attitude) as mediators and investigated the mechanism linking social capital and individual health among the elderly in rural China.

Current studies on social capital and individual health have paid little attention to the elderly in rural China. In fact, trust and reciprocity among neighbors in rural China have an important impact on people's lives. Research on their social capital can reveal further methods to improve individual health among the elderly and make the relevant departments pay more attention to the construction of social capital in rural China. Based on the theory of social capital, this study divided social capital into structural and cognitive social capital. Current scholars typically analyze social capital from the following four perspectives: the macro-level (national, state, regional and local government); the mid-level (streets and neighborhoods); the micro-level (social networks and social participants); the individual psychological level (trust and norms). Current studies mainly focused on the mid-level and micro-level of social capital. The structural social capital discussed in this study refers primarily to participation in social networks and social organizations, which occur at the micro-level. The cognitive social capital mainly consists of trust, mutual benefit and mutual assistance, which belong to the individual psychological level. Ordinary least squares (OLS) and two-stage least squares (2SLS) estimators were used to analyze the association between social capital and individual health in this study. We explored the mechanism linking structural and cognitive social capital and individual health through a mediation effect analysis.

Methods

Data source

The data came from the adult questionnaire of the 2016 China Family Panel Studies (CFPS), a longitudinal survey of Chinese communities, families, and individuals. The CFPS is a nationwide and large-scale social follow-up survey implemented by the Institute of Social Science Survey, Peking University (ISSS). The CFPS officially started with a baseline survey in 2010 and then carried out three rounds of full-sample tracking surveys in 2012, 2014 and 2016. This survey focused on the economic and noneconomic welfare of Chinese residents, including economic activities, access to education, family relations, family dynamics, population migration, physical and mental health. The target sample size of the CFPS was 16000 households, whose members were from 25 provinces/municipalities/autonomous regions of China (excluding Hong Kong, Macao, Taiwan, Xinjiang, Tibet, Qinghai, Inner Mongolia, Ningxia and Hainan) and covered 95% of the population of mainland China. The CFPS followed scientific sampling methods and guaranteed the randomness of the sample. The sampling used an implicit stratification, multi-stage, multi-level probability sampling method proportional to population size (PPS). The first two phases of CFPS sampling used official administrative divisions. The third stage used the map address method to construct the end sampling frame, and the households were selected by a circular isometric sampling method with random starting points [35]. In this study, the 2016 CFPS data were selected as the sample, and the research object was older adults in rural areas. Therefore, 3719 valid samples were obtained by screening the missing values and invalid samples for the population aged 60 and older.

Variables

Health evaluation index

The main dependent variable was the “health evaluation index”, which was based on self-rated health and others-rated health. Respondents assessed their health using a 5-point Likert scale (1=poor; 2=fair; 3=good; 4=very good; 5=excellent). In addition to self-rated health, this survey solicited visitors’ evaluation of the health status of the respondents. Visitors reported respondents’ health on a scale from 1 to 7. The higher the score was, the better the respondents’ health. Based on a previous study [36], we calculated self-rated health and others-rated health (mean of zero and a standard deviation of one) and gave each of them equal weight to obtain the health evaluation index, making the health assessment more objective.

To more objectively show the difference between self-rated health and others-rated health, this study conducted a further cross-analysis of these two variables. As shown in Table 1, the first column shows five self-rated health scores, and the first row shows seven others-rated health scores. With respect to self-rated health, the number of unhealthy (scored 1) older adults (n=1169) was the highest, while the number of very healthy (scored 5) older adults (n=255) was the lowest. Unhealthy (scored 1) and general (scored 2) older adults accounted for 53.24% of the total. With respect to others-rated health, the number of older adults (n=988) with a score of 5 was the highest, and the number of older adults (n=29) with a score of 1 was the lowest. The number of older adults with a score below 4 accounted for 33.69% of the total.

Table 1

Cross-analysis of self-rated health and others-rated health.

Self-rated health Others-rated health	1	2	3	4	5	6	7	Total
1	24	84	209	277	272	155	148	1169
2	1	12	75	182	219	187	135	811
3	3	14	60	192	317	320	172	1078
4	1	4	22	46	120	134	79	406
5	0	2	10	35	60	85	63	255
Total	29	116	376	732	988	881	597	3719

There was a significant difference between self-rated health and others-rated health (Table 1), which may be because respondents may exaggerate or underestimate their own health status. Interviewers may have evaluated the health status of respondents according to how they performed in the interviews, which would have caused measurement errors. Furthermore, the hidden health risks of respondents are unique and difficult to measure. For example, psychological health and emotional states change often and are difficult to observe. The health evaluation index was used as the dependent variable in this study to eliminate the disadvantage that respondents reported their health too subjectively and make effective use of the information observed by interviewers. We combined and calculated self-rated health and others-rated health to make the results more scientific and reasonable.

Structural and cognitive social capital

Structural social capital refers to externally observable aspects of social organization, such as roles, rules, procedures and precedents [3,9], for example, civic participation or group membership [17]. In accordance with the definition above, structural social capital was measured with relational network and group membership variables in this study. Relational network was used as a proxy of a person’s social network, which is an important element of structural capital. Individuals’ relational network was measured by the question, “In the past 12 months, what was the total amount of money, including material goods and cash, your family spent on banquets and ceremonies?” We included the amount of money (measured in RMB yuan) in logarithm form. Organization membership was determined based on the following three questions: “Are you a member of the Communist Party of China?”, “Are you a member of a religious group?” and “Are you a member of an association of individual workers?”. Respondents answered “Yes” (1) or “No” (0) to each question in the interview. Cognitive social capital is more internal and subjective, referring to shared norms, values, attitudes and beliefs [3]. In accordance with the definition above, cognitive social capital consists of trust, reciprocity and mutual assistance in this study. In the 2016 CFPS, respondents were asked to indicate their trust in different people: “Let ‘0’ be very untrustworthy and ‘10’ be very trustworthy. Please rate your degree of trust in the following groups of people”. The groups were parents, neighbors, Americans, strangers, government officials and doctors. Because Americans were not relevant to individual social capital in the study population and participants’ parents may have died long ago, the trust scores for Americans and parents were excluded. The trust scores for the other groups were summed and averaged to produce a general trust score. Reciprocity was assessed by the question “How is the relationship between neighbors in your community?” Responses to the question ranged from 1 to 5. The higher the score was, the better the neighborhood relationship. Mutual assistance was measured by the question “When you need any help from neighbors, do you think anyone will give a helping hand?” Responses to the question also ranged from 1 to 5. The higher the score was, the stronger the feeling. We calculated these indicators of structural and cognitive social capital (mean of zero and a standard deviation of one) and gave them the same weight to obtain the structural social capital index and cognitive social capital index.

Mediating variables

We selected two mediating variables: positive attitude and physical exercise. The 2016 CFPS provided some descriptions of respondents’ mental status in the past week. Positive attitude was indicated by two feelings: “I am happy in life” and “I feel pleasant.” The responses ranged from 1 to 4. We added the scores and averaged them to obtain a positive attitude. Physical exercise was measured by the question “How often do you exercise a week?”. Physical exercise is a numerical variable.

Control variables

We mainly selected demographic characteristics as control variables, including sex, age, education level, marital status and social status. Males were indicated by (1) and females by (0). Age was measured in years and centered on the mean (68.829). Education was measured by the highest level of education attained, in five categories: illiterate/semi-literate (0), primary school (1), junior high school (2), senior high school/vocational school (3), three-year college (4), 4-year college/bachelor’s degree (5) and master’s degree/doctoral degree (6). Marital status was assessed in two categories: never married/divorced/widowed (0) and married/cohabiting (1). Social status was measured by the question “What is your social status in your local area?” The answers ranged from low (1) to high (5).

