A study of the relationship between mutual aid and the mental health of older people

Abstract

Background

Focusing on the mental health of older people is of great significance to the implementation of a healthy China.

Methods

Based on individual-level tracking data from China (CHARLS) in 2013, 2015, 2018, and 2020, a quantitative study was conducted on the implementation of the mutual aid for older people policy and the mental health of older people using the DID double difference model.

Results

The study demonstrated that the implementation of mutual aid for older people care policies is significantly positively correlated with improvements in older people depression and mental health levels (p <0.01), with an estimated effect size of approximately 3.87%. Heterogeneity analysis revealed that this positive association varied across different groups: the strength of the correlation was higher among older people with disabilities than among healthy older people, and the association was stronger among males than among females.

Conclusions

The government should strengthen the top-level design of mutual aid-based older people care; At the social level, diverse resources should be connected to build a mutual aid platform for older people care; Community organizations should innovate service models and improve space creation; Family unit should strengthen internal support and improve emotional communication.

According to the World Health Organization (WHO), by 2050, the global population over 60 years of age is expected to reach 2.1 billion, accounting for 22% of the total global population. Against this backdrop, safeguarding the quality of life of older people, especially their mental health, has become a social issue that needs to be addressed urgently. Global studies indicate that 13.3% of adults aged ≥ 60 years experience depressive disorders, with prevalence reaching 25.6% in China and showing an increasing trend [1]. Mental health problems among older people are often closely related to insufficient social support and reduced socialization [2]. Social support is defined as the multidimensional assistance encompassing material, emotional, and informational resources that individuals obtain through their social networks (including but not limited to family members, peers, and community connections), which serves to mitigate life stressors and enhance mental health and quality of life. Empirical evidence demonstrates that social support plays a significant moderating role in alleviating depressive symptoms among older people [3], and reducing the level of depression occurrence.

The traditional family model of old-age care faces challenges in modern society, as children are unable to provide adequate care due to work and life pressures, resulting in some older people falling into emotional and psychological difficulties. Mutual aid has emerged as an innovative eldercare model that promotes resource sharing, emotional support, and cooperative engagement among older people. By facilitating “old-help-old” and intergenerational assistance, this approach addresses the limitations of traditional eldercare systems. Its core principle encourages older people to contribute to age-friendly services according to their abilities, fostering social integration and active participation [4]. Furthermore, mutual aid expands social support networks and enhances positive social interactions, which have been empirically linked to reduced depression risk, improved subjective well-being, and greater life satisfaction among older people [5].Mutual aid for older people has become an important form of social support by providing emotional, instrumental, and informational support and by reinforcing older people ' sense of social belonging. Therefore, this study adopts the perspective of mutual aid for older people’s care to investigate the association between the mutual aid model and the mental health status of older people. The research aims to evaluate the potential of mutual aid in alleviating depressive symptoms among older people population, address the existing research gap regarding the correlation between mutual aid and seniors’ psychological well-being, and provide an empirical foundation for formulating relevant policies and practical interventions in older people’s care services.

A review of research on the relationship between mutual aid and mental health

Among the research advances on mutual aid and the mental health of older people, Feng et al. [6] found that the mutual aid behavior of older people in the Huangshan Mutual Aid Commune had a positive effect on their mental health through the social support pathway. Deng et al. [7] used a controlled experimental method to instruct community-based older people in the study group to participate in mutual aid care, and the results showed that the anxiety and depression of older people in the study group were significantly improved. Some researchers have pointed out that a variety of social participation behaviors can increase the social capital of rural older people, whereby they can obtain a stronger sense of subjective well-being and social support and enhance their health status [8]. Huo et al. [9] analyzed data from the HRS longitudinal survey of the US Department of Health and found that formal or informal behaviors such as helping others can lead to positive perceptions of self-aging among older people, and the longer the time spent helping others, the lower the likelihood that older people will become depressed. There are also a few studies that have found that mutual aid aging behaviors, such as caring for the sick or disabled have a negative impact on the mental health status of older people [10]. Lieberman and Gourash [11] conducted a controlled study using traditional psychometrics on older people who participated in mutual aid groups versus those who did not, and through empirical analysis of the study, found that older people who participated in mutual aid for older people had higher levels of life satisfaction and better psychological well-being. José Alberto Orsi et al. [12] conducted a prospective study that assessed the clinical and rehabilitative outcomes of psychiatric patients using non-parametric tests, and the data showed that medication adherence increased more in the mutual aid group, and that continued participation in the mutual aid group was an important tool in the recovery process and in the treatment itself. Cipolletta et al. [13] conducted a controlled study of long-term caregivers and found that the establishment of mutual aid groups by caregivers facilitates the construction of new social networks and the activation of personal and social resources, which leads to increased well-being. According to activity theory, older people ‘s regular participation in social activities can improve their health, reduce the degree of depression, and lower the risk of death [14, 15]. Lee et al. [16] found that social participation has a positive effect on both physical health and mental health of older people, and that older people who participate in social activities have a higher self-assessed level of health than those who do not; some scholars believe that that older people are conducive to improving cognitive levels by participating in voluntary activities with certain cognitive requirements [17]. The more older people participate in recreational activities, the more conducive they are to reducing the incidence of disability and the risk of death [18].

