The Impact of Retirement and Grandchild Caregiving on Mental Health in China: The Role of Intergenerational Support

The Impact of Retirement and Grandchild Caregiving on Mental Health in China: The Role of Intergenerational Support | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The Impact of Retirement and Grandchild Caregiving on Mental Health in China: The Role of Intergenerational Support Yang Mou, Zijun Tan, Hongxian Nie This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6720729/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 08 Oct, 2025 Read the published version in BMC Public Health → Version 1 posted 18 You are reading this latest preprint version Abstract Background This study explores the impact of retirement, grandchild caregiving, and intergenerational support on the mental health of older adults in China, focusing on how these factors interact. Methods Using data from the 2018 China Health and Retirement Longitudinal Study (CHARLS), we employed Ordinary Least Squares (OLS) regression, Instrumental Variables (IV) regression, and Multiple Imputation regression to address potential endogeneity and missing data. Results Retirement is significantly associated with improved mental health, particularly among older adults with lower caregiving burdens (β = -0.435, p < 0.001). In contrast, higher caregiving intensity is linked to poorer mental health (β = 0.00094, p = 0.001). Satisfaction with parent-child relationships plays a protective role in mental health (β = -0.209, p < 0.001), and financial support to children is negatively associated with depressive symptoms (β= -0.023, p = 0.031). These findings underscore the interconnectedness of retirement, caregiving, and intergenerational support. Conclusion Extending the working age, without offering meaningful support, can lead to growing mental health issues among older adults—especially those already managing the demands of grandchild care. Any policy aimed at postponing retirement should be accompanied by tangible resources: local respite programs, training workshops tailored for elder caregivers, and even modest subsidies for childcare. These efforts not only provide relief—they also enable grandparents to emotionally recover while continuing to support their families without compromising their own well-being. Specifically, interventions such as local respite programs, targeted caregiver workshops, and even modest childcare subsidies could provide meaningful support. Grandchild Caregiving Mental Health Intergenerational Support Retirement Policies China Older Adults 1. Introduction China is undergoing a profound demographic shift. By the middle of this century, it is expected that nearly one-third of its citizens will be aged 60 or above.To address the growing economic strain posed by an aging labor force, policymakers have implemented delayed retirement strategies. While these measures aim to sustain fiscal balance, they may unintentionally increase mental burdens on older adults, particularly those who are also serving as caregivers for their grandchildren. Concurrently, family life in China is being reshaped by rapid urbanization and internal migration. These social shifts have resulted in more grandparents stepping in as the main providers of childcare. Although this role can offer emotional rewards and strengthen family ties, it also brings with it significant physical and psychological challenges for elderly caregivers. Past research has primarily treated retirement and caregiving as separate phenomena, often producing conflicting findings regarding their impact on mental health.However, few studies have examined how these two experiences interact, particularly in the context of China’s evolving economic policies and cultural norms.What remains underexplored is how stepping away from employment and assuming intensive caregiving duties intersect—particularly against China’s specific economic pressures and social policies.Moreover, while intergenerational support holds cultural significance, its influence on the intersection of retirement, caregiving, and mental health remains insufficiently explored. This study seeks to fill these gaps by addressing the following research questions: 1.How does retirement status influence the mental health of older adults in China? 2.What is the impact of caregiving intensity for grandchildren on psychological well-being? 3.Can intergenerational support buffer the mental health burdens associated with retirement and caregiving? Utilizing data from the China Health and Retirement Longitudinal Study (CHARLS) and an instrumental variable approach, this study provides empirical insights into the complex relationships between retirement, caregiving, and intergenerational support. Our findings offer valuable guidance for policymakers aiming to balance labor market objectives with the caregiving needs of families and the mental health of older populations. 2. Literature Review 2.1Retirement and Mental Health: Contradictory Findings The relationship between retirement and mental health has been widely debated, yielding mixed results. Some studies suggest that retirement can enhance psychological well-being, particularly in the initial years, as retirees often experience relief from work-related stress, leading to improvements in self-esteem, life satisfaction, and overall mental health( 1 , 2 ) . These benefits are most evident among those who retire voluntarily and view the transition positively ( 3 ). While the positive effects of retirement may initially seem evident, they tend to diminish over time.Many individuals who remain retired for extended periods often grapple with emotional difficulties such as loneliness, a diminished sense of purpose, and social withdrawal—each of which can take a toll on mental well-being ( 4 , 5 ). Notably, the mental health outcomes of retirement are closely linked to how that retirement comes about. People who exit the workforce involuntarily—for reasons like sudden job loss or chronic illness—tend to report greater levels of psychological distress, including heightened anxiety and depressive symptoms ( 6 , 7 ). In contrast, those who make the decision to retire on their own terms often fare better emotionally ( 8 , 9 ). Financial conditions add another layer of complexity. A lack of adequate income during retirement can intensify emotional struggles, particularly for those with minimal savings or unstable pensions (3,4). However, older adults who remain socially active and maintain strong interpersonal connections are often more resilient, underscoring the buffering effects of social engagement and community participation ( 10 ,6). Cultural expectations also shape the retirement experience. In contexts where professional life is deeply entwined with self-identity, stepping away from work may trigger a deep sense of disorientation or emotional void (5, 11 ). Taken together, while retirement can offer relief and psychological benefits in the short term, its longer-term effects are shaped by a combination of voluntariness, economic stability, and opportunities for meaningful social connection (1,2,11). 2.2Grandchild Caregiving and Mental Health: The Impact of Intensity The extent to which grandparents are involved in childcare significantly shapes their emotional well-being. When caregiving becomes too demanding, many older adults report heightened stress, emotional fatigue, and even physical symptoms like disrupted sleep or recurring headaches ( 12 ). In contrast, those who provide care in more moderate or occasional roles often maintain better psychological balance. These findings suggest that there may be a critical point at which caregiving shifts from fulfilling to burdensome ( 13 ). Nonetheless, grandchild caregiving is not inherently detrimental. For many seniors, participating in their grandchildren’s lives offers emotional satisfaction, a renewed sense of purpose, and a meaningful role within the family unit. Feeling appreciated can help reduce feelings of loneliness or sadness and bolster psychological resilience ( 14 ). However, when this role dominates daily life, the demands can become overwhelming. Caregivers in these situations frequently report emotional strain, including chronic anxiety and persistent fatigue—often intensified by ongoing concerns about their grandchildren’s future and a lack of personal time o r space ( 15 , 16 ). In such circumstances, access to strong social connections and effective personal coping strategies becomes especially important. These protective factors can act as emotional buffers, helping to mitigate the psychological costs of intense caregiving commitments.Grandparents who maintain strong relationships and receive emotional support from family and friends are more likely to report better mental health ( 17 , 18 ). Additionally, financial and reciprocal support from adult children can ease the psychological burden. When caregiving is coupled with both emotional and financial support from other family members, it helps reduce depressive symptoms and shifts caregiving from a solitary responsibility to a shared family duty( 19 ). While intergenerational support generally benefits caregivers, imbalances in these exchanges can lead to burnout. Open and honest communication between family members about caregiving roles, expectations, and boundaries is essential to ensure that caregiving remains a rewarding experience without leading to exhaustion ( 20 ). 2.3Delayed Retirement and Its Psychological Impact Delayed retirement has become an increasingly important topic of research, yet its mental health effects are multifaceted and often contradictory. For some individuals, staying in the workforce provides social connections and a continued sense of purpose. However, for others, working beyond the typical retirement age can lead to heightened stress and a diminished sense of social status, especially when one’s identity is strongly tied to their professional role( 21 ). The mental health outcomes associated with retirement are closely shaped by when and how individuals retire. Those who have time to anticipate and prepare for the transition generally navigate it more smoothly, often reporting greater emotional stability than peers forced into retirement due to layoffs or challenging work conditions( 22 ,2). Another important factor is cognitive health. Remaining in intellectually stimulating roles may help preserve mental agility, but continued exposure to high-pressure environments can wear down cognitive reserves, potentially leading to mental fatigue later on (11). For some, postponing retirement brings financial relief. Yet for others—especially those uncertain about their long-term income—delayed exit from the workforce can worsen anxiety and emotional distress ( 23 ). With more people now choosing to extend their careers, there is an urgent need to reconsider retirement frameworks that accommodate diverse needs and avoid a one-size-fits-all approach (22). Ultimately, while staying in the workforce may offer certain financial or social benefits, it can also come at a psychological cost. To support older adults through this complex transition, flexible retirement options that consider personal and economic realities are increasingly vital 2.4The Role of Intergenerational Support in Mitigating Psychological Burdens Support from younger generations can significantly lighten the emotional burden many older adults carry. Spending time with children or grandchildren—whether through daily activities, shared meals, or simply conversation—helps maintain a sense of connection that protects against social isolation, a key predictor of depression and anxiety in later life ( 24 , 25 ).In addition to emotional support, practical assistance such as help with everyday tasks and financial contributions can greatly alleviate the challenges of aging, especially for those facing physical limitations ( 26 ). Financial support from children is particularly important as older adults transition into retirement and face reduced income. This support alleviates economic stress and improves mental health outcomes ( 27 ,12). In addition to practical and financial support, caregiving roles can provide a sense of purpose, enhancing self-worth and reducing depression, especially when grandparents feel needed by their grandchildren (14). However, balance is crucial to prevent caregiver burnout. Effective communication between generations is key to ensuring caregiving remains fulfilling and does not lead to exhaustion (20). 