Statistical analysis

The characteristics of the participants were expressed as the mean and standard deviation for continuous variables and number (percentage) for categorical variables. Descriptive analysis (Table 2) showed that the structural social capital of the elderly was not high, with a mean relational network score of 6.75 and only 14.20% of older people participating in organizations. We found that the maximum (2.26) of the structural social capital index did not differ greatly from the minimum (-1.53). However, the cognitive social capital of the rural elderly was not low, as the means of reciprocity and mutual assistance were 3.90 and 4.56, respectively. It was surprising that the mean (5.38) of trust was not high, although the maximum was 10. We found that the maximum (1.20) of the cognitive social capital index was fairly distant from the minimum (-3.68). On the whole, the average of cognitive social capital (0.01) was higher than that of structural social capital (-0.01).

Table 2

Descriptive statistics of the variables (N=3719).

Variables	Max	Min	Mean	S.D.
Self-rated health	5	1	2.40	1.22
Others-rated health	7	1	5.03	1.37
Health evaluation index	1.78	-2.03	0.01	0.79
Relational network	12.71	0	6.75	2.66
Organization membership	Yes (14.20%); No (85.80%)
Structural social capital index	2.26	-1.53	-0.01	0.74
Trust	10	0	5.38	1.62
Reciprocity	5	1	3.90	0.81
Mutual assistance	5	1	4.56	0.89
Cognitive social capital index	1.20	-3.68	0.01	0.70
Positive attitude	4	1	2.93	0.89
Physical exercise	21	0	2.40	3.38
Sex	Males (51.82%); Females (48.18%)
Age	98	62	68.83	5.85
Education level	5	0	0.58	0.86
Marital status	Have a spouse (79.65); No spouse (20.35%)
Social status	5	1	3.14	1.15

Regarding the statistical analysis of the other variables (Table 2), the mean positive attitude among the elderly was not low, at 2.93. However, the mean physical exercise score was only 2.40, which indicated that some older people did not usually engage in physical activities. There was a large gap between the maximum (21) and the minimum (0) for physical exercise, indicating a large difference in physical exercise involvement. Males accounted for 51.82% of the sample in this study. The average age of respondents was 68.83. It is worth noting that the mean education level among the elderly in rural areas was very low, only 0.58. Respondents with a spouse accounted for 79.65%. There was no large gap in social status among the elderly, and the mean was 3.14.

Models

We established the following models to study the relationship between structural and cognitive social capital and individual health:

In Model 1, Yi was the health evaluation index; S was the structural social capital index; C was the cognitive social capital index; X were the control variables; and β1 and β2 were the estimated parameters reflecting the effect of structural and cognitive social capital on individual health, respectively. The dependent variable in this study was the health evaluation index, a numerical variable. Therefore, we used a linear regression model.

The main problem with Model 1 was the endogeneity between structural social capital and individual health. One’s health condition may in turn affect one’s structural social capital. It is possible that older people with better health were more inclined to expand their relational networks and participate in various organizations. Endogeneity can result in bias in the estimation results of Model 1. The best way to solve endogeneity bias is to find suitable instrumental variables. In this study, the variable of lag one phase of the endogeneity explanatory variable (structural social capital) was selected as the instrumental variable. On the one hand, the endogenous explanatory variable (structural social capital) was related to the lag variable. On the other hand, the variable of lag one phase had already occurred and may not be related to the current period. The disturbance was irrelevant. We used the relational network and organization membership of the same cohort of older adults from the 2014 CFPS data as instrumental variables.

Using two-stage least squares (2SLS) estimation, we reset the regression equation to:

In Model 2, Z1 and Z2 were instrumental variables; X represented other control variables; vi was the linear random error; λ1 and λ2 were the estimated parameters, which respectively indicated the influence of the relationship network and organization membership on structural social capital. In Model 3, was the predictive effect of the first-stage regression result; C was cognitive social capital; X represented other control variables; and εi was the linear random error. β3 and β4 were estimated parameters that represented the predictive effect of the first-stage regression result and the relationship between social capital and individual health, respectively.

According to the method of causal stepwise regression proposed by Baron and Kenny [37], the procedure for testing the mediation effect is as follows: first, the independent variable is regressed on the dependent variable, and the regression coefficient c must be significant. The existence of main effect is the premise of the mediation effect (Model 4). Then, the independent variable is regressed on the intermediate variable. The regression coefficient a should be significant, which indicates that the independent variable has an effect on the intermediary variable (Model 5). Finally, the independent variable and the intermediate variable are simultaneously regressed on the dependent variable. The variable regression coefficient b should be significant (Model 6), while the independent variable regression coefficient c' is not significant or the effect size is significantly reduced compared to c. Satisfying all the above three conditions is indicative of a mediating effect. In addition, the coefficient c' in Model 6 is used to determine whether the mediation effect is partial (c' is significant) or complete (c' is not significant).

In Model 4, yi was the health evaluation index; x was structural and cognitive social capital and other control variables; and c was the estimated parameters. In Model 5, mi represented positive attitude and physical exercise; x was structural and cognitive social capital and other control variables; and a was the estimated parameters. In Model 6, yi was the health evaluation index; x was structural and cognitive social capital and other control variables; mi was positive attitude or physical exercise; and c and b were the estimated parameters.

We estimated the models with the statistical software package Stata 13.0.

Results

Investigating the relationship between social capital and individual health

In Model 1 of Table 3, only the mediating variables and control variables were included to examine the relationship between them and individual health. Both mediating variables had a significant correlation with individual health among the elderly, with coefficients of 0.221 (p<0.001) and 0.011 (p<0.001). Age had a negative association with individual health, as the coefficient was -0.011 (p<0.001). Males were more likely to reporter better health than females were. Education and social status had a positive correlation with individual health; the coefficients were both 0.067 (p<0.001). However, marriage did not have a close relationship with individual health. In Model 2, after controlling for other factors affecting the health among the elderly, structural and cognitive social capital were added to the estimated model. Structural social capital played a significant role in promoting individual health among the elderly, with a coefficient of 0.068 (P<0.01). Cognitive social capital also had a significant correlation with individual health; the coefficient was 0.097 (P<0.001).

Table 3

Regression results of the relationship between social capital and individual health.

Variables		Model 1	Model 2	Model 3	Model 4
		Coef.	Coef.	Coef.	Coef.
Structural social capital index		/	0.068***	/	0.062**
		/	(0.035-0.102)	/	(0.020-0.104)
Cognitive social capital index		/	0.097***	-0.003	0.097***
		/	(0.061-0.133)	(-0.022,0.017)	(0.060-0.135)
Mediating variables	Positive attitude	0.221***	0.203***	0.001	0.203***
		(0.193-0.248)	(0.175-0.231)	(-0.013-0.016)	(0.174-0.232)
	Physical exercise	0.011**	0.009**	0.010***	0.009*
		(0.004-0.018)	(0.002-0.016)	(0.006-0.014)	(0.002-0.016)
Control variables	Sex	0.123***	0.120***	0.041**	0.121***
		(0.072-0.174)	(0.069-0.171)	(0.013-0.069)	(0.070-0.171)
	Age	-0.011***	-0.011***	0.004**	-0.011***
		(-0.016--0.007)	(-0.016--0.007)	(0.001-0.006)	(-0.016--0.007)
	Education level	0.067***	0.059***	0.029***	0.060***
		(0.037-0.097)	(0.029-0.089)	(0.011-0.048)	(0.031-0.088)
	Marital status	0.055	0.052**	0.015	0.053
		(-0.009-0.119)	(-0.012-0.116)	(-0.018-0.047)	(-0.012-0.117)
	Social status	0.067***	0.054***	0.010	0.055***
		(0.046-0.088)	(0.033-0.076)	(-0.002-0.022)	(0.033-0.077)
Instrumental variables	Relational network in lag period	/	/	0.195***	/
		/	/	(0.190-1.999)	/
	Organization membership in lag period	/	/	0.972***	/
		/	/	(0.917-1.028)	/
Constant		-0.251	-0.145	-1.793***	-0.146
		(-0.579-0.076)	(-0.472-0.183)	(-1.986--1.600)	(-0.472-0.180)
R2		0.117	0.126	0.715	0.128
Number of observations		3719	3719	3719	3719

The 95% confidence intervals are in parentheses.