Mutual aid affects the level of mental health of older people through a number of mechanisms. This mainly includes three aspects: the health dimension, life satisfaction, and the economic dimension. Firstly, in the health dimension, as mutual aid aging involves activities such as housework collaboration and group exercise, the amount of physical activity can be increased in the process of participation, while regular exercise can promote endocannabinoid secretion and improve mood, and encouraging moderate-intensity physical activity may help to protect older people from depression-related cognitive under-performance [19], thus resisting depressive mood. On the other hand, individuals are more likely to receive preventive health services from time banks by participating in them [20]. Secondly, in terms of life satisfaction, according to role theory, the loss of social roles may lead to psychological problems. Through mutual aid activities, older people are provided with better opportunities to contact social networks [21]. And older people act as “helpers” in mutual aid, which helps them to make friends through serving others and makes them more active [22, 23], and in this process, older people’s self-worth is rebuilt, and older people regain their role identity, thus improving self-esteem and efficacy and reducing loneliness. Role identity, thus enhancing self-esteem and sense of efficacy, which in turn reduces loneliness. Again, in the economic dimension, financial difficulties are an important trigger of depression in older people [24], and the mutual aid model reduces the financial burden through resource sharing, indirectly alleviating the anxiety and depression caused by financial problems in older people.

Research innovations

To summarize, the previous studies targeted a specific population and used cross-sectional data to analyze the effects of mutual aid on mental health; therefore, based on the previous studies, this article uses CHARLS’s 4 years of panel data, expands the population sample size, and verifies the evaluation of the effects of implementing mutual aid in China through the empirical model of the DID study to provide suggestions for the implementation of mutual aid policies.

Objects and methods

Data sources

Microdata from Peking University’s China Health and Aging Tracking Survey (CHARLS) for 4 periods in 2013, 2015, 2018, and 2020 were used to assess the health effects of the implementation of the mutual aid for older people policy in China. The data used random stratified sampling to survey the middle-aged and older people population aged 45 years and older in 450 villages (communities) in 150 counties across 28 provinces (autonomous regions and municipalities directly under the central government). In this paper, older people aged 60 and above were selected as the sample size, and the final sample data of 69,030 were obtained after the missing values were supplemented by the method of multiple interpolation borrowed from Pang Xin Sheng [25].

Statistical methods

The CHARLS database was screened and organized using StataMP17 with a test criterion of α = 0.05. The data were analyzed using the DID double difference method, and the data were further validated and analyzed using the PSM propensity score matching method to prevent multiple influences on the data, such as the development of productivity and the gradual increase in income.

Description of variables

Explained variables

The dependent variable in this study is the mental health of older people. The concept of “mental health” originated in the early 20th century and was defined as “the absence of physical and mental illness and the ability to adapt happily to the environment.” With the development of the times, this concept has gradually emphasized the ability of individuals to achieve self-actualization and the release of their potential [26]. There are various indicators for measuring the mental health of older people, including but not limited to assessing cognitive function, loneliness, social skills, and depression. Among these, depression is the most common mental health issue among older people in their later years [27]. The global prevalence of depression among older people ranges from 10% to 20% [28], while the proportion of Chinese older people in a depressive state reaches as high as 19.05% [29]. Currently, the academic community unanimously recognizes depression levels as a key indicator for measuring mental health status, widely applied in various empirical studies [30]. Therefore, this article draws on existing research [31] and uses the 10-item abbreviated version of the Center for Epidemiologic Studies Depression Scale (CES-D) from the CHARLS questionnaire to measure the mental health status of older people in China [32].The scale had 8 negatively oriented questions and 2 positively oriented questions. All 10 questions of the scale were scored on a 4-point scale (rarely or not at all = 0, not too much = 1, sometimes or half the time = 2, most of the time = 3), and the two positive mood questions were reverse scored, and then the scores of the 10 questions were summed to construct a composite index of depressed mood in older people, which was scored in the range of 0 to 30 points. Defining a score of 10 or less as no depressive mood and 10 and above as having depressive mood, the higher the score, the more serious the degree of depression, and the lower the level of mental health of older people [33].