2.5Research Gaps and Future Directions Although considerable research has been conducted on the individual effects of retirement and caregiving on mental health, studies exploring their combined effects remain limited, especially in China. Older adults often face the dual pressures of retirement and caregiving, yet little is understood about how these roles intersect and impact mental health. Future research should investigate the combined effects of retirement and caregiving, particularly within China’s rapidly evolving socio-economic context. Further studies on formal support services, such as respite care and community-based programs, could offer valuable insights into alleviating caregiver burden. Longitudinal studies would be instrumental in understanding the long-term mental health impacts of retirement and caregiving intensity. As China’s demographic landscape continues to shift and family structures evolve, more research is needed to understand how intergenerational support systems adapt. Future studies should explore how different types of intergenerational support influence mental health, considering cultural and policy factors. 3. Method 3.1Sample Selection and Attrition This study uses data from the 2018 China Health and Retirement Longitudinal Study (CHARLS), with an initial sample of 17,814 observations. After excluding samples with missing key variables, 7,825 valid observations remained. To address missing data, we employed the Multiple Imputation (MI) method to impute key variables, such as self-rated health (srh) and depressive symptoms (cesd10_std). The detailed process of multiple imputation is presented in Table 1.After further excluding observations with excessive missing data, the final sample for regression analysis comprised 2,657 individuals. Table 1: Imputation Process for Multiple Imputed Datasets Step Description Notes Step 1 Registering variables for imputation age, srh, cesd10_std marked for imputation Step 2 Performing multivariate normal imputation 5 imputed datasets generated Step 3 Checking convergence and stability Convergence achieved with EM optimization Step 4 Analyzing imputed datasets Imputed datasets used for further analysis 3.2Results of Sample Loss Comparison A comparison of the excluded and retained samples based on key characteristics, including age, self-rated health (srh), and depressive symptoms (cesd10_std), showed no significant differences in depressive symptoms (p = 0.4898), indicating that sample attrition did not introduce substantial bias. Descriptive statistics and p-values for this comparison are reported in Table 2.However, older individuals and those with poorer health were more likely to be excluded (p < 0.001). Table 2: Sample Loss Comparison Variable Deleted Sample (N = 43) Retained Sample (N = 7,782) p-value Age 67.07 (± 6.00) 61.59 (± 7.56) < 0.001 Self-Rated Health (srh) 2.73 (± 1.06) 3.00 (± 1.03) 0.0876 Depressive Score (cesd10_std) 0.45 (± 1.36) 0.05 (± 1.01) 0.4898 Note: *** p < 0.01, ** p < 0.05, * p < 0.10. The imputation process generated five datasets, ensuring that the results accurately reflected the full population. Sensitivity analysis confirmed the consistency of the results, comparing multiple imputation with listwise deletion methods. 3.3Statistical Analysis and Model Specification Regression analyses were performed using Ordinary Least Squares (OLS) and Instrumental Variables (IV) methods to examine the relationships between retirement, caregiving, and mental health. Key variables, such as retirement status, caregiving intensity, and mental health outcomes (self-rated health and depressive symptoms), were included as explanatory variables in the models. To address potential endogeneity, we employed instrumental variables, specifically urban-rural residence and retirement age, as instruments for retirement status. 3.4Variable Definitions 3.4.1Dependent Variable The primary outcome variable is mental health, measured using the standardized CES-D 10 score. This scale assesses depressive symptoms in older adults, including feelings of sadness, anxiety, and loss of interest. The score is standardized, with a range of -1.325 to 3.24, where higher values indicate more severe depressive symptoms. 3.4.2Independent Variables Retirement Status (Retire): Retirement Status (Retire): A binary variable indicating whether the individual is retired (1 = retired, 0 = not retired). This variable captures the respondent's labor market exit status and serves as a key explanatory variable. Grandchild Caregiving (WPC):A binary variable representing whether the respondent provides caregiving for grandchildren (1 = provides care, 0 = does not provide care). This variable reflects the respondent's involvement in family caregiving roles. Caregiving Intensity (HTSPCH): A continuous variable indicating the number of hours per week the respondent spends caring for grandchildren. The variable ranges from 0 to 336 hours, with extreme values above 168 hours capped at 168 to maintain data plausibility and reduce outlier effects. Intergenerational Support Parent-Child Relationship Satisfaction (Sati_Child): An ordinal variable measuring satisfaction with the parent-child relationship, ranging from 1 (very dissatisfied) to 5 (very satisfied), reflecting the intensity of emotional support. Financial Support Financial Support to Children (Log_Fcamt): A continuous variable representing the amount of financial support provided to children, measured as the log-transformed financial amount. Financial Support from Children (Log_Tcamt):A continuous variable representing the amount of financial support received from children, measured as the log-transformed financial amount. 3.4.3Control Variables Squared Age Term (c.age#c.age): The square of the individual’s age (c.age) is included as a continuous variable to account for non-linear age effects in the regression models. Self-Rated Health (SRH): An ordinal variable reflecting the individual’s subjective health perception, with a scale from 1 (very poor) to 5 (excellent). Pension Status (Pension):A binary variable indicating whether the individual receives a pension (1 = receives pension, 0 = does not receive pension). Marital Status (Marry):A binary variable indicating whether the individual is married (1 = married, 0 = not married). Family Size (Family_Size):A continuous variable indicating the number of household members, ranging from 1 to 13. Number of Grandchildren under 16 (grandchildu16): A continuous variable indicating the number of grandchildren under the age of 16, with a range from 0 to 14. The definitions and measurement methods of all variables used in the analysis are summarized in Table 3. Table 3: Variable Definitions Variable Definition Measurement cesd10_std Standardized CES-D 10 score, a measure of depressive symptoms in older adults. Assesses symptoms such as sadness, anxiety, and loss of interest. Continuous variable, range: -1.325 to 3.24. Higher scores indicate more severe depressive symptoms. retire Indicator variable for whether the individual is retired. Binary variable (1 = retired, 0 = not retired). sati_child Satisfaction with the parent-child relationship. Ordinal variable (1 = very dissatisfied, 5 = very satisfied). wpc Indicates whether the individual provides caregiving to grandchildren. Binary variable (1 = provides care, 0 = does not provide care). htspch Number of hours per week spent providing caregiving to grandchildren, reflecting caregiving intensity. Continuous variable, range: 0 to 336 hours. Extreme values adjusted to 168 hours if exceeding this threshold. srh Self-rated health status, reflecting the individual’s subjective health perception. Ordinal variable (1 = very poor, 5 = excellent). pension Indicates whether the individual receives a pension. Binary variable (1 = receives pension, 0 = does not receive pension). log_fcamt Log-transformed amount of financial support provided to children. Continuous variable, log-transformed financial support amount. log_tcamt Log-transformed amount of financial support received from children. Continuous variable, log-transformed financial support amount. c.age The individual’s age. Continuous variable, measured in years. c.age#c.age Squared term of age to account for non-linear age effects. Continuous variable, the square of c.age. marry Marital status. Binary variable (1 = married, 0 = not married). family_size Number of household members. Continuous variable, range: 1 to 13. grandchildu16 Number of grandchildren under the age of 16. Continuous variable, range: 0 to 14. 3.5Empirical Strategy To address potential endogeneity concerns in the relationship between retirement and mental health, we employ an Instrumental Variables (IV) approach. The instrumental variables used in this study are urban/rural household registration and retirement age policy, which are both exogenous to individuals' health and psychological factors but influence retirement timing. 3.5.1Instrumental Variables Urban/Rural Household Registration (Urban/Rural):The urban-rural divide in China has a notable impact on access to critical services like pensions and healthcare. However, this divide does not directly affect mental health outcomes, except through its influence on retirement status. Therefore, urban-rural classification can be used as a valid instrumental variable for retirement status. Zhang et al. (2022) have pointed out that disparities in mental well-being between older adults in rural and urban areas stem mainly from unequal access to financial resources and medical services, rather than the urban-rural status itself( 28 ). This observation reinforces the validity of using household registration type as an instrument to isolate the causal effect of retirement on psychological outcomes. 2.Retirement Age Policy (Retirement_Age2):China’s official retirement regulations stipulate different age thresholds for men and women—62 for men and 58 for women. These age-based rules create an exogenous source of variation in retirement behavior that is not tied to individuals’ mental health or personal choices. In this analysis, we employ the male retirement age (62) as an instrumental variable. By that point, most women in the sample have already exited the workforce, which helps isolate the effect of retirement timing for men without conflating it with gender-based differences.This policy-driven variation offers a unique analytical advantage: it influences when people retire, but not their psychological condition directly. Huang (2024) corroborates this logic, arguing that while mandatory retirement policies strongly determine retirement timing, they remain unrelated to mental health outcomes. This makes the male retirement age a suitable instrument for examining causal links between retirement and mental well-being( 29 ). 3.5.2First-Stage Regression and Instrument Validity We conduct a first-stage regression of retirement on the instruments (urban/rural registration and retirement age policy). The regression results show that both instruments are strongly significant in explaining retirement: Urban/Rural: Coefficient = - 0.4383, p-value < 0.001. Retirement Age Policy: Coefficient = -0.1145, p-value < 0.001. Where Urban/Rural represents the urban/rural household registration, Retirement Age Policy refers to the retirement age threshold of 62 years for men (used as the instrument for retirement status), X_i includes control variables (such as age, marital status, etc.), and ε_i is the error term. Endogeneity Test To check for endogeneity in the retirement variable, we perform the Durbin-Wu-Hausman test, which yields a p-value = 0.0003. This result confirms that retirement is endogenous, thus necessitating the use of the IV approach to address potential bias in the relationship between retirement and mental health. Overidentification Test We conduct the Hansen J test to test the exogeneity of our instruments. The test results show a p-value = 0.6866, indicating that both instruments are valid and do not directly affect mental health, confirming their appropriateness for this study.Table 4 presents the results of the first-stage regression, demonstrating the strong significance of both instruments in predicting retirement status, as well as the relevant test statistics for instrument validity. 