*p<0.05

**p<0.01

***p<0.001.

According to the above findings, there was an obvious correlation between structural and cognitive social capital and individual health, but any observed causal relationship must be interpreted carefully. We used 2SLS estimation, Model 3 shows the correlated factors of structural social capital. The relationship network and organizational membership in the first phase had a significant positive correlation with structural social capital. The first phase F statistic was 914.48, which was far greater than 10. There was no problem with weak instruments. The model was subjected to an over-identification test (p=0.6647); the relationship between networks and organization membership in the lag phase was not related to the disturbance term. In Model 4, both structural and cognitive social capital had a positive correlation with individual health, and the coefficients were 0.062 (p<0.01) and 0.097 (p<0.001), respectively. Compared with those in Model 2, the direction of each control variable did not change, and the coefficient of structural social capital on individual health dropped from 0.068 to 0.062. After correcting for endogeneity bias, the general results showed a causal relationship from structural and cognitive social capital with individual health. The estimation results also revealed that the effect of structural social capital on health was larger than that of cognitive social capital, consistent with the current academic perspective [38,39].

Analysis of the mechanism linking social capital and individual health

The theoretical analysis and empirical results mentioned above showed relationships between structural and cognitive social capital and individual health among the elderly in rural China. To further clarify the mechanism by which social capital is correlated with individual health, we examined the mediating effects of physical exercise and positive attitude on the relationship between social capital and individual health. The results are shown in Table 4. In Model 5, structural social capital had a significant correlation with individual health (β=0.085, P<0.001), and cognitive social capital was significantly associated with individual health (β=0.150, P<0.001).

Table 4

Mediation analysis results for physical exercise and positive attitude.

Variables		Health evaluation index	Physical exercise	Positive attitude	Health evaluation index	Health evaluation index
		Model 5	Model 6	Model 7	Model 8	Model 9
Independent variable	Structural social capital index	0.085***	0.385***	0.064**ss	0.080***	0.072***
		(0.051-0.119)	(0.238-0.532)	(0.026-0.102)	(0.046-0.114)	(0.038-0.105)
	Cognitive social capital index	0.150***	0.249**	0.251***	0.147***	0.099***
		(0.114-0.186)	(0.094-0.405)	(0.211-0.292)	(0.111-0.183)	(0.063-0.135)
Instrumental variables	Physical exercise				0.012**
					(0.004-0.019)
	Positive attitude					0.205***
						(0.0177-0.233)
R2		0.078	0.045	0.074	0.080	0.126
Number of samples		3719	3719	3719	3719	3719

The 95% confidence intervals are in parentheses.

*p<0.05

**p<0.01

***p<0.001. The effect of control variables was omitted.

In Model 6, we examined the relationship between social capital and the mediating variable (physical exercise). Structural social capital had a positive correlation with physical exercise (β=0.385, P<0.001). Cognitive social capital also had a significant association with physical exercise (β=0.249, P<0.01). In Model 8, the independent variables and mediating variable (physical exercise) were added to the model to investigate the relationship between them and individual health among the elderly. The coefficient of structural social capital and individual health was 0.080 (P<0.001), and the coefficient of cognitive social capital and individual health was 0.147 (P<0.001). The mediating variable (physical exercise) had a significantly positive correlation with individual health (β=0.012, P<0.01). Physical exercise played an intermediary role in the relationship between structural social capital and individual health, and the mediating effect was 0.005 (0.385*0.012=0.00462). It also played an intermediary role in the relationship between cognitive social capital and individual health, and the mediating effect was 0.003 (0.249*0.012=0.002988).

In Model 7, we examined the relationship between social capital and the mediating variable (positive attitude). Structural social capital had a positive correlation with positive attitude (β=0.064, P<0.01). Cognitive social capital also had a significant association with positive attitude (β=0.251, P<0.001). In Model 9, the independent variables and mediating variable (positive attitude) were added to the model to investigate the relationship between them and individual health among the elderly. The coefficient of structural social capital and individual health was 0.072 (P<0.001). The coefficient of cognitive social capital and individual health was 0.099 (P<0.001). The mediating variable (positive attitude) had a significantly positive correlation with individual health (β=0.205, P<0.001). Positive attitude played an intermediary role in the relationship between structural social capital and individual health, with a mediating effect of 0.013 (0.064*0.205=0.01312). It also played an intermediary role in the relationship between cognitive social capital and individual health; the mediating effect was 0.051 (0.251*0.205=0.051455).

The above analysis results indicate that both physical exercise and positive attitude played an intermediary role in the relationship between structural and cognitive social capital and individual health. As shown in Table 5, the percentage of the mediating effect of physical exercise on the relationship between structural social capital and individual health (5.88%) was greater than the that on the relationship between cognitive social capital and individual health among the elderly (2%). The percentage of the mediating effect of positive attitude on the relationship between structural social capital and individual health (15.29%) was less than that on the relationship between cognitive social capital and individual health among the elderly (34%). In general, the mediating effect of positive attitude was stronger than that of physical exercise.

Table 5

Effect of physical exercise and positive attitude.

Instrumental variables	Independent variables	Direct effect	Mediation effect	Total effect	Percentage of the mediation effect (%)
Physical exercise	Structural social capital index	0.080	0.005	0.085	5.88
	Cognitive social capital index	0.147	0.003	0.150	2
Positive attitude	Structural social capital index	0.072	0.013	0.085	15.29
	Cognitive social capital index	0.099	0.051	0.150	34

Discussion

Differential relationship of structural and cognitive social capital with individual health

The key findings showed that structural and cognitive social capital had a positive correlation with individual health among the elderly in rural China. This result is consistent with the large body of research emphasizing the strong link between social capital and individual health. People who have structural social capital could get available public spaces and access to mutual support, and rapid diffusion of health information and healthy norms of behavior through their clubs and associations [40]. Social groups, even those whose focus is not directly about health (e.g. religious organizations and other interest groups), tend to provide opportunities for its members to know and stay alert about health-related issues. Structural social capital could induce more collective actions, which hold promise for improving the health and well-being of the Chinese population by promoting healthy behavior [41]. As for cognitive social capital, it indicates the ability to seek for information, material, and emotional support networks, comply with social norms and peer control, trust and work closely with others in their daily activities, all of which could lead to receive adequate medical services and psychological support to buffer sufferings caused by illness. The impact of cognitive social capital is mainly psychological support through interpersonal trust and mutual assistance, which generally predicts good self-rated health [42], and similar findings were also identified in Chinese studies [43]. Therefore, cognitive social capital has different association with individual health from structural social capital. While structural aspects provide support through formal and informal institutions, cognitive social capital may increase the sense of belonging in one’s community, which is beneficial for mental health [44].

Interestingly, cognitive social capital had a stronger association with individual health promotion than structural social capital in this finding. In individual health studies, cognitive social capital, i.e., trust and reciprocity, seems to have a stronger impact on health than structural aspects [45]. The elderly in rural China participated in far fewer social organizations than old adults in urban China [27,46]. The only organizations in their vicinity may be workers’ associations, party organizations, etc. However, trust and mutual assistance between neighbors are common in the social life of the elderly, and their perception of neighborhood relations is stronger than that of organizational participation. The elderly in rural China have less structural social capital than cognitive social capital, so cognitive social capital has a stronger regulating effect on their health. In particular, females have difficulty participating in social organizations because of their limited education and domestic work [20]; they have less structural social capital than males, and they receive health promotion more from cognitive social capital.