Core explanatory variables

The explanatory variable in this paper is mutual policy implementation, measured using the cross-multiplier term of the mutual pension policy implementation time dummy variable and the group dummy variable. The specific settings are as follows: The explanatory variable in this paper is mutual policy implementation, measured using the cross-multiplier term of the mutual pension policy implementation time dummy variable and the group dummy variable. The specific settings are as follows: (1) Time virtual variable settings: Mutual aid for older people refers to a community or village-based approach to old age in which low-collar older people, volunteers, or neighbors provide basic care, companionship, and living assistance to older people or older people with disabilities through time exchange and resource sharing. Implementation is coordinated by community neighborhood committees, village committees, or social organizations, which form twinning support groups to provide services such as meal delivery, cleaning, and shopping on behalf of older people. The frequency of services is arranged 1-3 times a week according to the proportion of communities or villages participating in community mutual assistance for older people.

The term “mutual aid for older people " officially appeared in the 2018 State Council government work report, which marks the rise of mutual aid for older people from a local practice to a national strategy [34]. This is a breakthrough in China’s pension policy. Therefore, this study sets 2018 as the policy time point, with years 2018 and thereafter set to 1 and previous years set to 0.

(2) The assignment method for group-level dummy variables is as follows: Mutual aid for older people is closely related to social support. The role of social support lies in maintaining individuals’ positive psychological and emotional experiences, thereby enabling the older people to have both the willingness and ability to participate in mutual aid for older people [35]. Different scholars define the providers of social support as “significant others such as family members, friends, colleagues, relatives, and neighbors [36]” or “social networks” [37]. Therefore, based on the above discussion and considering the availability of indicators in the CHARLS database, proxy indicators for mutual aid in old age were selected. The corresponding question in the questionnaire is “Have you engaged in any of the following social activities in the past month?” The five options are “Visiting neighbors, socializing with friends,” “Playing mahjong, chess, cards, or attending community activity rooms,” “Providing assistance to relatives, friends, or neighbors who do not live with you,” “Participating in volunteer activities,” and “Caring for sick or disabled individuals who do not live with you.” If any of the five mutual aid behaviors were participated in, it was assigned a value of 1; if not, it was assigned a value of 0. Respondents who participated in mutual aid for older people were assigned a value of 1, and those who did not were assigned a value of 0.

Control variables

This paper uses six variables as control variables: age, gender, marital status, self-care ability, pension status, life satisfaction, and health satisfaction. (1) age: the year of birth of the respondent is taken as the starting point, and the year of the survey is the cutoff date, and the age of older people is divided into three groups: 60–69 years old, 70–79 years old, and ≥ 80 years old; (2) gender: males are assigned a value of 0, and females are assigned a value of 1; (3) marital status: the definition of married with a partner spouse is defined as married with the value of 0, and the definition of the other conditions is defined as unmarried with the value of 1; (4) self-care ability. The Autonomy of Daily Living (ADL) scale was used as a measure of the physical health of older people. The scale includes six items: dressing, bathing, eating, getting in and out of bed, using the toilet, and controlling bowel movements. Each item is divided into four levels based on the degree of self-care (0 = completely self-sufficient, 1 = has difficulty but can complete independently, 2 = needs assistance, 3 = completely unable to care for oneself). A total score > 0 is defined as an older people with disabilities, and a score of 0 is defined as a healthy older people person; (5) Pension status: Using the question “Do you have a pension?” in the questionnaire, assign a value of 1 to those who do not have a pension and a value of 0 to those who do; (6) Life satisfaction: “Are you satisfied with your life?”. Satisfaction decreases from 1 to 5; the level of satisfaction gradually decreased from 1 to 5. For the purpose of the study, life satisfaction was regrouped and assigned a value, extremely satisfied and very satisfied were divided into satisfied and assigned a value of 1, more satisfied was divided into average and assigned a value of 2, and the rest of the cases were assigned a value of 3; (7) Health satisfaction: “Are you satisfied with your own health?”, the degree of satisfaction gradually decreases from 1 to 5. For the study, life satisfaction was regrouped and assigned a value; extremely satisfied and very satisfied were categorized as satisfied and assigned a value of 1, more satisfied were categorized as average and assigned a value of 2, and the rest of the cases were assigned a value of 3. Specific variables are detailed in Table 1.

Table 1.

Descriptive statistics of variables

Variable Name		Assignment of values	Average value	Standard deviation
Mental health	Not in a depressive mood	0	0.143	0.350
	Depressed mood	1
Involvement in mutual aid	Not Engaged	0	0.412	0.492
	Involved	1
Age	60–69 years old	1	2.173	0.888
	70-79years old	2
	80 years old and above	3
Gender	Male	0	0.509	0.5
	Female	1
Marital status	Married	0	0.157	0.364
	Unmarried	1
Self-care ability	Healthy	0	0.241	0.428
	Disabled	1
Life Satisfaction	Satisfactory	1	1.703	0.741
	Average	2
	Unsatisfactory	3
Health Satisfaction	Satisfactory	1	1.770	0.675
	Average	2
	Dissatisfied	3
Pension Status	Pension	0	0.143	0.35
	No pension	1

Empirical results

Endogeneity test

Multicollinearity test

Before carrying out the empirical analysis of PSM-DID, in order to prevent the existence of serious multicollinearity problems among the variables, which will have an impact on the regression results of this paper, this paper carried out the Variance Inflation Factor (VIF) test for each variable, and the test results are shown in Table 2. The results show that the VIF values of all variables are below 2 (with a maximum value of 1.16), which is far below the standard threshold of 10. This result indicates that there is no serious multicollinearity problem among the explanatory variables in the research model, thereby ensuring the reliability of the subsequent hypothesis testing results.