3.5.3Second-Stage Regression In the second stage, we estimate the effect of retirement on mental health (measured by the CES-D score), using the predicted retirement status from the first-stage regression as the instrumental variable. The IV regression results reveal a significant negative effect of retirement on mental health, showing that retirement contributes to a decrease in depressive symptoms. Where Retire is the predicted retirement status from the first stage, Caregiving Intensity is the number of caregiving hours per week, Parent-Child Satisfaction is the measure of satisfaction with the parent-child relationship, and X_i represents control variables. The IV regression results indicate a significant reduction in depressive symptoms due to retirement, with a coefficient of -0.6409 and p < 0.001. This finding suggests that retirement has a positive impact on mental health by alleviating depressive symptoms. The Durbin-Wu-Hausman test confirms that retirement is endogenous, while the Hansen J test supports the validity of our instruments, ensuring that urban-rural registration and retirement age policy are appropriate instruments for examining the impact of retirement on mental health. Statistical Methods All statistical analyses were conducted using Stata 16. Standard errors were clustered at the individual level to account for potential intra-individual correlation. Robust standard errors are reported throughout the analysis. Table 4: First-Stage Regression Results Variable Coefficient Standard Error Z-Statistic p-Value 95% Confidence Interval Urban/Rural -0.4383 0.0179 -24.49 < 0.001 [-0.4734, -0.4032] Retirement Age Policy -0.1145 0.0261 -4.39 < 0.001 [-0.1657, -0.0634] F-statistic 326.72 - - < 0.001 - Hansen J Test 0.163 - - 0.6866 - Note: *** p < 0.01, ** p < 0.05, * p < 0.10. 4 Results 4.1Descriptive Statistics Table 5 presents the descriptive statistics for the variables included in the analysis. The initial sample consisted of 7,825 respondents. After excluding individuals with missing data or inconsistent responses, the final sample for regression analysis was reduced to 2,657 respondents. The mean age of the sample is 61.6 years, with ages ranging from 40 to 75. Among the 7,825 respondents, 14% are retired, and 53.5% report providing care for grandchildren. The average caregiving intensity is 53.2 hours per week, indicating a significant caregiving burden for many older adults. It is important to note that the descriptive statistics reflect the full sample of 7,825 respondents, while the regression analysis is based on the reduced sample of 2,657 after excluding observations with missing values for key variables. The average CES-D 10 score is 0.047, indicating a moderate level of depressive symptoms across the sample. Self-rated health, measured on a 5-point scale, has an average score of 3.0, suggesting that respondents generally rate their health as moderate. 64.1% of the respondents receive a pension, while 35.9% do not. Financial transfers between family members are notable: the average value of financial support provided to children (log-transformed) is 7.89, and the average value of financial support received from children is 7.43. The sample is nearly balanced in terms of gender, with 46.8% of respondents being female. The average family size is 2.82 members, and 81% of respondents are married. These descriptive statistics provide an overview of the sample's demographic and socio-economic characteristics, offering essential context for understanding the relationship between retirement, caregiving, and mental health among older adults in China. Table 5: Descriptive Statistics of the Sample Variable Obs Mean Std. Dev. Min Max cesd10_std 7,785 0.047 1.0116 -1.3251 3.2404 retire 7,773 0.14 0.347 0 1 sati_child 7,825 3.5972 0.7396 1 5 wpc 7,825 0.5355 0.4988 0 1 htspch 7,825 46.5507 60.7963 0 168 srh 7,822 3.0015 1.0266 1 5 pension 7,022 0.6413 0.4797 0 1 log_fcamt 6,205 7.8905 1.3835 0 12.6115 log_tcamt 3,292 7.4312 1.9129 1.0986 13.9995 age 7,825 61.6157 7.5615 40 75 gender 7,825 0.4682 0.499 0 1 edu 7,825 2.0235 1.0239 1 4 marry 7,825 0.8096 0.3927 0 1 family_size 7,825 2.822 1.5504 1 13 grandchildu16 7,825 0.2276 0.912 0 14 4.2Regression Results Table 6 presents the regression results examining the relationship between retirement, grandchild caregiving, intergenerational support, and mental health outcomes. The effects of retirement, caregiving intensity, and intergenerational support on mental health are estimated, controlling for various socio-demographic variables. 4.2.1OLS Regression Results The Ordinary Least Squares (OLS) regression results show a significant negative association between retirement and depressive symptoms (β= - 0.276, p < 0.001), indicating that retirement is linked to a substantial reduction in depressive symptoms. Satisfaction with parent-child relationships is also negatively associated with depressive symptoms (β= -0.210, p < 0.001), suggesting that greater satisfaction with these relationships correlates with fewer depressive symptoms. Grandchild caregiving is associated with improved mental health outcomes, with caregiving linked to a reduction in depressive symptoms (β= -0.113, p = 0.010). However, this effect is weaker compared to retirement and parent-child relationship satisfaction. Caregiving intensity shows a contrasting pattern. Higher caregiving intensity is associated with increased depressive symptoms (β = 0.00094, p= 0.001), suggesting that while caregiving for grandchildren can enhance emotional well-being, more intensive caregiving is linked to greater psychological strain. Self-rated health (SRH) is strongly associated with depressive symptoms, with poorer self-reported health corresponding to higher levels of depressive symptoms (β = -0.304, p < 0.001). Financial support to children is negatively associated with depressive symptoms (β= -0.032, p= 0.001), indicating that older adults who provide financial support to their children experience fewer depressive symptoms. However, financial support received from children does not significantly affect mental health (β= -0.013, p = 0.306).Detailed OLS regression results, including coefficient estimates, standard errors, and significance levels, are presented in Table 6. Table6: Regression Results - OLS Regression Variable Coefficient Standard Error t-Statistic p-Value 95% Confidence Interval Retire -0.2756 0.0479 -5.75 <0.001 [-0.3697,-0.1816] Sati_Child -0.2106 0.0241 -8.74 <0.001 [-0.2579, -0.1633] WPC -0.1198 0.0476 -2.52 0.012 [-0.2132, -0.0264] HTSPCH 0.0011 0.0004 2.91 0.004 [0.0004, 0.0019] SRH -0.304 0.0168 -18.1 <0.001 [-0.3369, -0.2711] Pension -0.0151 0.0491 -0.31 0.758 [-0.1114, 0.0812] Log_Fcamt -0.0125 0.0128 -0.98 0.329 [-0.0376, 0.0126] Log_Tcamt -0.0318 0.0098 -3.25 0.001 [-0.0510, -0.0126] Age^2 -0.0001 0.00003 -2.78 0.005 [-0.0001, -0.00002] Married -0.2156 0.0511 -4.22 <0.001 [-0.3158, -0.1154] Family Size -0.0396 0.0135 -2.93 0.003 [-0.0661, -0.0131] Grandchildu16 0.0364 0.0234 1.55 0.121 [-0.0096, 0.0824] Constant 2.615 0.177 14.77 <0.001 [2.2679, 2.9620] Note: *** p < 0.01, ** p < 0.05, * p < 0.10. 4.2.2Instrumental variable regression results To address endogeneity concerns, we conducted Instrumental Variable (IV) regression using urban-rural registration and mandatory retirement age as instruments for retirement status. The IV regression results largely confirm the findings from the OLS model.Financial support to children continues to show a significant negative association with depressive symptoms (β = -0.023, p = 0.031). In contrast, financial support received from children does not significantly affect mental health outcomes (β= -0.014, p = 0.268). The Instrumental Variable (IV) regression results largely confirm the findings from the OLS model. Retirement remains significantly associated with improved mental health (β = -0.435, p < 0.001), consistent with the OLS results. Parent-child satisfaction continues to have a protective effect on mental health (β= -0.209, p < 0.001), and grandchild caregiving is linked to fewer depressive symptoms (β= -0.109, p = 0.012). However, caregiving intensity remains a significant risk factor for mental health deterioration (β = 0.00094, p = 0.001). Financial support to children continues to show a significant negative association with depressive symptoms (β = -0.023, p= 0.031), whereas financial support received from children does not significantly affect mental health outcomes (β = -0.014, p = 0.268).The full results of the IV regression, including coefficient estimates, standard errors, and confidence intervals, are presented in Table 7. Table 7: Regression Results - IV Regression Variable Coefficient Standard Error Z-Statistic p-Value 95% Confidence Interval Retire -0.4352 0.0886 -4.91 <0.001 [-0.6088, -0.2616] Sati_Child -0.2093 0.0271 -7.72 <0.001 [-0.2624, -0.1561] WPC -0.1149 0.0465 -2.47 0.013 [-0.2060, -0.0238] HTSPCH 0.0011 0.0004 3.06 0.002 [0.0004, 0.0019] SRH -0.3017 0.0165 -18.24 <0.001 [-0.3342, -0.2693] Pension 0.0235 0.0533 0.44 0.66 [-0.0810, 0.1280] Log_Fcamt -0.0136 0.0128 -1.06 0.288 [-0.0387, 0.0115] Log_Tcamt -0.023 0.0105 -2.18 0.029 [-0.0436, -0.0023] Age^2 -0.0001 0.00003 -2.48 0.013 [-0.0001, -0.00002] Married -0.2159 0.0548 -3.94 <0.001 [-0.3234, -0.1085] Family Size -0.0411 0.013 -3.17 0.002 [-0.0665, -0.0157] Grandchildu16 0.0302 0.0262 1.15 0.249 [-0.0211, 0.0816] Constant 2.5244 0.1887 13.38 <0.001 [2.1546, 2.8943] Note: *** p < 0.01, ** p < 0.05, * p < 0.10. 4.3Robustness Checks To test the robustness of our findings, we added additional control variables, such as gender and education level, to the regression model. The results remained consistent with the main model, showing no significant changes. Specifically, retirement continued to be significantly negatively associated with depressive symptoms (β = -0.355, p < 0.001). Similarly, parent-child relationship satisfaction (β = -0.219, p < 0.001) and grandchild caregiving (β = -0.103, p = 0.015) maintained their significant protective effects. In contrast, caregiving intensity remained significantly associated with increased depressive symptoms (β = 0.00100, p < 0.001), indicating that excessive caregiving exacerbates mental health challenges. Additionally, both gender and education level were found to influence mental health outcomes. Female respondents reported fewer depressive symptoms (β = -0.267, p < 0.001), while higher education attainment was associated with better mental health (β = -0.058, p = 0.005). This suggests that higher education may improve mental health by increasing social participation and access to resources. To further examine the consistency of our results across different demographic groups, we conducted heterogeneity analyses based on gender and urban-rural residence. While some background differences were observed, the regression results showed no significant variations, indicating that the relationships between retirement, caregiving, and mental health are generally consistent across these groups. This suggests that our conclusions are broadly applicable to different subgroups. Nonetheless, future research could explore the impact of these variables within specific groups using larger subgroup samples, which would provide further insights into the mechanisms through which retirement and caregiving influence mental health.The detailed results of the robustness checks, including all additional control variables, are presented in Table 8. Table 8: Robustness Check Results Variable Coefficient Std. Error Z-Statistic p-value 95% Confidence Interval Retirement Status (Instrumented) -0.355*** 0.101 -3.51 <0.001 [-0.552, -0.158] Parent-Child Satisfaction -0.219*** 0.027 -8.14 <0.001 [-0.271, -0.167] Grandchild Caregiving -0.103* 0.042 -2.46 0.015 [-0.186, -0.020] Caregiving Intensity (hrs/week) 0.00100*** 0.00027 3.7 <0.001 [0.00047, 0.00153] Self-Rated Health -0.292*** 0.016 -18.24 <0.001 [-0.323, -0.261] Financial Support to Children -0.018 0.01 -1.8 0.089 [-0.037, 0.001] Financial Support from Children -0.013 0.013 -0.99 0.301 [-0.038, 0.013] Pension Status -0.007 0.053 -0.13 0.887 [-0.111, 0.096] Age Squared -0.000042 0.000028 -1.5 0.132 [-0.000097, 0.000012] Gender -0.267*** 0.036 -7.41 < 0.001 [-0.338, -0.197] Education Level -0.058** 0.021 -2.77 0.005 [-0.099, -0.018] Marital Status -0.135* 0.055 -2.46 0.015 [-0.244, -0.027] Family Size -0.039** 0.013 -2.88 0.002 [-0.064, -0.014] Number of Grandchildren <16 0.011 0.025 0.45 0.651 [-0.037, 0.059] Constant 2.571*** 0.194 13.27 < 0.001 [2.185, 2.948] Note: *** p < 0.01, ** p < 0.05, * p < 0.10. 