Not only in rural China, but also in other rural areas, the promotion of cognitive social capital on health is greater than that of structural social capital. Cognitive social capital indicators such as trust and reciprocity were found to have larger effects on self-rated health than structural social capital indicators (e.g., social participation) [47]. Rural respondents reported higher levels of civic cohesion and trust than their urban counterparts, but they reported lower levels of community support [48]. A study in South Australia documented that rural respondents had significantly higher levels of community trust than their urban counterparts [49]. Rural elderly tend to develop cognitive social capital from community norm and trust, and they do not have a number of recreation facilities (e.g., exercise facilities, room for card games, etc) and organizations (e.g., elderly association, activity center for the elderly, etc) to get structural social capital. This finding showed that rural resident have more cognitive social capital than structural social capital, which lead to different relationship between them with individual health.

How structural and cognitive social capital connect with individual health

It is worth further exploring how the two types of social capital are associated with individual health. The effect of structural social capital on health promoted an interactive function in which the network may promote the exchange of health information and the sharing of resources. Larger structural social capital (e.g., social networks) can expand the source of information and promote the spread of health information among network members. Structural social capital also can be a source of self efficacy belief for finding, understanding, and using health information [50]. The elderly in rural areas who joined various organizations had common goals and norms, which may promote the cultivation and persistence of healthy behavior. This finding supports the idea that structural social capital promotes individual health, which is consistent with the academic perspective [51,52].

With respect to cognitive social capital, trust played a role primarily in reducing social conflict, promoting social harmony and improving individuals’ mood. Reciprocity and mutual assistance may promote social bonds between individuals, which could reduce the cost of health information exchange for the elderly and play a significant role in promoting individual health among older adults in rural China. People who obtain relevant health information from their interpersonal networks, the media, or their government may decide to engage in health-protective action only if they trust that particular information source [53]. Cognitive social capital is significantly associated with reduced morbidity and pain perception, which is beneficial to physical health. This result is consistent with the recent finding that enhancing cognitive social capital is an effective way to alleviate pain [54].

Furthermore, our results also showed that physical activity and positive attitude play different mediation on the relationship between social capital and individual health. This is probably because structural and cognitive social capital connect with health through different ways, and they affect different aspects of health. Structural social capital mainly provides the dissemination of health information and the imitation of health behaviors for the elderly, which demonstrated the mediating effect of physical exercise on the relationship between structural social capital and individual health was greater than positive attitude. Structural social capital is mainly beneficial to physical health, and cognitive social capital play a positive role in mental health. For example, individual-level measures of social capital, including social engagement, trust, neighborhood attachment, and sense of belonging, have been shown to be negatively associated with common mental disorders [55]. Previous research found that trust and reciprocity, as components of cognitive social capital, have a stronger relationship with the mental health of older immigrants from China than does structural social capital [14]. Cognitive mainly relieve individual stress and regulate mindset to promote health, which demonstrated the mediating effect of positive attitude on the relationship between cognitive social capital and individual health was larger than physical activity. According to previous Chinese studies, trust is the main social capital element associated with health; norms of reciprocity and social networks have little influence on population health [56]. However, this study did not examine the effects of trust, reciprocity, and mutual assistance in detail, and it did not assess the degree of influence that different forms of cognitive social capital have.

Strengths and limitations

The first strength of this study is the use of a health evaluation index based on self-rated health and others-rated health to investigate the potential relationship between two types of social capital and individual health. Current scholars primarily focus on the physical health, mental health or self-rated health among the elderly but have not used comprehensive evaluation indicators to measure health. The second strength is the study’s solving of the endogeneity problem by finding instrumental variables and using 2SLS estimation to analyze the causal relationship between social capital and individual health. The third strength is exploring the mechanism linking social capital and individual health via two important mediating variables, physical exercise and positive attitude, which was beneficial for exploring how social capital affects individual health.

There are several limitations to this study. Firstly, the measurement of health is not an easy task. At present, the common health measurement indicators for the elderly include self-rated health, activities of daily living scales, BMI index and chronic disease types. This paper obtains a health evaluation index through the exponentiation of self-rated health and others-rated health, which corrects for the subjectivity of self-rated health. However, more accurate and scientific measurements of individual health should be used in future research. The second shortcoming is that other factors affecting health are not included in the model as control variables. The factors affecting the health status of individuals go beyond living habits, economic conditions, the social environment, genetics, family status, etc. Our variables were limited according to the existing indicators of the 2016 CFPS, which meant that we could not include other variables that affect individual health. The third limitation is the handling of endogeneity bias. Although the CFPS has three periods of survey data, the social capital variables from the three phases are not exactly the same. It is impossible to use panel data. Therefore, we used the structural capital in the lag phase as an instrumental variable to overcome endogeneity bias. More research is needed to explain the dynamics of social capital and its relationship with health over multiple time periods.

Conclusions

This study divided social capital into structural and cognitive social capital and focused on the elderly in rural China. Using the 2016 CFPS data, we drew three conclusions. Firstly, when controlling other variables affecting health, both structural and cognitive social capital had a significant positive correlation with individual health. Secondly, after resolving the endogeneity bias of structural social capital and health, structural social capital still had a significant association with health, and the net effect was smaller than before. Thirdly, physical exercise and positive attitude played mediating roles in the relationship between social capital and individual health among elderly rural residents. The mediating effect of positive attitude was greater than that of physical exercise.

This study has several policy implications. The health status of elderly persons in rural China is poor, and the relevant government entities should carry out regular physical examinations of these individuals and follow up their physical condition in a timely manner. In particular, the health department must appoint psychologists to implement interventions for this population so that the elderly can maintain a positive attitude. The medical department should assign more medical resources to villages to build a more comprehensive rural medical and health system, actively promote health knowledge among the elderly, and encourage the elderly to correct poor diet and hygiene habits.

Our findings implied that accumulating a higher stock of social capital is beneficial to health. It is necessary to take the needs of the elderly as the starting point to establish rural organizations based on local characteristics and demands and enhance enthusiasm for participation in organizations among elderly rural residents. More public resources should be devoted to supporting local activities and local organizations in order to encourage more elderly rural residents to join neighborhood social groups. To encourage reciprocity among neighbors, policymakers should create and develop an ethos of community support by praising and rewarding those who help their neighbors. Efforts should be made to promote a sense of belonging among neighborhood residents. By strengthening community building and valuing the importance of trust and social norms, we can make cognitive social capital play a positive role in the lives of the elderly in rural areas.

Further analyses are needed to identify which particular elements of social capital may be driving the results and the effect size. The elderly in rural China are highly heterogenous, and there are different relationships between social capital and individual health in different populations of old people. For example, left-behind old adults and others have different mental health needs, and male and female elderly have differential structural social capital. The impact of social capital on the health of different groups among the elderly is worth further exploration. Future research should also explore other mediating variables in addition to physical exercise and positive attitude that contribute to the mechanism linking social capital and individual health.

(RAR)

Click here for additional data file.

18 Dec 2019

PONE-D-19-29520

How does social capital affect individual health among the rural elderly in China?--mediating effect analysis of physical exercise and positive attitude

PLOS ONE

Dear Dr Yue,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

1) Manuscripts submitted to  are expected to report statistical methods in sufficient detail for others to replicate the analysis performed. Ensure that results are rigorously reported in accordance with community standards and that the statistical methods employed are appropriate for the study design. Please include reliability analyses for all measures.

2) English language -  We may recommend that authors seek independent editorial help before submitting a revision. These services can be found on the web using search terms like “scientific editing service” or “manuscript editing service.”

==============================

We would appreciate receiving your revised manuscript by 17 January 2020. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Rosemary Frey

Academic Editor

PLOS ONE

Journal Requirements:

1.