Table 2.

Multiple covariance test

Variable Name	VIF	1/VIF
Self-care ability	1.16	0.86
Health Satisfaction	1.13	0.88
Life Satisfaction	1.13	0.88
Age	1.11	0.90
Mental health	1.08	0.92
Marital status	1.07	0.93
Gender	1.04	0.96
Pension Status	1.02	0.97
Mean VIF	1.09

Two-stage regression

This study employed a two-stage regression method to examine the impact of mutual aid for older people on the mental health. As shown in Table 3, Step 1 constructed a basic regression model using a time-individual two-way fixed effects estimation method, with the mental health status of older people as the dependent variable and participation in mutual aid for older people as the core explanatory variable, without including other control variables. The regression results indicate that participation in mutual aid for older people has a positive and significant impact on improving older people’s depressive symptoms and enhancing their mental health levels (B = −0.024, p < 0.01).In Step2, a series of control variables were added, including age, gender, marital status, self-care ability, pension status, health satisfaction, and life satisfaction, and regression analysis was conducted using time-individual two-way fixed effects, and the results showed that mutual aid still significantly improved the depression status of Chinese older people at the level of P < 0.01. In addition, research data shows that participation in mutual aid for older people is significantly negatively correlated with depressive tendencies among older people population in China. With increasing age (B=−0.021, p < 0.01), increasing self-care ability (B=−0.134, p < 0.01), increasing health satisfaction (B = 0.032, p < 0.01), and enhanced life satisfaction (β = 0.008, p < 0.01), the incidence of their depressive symptoms showed a significant downward trend. This may be because mutual aid for older people care has a distinct social attribute. Through geographical ties, participating older people are able to break out of social isolation and reestablish emotional connections with each other. The participation of older people, who have strong self-care abilities, high life satisfaction, and high health satisfaction, in mutual aid programs enhances their sense of happiness. This occurs both when they actively provide care services and when they receive them. Such participation not only fulfills their desire for self-actualization but also changes the situation of feeling idle in old age. It enriches spiritual interactions among older people and effectively reduces depressive symptoms. In essence, mutual aid for older people establishes a benign psychological-social-physical cycle. Participation in mutual aid has a significant negative effect on the mental health of Chinese older people with pension status, unmarried marital status, and female gender.

Table 3.

Regression results

		Time-individual fixed effects
		P	B	Std. err	95%CI	R2
step1	Involvement in mutual aid	< 0.01	−0.024	0.004	[−0.03 ~−0.01]	0.22
Step2	Involvement in mutual aid	< 0.01	−0.021	0.004	[−0.03 ~−0.01]	0.23
	Age	<0.05	−0.042	0.003	[−0.02 ~−0.03]
	Gender	0.75	0.004	0.013	[−0.02 ~ 0.06]
	Marital status	< 0.01	0.040	0.009	[0.02 ~−0.06]
	Self-care ability	< 0.01	−0.134	0.004	[−0.02 ~−0.04]
	Pension Status	< 0.05	0.012	0.005	[0.01 ~ 0.02]
	Life Satisfaction	< 0.01	0.008	0.002	[0.03 ~ 0.01]
	Health Satisfaction	< 0.01	0.032	0.002	[0.02 ~−0.03]

The results of Step 1 and Step 2 indicate that, regardless of whether the analysis is based on benchmark regression or includes control variables, the implementation of mutual aid for older people is significantly associated with improvements in depression among older people. Specifically, step 2 shows that after the implementation of the mutual aid policy, depression among older people in China significantly improves their mental health by about 2.1%, and the estimation is significant at the 1% level.

Model construction

Difference-in-Difference (DID) modeling

The difference-in-difference (DID) method is a natural experimental method that is widely used in the evaluation of policy effects. The DID method can control for differences between the treatment and control groups before and after the implementation of the individual-level mutual aid pension policy, as well as other systematic differences. Therefore, this study is divided into two steps: in the first step, the data are matched for PSM propensity scores, and in the second step, the matched samples are analyzed by DiD regression.

Propensity score estimation: first estimate the probability (propensity score) that an individual will receive the treatment (policy intervention) and match the control group based on the score, modeled as follows: , where = 1 indicates that individual belongs to the treatment group, i.e., subject to policy intervention; represents control variables (characteristics that influence whether treatment is received, which in this study refer to gender, age, marital status, pension status, life satisfaction, and health satisfaction); represents the propensity score of individual , i.e., the probability of receiving treatment.