5 Discussion 5.1Overview of Main Findings The key findings of this study indicate that retirement, satisfaction with parent-child relationships, and caregiving intensity significantly impact the mental health of older adults. Specifically, retirement is negatively associated with mental health, suggesting that it may alleviate mental health burdens among older adults. In contrast, higher caregiving intensity is negatively correlated with mental health, indicating that excessive caregiving can exacerbate psychological stress. Additionally, satisfaction with parent-child relationships has a protective effect on mental health, highlighting the importance of family support for well-being. 5.2Discussion of Regression Model Comparisons To validate the robustness of our findings, we employed three different regression models:Ordinary Least Squares (OLS) Regression, Instrumental Variables (IV) Regression, and Multiple Imputation Regression. These models allowed us to examine the relationship between retirement, caregiving, and mental health from multiple perspectives, addressing concerns such as endogeneity and missing data. 5.2.1Comparison between OLS and IV Regression In the OLS Regression, we observed a significant negative relationship between retirement and mental health (β = -0.372, p < 0.001), suggesting that retirement improves mental health in older adults. However, this result may be subject to endogeneity issues, such as reverse causality or omitted variable bias (e.g., health status or social support). To overcome this issue, we applied an Instrumental Variables (IV) approach, utilizing urban-rural household registration status and mandatory retirement age policies as instrumental variables.The IV results confirmed a stronger negative effect of retirement on mental health (β = -0.572, p < 0.001), consistent with the OLS results but with a larger magnitude. This indicates that the IV approach yields a more cautious and dependable estimate of the causal link between retirement and mental health. 5.2.2Impact of Multiple Imputation Regression Because some variables contained missing values, we applied multiple imputation techniques to address the data gaps, resulting in five completed datasets. Analysis using these imputed datasets confirmed that retirement continued to have a significant negative association with mental health (β = -0.357, p < 0.001).Although there were slight variations in the coefficients compared to the OLS regression, the imputation process had minimal impact on the results, further enhancing the robustness of the estimates.Table 9 presents a side-by-side comparison of the key regression results across OLS, IV, and multiple imputation models, demonstrating the consistency and robustness of our main findings. Table 9: Comparison of Regression Results Variable OLS Regression (Complete Case) IV Regression Multiple Imputation Regression Retirement (retire) -0.372 (p < 0.001) -0.572 (p < 0.001) -0.357 (p < 0.001) Satisfaction with Child Relationships (sati_child) -0.196 (p < 0.001) -0.198 (p < 0.001) -0.200 (p < 0.001) Caregiving Intensity (wpc) -0.099 (p < 0.001) -0.090 (p < 0.001) -0.106 (p < 0.001) Self-Rated Health (srh) -0.363 (p < 0.001) -0.359 (p < 0.001) -0.349 (p < 0.001) Pension (pension) -0.049 (p < 0.001) -0.009 (p = 0.543) -0.049 (p = 0.082) Financial Support from Parents (log_fcamt) -0.041 (p < 0.001) -0.041 (p < 0.001) -0.034 (p < 0.001) Financial Support from Children (log_tcamt) -0.012 (p < 0.001) -0.005 (p = 0.168) -0.016 (p = 0.080) Constant (_cons) 2.416 (p < 0.001) 2.349 (p < 0.001) 2.340 (p < 0.001) Note: *** p < 0.01, ** p < 0.05, * p < 0.10. 5.4Rationale for Model Selection Different regression models offer unique advantages in addressing various data issues: OLS Regression provides basic estimates of relationships between variables. However, it assumes no endogeneity, which may introduce bias when analyzing causal relationships, particularly between retirement and mental health. IV Regression addresses endogeneity by using valid instruments—urban-rural differences and the retirement age policy—to generate more reliable causal estimates. These instruments influence retirement decisions but do not directly affect mental health, avoiding biases inherent in OLS regression. Multiple Imputation Regression deals with missing data by imputing missing values, reducing bias and improving sample completeness. While the coefficients from multiple imputation are similar to those from OLS, the imputation process strengthens the reliability of the analysis. Although the coefficients slightly vary across models, all consistently show that retirement negatively affects mental health. We consider IV regression the most appropriate model, as it effectively addresses endogeneity and provides more accurate causal inferences, while Multiple Imputation Regression ensures the robustness and completeness of the analysis. 5.5Robustness of Results By comparing different regression models, we have validated the robustness of our findings. Despite slight variations in the coefficients, the main result—that retirement significantly negatively affects mental health—remains consistent across all models. This consistency reinforces the reliability of our results, suggesting that retirement plays a crucial role in improving mental health among older adults in China. 5.6Policy Recommendations 5.6.1Specific and Actionable Recommendations To better support older adults who shoulder the demands of caregiving—particularly for their grandchildren—local governments should consider introducing targeted financial and community-based interventions. One practical step would be to offer a monthly subsidy of ¥500 per child to help cover the costs of daycare services. This could provide much-needed relief for families where grandparents spend over 40 hours each week on childcare responsibilities. In fact, pilot programs in cities like Shanghai have already shown that such financial support, coupled with accessible, high-quality childcare, can ease the strain on aging caregivers. Beyond financial aid, municipal authorities could develop neighborhood-level caregiver support hubs. These might include practical training programs, temporary respite care options, and psychological counseling services. Dedicated funding for these initiatives could help create spaces where caregivers receive both emotional and logistical support. Additionally, tax breaks or direct subsidies for family caregivers may help them better manage the competing pressures of work and caregiving—a balance that is increasingly difficult to maintain without institutional support. 5.6.2International Experiences as Reference Lessons from other countries can offer valuable guidance for designing effective caregiving support systems. In Japan, the Long-Term Care Insurance program provides extensive resources to elderly individuals and their families who are involved in intensive caregiving arrangements. This approach has demonstrated measurable improvements in the well-being of both caregivers and care recipients. Given China’s aging demographics, drawing inspiration from such structured systems may be a promising direction. The Nordic countries present another compelling model. Their integrated approach combines public funding, community services, and in-home care, ensuring that older adults receive appropriate support while caregivers benefit from workplace flexibility and strong institutional backing. One of the most notable features of this model is its emphasis on universal accessibility. Adapting elements of this system could help Chinese local governments build a more inclusive, sustainable, and family-centered framework for elder care and caregiver assistance. 5.6.3Policy Implementation and Evaluation Budget Allocation:Local authorities should allocate resources to support community daycare subsidies, caregiver support programs, and long-term care initiatives, with funding distributed based on regional demand to ensure fairness and accessibility. Additional allocations should cover caregiver training activities and the development of caregiver resource centers in community settings. Monitoring and Evaluation:Continuous evaluation is essential to gauge policy effectiveness. Local governments should perform annual assessments to measure caregiver satisfaction, monitor reductions in caregiving-related financial stress, and evaluate improvements in older adults' overall well-being. Regular feedback will allow policymakers to refine their strategies to effectively address caregiving challenges and enhance quality of life. Partnerships with NGOs:Building partnerships with non-governmental organizations and community-based groups can effectively expand service accessibility, particularly for underserved rural areas. Such collaborative efforts can enhance caregiver and elder support networks, boost public awareness of available caregiving resources, and help families better navigate existing support systems. Workplace Flexibility:Workplace flexibility initiatives should be implemented to better support family caregivers.For example, caregivers could be granted time off or flexible working hours to manage caregiving duties. Companies could be incentivized to offer such benefits, ensuring caregivers are supported both in the workplace and at home. 5.6.4Limitations and Future Directions This study has several limitations that should be considered when interpreting the results.First, although the study uses a large, nationally representative sample from the CHARLS dataset, it relies on cross-sectional data, which limits the ability to draw causal conclusions about the relationship between retirement and mental health. Despite using instrumental variables to address potential endogeneity, the results may still be influenced by unobserved confounders affecting both retirement and mental health. Second, while multiple imputation was employed to handle missing data, residual bias may remain due to sample attrition. Specifically, excluded participants were older and more likely to report poorer health. Although the imputation process helped reduce bias and ensured a balanced final sample with respect to key covariates, unobserved differences—especially related to age and health—may still affect the results. Finally, while the study focuses on key variables such as self-rated health and depressive symptoms, it does not consider other potential factors influencing mental health, such as social support, economic status, and personal coping mechanisms. Future research could explore these additional factors and examine the long-term effects of retirement on mental health using longitudinal data, providing a deeper understanding of causal pathways and long-term impacts. Declarations Data Availability Statement The CHARLS dataset analysed during the current study is publicly available from the official CHARLS repository (http://charls.pku.edu.cn/en). Conflict of Interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest . Author Contributions ZT and HN contributed equally to this work and share first authorship. YM served as the corresponding author. ZT and YM were responsible for the conceptualization, research design, and methodological framework. YM performed the formal analysis, data curation, and supervised the overall research process. ZT and HN collected the data and drafted the initial manuscript. HN also contributed to the translation, language refinement, and final quality control of the article. All authors contributed to manuscript review and editing and approved the final version for submission. Funding The authors received no specific funding for this work. Ethics Statement This study utilized data from the China Health and Retirement Longitudinal Study (CHARLS), which obtained ethical approval from the Institutional Review Board at Peking University. The IRB approval number for the main household survey, including anthropometrics, is IRB00001052-11015, and the IRB approval number for biomarker collection is IRB00001052-11014. All participants provided informed consent prior to data collection. References Fleischmann, M., Xue, B., & Head, J. (2020). Mental Health Before and After Retirement—Assessing the Relevance of Psychosocial Working Conditions: The Whitehall II Prospective Study of British Civil Servants. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 75(2), 403-413. https://doi.org/10.1093/geronb/gbz042. Vo, T.T., & Phu-Duyen, T.T. (2023). Mental health around retirement: evidence of Ashenfelter's dip. Global Health Research and Policy, 8(1), 35. https://doi: 10.1186/s41256-023-00320-3. Kolodziej, I.WK, & García-Gómez, P. (2019). Saved by retirement: Beyond the mean effect on mental health. Social Science & Medicine (1982), 225, 85-97. https://doi: 10.1016/j.socscimed.2019.02.003. Epub 2019 Feb 4. Olesen, K., Rod, N.H., Madsen, I.E., Bonde, J.P., & Rugulies, R. (2015). Does retirement reduce the risk of mental disorders? A national registry-linkage study of treatment for mental disorders before and after retirement of 245,082 Danish residents. Occupational and Environmental Medicine, 72(5), 366-72. https://doi: 10.1136/oemed-2014-102228. Epub 2015 Mar 26. Mosconi, G., Vigezzi, G.P., Bertuccio, P., Amerio, A., & Odone, A. (2023). Transition to retirement impact on risk of depression and suicidality: results from a longitudinal analysis of the Survey of Health, Ageing and Retirement in Europe (SHARE). Epidemiology and Psychiatric Sciences, 32, e34. doi: 10.1017/S2045796023000239. Oltmans, E., Confectioner, K., Jonkers, R., Kerkhoffs, G. MMJ, Moen, M., Verhagen, E., Wylleman, P., & Gouttebarge, V. (2022). A 12-month prospective cohort study on symptoms of mental health disorders among Dutch former elite athletes. The Physician and Sportsmedicine, 50(2), 123-131.doi: 10.1080/00913847.2020.1868276. Åhlin, J.K., Peristera, P., Westerlund, H., & Magnusson Hanson, L.L. (2020). Psychosocial working characteristics before retirement and depressive symptoms across the retirement transition: a longitudinal latent class analysis. Scandinavian Journal of Work, Environment & Health, 46(5), 488-497. https:// doi: 10.5271/sjweh.3889. Odone, A., Gianfredi, V., Vigezzi, G.P., Amerio, A., Ardito, C., d'Errico, A., Stuckler, D., Costa, G., & (2021). Does retirement trigger depressive symptoms? A systematic review and meta-analysis. Epidemiology and Psychiatric Sciences, 30, 77. https://10.1017/S2045796021000627 .Noh, J.W., Kwon, Y.D., Lee, L.J., Oh, I.H., & Kim, J. (2019). Gender differences in the impact of retirement on depressive symptoms among middle-aged and older adults: A propensity score matching approach. PLoS One, 14(3), e0212607. https://10.1371/journal.pone.0212607 Huijs, J.JJM, Houtman, I.LD, & Blonk, R.WB. (2019). Development of a new measure for mental retirement; testing of a three-factor structure of mental retirement in different subgroups. BMC Public Health, 19(1), 1604. https://10.1186/s12889-019-7649-5. Denier, N., Clouston, S. A. P., Richards, M., & Hofer, S. M. (2017). Retirement and Cognition: A Life Course View. Advances in Life Course Research, 31, 11-21. doi: 10.1016/j.alcr.2016.10.004. Barman, P., & Sahoo, H. (2024). Through the eyes of grandparents: an in-depth exploration of the nexus between grandchild caring and the psychological well-being of older grandparents. BMC Geriatrics, 24(1), 399. doi: 10.1186/s12877-024-04998-z. Hong, Y., & Xu, W. (2023). Continuity and changes in grandchild care and the risk of depression for Chinese grandparents: new evidence from CHARLS. Frontiers in Public Health, 11, 1217998. https://doi: 10.3389/fpubh.2023.1217998. Bangerter, L.R., Kim, K., Zarit, S.H., Birditt, K.S., & Fingerman, K.L. (2015). Perceptions of Giving Support and Depressive Symptoms in Late Life. The Gerontologist, 55(5), 770-779. https://doi.org/10.1093/geront/gnt210. Kim, K.M., Lee, J.J., & Chung, U.S. (2020). Perceived Health Status of and Moderating Factors in Elderly People Caring for Their Grandchildren. Psychiatry Investigation, 17(4), 275-282. https://DOI: https://doi.org/10.30773/pi.2019.0115. Tang, F., Xu, L., Chi, I., & Dong, X. (2016). Psychological Well-Being of Older Chinese-American Grandparents Caring for Grandchildren. Journal of the American Geriatrics Society, 64(11), 2356-2361. doi: 10.1111/jgs.14455. Xia, C., Wei, T., Tang, Q., Zheng, H., Chen, G., & Lu, J. (2023). Depression and Associated Factors among Family Caregivers of Children with Disabilities: Analysis of Intergenerational Differences. Healthcare (Basel, Switzerland), 11(19), 2693. https://doi.org/10.3390/healthcare11192693. Lee, J., Lee, A., Lee, D., Jung, H.Y., Kim, S.G., & Lee, S.I. (2019). Suicidal Ideation of the Elderly According to Their Involvement in Grandchild Care. Psychiatry Investigation, 16(8), 625-628. https://oi: 10.30773/pi.2019.06.06. Wang, H., & Huang, J. (2023). Impacts of grandparenting on older Chinese adults' mental health: a cross-sectional study. BMC Geriatrics, 23(1), 660. https://doi: 10.1186/s12877-023-04396-x. Nogueira, A.J., & Ribeiro, M.T. (2023). "I'm Afraid If This Goes Wrong… What Will Become of Me?": The Psychological Experience of Grandparents in Pediatric Palliative Care. Healthcare (Basel, Switzerland), 11(17). https://doi.org/10.3390/healthcare11172391. Damman, M., Henkens, K., & Kalmijn, M. (2015). Missing Work After Retirement: The Role of Life Histories in the Retirement Adjustment Process. The Gerontologist, 55(5), 802-813. https://doi: 10.1093/geront/gnt169. Mannes, Z. L., Waxenberg, L. B., Cottler, L. B., Perlstein, W. M., Burrell, L. E., Ferguson, E. G., Edwards, M. E., & Ennis, N. (2019). Prevalence and Correlates of Psychological Distress among Retired Elite Athletes: A Systematic Review. International Review of Sport and Exercise Psychology, 12(1), 265-294. https:// doi: 10.1080/1750984X.2018.1469162. Aabo, M. G., Mølgaard, K., & Lassen, A. J. (2023). The worn-out syndrome: Uncertainties in late working life triggering retirement decisions. PLoS One, 18(3), e0282905.doi: 10.1371/journal.pone.0282905. Cohen, L., Shiovitz-Ezra, S., & Erlich, B. (2023). Support for Older Parents in Need in Europe: The Role of the Social Network and of Individual and Relational Characteristics. Innovation in Aging, 7(4), igad032. https://doi: 10.1093/geroni/igad032. Hoffman, Y., & Shrira, A. (2019). Variables Connecting Parental PTSD to Offspring Successful Aging: Parent-Child Role Reversal, Secondary Traumatization, and Depressive Symptoms. Frontiers in Psychiatry, 10, 718. https://doi.org/10.3389/fpsyt.2019. Earl, E.J., & Marais, D. (2023). The experience of intergenerational interactions and their influence on the mental health of older people living in residential care. PLoS One, 18(7), e0287369. https://doi: 10.1371/journal. Han, S., Guo, J., & Xiang, J. (2024). Is intergenerational care associated with depression in older adults?. Frontiers in Public Health, 12, 1325049. https://doi.org/10.3389. Zhang, J., Chandola, T., & Zhang, N. (2022). Understanding the longitudinal dynamics of rural–urban mental health disparities in later life in China. Aging & Mental Health, 27(7), 1419–1428. https://doi.org/10.1080/13607863.2022.2098912. Huang, W. (2024). Economic impact of retirement on the elderly population: A literature review. China Economic Journal, 38(2), 317–337.10.1080/17538963.2024.2425893. Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterial.docx Cite Share Download PDF Status: Published Journal Publication published 08 Oct, 2025 Read the published version in BMC Public Health → Version 1 posted Editorial decision: Revision requested 22 Jul, 2025 Reviews received at journal 20 Jul, 2025 Reviews received at journal 18 Jul, 2025 Reviews received at journal 17 Jul, 2025 Reviews received at journal 16 Jul, 2025 Reviews received at journal 14 Jul, 2025 Reviewers agreed at journal 12 Jul, 2025 Reviewers agreed at journal 10 Jul, 2025 Reviewers agreed at journal 09 Jul, 2025 Reviewers agreed at journal 08 Jul, 2025 Reviewers agreed at journal 07 Jul, 2025 Reviewers agreed at journal 07 Jul, 2025 Reviewers agreed at journal 07 Jul, 2025 Reviewers invited by journal 07 Jul, 2025 Editor invited by journal 24 Jun, 2025 Editor assigned by journal 22 May, 2025 Submission checks completed at journal 22 May, 2025 First submitted to journal 21 May, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6720729","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":462120984,"identity":"be1374cc-d7fb-46ca-a75d-fe164722da60","order_by":0,"name":"Yang Mou","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0klEQVRIiWNgGAWjYJACxg8G/+X4QayEAiK1MEtUMBtLNoC0GBBrDc8Z5sQNB0AsYrQYXDt8TEKyjS1x8/nViR8eGDDI84sdIKDldlqaRGEbj/G2G283SwAdZjhzdgIhLTlmQFskZLfdOLsBpCXB4DYxWnjbDBg3zzi7+QfxWnjOJChu4O/dRpwtkrfTkq0lKg4YS9zg3WaRYCBB2C98t5MP3vxgcECOv//s5ps/Kmzk+aUJaFE4AGNJgFVK4FcOAvINMBb/AdyqRsEoGAWjYGQDAGFkRx2oyVGFAAAAAElFTkSuQmCC","orcid":"","institution":"Kunming Preschool Education College","correspondingAuthor":true,"prefix":"","firstName":"Yang","middleName":"","lastName":"Mou","suffix":""},{"id":462120986,"identity":"428f91bb-697e-4930-bd94-1c37829cde27","order_by":1,"name":"Zijun Tan","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Zijun","middleName":"","lastName":"Tan","suffix":""},{"id":462120990,"identity":"61aa513f-8058-4bca-b23f-eb2b891227ce","order_by":2,"name":"Hongxian Nie","email":"","orcid":"","institution":"Kunming Preschool Education College","correspondingAuthor":false,"prefix":"","firstName":"Hongxian","middleName":"","lastName":"Nie","suffix":""}],"badges":[],"createdAt":"2025-05-22 03:38:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6720729/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6720729/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12889-025-24774-x","type":"published","date":"2025-10-08T15:57:23+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":93419665,"identity":"1e2dff62-6c58-4904-bf8d-2fdfe370c06a","added_by":"auto","created_at":"2025-10-13 16:05:35","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1240461,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6720729/v1/d5fb89a6-871a-4244-bbb4-6bae3b568e32.pdf"},{"id":83762383,"identity":"8f916d22-78be-4a36-88b9-270e72accf08","added_by":"auto","created_at":"2025-06-02 09:56:22","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":40023,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-6720729/v1/44196c5130fd04dcd84032c1.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"The Impact of Retirement and Grandchild Caregiving on Mental Health in China: The Role of Intergenerational Support","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eChina is undergoing a profound demographic shift. By the middle of this century, it is expected that nearly one-third of its citizens will be aged 60 or above.To address the growing economic strain posed by an aging labor force, policymakers have implemented delayed retirement strategies. While these measures aim to sustain fiscal balance, they may unintentionally increase mental burdens on older adults, particularly those who are also serving as caregivers for their grandchildren.\u003c/p\u003e\n\u003cp\u003eConcurrently, family life in China is being reshaped by rapid urbanization and internal migration. These social shifts have resulted in more grandparents stepping in as the main providers of childcare. Although this role can offer emotional rewards and strengthen family ties, it also brings with it significant physical and psychological challenges for elderly caregivers.\u003c/p\u003e\n\u003cp\u003ePast research has primarily treated retirement and caregiving as separate phenomena, often producing conflicting findings regarding their impact on mental health.However, few studies have examined how these two experiences interact, particularly in the context of China\u0026rsquo;s evolving economic policies and cultural norms.What remains underexplored is how stepping away from employment and assuming intensive caregiving duties intersect\u0026mdash;particularly against China\u0026rsquo;s specific economic pressures and social policies.Moreover, while intergenerational support holds cultural significance, its influence on the intersection of retirement, caregiving, and mental health remains insufficiently explored.\u003c/p\u003e\n\u003cp\u003eThis study seeks to fill these gaps by addressing the following research questions:\u003c/p\u003e\n\u003cp\u003e1.How does retirement status influence the mental health of older adults in China?