When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please clarify if the database used is anonymized. If it is not please indicate how you ensured anonymity of research subjects. Please indicate if the original database was approved by an IRB and if so, what the name of this IRB is.

3. Please clarify your financial disclosure statement to fully explain the role that the funding agencies had in the study. In particular please describe any oversight they had and how this impacted the design and execution of the study.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This paper purports to examine the associations of structural and cognitive social capital with subjective health - an index based on a person's self-rated health and a rating of that person health conducted by a third person. Additionally, they assessed physical exercise and attitude as mediators. A strength of the paper is the use of 2SLS, which takes into account bidirectional relationships.

The authors have not sufficiently demonstrated an understanding of the concept of cognitive and structural social capital and their connections with the health of individuals. The authors make claims such as "scholars have had a heated discussion" but do not cite evidence of those scholars and their heated discussions. I am concerned about the validity of the structural and cognitive social capital variables. Some information on social capital this is presented in the discussion would be better suited to the introduction, and ideally, statistics should not appear in the conclusion. Policy implications for the paper are weak, for "relevant government departments should pay attention to the health status of rural elderly".

Reviewer #2: The article " How does social capital affect individual health among the rural elderly in China? --mediating effect analysis of physical exercise and positive attitude " makes an effort to advance our understanding among individual health, social capital, physical exercise and positive attitude specifically in rural elderly which shows novelty of work. The literature review is decent, the method is acceptable, English language and style are fine but minor spell check is required.

Its better to explain the theoretical approach in the last part of introduction and it should lead the reader to believe that the current study is necessary.

The methodology part should tell the reader directly whether the sample is random or not.

The research has used the scales like Health evaluation index, Structural social capital, Cognitive Social Capital, positive attitude and physical exercise, the authors are recommended to conduct reliability test on these scales for the current sample and mention reliability scores.

Authors mentioned that one question was used to measure physical exercise but its not clear which type of scale was used and what was the standard of good/poor physical exercise.

The authors analyzed the mediating effect but the model used for mediating effect is not properly described. It is advised to make it clear that which model was used.

Its better to mention study objectives again in conclusion part.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Azam Tariq

College of Humanities and Social Sciences

Huazhong Agricultural University, Wuhan, China

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

13 Jan 2020

Response to Reviewers

Dear Dr./Prof. Frey,

On behalf of my co-authors, we thank you very much for giving us an opportunity to revise our manuscript, and we also appreciate the positive and constructive comments and suggestions, given by you and other reviewers, on our manuscript entitled “How does social capital affect individual health among the elderly in rural China?--mediating effect analysis of physical exercise and positive attitude” (ID:PONE-D-19-29520).

Having studied the reviewer’s comments carefully, we have tried out best to make revisions which were marked in red in this paper. Please find the revised version enclosed, which we would like to submit for your kind consideration.

Point 1:Manuscripts submitted to PLOS ONE are expected to report statistical methods in sufficient detail for others to replicate the analysis performed. Ensure that results are rigorously reported in accordance with community standards and that the statistical methods employed are appropriate for the study design. Please include reliability analyses for all measures.

Response 1:

We appreciate the comments of the reviewers. In the previous manuscript, we did not describe the mediation effect analysis method in detail. In the revised manuscript, we explained this method in detail (please see line 338-361). In the previous model analysis, we reported standard errors. After studying the community standards in detail, we reported 95% confidence intervals (please see Table 3 and Table 4). Regarding the data of this article, we have uploaded the data in the previous manuscript, and filled in the website link of downloading and viewing database. We added the command data in a file uploaded to ensure that the experiment can be repeated operated. We proofread the format in this study again, responded to the comments of reviewers, and revised in accordance with their suggestion. (please see below response for comments to reviewers).

Point 2:English language --We may recommend that authors seek independent editorial help before submitting a revision. These services can be found on the web using search terms like “scientific editing service” or “manuscript editing service.”

Response 2:

Thanks a lot for the editor's suggestion. We apologize for not using the standard English language before. We have seek independent editorial help before submitting a revision.

We have invited experts whose native language is English to modify the presentation of this article, considering that these modifications may not fully meet the requirements of PLOS ONE. At the same time, we are afraid that the paper editing service organization cannot complete the paper editing within the deadline. Therefore, we promise that, in addition to the English expression, if the entire article can be officially accepted, we will use your company’s English editing service to make the manuscript fully meet the publishing requirements.

We would like to express our great gratitude to you and other reviewers for the comments on our paper. Looking forward to hearing from you.

Thank you and best regards.

Yours sincerely,

Zhang Yue

Corresponding author:

Name: Zhang Yue

E-mail: yuezhang.znufe@163.com

Response to Reviewer 1 Comments

Dear Editors and Reviewers,

Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled “How does social capital affect individual health among the elderly in rural China?--mediating effect analysis of physical exercise and positive attitude” (ID: PONE-D-19-29520). Those comments are all valuable and very helpful for revising and improving our paper, and simultaneously they are of great significance to give instructions to our research. We have studied the comments carefully and made revisions which we hope to meet with approval. Revised portions are marked in red in the original paper. The main revisions on this paper, along with our responses to the reviewer’s comments, are as follows:

Point 1: The authors have not sufficiently demonstrated an understanding of the concept of cognitive and structural social capital and their connections with the health of individuals.

Response 1:

We appreciate the comments of the reviewers. We apologize for ignoring detailed explain the concept of cognitive and structural. In our study, cognitive social capital mainly includes an individual's moral norms, value, attitude and trust. Structural social capital is mainly an individual's social networks, his or her social engagement and other social structural factors. In this article, we further point out the characteristics and differences between structured social capital and cognitive social capital (see line 89-95). As for the concept of the two types of social capital, we mainly borrowed academic points of these references as follows:

1.Park M. Impact of social capital on depression trajectories of older women in Korea. Aging & Mental Health, 2017,21(4):354-361. https://doi.org/10.1080/13607863.2015.1088511 PMID: 26404493

Structural social capital acts as network resources to older women such as the quantity of social network, family and friends support. The cognitive component of social capital includes perception of support, reciprocity, sharing and trust.

2.Harpham T, Grant E, Thomas E. Measuring social capital within health surveys: key issues. Health Policy Plan. 2002; 17(1):106-11. https://doi.org/10.1093/heapol/17.1.106 PMID: 11861592

The structural component includes extent and intensity of association links or activity ,and the cognitive component covers perceptions of support, reciprocity, sharing and trust.

3.Rostila, M. Social capital and health in European welfare regimes: a multilevel approach. Journal of European Social Policy. 2007, 17, 223-239. https://doi.org/10.1177/0958928707078366

The structural component of social capital is the extent and intensity of participation in associations and other forms of social activities (e.g. density of civic associations, measures of informal social participation) whereas the cognitive component is about peoples perceptions of interpersonal trust, sharing and reciprocity.

4.Harpham T. The measurement of community social capital trough Surveys. In: Kawachi I, Subramanian SV, Kim D, editors. Social Capital and Health. New York: Springer; 2008. p51-62.

Structural social capital refers to what people do (associational links, networks) which could be objectively verified (by observation or records). Cognitive social capital refers to what people feel (values and perceptions) and is thus subjective.

Point 2: The authors have not sufficiently demonstrated cognitive and structural social capital and their connections with the health of individuals.

Response 2:

In the previous manuscripts, we mainly focused on the literature review of the relationship between social capital and health and did not review the literature of structured social capital, cognitive social capital’s connection with health in more detail. We reorganized the relevant literature and focused on solving this problem. Please see line105-118.

List of newly added related references as follows:

1.Ichida Y, Kondo K, Hirai H, et al. Social capital, income inequality and self-rated health in Chita Peninsula, Japan: a multilevel analysis of older people in 25 communities. Soc Sci Med 2009;69:489-99 https://doi.org/10.1016/j.socscimed.2009.05.006 PMID: 19523728

People who live in conditions of high-income inequality tend to exhibit low trust levels, and that trust of cognitive social capital mediates the relation between income inequality and health.