DID double difference model: Construct a double difference model to estimate the policy effect on the matched sample.

is the dependent variable in this paper; Treat is a treatment group dummy variable, taking the value of 1 whenis in the experimental group (participating in mutual aid for older people) and 0 when it is in the control group (not participating in mutual aid for older people); is a treatment period dummy variable, the time point of the policy implementation in this paper is 2018, so that is in the 2018 pre-treatment period taking the value of 0, and in the 2018 post-treatment period taking the value of 1; is the interaction term of the policy and time intervention, and its coefficient is the effect value of this paper’s concern. After taking the value of 1, is the interaction term between policy and time intervention, and its coefficient is the effect value of this paper’s concern.

PSM test

Since the test for the presence of selective bias in the sample and to prevent endogeneity, the PSM statistic was chosen to be used for the test. PSM was first proposed by Rosenbaum and Rubin (1983) and is mainly used to reduce the effects of data bias and confounding variables in observational studies and to make the experimental group treatment effects more pronounced. The basic principle of PSM: it is assumed that individuals choose whether to participate in a program solely based on the observable characteristics to choose whether or not to participate in a program, and that unobservable characteristics do not influence whether or not an individual participates in a program [12]. Matching between the control group and the treatment group based on propensity scores is used to solve the sample self-selectivity problem by matching individuals in the treatment group with individuals in the control group who have the highest possible degree of similarity.

In this paper, we use nearest neighbor matching (1:1), whose standard deviation and deduction error after matching can be reflected in the following table. In general, as can be seen from the table, the standard error after matching is smaller and the matching effect is better. Except for the standard deviation of the two indicators of marital status and health satisfaction, which are increasing, the standard deviation of all other indicators is less than 10%. After matching, the difference between the experimental group and the control group is statistically significant. At the same time, the t-test results show that the samples after matching of the two groups are no longer significantly different on each control variable, thus removing part of the inherent problems within the study sample. The results are shown in Table 4.

Table 4.

Propensity score matching (PSM) propensity score matching results

Variable name		Mean Value		Standard Error	Error Reduction	T-Test
		Processing group	Control group			T-value	P-value
Sex	Before Matching	0.5204	0.5003	4.0	94.0	5.2	0.01
	After Match	0.5204	0.5216	−0.2		−0.29	0.775
Age	Pre-Match	2.1614	2.1816	−2.3	92.3	−2.93	0.003
	Match After	2.1614	2.1599	0.2		0.21	0.836
Marital status	Match Before	0.1555	0.15766	−0.6	−73.9	−0.77	0.441
	Match After	0.1555	0.15174	1		1.24	0.213
Pension status	Match Before	0.13943	0.14495	−1.6	30	−2.04	0.041
	Match After	0.13943	0.1433	−1.1		−1.32	0.185
Life Satisfaction	Match Before	1.8119	1.7413	10.5	96.4	13.53	0.01
	Match After	1.8118	1.8144	−0.4		−0.46	0.644
Health Satisfaction	Match Before	1.7018	1.71036	−0.2	−409.5	−0.31	0.754
	Match After	1.7018	1.7109	−1.2		−1.48	0.138

PSM-DID benchmark regression results

In accordance with the sample selection and empirical methods described in the previous section, this part uses CHARLS data from 2013, 2015, 2018, and 2020 to identify the effect of the implementation of mutual aid on the depression level of Chinese older people according to the formula. The results are shown in Table 5.

Table 5.

PSM-DID benchmark regression results

	(1)
Variable name	mental health
did	−0.021*** (0.004)
Age	−0.042*** (0.0031)
Gender	0.004 (0.013)
Marital status	0.040*** (0.009)
Self-care ability	−0.013*** (0.004)
Pension status	0.012** (0.005)
Health Satisfaction	0.008*** (0.002)
Life Satisfaction	0.032*** (0.002)
time-fixed effect	yes
Individual-fixed effect	yes
N	69,030
R2	0.222

注：*** p<0.01, ** p<0.05, * p<0.1

Table 5 demonstrates the double-difference effect of the effect of the implementation of mutual aid pension policy on the level of depression among Chinese older people, it can be seen that age, marital status, self-care ability, health satisfaction, life satisfaction, and pension status are statistically significant, the first five are significant at the 1% level, and the latter is significant at the 5% level. This indicates that the implementation of mutual aid older people’s care policies is associated with a reduction in depression among older people and an improvement in their mental health, with a DID result coefficient of −0.021.