\u003c/p\u003e\n\u003cp\u003e2.What is the impact of caregiving intensity for grandchildren on psychological well-being?\u003c/p\u003e\n\u003cp\u003e3.Can intergenerational support buffer the mental health burdens associated with retirement and caregiving?\u003c/p\u003e\n\u003cp\u003eUtilizing data from the China Health and Retirement Longitudinal Study (CHARLS) and an instrumental variable approach, this study provides empirical insights into the complex relationships between retirement, caregiving, and intergenerational support. Our findings offer valuable guidance for policymakers aiming to balance labor market objectives with the caregiving needs of families and the mental health of older populations.\u003c/p\u003e"},{"header":"2. Literature Review","content":"\u003cp\u003e\u003cstrong\u003e2.1Retirement and Mental Health: Contradictory Findings\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe relationship between retirement and mental health has been widely debated, yielding mixed results. Some studies suggest that retirement can enhance psychological well-being, particularly in the initial years, as retirees often experience relief from work-related stress, leading to improvements in self-esteem, life satisfaction, and overall mental health(\u003csup\u003e1\u003c/sup\u003e,\u003csup\u003e2\u003c/sup\u003e) . These benefits are most evident among those who retire voluntarily and view the transition positively (\u003csup\u003e3\u003c/sup\u003e).\u003c/p\u003e\n\u003cp\u003eWhile the positive effects of retirement may initially seem evident, they tend to diminish over time.Many individuals who remain retired for extended periods often grapple with emotional difficulties such as loneliness, a diminished sense of purpose, and social withdrawal\u0026mdash;each of which can take a toll on mental well-being (\u003csup\u003e4\u003c/sup\u003e,\u003csup\u003e5\u003c/sup\u003e). Notably, the mental health outcomes of retirement are closely linked to how that retirement comes about. People who exit the workforce involuntarily\u0026mdash;for reasons like sudden job loss or chronic illness\u0026mdash;tend to report greater levels of psychological distress, including heightened anxiety and depressive symptoms (\u003csup\u003e6\u003c/sup\u003e,\u003csup\u003e7\u003c/sup\u003e). In contrast, those who make the decision to retire on their own terms often fare better emotionally (\u003csup\u003e8\u003c/sup\u003e,\u003csup\u003e9\u003c/sup\u003e).\u003c/p\u003e\n\u003cp\u003eFinancial conditions add another layer of complexity. A lack of adequate income during retirement can intensify emotional struggles, particularly for those with minimal savings or unstable pensions (3,4). However, older adults who remain socially active and maintain strong interpersonal connections are often more resilient, underscoring the buffering effects of social engagement and community participation (\u003csup\u003e10\u003c/sup\u003e,6).\u003c/p\u003e\n\u003cp\u003eCultural expectations also shape the retirement experience. In contexts where professional life is deeply entwined with self-identity, stepping away from work may trigger a deep sense of disorientation or emotional void (5,\u003csup\u003e11\u003c/sup\u003e). Taken together, while retirement can offer relief and psychological benefits in the short term, its longer-term effects are shaped by a combination of voluntariness, economic stability, and opportunities for meaningful social connection (1,2,11).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2Grandchild Caregiving and Mental Health: The Impact of Intensity\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe extent to which grandparents are involved in childcare significantly shapes their emotional well-being. When caregiving becomes too demanding, many older adults report heightened stress, emotional fatigue, and even physical symptoms like disrupted sleep or recurring headaches (\u003csup\u003e12\u003c/sup\u003e). In contrast, those who provide care in more moderate or occasional roles often maintain better psychological balance. These findings suggest that there may be a critical point at which caregiving shifts from fulfilling to burdensome (\u003csup\u003e13\u003c/sup\u003e).\u003c/p\u003e\n\u003cp\u003eNonetheless, grandchild caregiving is not inherently detrimental. For many seniors, participating in their grandchildren\u0026rsquo;s lives offers emotional satisfaction, a renewed sense of purpose, and a meaningful role within the family unit. Feeling appreciated can help reduce feelings of loneliness or sadness and bolster psychological resilience (\u003csup\u003e14\u003c/sup\u003e). However, when this role dominates daily life, the demands can become overwhelming. Caregivers in these situations frequently report emotional strain, including chronic anxiety and persistent fatigue\u0026mdash;often intensified by ongoing concerns about their grandchildren\u0026rsquo;s future and a lack of personal time o r space (\u003csup\u003e15\u003c/sup\u003e,\u003csup\u003e16\u003c/sup\u003e).\u003c/p\u003e\n\u003cp\u003eIn such circumstances, access to strong social connections and effective personal coping strategies becomes especially important. These protective factors can act as emotional buffers, helping to mitigate the psychological costs of intense caregiving commitments.Grandparents who maintain strong relationships and receive emotional support from family and friends are more likely to report better mental health (\u003csup\u003e17\u003c/sup\u003e,\u003csup\u003e18\u003c/sup\u003e). Additionally, financial and reciprocal support from adult children can ease the psychological burden. When caregiving is coupled with both emotional and financial support from other family members, it helps reduce depressive symptoms and shifts caregiving from a solitary responsibility to a shared family duty(\u003csup\u003e19\u003c/sup\u003e).\u003c/p\u003e\n\u003cp\u003eWhile intergenerational support generally benefits caregivers, imbalances in these exchanges can lead to burnout. Open and honest communication between family members about caregiving roles, expectations, and boundaries is essential to ensure that caregiving remains a rewarding experience without leading to exhaustion (\u003csup\u003e20\u003c/sup\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.3Delayed Retirement and Its Psychological Impact\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDelayed retirement has become an increasingly important topic of research, yet its mental health effects are multifaceted and often contradictory. For some individuals, staying in the workforce provides social connections and a continued sense of purpose. However, for others, working beyond the typical retirement age can lead to heightened stress and a diminished sense of social status, especially when one\u0026rsquo;s identity is strongly tied to their professional role(\u003csup\u003e21\u003c/sup\u003e).\u003c/p\u003e\n\u003cp\u003eThe mental health outcomes associated with retirement are closely shaped by when and how individuals retire. Those who have time to anticipate and prepare for the transition generally navigate it more smoothly, often reporting greater emotional stability than peers forced into retirement due to layoffs or challenging work conditions(\u003csup\u003e22\u003c/sup\u003e,2).\u003c/p\u003e\n\u003cp\u003eAnother important factor is cognitive health. Remaining in intellectually stimulating roles may help preserve mental agility, but continued exposure to high-pressure environments can wear down cognitive reserves, potentially leading to mental fatigue later on (11).\u003c/p\u003e\n\u003cp\u003eFor some, postponing retirement brings financial relief. Yet for others\u0026mdash;especially those uncertain about their long-term income\u0026mdash;delayed exit from the workforce can worsen anxiety and emotional distress (\u003csup\u003e23\u003c/sup\u003e). With more people now choosing to extend their careers, there is an urgent need to reconsider retirement frameworks that accommodate diverse needs and avoid a one-size-fits-all approach (22).\u003c/p\u003e\n\u003cp\u003eUltimately, while staying in the workforce may offer certain financial or social benefits, it can also come at a psychological cost. To support older adults through this complex transition, flexible retirement options that consider personal and economic realities are increasingly vital\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.4The Role of Intergenerational Support in Mitigating Psychological Burdens\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSupport from younger generations can significantly lighten the emotional burden many older adults carry. Spending time with children or grandchildren\u0026mdash;whether through daily activities, shared meals, or simply conversation\u0026mdash;helps maintain a sense of connection that protects against social isolation, a key predictor of depression and anxiety in later life (\u003csup\u003e24\u003c/sup\u003e,\u003csup\u003e25\u003c/sup\u003e).In addition to emotional support, practical assistance such as help with everyday tasks and financial contributions can greatly alleviate the challenges of aging, especially for those facing physical limitations (\u003csup\u003e26\u003c/sup\u003e).\u003c/p\u003e\n\u003cp\u003eFinancial support from children is particularly important as older adults transition into retirement and face reduced income. This support alleviates economic stress and improves mental health outcomes (\u003csup\u003e27\u003c/sup\u003e,12). In addition to practical and financial support, caregiving roles can provide a sense of purpose, enhancing self-worth and reducing depression, especially when grandparents feel needed by their grandchildren (14). However, balance is crucial to prevent caregiver burnout. Effective communication between generations is key to ensuring caregiving remains fulfilling and does not lead to exhaustion (20).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.5Research Gaps and Future Directions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAlthough considerable research has been conducted on the individual effects of retirement and caregiving on mental health, studies exploring their combined effects remain limited, especially in China. Older adults often face the dual pressures of retirement and caregiving, yet little is understood about how these roles intersect and impact mental health. Future research should investigate the combined effects of retirement and caregiving, particularly within China\u0026rsquo;s rapidly evolving socio-economic context.\u003c/p\u003e\n\u003cp\u003eFurther studies on formal support services, such as respite care and community-based programs, could offer valuable insights into alleviating caregiver burden. Longitudinal studies would be instrumental in understanding the long-term mental health impacts of retirement and caregiving intensity.\u003c/p\u003e\n\u003cp\u003eAs China\u0026rsquo;s demographic landscape continues to shift and family structures evolve, more research is needed to understand how intergenerational support systems adapt. Future studies should explore how different types of intergenerational support influence mental health, considering cultural and policy factors.\u003c/p\u003e"},{"header":"3. Method","content":"\u003cp\u003e\u003cstrong\u003e3.1Sample Selection and Attrition\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study uses data from the 2018 China Health and Retirement Longitudinal Study (CHARLS), with an initial sample of 17,814 observations. After excluding samples with missing key variables, 7,825 valid observations remained. To address missing data, we employed the Multiple Imputation (MI) method to impute key variables, such as self-rated health (srh) and depressive symptoms (cesd10_std). The detailed process of multiple imputation is presented in Table 1.After further excluding observations with excessive missing data, the final sample for regression analysis comprised 2,657 individuals.