2.Pollack CE and von dem Knesebeck O. Social capital and health among the aged: Comparisons between the United States and Germany. Health Place. 2004; 10(4):383-91. https://doi.org/10.1016/j.healthplace.2004.08.008 PMID: 15491897

social capital including both norms (reciprocity and civic trust) and behaviors (participation) were associated with self-reported health (indicators-overall health, depression (CES-D) and functional limitations).

3.Li Q, Zhou X, Ma S and etc. The effect of migration on social capital and depression among older adults in China. Soc Psychiatry Psychiatr Epidemiol. 2017 Dec;52(12):1513-1522. https://doi.org/10.1007/s00127-017-1439-0.

Social capital measurements included cognitive (generalized trust and reciprocity) and structure (support from individual and social contact) aspects.The depression disadvantage is partly accounted for by lower level of cognitive social capital (trust and reciprocity).

4.Brinkhues S, Dukers-Muijrers NHTM, Hoebe CJPA, et al. Socially isolated individuals are more prone to have newly diagnosed and prevalent type 2 diabetes mellitus-the Maastricht study-. BMC Public Health. 2017; 17:955. https://doi.org/10.1186/s12889-017-4948-6 PMID: 29254485

Several aspects of structural and functional characteristics of the social network were associated with newly and previously diagnosed T2DM, Men and women who were more socially isolated, and who received less emotional and practical support, more frequently had newly and previously diagnosed T2DM.

5.Liu G, Xue X, et al. How does social capital matter to the health status of older adults? Evidence from the China Health and Retirement Longitudinal Survey. Econ Hum Biol. 2016; 22:177-189. https://doi.org/10.1016/j.ehb.2016.04.003 PMID: 27235837

We obtain evidence indicating that structural social capital has a significant and positive effect on general and physical health but not on mental health.

6.Norstrand JA, Xu Q. Social capital and health outcomes among older adults in China: the urban-rural dimension. Gerontologist. 2012; 52(3):325-34. https://doi.org/10.1093/geront/gnr072 PMID: 21746837

Trust was significantly associated with physical and emotional health, and participation of organization was significantly associated with physical health among the urban older Chinese.

Point 3: The authors make claims such as "scholars have had a heated discussion" but do not cite evidence of those scholars and their heated discussions.

Response 3:

We appreciate the comments of the reviewers. We apologize for exaggerating current research on the mechanism that correlates social capital with individual health. We have corrected the previous statement. Please see line 124-125.

We have combed through related literature and found that there are not enough articles on the mechanism of how social capital affects health. We found four articles which propose the mechanism that connect social capital with individual health. Please see references 29,30,31,32. We think that instead of finding their argument here, we found some common points of the predecessors, which can contribute to the mechanism in this study. Please see line 138-139.

Point 4: I am concerned about the validity of the structural and cognitive social capital variables.

Response 4:

We appreciate the comments of the reviewers. In the part of variable selection, we did not introduce the basis of selecting variables of structural and cognitive social capital in detail, which led to the lack of effectiveness in selecting related variables. Structural social capital refers to externally observable aspects of social organization, such as roles, rules, procedures and precedents. Cognitive social capital is more internal and subjective, referring to shared norms, values, attitudes and beliefs. We have added the selection basis of these two types of social capital variables in the variables section to ensure the validity of variable selection. Please see line 229-231, 242-243. The references are listed as follows:

1.Moore S, Kawachi I. Twenty years of social capital and health research: a glossary. J Epidemiol Community Health. 2017; 71(5):513-7. https://doi.org/10.1136/jech-2016-208313 PMID: 28087811

2.Putnam RD. Bowling Alone: America's Declining Social Capital. Journal of Democracy. 1995; (January): 65-78.

3.Dauner KN, Wilmot NA, Schultz JF. Investigating the temporal relationship between individual-level social capital and health in fragile families. BMC Public Health. 2015; (15): 1130. https://doi.org/10.1186/s12889-015-2437-3 PMID: 26572491

Point 5: Some information on social capital this is presented in the discussion would be better suited to the introduction.

Response 5:

We agree with the reviewer’s suggestion that some information on social capital presented in the discussion would be better suited to the introduction. We found that analysis perspectives of social capital are more suitable for the introduction. Please see line 158-166.

Point 6: Statistics should not appear in the conclusion.

Response 6:

We appreciate the comments of the reviewers. We apologize for Putting the statistics in the conclusion part of this article. We have deleted the statistics in the conclusion section, please see line 544-548.

Point 7: Policy implications for the paper are weak, for "relevant government departments should pay attention to the health status of rural elderly".

Response 7:

We agree with the comments of the reviewers. We have rewritten the policy recommendations section. It is mainly carried out from two aspects. One is that how the governments effectively manage and timely detect individual health among the elderly in rural China. The other is how the relevant departments effectively promote the increase of rural elderly's structural and cognitive social capital. In this regard, we make the following suggestions:

The health status of elderly persons in rural China is poor, and the relevant government entities should carry out regular physical examinations of these individuals and follow up their physical condition in a timely manner. In particular, the health department must appoint psychologists to implement interventions for this population so that the elderly can maintain a positive attitude. The medical department should assign more medical resources to villages to build a more comprehensive rural medical and health system, actively promote health knowledge among the elderly, and encourage the elderly to correct poor diet and hygiene habits.

More public resources should be devoted to supporting local activities and local organizations in order to encourage more elderly rural residents to join neighborhood social groups. To encourage reciprocity among neighbors, policymakers should create and develop an ethos of community support by praising and rewarding those who help their neighbors. Efforts should be made to promote a sense of belonging among neighborhood residents. By strengthening community building and valuing the importance of trust and social norms, we can make cognitive social capital play a positive role in the lives of the elderly in rural areas.

For the part we modified, please see line 552-560, 573-582.

Response to Reviewer 2 Comments

Dear Editors and Reviewers,

Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled “How does social capital affect individual health among the elderly in rural China?--mediating effect analysis of physical exercise and positive attitude” (ID: PONE-D-19-29520). Those comments are all valuable and very helpful for revising and improving our paper, and simultaneously they are of great significance to give instructions to our research. We have studied the comments carefully and made revisions which we hope to meet with approval. Revised portions are marked in red in the original paper. The main revisions on this paper, along with our responses to the reviewer’s comments, are as follows:

Point 1: Its better to explain the theoretical approach in the last part of introduction and it should lead the reader to believe that the current study is necessary.

Response 1:

We agree with the comments of the reviewers. In the previous manuscript, we failed to fully explain the theoretical method and necessity of the research in the last paragraph of introduction. In this regard, we have seriously revised and added the research significance, the analysis perspective of social capital, and the research method of this article, as detailed below:

Current studies on social capital and individual health have paid little attention to the elderly in rural China. In fact, trust and reciprocity among neighbors in rural China have an important impact on people's lives. Research on their social capital can reveal further methods to improve individual health among the elderly and make the relevant departments pay more attention to the construction of social capital in rural China. Based on the theory of social capital, this study divided social capital into structural and cognitive social capital. Current scholars typically analyze social capital from the following four perspectives: the macro-level (national, state, regional and local government); the mid-level (streets and neighborhoods); the micro-level (social networks and social participants); the individual psychological level (trust and norms). Current studies are mainly focused on the mid-level and micro-level of social capital. The structural social capital discussed in this study refers primarily to participation in social networks and social organizations, which occurs at the micro-level. The cognitive social capital mainly consists of trust, mutual benefit and mutual assistance, which belong to the individual psychological level. Ordinary least squares (OLS) and two-stage least squares (2SLS) estimators were used to analyze the association between social capital and individual health in this study. We explored the mechanism linking structural and cognitive social capital and individual health through a mediation effect analysis.