Heterogeneity test

Heterogeneity test of self-care ability

There may be variability in the impact of the mutual aid older people policy on the mental health level of different Chinese residents, and there is an important dynamic bidirectional effect between physical health and mental health [38]. It has been pointed out that there is a significant positive correlation between the degree of incapacity and the current depression in older people. Therefore, in this paper, we will analyze the heterogeneity in terms of self-care ability, and do time-fixed effects group regression with self-care ability grouping to examine the effect of different self-care abilities and mutual aid on the mental health of older people. According to the self-care ability scale, older people are categorized as either disabled or healthy. The coefficient of did in Eq. (1) is −0.0153, which is significantly negative at the 1% level, controlling for other relevant variables, and time- fixed effects. The coefficient of did in Eq. (2) is −0.0594, which is significantly negative at the 1% level. It indicates that, regardless of their level of self-care ability, participation in mutual aid is associated with a reduced likelihood of depressive symptoms and improved mental health among older people. However, comparing the regression coefficients of the cross term did in Eq. (1) and Eq. (2), it can be found that the implementation of mutual aid for older people has a stronger effect on the enhancement of the mental health level of older people with disabilities. The results are shown in Table 6.

Table 6.

Heterogeneity test of the self-care ability of older people

	Mental Health
	(1) Health	(2) Disability
did	−0.0153*** (−3.42)	−0.0594*** (−4.01)
Age	−0.0292 (−9.1)	−0.0449*** (−3.88)
Sex	0.0110 (0.84)	0.0383 (0.56)
Marital Status	0.0396*** (3.66)	0.0232 (0.85)
Life Satisfaction	0.0181***	0.0476***
Health Satisfaction	0.00546* (1.95)	0.00689 (0.96)
Pension Status	0.0000945 (0.02)	0.0506 (2.13)
Time-fixed effect	yes	yes
Individual-fixed effect	yes	yes
N	69,030	69,030
R2	0.453	0.607

注：*** p<0.01, ** p<0.05, * p<0.1

Analysis of gender heterogeneity

After grouping older people by gender and controlling for other relevant variables, the data were analyzed using the time-individual double fixed-effects method, resulting in a coefficient of −0.0301 for did in Eq. (1), which is significantly negative at the 1% level. The coefficient of did in Eq. (2) is −0.0136, which is significantly negative at the 5% level. It indicates that, regardless of gender among the older people, participation in mutual aid is associated with a reduced likelihood of depressive symptoms and improved mental health among older people. However, comparing the regression coefficients of the cross-term in Eqs. (1) and (2), we find that mutual aid programs have a stronger positive effect on mental health among older males than among older females. The results are shown in Table 7.

Table 7.

Analysis of gender heterogeneity

Variable Name	Mental Health
	(1)Males	(2)Females
did	−0.0301*** (−5.04)	−0.0136** (−2.09)
Age	−0.0455*** (−10.95)	−0.0376*** (−7.76)
Self-care ability	−0.0186** (−2.48)	−0.0158** (−2.46)
Marital Status	0.0223 (1.48)	0.0503*** (3.96)
Life Satisfaction	0.0311*** (7.53)	0.0344*** (8.64)
Health Satisfaction	0.0000441 (0.01)	0.0120*** (3.49)
Pension Status	0.00870 (1.25)	0.000103 (0.01)
time-fixed effect	yes	yes
Individual-fixed effect	yes	yes
N	69,030	69,030
R2	0.428	0.477

* p<0.1, ** p<0.05, *** p<0.01

robustness check

Placebo test

Considering that the occurrence of non-accidental events may lead to biased estimation results, this study proposes to eliminate the effect of small probability events by using a placebo test with random sampling at different time points. The placebo test was conducted by randomly selecting the treatment group with 500 cycles, and the results are shown in Fig. The β-value is −0.024, the mean of the coefficients is very close to 0, and the probability of the simulated coefficients exceeding the true coefficients is less than 1%, which further confirms the reliability of the estimation results in this paper. The results are shown in Fig. 1.

Fig. 1.

Placebo test

Parallel trend test

The important premise assumption for the validity of the double-difference method is the convergence assumption. In order to verify the reliability of the DID estimation results, the key assumption of the DID model is that the control group and the treatment group satisfy the parallel trend assumption, i.e., in the absence of the implementation of the mutual aid old-age pension policy, the mental health level of the Chinese residents in the control group and the treatment group have the same trend of change. This paper draws on the practice of using event analysis to test for parallel trends. Specifically, the following estimation equation is used:

where represents a series of policy dummy variables for the two years before, the year of, and the year after the implementation of the mutual pension policy. Where the first two years of policy implementation are denoted in Fig. 2 as before 1, before 2, the year of policy implementation is denoted as current, and 1 year after policy implementation is denoted as after1.

Fig. 2.

Parallel trend test

Conclusion

This study utilized a sample of 69,030 adults aged 60 and above from the 2013–2020 waves of the China Health and Retirement Longitudinal Study (CHARLS) to examine the relationship between mutual aid policies and older people’s mental health. The following conclusions were obtained: First, research has demonstrated that the implementation of mutual aid for older people’s policies is associated with improvements in older people’s depressive moods and psychological well-being. Secondly, Heterogeneity analysis shows that participants of mutual aid for older people policies generally report good psychological well-being. This positive association is more pronounced among older people with disabilities than among their healthy counterparts and among males than females.