\u003c/p\u003e\n\u003cp\u003eTable 1: Imputation Process for Multiple Imputed Datasets\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eStep\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eDescription\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eNotes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eStep 1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eRegistering variables for imputation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eage, srh, cesd10_std marked for imputation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eStep 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003ePerforming multivariate normal imputation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e5 imputed datasets generated\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eStep 3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eChecking convergence and stability\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eConvergence achieved with EM optimization\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eStep 4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eAnalyzing imputed datasets\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eImputed datasets used for further analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003e3.2Results of Sample Loss Comparison\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA comparison of the excluded and retained samples based on key characteristics, including age, self-rated health (srh), and depressive symptoms (cesd10_std), showed no significant differences in depressive symptoms (p = 0.4898), indicating that sample attrition did not introduce substantial bias. Descriptive statistics and p-values for this comparison are reported in Table 2.However, older individuals and those with poorer health were more likely to be excluded (p \u0026lt; 0.001).\u003c/p\u003e\n\u003cp\u003eTable 2: Sample Loss Comparison\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"590\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eDeleted Sample (N = 43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eRetained Sample (N = 7,782)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003ep-value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e67.07 (\u0026plusmn; 6.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e61.59 (\u0026plusmn; 7.56)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eSelf-Rated Health (srh)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e2.73 (\u0026plusmn; 1.06)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e3.00 (\u0026plusmn; 1.03)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e0.0876\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eDepressive Score (cesd10_std)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e0.45 (\u0026plusmn; 1.36)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e0.05 (\u0026plusmn; 1.01)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e0.4898\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eNote: *** p \u0026lt; 0.01, ** p \u0026lt; 0.05, * p \u0026lt; 0.10.\u003c/p\u003e\n\u003cp\u003eThe imputation process generated five datasets, ensuring that the results accurately reflected the full population. Sensitivity analysis confirmed the consistency of the results, comparing multiple imputation with listwise deletion methods.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.3Statistical Analysis and Model Specification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRegression analyses were performed using Ordinary Least Squares (OLS) and Instrumental Variables (IV) methods to examine the relationships between retirement, caregiving, and mental health. Key variables, such as retirement status, caregiving intensity, and mental health outcomes (self-rated health and depressive symptoms), were included as explanatory variables in the models. To address potential endogeneity, we employed instrumental variables, specifically urban-rural residence and retirement age, as instruments for retirement status.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.4Variable Definitions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.4.1Dependent Variable\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe primary outcome variable is mental health, measured using the standardized CES-D 10 score. This scale assesses depressive symptoms in older adults, including feelings of sadness, anxiety, and loss of interest. The score is standardized, with a range of -1.325 to 3.24, where higher values indicate more severe depressive symptoms.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.4.2Independent Variables\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRetirement Status (Retire): Retirement Status (Retire): A binary variable indicating whether the individual is retired (1 = retired, 0 = not retired). This variable captures the respondent\u0026apos;s labor market exit status and serves as a key explanatory variable.\u003c/p\u003e\n\u003cp\u003eGrandchild Caregiving (WPC):A binary variable representing whether the respondent provides caregiving for grandchildren (1 = provides care, 0 = does not provide care). This variable reflects the respondent\u0026apos;s involvement in family caregiving roles.\u003c/p\u003e\n\u003cp\u003eCaregiving Intensity (HTSPCH): A continuous variable indicating the number of hours per week the respondent spends caring for grandchildren. The variable ranges from 0 to 336 hours, with extreme values above 168 hours capped at 168 to maintain data plausibility and reduce outlier effects.\u003c/p\u003e\n\u003cp\u003eIntergenerational Support\u003c/p\u003e\n\u003cp\u003eParent-Child Relationship Satisfaction (Sati_Child): An ordinal variable measuring satisfaction with the parent-child relationship, ranging from 1 (very dissatisfied) to 5 (very satisfied), reflecting the intensity of emotional support.\u003c/p\u003e\n\u003cp\u003eFinancial Support\u003c/p\u003e\n\u003cp\u003eFinancial Support to Children (Log_Fcamt): A continuous variable representing the amount of financial support provided to children, measured as the log-transformed financial amount.\u003c/p\u003e\n\u003cp\u003eFinancial Support from Children (Log_Tcamt):A continuous variable representing the amount of financial support received from children, measured as the log-transformed financial amount.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.4.3Control Variables\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSquared Age Term (c.age#c.age): The square of the individual\u0026rsquo;s age (c.age) is included as a continuous variable to account for non-linear age effects in the regression models.\u003c/p\u003e\n\u003cp\u003eSelf-Rated Health (SRH): An ordinal variable reflecting the individual\u0026rsquo;s subjective health perception, with a scale from 1 (very poor) to 5 (excellent).\u003c/p\u003e\n\u003cp\u003ePension Status (Pension):A binary variable indicating whether the individual receives a pension (1 = receives pension, 0 = does not receive pension).\u003c/p\u003e\n\u003cp\u003eMarital Status (Marry):A binary variable indicating whether the individual is married (1 = married, 0 = not married).\u003c/p\u003e\n\u003cp\u003eFamily Size (Family_Size):A continuous variable indicating the number of household members, ranging from 1 to 13.\u003c/p\u003e\n\u003cp\u003eNumber of Grandchildren under 16 (grandchildu16): A continuous variable indicating the number of grandchildren under the age of 16, with a range from 0 to 14.\u003c/p\u003e\n\u003cp\u003eThe definitions and measurement methods of all variables used in the analysis are summarized in Table 3.\u003c/p\u003e\n\u003cp\u003eTable 3: Variable Definitions\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eDefinition\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eMeasurement\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003ecesd10_std\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eStandardized CES-D 10 score, a measure of depressive symptoms in older adults. Assesses symptoms such as sadness, anxiety, and loss of interest.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eContinuous variable, range: -1.325 to 3.24. Higher scores indicate more severe depressive symptoms.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003eretire\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eIndicator variable for whether the individual is retired.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eBinary variable (1 = retired, 0 = not retired).\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003esati_child\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eSatisfaction with the parent-child relationship.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eOrdinal variable (1 = very dissatisfied, 5 = very satisfied).\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003ewpc\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eIndicates whether the individual provides caregiving to grandchildren.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eBinary variable (1 = provides care, 0 = does not provide care).\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003ehtspch\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eNumber of hours per week spent providing caregiving to grandchildren, reflecting caregiving intensity.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eContinuous variable, range: 0 to 336 hours. Extreme values adjusted to 168 hours if exceeding this threshold.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003esrh\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eSelf-rated health status, reflecting the individual\u0026rsquo;s subjective health perception.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eOrdinal variable (1 = very poor, 5 = excellent).\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003epension\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eIndicates whether the individual receives a pension.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eBinary variable (1 = receives pension, 0 = does not receive pension).\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003elog_fcamt\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eLog-transformed amount of financial support provided to children.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eContinuous variable, log-transformed financial support amount.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003elog_tcamt\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eLog-transformed amount of financial support received from children.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eContinuous variable, log-transformed financial support amount.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003ec.age\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eThe individual\u0026rsquo;s age.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eContinuous variable, measured in years.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003ec.age#c.age\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eSquared term of age to account for non-linear age effects.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eContinuous variable, the square of c.age.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003emarry\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eMarital status.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eBinary variable (1 = married, 0 = not married).\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003efamily_size\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eNumber of household members.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eContinuous variable, range: 1 to 13.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 86px;\"\u003e\n \u003cp\u003egrandchildu16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 289px;\"\u003e\n \u003cp\u003eNumber of grandchildren under the age of 16.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 193px;\"\u003e\n \u003cp\u003eContinuous variable, range: 0 to 14.