As for specific modifications, please see line 152-170.

Point 2: The methodology part should tell the reader directly whether the sample is random or not.

Response 2:

We strongly agree with the reviewer’s recommendations. We apologize for ignoring introducing the selection process of the sample in the part of data source.

CFPS followed scientific sampling methods and guaranteed the randomness of the sample. CFPS sampling used Implicit stratification, multi-stage, multi-level, probability sampling method proportional to population size (PPS). The first two phases of CFPS sampling used official administrative divisions. The third stage used maps address method to construct the end sampling frame, and the households were selected by using a circular isometric sampling method with random starting points. More detailes about sampling of CFPS see http://www.isss.pku.edu.cn/cfps/

In PLoS ONE, there is also study using CFPS data. (Chen H, Meng T. Bonding, Bridging, and Linking Social Capital and Self-Rated Health among Chinese Adults: Use of the Anchoring Vignettes Technique. PLoS One. 2015; 10(11):e0142300. https://doi.org/10.1371/journal.pone.0142300 PMID: 26569107）

In the revised draft, we clearly stated whether the sample selection was random. Please see line 186-191.

Point 3: The research has used the scales like Health evaluation index, Structural social capital, Cognitive Social Capital, positive attitude and physical exercise, the authors are recommended to conduct reliability test on these scales for the current sample and mention reliability scores.

Response 3:

Thanks to the reviewer’s comments. We apologize for not providing for reliability scores of these five variables. The selection of these five indicators is based on relevant definitions rather than scales. These indicators were obtained via exponentiation of variables (mean of zero and a standard deviation of one) and by giving them equal weight.Regarding the indexing process, we are mainly for the convenience of the intermediary analysis later. For specific methods, we have also referred to relevant literature (please see reference 34. Ho CY, Better Health With More Friends: The Role of Social Capital in Producing Health. Health Econ. 2016; 25(1):91-100. https://doi.org/10.1002/hec.3131 PMID: 25431183).

The health evaluation index is obtained by self-rated health and others-rated health through operationalization and giving them equal weight. Both self-rated health and others-rated health are measures of individual health.

The structural social capital is obtained by relationship network and organization members through operationalization and giving them equal weight. The organization members and the relationship network are selected according to the traditional definition of structural social capital (please see line 229-231).

The cognitive social capital is obtained by trust, reciprocity, and mutual assistance through operationalization and giving them equal weight. These three variables are also selected according to the traditional definition of cognitive social capital (please see line 242-243).

Positive mentality is summed up through the equal empowerment of two questions, "I am happy in life" and "I am pleasant". One of these two questions asks the mentality in life, the other is the current subjective feeling, which measures the positive attitude in different aspects.

Physical exercise is judged by "How often do you exercise a week". There is one indicator of physical exercise.

Because the measurement of these variables is not from a scale, but specific problems. These questions are for different aspects of the indicators. We are sorry that we could not get reliability scores.

Point 4: Authors mentioned that one question was used to measure physical exercise but its not clear which type of scale was used and what was the standard of good/poor physical exercise.

Response 4:

“Physical exercise”is judged by "How often do you exercise a week". This question is asking how many times the elderly exercise a week in rural China. But the variable does not measure good or bad of the elderly’s exercise. It is difficult to judge the standard of physical activity among the elderly in rural China. The elderly may exercise a lot, but they are doing farm work and may not have good health. So this question can only objectively reflect the number of times that the elderly exercise and can not provide measure standard. As for how it relates to health, it needs to be objectively put into the model for further investigation.

Point 5: The authors analyzed the mediating effect but the model used for mediating effect is not properly described. It is advised to make it clear that which model was used.

Response 5:

We agree with the reviewer’s suggestion and we added analysis method of mediation effect in the part of model introduction. This method mainly refers to the practice of Baron and Kenny (Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol. 1986；51(6):1173-82. https://doi.org/10.1037//0022-3514.51.6.1173 PMID: 3806354). We performed stepwise regression on related variables, observes the change of coefficients, and judges whether there is a mediation effect and the size according to the significance of the variable and the size of the coefficient. The main method is shown in line 338-361.

Point 6: Its better to mention study objectives again in conclusion part.

Response 6:

We strongly agree with this comment from the reviewers. In the conclusion part of this study, we have proposed study objectives, mainly to further subdivide Chinese rural elderly, such as examining the different impact of social capital on the health of left-behind and non-left-behind elderly in rural China, etc. And we will continue to explore the mediating variables that social capital affects the health of the elderly. Explained as follows:

Further analyses are needed to identify which particular elements of social capital may be driving the results and the effect size. The elderly in rural China are highly heterogenous, and there are different relationships between social capital and individual health in different populations of old people. For example, left-behind old adults and others have different mental health needs, and male and female elderly persons have differential structural social capital. The impact of social capital on the health of different groups among the elderly is worth further exploration. Future research should also explore other mediating variables in addition to physical exercise and positive attitude that contribute to the mechanism linking social capital and individual health.

The specific content of the modification, please see line 573-582.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

25 Feb 2020

PONE-D-19-29520R1

How does social capital affect individual health among the elderly in rural China?--mediating effect analysis of physical exercise and positive attitude

PLOS ONE

Dear Zhang,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

Please address the minor issues raised by reviewer 3 regarding:

the relationship between social capital and health outcomes

greater detail about the population under study

==============================

We would appreciate receiving your revised manuscript by 24 March 2020. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Rosemary Frey

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Yes

Reviewer #3: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

Reviewer #3: I Don't Know

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: the authors have addressed all the questions and suggestions. and it is suggested to accept the manuscript.

Reviewer #3: This submission is a review from a original previous submission on social capital and depressive syntoms in Chinese elderly which I had not reviewed.

I have carefully read the R1 version of the paper as well as the comments by the previous reviewers and the authors answers to them. The vast majority of their comments have been addressed and my only suggestion would be to be more specific in the section describing the evidence on the relationship between social capital in its structural and cognitive dimensions and different health outcomes. This part would benefit greatly if the authors could specify the population in which the studies had been conducted - epsecially taking into account that they argue that it has been little explored in elderly popultion from China.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: Yes: Azam Tariq, Department of Sociology, College of Humanities and Social Sciences, Huazhong Agricultural University,

Wuhan 430070, China; azam_tariq@webmail.hzau.edu.cn

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

13 Mar 2020

Response to Reviewers

Dear Dr./Prof. Frey,

On behalf of my co-authors, we thank you very much for giving us an opportunity to revise our manuscript, and we also appreciate the positive and constructive comments and suggestions, given by you and other reviewers, on our manuscript entitled “How does social capital affect individual health among the elderly in rural China?--mediating effect analysis of physical exercise and positive attitude” (ID:PONE-D-19-29520).

Having studied the reviewer’s comments carefully, we have considered your suggestions and comments, trying our best to make revisions which were marked in red in this paper. Please find the revised version enclosed, which we would like to submit for your kind consideration.

Point 1:Please address the minor issues raised by reviewer 3 regarding:the relationship between social capital and health outcomes.

Response 1:

We appreciate the comments of the reviewers. In our previous manuscript, we only discussed the relationship between structural and cognitive social capital during the literature review and data analysis process. In the discussion section of this revised manuscript, based on the results of empirical analysis, we conducted a more detailed discussion on the relationship between structural and cognitive social capital and health.

Firstly, we find that both structural social capital and cognitive social capital have a significant promotion effect on health, and the coefficients were 0.062 (p<0.01) and 0.097 (p<0.001), respectively. We explain the role of structural and cognitive social capital in promoting health (please see line 475-493).