Discussion

Mutual aid can significantly improve the mental health of older people

The findings of this study indicate that engagement in mutual aid activities is associated with significant improvements in mental health and a reduced risk of depression among older people, which is consistent with the results reported by Jiang et al. [39] and Zhao et al. [40]. The possible reason for this is that mutual aid is a way for older people to support each other and help each other through participation in social activities, to bring their own strength into play, to satisfy themselves by serving others, and to achieve the pursuit of self-fulfillment [41]. There is a positive correlation between loneliness and depression, and loneliness is the influencing factor that has the greatest impact on the experience of depression [42]. And older people in the process of participating in mutual aid, implementation of older people services to expand the size of their social network,, some studies have shown that the increase in the size of the social network can significantly reduce the degree of depression in older people [43], On the other hand, mutual aid is a comprehensive pension model, including daily care services, medical security services, spiritual comfort services and legal support services, etc., the core of which lies in the integration of government, social and family resources to form a sustainable network, and with the improvement of the national policy, the living environment of mutual aid has been significantly improved, with an emphasis on meeting the needs of groups with different interests, for example, square dance venues, fitness trails, etc. Multi-functional activity areas, older people through the participation of mutual aid for older people, increased opportunities for physical exercise for older people, while some studies have found that physical exercise can reduce the symptoms of depression in older people [44]. In addition, when older people participate in mutual aid for older people, volunteering gives people a sense of meaning in life, and for older volunteers, learning and making social connections are key factors in gaining meaning [45], which can improve older people’s sense of purpose, control, and overall self-efficacy [46]. Older people’s sense of accomplishment and self-worth can effectively advance successful aging [47]. On the other hand, mutual aid rehabilitation programs serve as an innovative and effective approach in mutual aid aging services [48]. Mutual aid relationships can effectively reduce the fear of future loneliness [49]. In turn, it improves the mental health of older people.

The implementation of mutual assistance in old age has different impacts on different groups of older people

Heterogeneity analysis revealed that participation in mutual aid programs was associated with improved mental health among older people, with a stronger association observed for those with disabilities. Additionally, the mental health benefits were more pronounced among older men than among older women. Some studies have shown that the initial level of depression in older people with disabilities is significantly higher than that in older people without disability [50], and women tend to be more likely than men to suffer from the stigma of mental health problems and to be more vulnerable to inequality, discrimination, and social exclusion [51]. And thus resulting in a greater role of mutual aid in enhancing the mental health of older people with disabilities and male older people. As estrogen exerts its antidepressant effects through the 5-HT system [52], female older people individuals experience a decline in estrogen levels due to age-related factors, which affects neurotransmitters such as dopamine and leads to a decrease in mood regulation. Because of the factor of incapacity, older people’s self-activity decreases, the scope of activities is limited, in the family, they did not lose their ability before, they relieve the pressure of their children’s lives, take the initiative to assume the responsibility of family services, and when older people after the incapacitation, the role of the family changes, older people become a passive recipient of the role of the service, the emergence of a self-identity crisis so that the older people gradually lose their sense of self-worth, at the same time, coupled with the child. At the same time, coupled with children’s long-term care triggered intergenerational relationship dependence tension, further strengthening the passive dependence of older people, exacerbating the chances of depression in older people, while through the use of mutual aid in old-age care and older people services provided by mutual aid in old-age care, on the one hand, it is convenient for older people with disabilities to obtain health, life, entertainment and other services in their homes or in the community, which, to a certain extent, improves the quality of life of older people, reduces the sense of loneliness of older people, and reduces the depression occurrence. On the other hand, mutual assistance in old-age care reduces the pressure on children in the older people’s homes to take care of them and eases intergenerational conflicts; mutual assistance in old-age care replaces some of the physical labor of family members through services such as community canteens, janitorial assistance, and health management services. In addition, mutual aid has reshaped the connotation of filial piety. While traditional filial piety emphasizes “personal care”, mutual aid expands it to “supporting older people to live independently”, and they gain a sense of accomplishment in community activities, which reduces the moral pressure on their children and is more in line with the needs of modern society. Children are more proactive in communicating with each other as a result of the positive state of older people, forming a virtuous cycle in the relationship between children and their parents, which is conducive to promoting the well-being of family relations.

Recommendations

Based on the findings of this paper, this paper makes four policy points:

• First: Government organization: strengthening the top-level design of the mutual-help approach to ageing in place.

  Introducing the Regulations on the Promotion of Mutual Aid for Older People, clarifying the legal status, organizational forms, rights, and responsibilities of mutual aid for older people, and regulating the content of services and operational mechanisms. Provide operating subsidies to social organizations, communities, and individuals involved in mutual aid for older people, such as rent reductions for venues and service subsidies, and provide tax incentives for income from mutual aid for them. Raise the level of old-age pensions and ease the financial pressure on the older people. Encourage cooperation between medical institutions, social organizations, and mutual aid institutions for older people, thereby promoting greater participation in mutual aid. Summarize the excellent experiences and models that have emerged in the implementation of the mutual aid for older people’s policy and promote them nationwide.

• Second: Social organizations: linking diversified resources and building a mutual-help platform for the older people.

  Social organizations should join forces with local enterprises to provide funds, materials (such as health monitoring equipment), or door-to-door medical services, etc., so as to reduce the cost of mutual aid for older people. Cooperate with medical institutions and social work organizations to provide the older people with professional support, such as mental health management and psychological counseling, and establish a platform for hormone testing for the older female people to alleviate their emotional fluctuations caused by hormone levels. Establishing an online or offline psychological testing platform, develop a psychological testing applet for the older people, integrate the psychological needs of the older people in a supply-demand matching manner, promptly and effectively identify older people with high risks to their mental health, and arrange for guidance by exhibition personnel. For the older people with disabilities, develop a remote monitoring platform, guide young children to teach them to use smart bracelets and emergency call equipment through technological empowerment, so as to provide one-to-one care in a timely manner and improve the quality of the implementation of mutual aid.

• Third: community organization: innovating service models and improving space creation.

  When creating mutual aid and happiness homes for older people, it is necessary to provide multi-level, precise, and comprehensive care services based on their personalities, living habits, and individualized needs. This approach enables all-around support, ranging from basic daily care to spiritual comfort. Community mutual aid and happiness organizations carry out joint efforts with community medical resources, rural doctors, and other institutions to introduce customized care and medical services, develop the spirit of mutual assistance, and carry out activities to help them. Regular psychological training has been conducted to set up “mutual aid corners for older people”: shared tool cabinets (wheelchairs, crutches), medicine boxes, and message boards have been set up in the community, and items have been placed at a lower level for the convenience of older people in view of their limited mobility. Community mutual aid organizations for older people should focus on capacity building, cultivate community leaders, and improve support mechanisms. By conducting educational lectures, they should promote first aid and healthcare knowledge among seniors while establishing health records to monitor their physical and mental well-being, particularly for those with disabilities, and provide specialized care. Additionally, organizations should regularly host recreational activities and fitness programs, encourage seniors to participate in volunteer assistance and mutual aid initiatives, and enrich their spiritual and cultural lives. Strengthen the mental health screening of the older people, and provide psychological comfort to them on the verge of depression in a timely manner. Focus on the mental health of older people and the objects around the older people to help them master the skills and methods of psychological relaxation. Pay attention to the physical health of older people and guide them to develop and maintain good living habits and dietary habits.

• Fourth: family organization: strengthening internal family assistance and improving emotional communication.

  Children should take the initiative to study the policies on mutual aid for older people, gaining insight into their implementation advantages through online platforms, participating in community mutual aid for older people publicity activities, etc. This approach helps them to transform traditional notions and practices of filial piety. Carry out the family honor roll activities within the family, and commend older people and young children who actively participate in mutual aid, so as to give encouragement on the spiritual level.

Research perspectives

Due to limitations in sample selection and measurement methods, differences in mental health among older people may be influenced by other factors. Future studies with more rigorous designs are needed to further validate these findings. First: Increasing the control variables in order to study the effect of mutual aid on the mental health of older people in more depth. Second: The implementation effect of mutual aid can be analyzed from multiple perspectives, such as physical health, social health, and life satisfaction of older people. Third: Increasing research on regional distribution to understand the impact of mutual aid on the mental health of older people in different regions by analyzing regional heterogeneity. Fourth: Due to limitations in secondary data, this paper uses the Center for Epidemiologic Studies Depression Scale (CSE-D) to measure the mental health of older people, and there is still room for improvement in measurement accuracy. Fifth: Expanding the geographical scope of mutual aid for older people’s care policies will further increase the effectiveness of the DID model method.

Authors’ contributions

Author 1(First Author): Liu XH is responsible for:1. Collecting, organizing, and analyzing data for the paper; 2. Drafting and revising the paper content; 3. Collecting and organizing literature. Author 2: Zheng L is responsible for collecting and organizing literature. Author 3: Li M is responsible for collecting and organizing literature. Author 4:(Corresponding Author): Hu B is responsible for:1. Proposing research ideas; 2. Provide data methodology; 3. Funding support for paper publication.

Funding

None.

Data availability

We extracted the data required for this study from the CHARLS public database through a selection process. The data acquisition link is https://charls.pku.edu.cn/.

Declarations

Ethics approval and consent to participate

This round of investigation was approved by the Biomedical Ethics Committee of Peking University under the approval number: IRB00001052-11015. All authors have agreed to participate in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.