Secondly, we find that the structural social capital's promotion of individual health among rural elderly is less than the cognitive social capital. We have explored this phenomenon in more depth. It may be because rural elderly rely more on mutual help and interpersonal trust in the community. They have more cognitive social capital than structural social capital, which has led to the different promotion of two types of social capital on individual health (please see line 496-509). We also find this difference exist in research from other regions. We analyzed our finding and results of other scholars, summarizing the reason. Rural elderly tend to develop cognitive social capital from community norm and trust, and they do not have a number of recreation facilities (e.g., exercise facilities, room for card games, etc) and organizations (e.g., elderly association, activity center for the elderly, etc) to get structural social capital (please see line 510-523).

Finally, we explore how two aspects of social capital connect with individual health (please see line 526-547). We found positive mentality and physical activity both play a mediating role on the relationship between social capital and health based on previous empirical research, but their mechanisms are different. This is probably because structural and cognitive social capital connect with health through different ways, and they affect different aspects of health. Structural social capital mainly promotes health through team members' access to health information and imitating health behaviors. Cognitive mainly relieve individual stress and regulate mindset to promote health (please see line 548-570)

We hope to discuss the above three parts to further analyze the relationship between structural social capital and individual health.

Point 2:Please address the minor issues raised by reviewer 3 regarding: greater detail about the population under study

Response 2:

Thanks a lot for the editor's suggestion. We apologize for not give detail introduction about the elderly in rural China. Older adults in rural areas constitute the majority of elderly persons in China, and a large number of them do not have much financial income, do not get adequate medical and health services, and often take care of themselves. The main problems rural elderly face are economic poverty, poor physical health, mental loneliness and etc. With the aging of rural areas, the health problem of rural elderly is more prominent than that of urban elderly, which need significant concern of the whole society (please see line 54-65).

The living environment of the elderly in rural China is special. The culture of rural China values trust, mutual assistance and reciprocal exchange, which provide cultural soil for cultivating social capital. Rural residents tend to be more altruistic, honest, and trusting of others, and they reported higher levels of civic cohesion and interpersonal trust than their urban counterparts in China. Under such circumstances, it is necessary and important to study social capital and individual health among the elderly.

We would like to express our great gratitude to you and other reviewers for the comments on our paper. Looking forward to hearing from you.

Thank you and best regards.

Yours sincerely,

Zhang Yue

Corresponding author:

Name: Zhang Yue

E-mail: yuezhang.znufe@163.com

Response to Reviewer 3 Comments

Dear Reviewers,

Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled “How does social capital affect individual health among the elderly in rural China?--mediating effect analysis of physical exercise and positive attitude” (ID: PONE-D-19-29520). Those comments are all valuable and very helpful for revising and improving our paper, and simultaneously they are of great significance to give instructions to our research. We have studied the comments carefully and made revisions which we hope to meet with approval. Revised portions are marked in red in the original paper. The main revisions on this paper, along with our responses to the reviewer’s comments, are as follows:

Point 1:More specific in the section describing the evidence on the relationship between social capital in its structural and cognitive dimensions and different health outcomes.

Response 1:

We appreciate the comments of the reviewer. In previous manuscripts, we only introduced the relationship between social capital and health in the literature review and empirical research sections. We apologize for ignoring detailed explaining the relationship between social capital and individual health. We have improved the manuscript and focused on the discussion to further explain the relationship between social capital and individual health. Our analysis follows the following lines: Is there a relationship between structural social capital and individual health? What are the coefficients of structural and cognitive social capital on individual health? How do structural and cognitive social capital connect with individual health?

Firstly, we find that both structural social capital and cognitive social capital have a significant promotion effect on health, and the coefficients were 0.062 (p<0.01) and 0.097 (p<0.001), respectively. We explain the promotion of structural social capital. People who have structural social capital could get available public spaces and access to mutual support, and rapid diffusion of health information and healthy norms of behavior through their clubs and associations (please see line 475-487). We also explain the role of cognitive social capital in promoting health. The impact of cognitive social capital is mainly psychological support through interpersonal trust and mutual assistance, which generally predicts good self-rated health (please see line 485-493).

Secondly, we explained why the coefficient of cognitive social capital on individual health is larger than structural social capital. The elderly in rural China participated in far fewer social organizations than old adults in urban China. Trust and mutual assistance between neighbors are common in the social life of rural elderly, and their perception of neighborhood relations is stronger than that of organizational participation (please see line 496-509). We also find this difference exist in research from other regions and summarized the reason: rural elderly tend to develop cognitive social capital from community norm and trust, and they do not have a number of recreation facilities (e.g., exercise facilities, room for card games, etc) and organizations (e.g., elderly association, activity center for the elderly, etc) to get structural social capital (please see line 510-523).

Finally, we explored that how structural and cognitive social capital connect with individual health. We focused on two different ways in which social capital affects health. Structural social capital mainly provides the dissemination of health information and the imitation of health behaviors for the elderly, which demonstrated the mediating effect of physical exercise on the relationship between structural social capital and individual health was greater than positive attitude. Structural social capital is mainly beneficial to physical health, and cognitive social capital play a positive role in mental health. Cognitive mainly relieve individual stress and regulate mindset to promote health, which demonstrated the mediating effect of positive attitude on the relationship between cognitive social capital and individual health was larger than physical activity (please see line 548-570).

We hope to explain the different relationship between structural and cognitive social capital and individual health through discussion of these three aspects above.

Point 2:This part would benefit greatly if the authors could specify the population in which the studies had been conducted - epsecially taking into account that they argue that it has been little explored in elderly popultion from China.

Response 2:

We agree with the reviewer’s suggestion. In previous manuscripts, we did not give a detailed introduction to the group of elderly people in rural China. In fact, the health problems of this group are worthy of attention (please see line 53-65). We introduce this group as follows:

Older adults in rural areas constitute the majority of elderly persons in China, and a large number of them live alone at home because their family numbers go out to work in cities with the acceleration of urbanization. These old people do not have much financial income, do not get adequate medical and health services, and often take care of themselves. The main problems rural elderly face are economic poverty, poor physical health, mental loneliness and etc. With the aging of rural areas, the health problem of rural elderly is more prominent than that of urban elderly, which need significant concern of the whole society. Currently, Chinese government has carried out “Healthy China” strategy to meet the challenge of aging. Under such circumstances, knowledge of social determinants of healthy aging are crucial for the development of evidence-based policies and interventions and the sustainable development of Chinese society.

The environment in which they live is rich in social capital, and social capital is closely related to the lives of these elderly people. We explain the importance of studying the health and social capital issues of this group as follows :

China is a typical Guanxi-based society, and evidence shows Guanxi and social capital have similar connotations and effects. Chinese tend to seeking for social support and maintain social status in the social structure in which they live. In addition, the culture of rural China values trust, mutual assistance and reciprocal exchange, which provide cultural soil for cultivating social capital. Rural residents tend to be more altruistic, honest, and trusting of others, and they reported higher levels of civic cohesion and interpersonal trust than their urban counterparts in China. Some relevant departments do not provide sufficient or formal credit systems for rural elderly, and the elderly often rely more heavily on the development of social capital in daily life.

For more details, please see line 68-81.

We hope to explore the association between social capital and individual health among the elderly in rural area and make a reference for the implementation of the “Healthy China” strategy.

Thanks again for the reviewers' comments and suggestions!

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

23 Mar 2020

How does social capital affect individual health among the elderly in rural China?--mediating effect analysis of physical exercise and positive attitude

PONE-D-19-29520R2

Dear Dr. Yue,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Rosemary Frey

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #3: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #3: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #3: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #3: (No Response)

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #3: I commend the authors for their throughness in addressing my comments and suggestions, which have been all considered. I would only recommend a brief explanation of what a "guanxi" society means, as not all readers may be familiar with this term.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #3: No

14 Jul 2020

PONE-D-19-29520R2

How does social capital affect individual health among the elderly in rural China?--mediating effect analysis of physical exercise and positive attitude

Dear Dr. Yue:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Rosemary Frey

Academic Editor

PLOS ONE