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003e3.5Empirical Strategy\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo address potential endogeneity concerns in the relationship between retirement and mental health, we employ an Instrumental Variables (IV) approach. The instrumental variables used in this study are urban/rural household registration and retirement age policy, which are both exogenous to individuals\u0026apos; health and psychological factors but influence retirement timing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.5.1Instrumental Variables\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUrban/Rural Household Registration (Urban/Rural):The urban-rural divide in China has a notable impact on access to critical services like pensions and healthcare. However, this divide does not directly affect mental health outcomes, except through its influence on retirement status. Therefore, urban-rural classification can be used as a valid instrumental variable for retirement status. Zhang et al. (2022) have pointed out that disparities in mental well-being between older adults in rural and urban areas stem mainly from unequal access to financial resources and medical services, rather than the urban-rural status itself(\u003csup\u003e28\u003c/sup\u003e). This observation reinforces the validity of using household registration type as an instrument to isolate the causal effect of retirement on psychological outcomes.\u003c/p\u003e\n\u003cp\u003e2.Retirement Age Policy (Retirement_Age2):China\u0026rsquo;s official retirement regulations stipulate different age thresholds for men and women\u0026mdash;62 for men and 58 for women. These age-based rules create an exogenous source of variation in retirement behavior that is not tied to individuals\u0026rsquo; mental health or personal choices. In this analysis, we employ the male retirement age (62) as an instrumental variable. By that point, most women in the sample have already exited the workforce, which helps isolate the effect of retirement timing for men without conflating it with gender-based differences.This policy-driven variation offers a unique analytical advantage: it influences when people retire, but not their psychological condition directly. Huang (2024) corroborates this logic, arguing that while mandatory retirement policies strongly determine retirement timing, they remain unrelated to mental health outcomes. This makes the male retirement age a suitable instrument for examining causal links between retirement and mental well-being(\u003csup\u003e29\u003c/sup\u003e).\u003c/p\u003e\n\u003cp\u003e3.5.2First-Stage Regression and Instrument Validity\u003c/p\u003e\n\u003cp\u003eWe conduct a first-stage regression of retirement on the instruments (urban/rural registration and retirement age policy). The regression results show that both instruments are strongly significant in explaining retirement:\u003c/p\u003e\n\u003cp\u003eUrban/Rural: Coefficient = - 0.4383, p-value \u0026lt; 0.001.\u003c/p\u003e\n\u003cp\u003eRetirement Age Policy: Coefficient = -0.1145, p-value \u0026lt; 0.001.\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003c/p\u003e\n\u003cp\u003eWhere Urban/Rural represents the urban/rural household registration, Retirement Age Policy refers to the retirement age threshold of 62 years for men (used as the instrument for retirement status), X_i includes control variables (such as age, marital status, etc.), and \u0026epsilon;_i is the error term.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEndogeneity Test\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo check for endogeneity in the retirement variable, we perform the Durbin-Wu-Hausman test, which yields a p-value = 0.0003. This result confirms that retirement is endogenous, thus necessitating the use of the IV approach to address potential bias in the relationship between retirement and mental health.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOveridentification Test\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe conduct the Hansen J test to test the exogeneity of our instruments. The test results show a p-value = 0.6866, indicating that both instruments are valid and do not directly affect mental health, confirming their appropriateness for this study.Table 4 presents the results of the first-stage regression, demonstrating the strong significance of both instruments in predicting retirement status, as well as the relevant test statistics for instrument validity.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.5.3Second-Stage Regression\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the second stage, we estimate the effect of retirement on mental health (measured by the CES-D score), using the predicted retirement status from the first-stage regression as the instrumental variable. The IV regression results reveal a significant negative effect of retirement on mental health, showing that retirement contributes to a decrease in depressive symptoms.\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Where Retire is the predicted retirement status from the first stage, Caregiving Intensity is the number of caregiving hours per week, Parent-Child Satisfaction is the measure of satisfaction with the parent-child relationship, and X_i represents control variables.\u003c/p\u003e\n\u003cp\u003eThe IV regression results indicate a significant reduction in depressive symptoms due to retirement, with a coefficient of -0.6409 and p \u0026lt; 0.001. This finding suggests that retirement has a positive impact on mental health by alleviating depressive symptoms.\u003c/p\u003e\n\u003cp\u003eThe Durbin-Wu-Hausman test confirms that retirement is endogenous, while the Hansen J test supports the validity of our instruments, ensuring that urban-rural registration and retirement age policy are appropriate instruments for examining the impact of retirement on mental health.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical Methods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll statistical analyses were conducted using Stata 16. Standard errors were clustered at the individual level to account for potential intra-individual correlation. Robust standard errors are reported throughout the analysis.\u003c/p\u003e\n\u003cp\u003eTable 4: First-Stage Regression Results\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"590\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 106px;\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003eCoefficient\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003eStandard Error\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 94px;\"\u003e\n \u003cp\u003eZ-Statistic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 67px;\"\u003e\n \u003cp\u003ep-Value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e95% Confidence Interval\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 106px;\"\u003e\n \u003cp\u003eUrban/Rural\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003e-0.4383\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.0179\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 94px;\"\u003e\n \u003cp\u003e-24.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 67px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e[-0.4734, -0.4032]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 106px;\"\u003e\n \u003cp\u003eRetirement Age Policy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003e-0.1145\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.0261\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 94px;\"\u003e\n \u003cp\u003e-4.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 67px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e[-0.1657, -0.0634]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 106px;\"\u003e\n \u003cp\u003eF-statistic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003e326.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 94px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 67px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 106px;\"\u003e\n \u003cp\u003eHansen J Test\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003e0.163\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 94px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 67px;\"\u003e\n \u003cp\u003e0.6866\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNote: *** p \u0026lt; 0.01, ** p \u0026lt; 0.05, * p \u0026lt; 0.10.\u003c/p\u003e"},{"header":"4 Results","content":"\u003cp\u003e\u003cstrong\u003e4.1Descriptive Statistics\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTable 5 presents the descriptive statistics for the variables included in the analysis. The initial sample consisted of 7,825 respondents. After excluding individuals with missing data or inconsistent responses, the final sample for regression analysis was reduced to 2,657 respondents.\u003c/p\u003e\n\u003cp\u003eThe mean age of the sample is 61.6 years, with ages ranging from 40 to 75. Among the 7,825 respondents, 14% are retired, and 53.5% report providing care for grandchildren. The average caregiving intensity is 53.2 hours per week, indicating a significant caregiving burden for many older adults.\u003c/p\u003e\n\u003cp\u003eIt is important to note that the descriptive statistics reflect the full sample of 7,825 respondents, while the regression analysis is based on the reduced sample of 2,657 after excluding observations with missing values for key variables.\u003c/p\u003e\n\u003cp\u003eThe average CES-D 10 score is 0.047, indicating a moderate level of depressive symptoms across the sample. Self-rated health, measured on a 5-point scale, has an average score of 3.0, suggesting that respondents generally rate their health as moderate. 64.1% of the respondents receive a pension, while 35.9% do not.\u003c/p\u003e\n\u003cp\u003eFinancial transfers between family members are notable: the average value of financial support provided to children (log-transformed) is 7.89, and the average value of financial support received from children is 7.43. The sample is nearly balanced in terms of gender, with 46.8% of respondents being female. The average family size is 2.82 members, and 81% of respondents are married.\u003c/p\u003e\n\u003cp\u003eThese descriptive statistics provide an overview of the sample\u0026apos;s demographic and socio-economic characteristics, offering essential context for understanding the relationship between retirement, caregiving, and mental health among older adults in China.\u003c/p\u003e\n\u003cp\u003eTable 5: Descriptive Statistics of the Sample\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003eObs\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003eStd. Dev.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003eMin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003eMax\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003ecesd10_std\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e7,785\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0.047\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1.0116\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e-1.3251\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e3.2404\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003eretire\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e7,773\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0.347\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003esati_child\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e7,825\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e3.5972\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0.7396\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003ewpc\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e7,825\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0.5355\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0.4988\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003ehtspch\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e7,825\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e46.5507\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e60.7963\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e168\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003esrh\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e7,822\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e3.0015\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1.0266\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003epension\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e7,022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0.6413\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0.4797\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003elog_fcamt\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e6,205\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e7.8905\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1.3835\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e12.6115\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003elog_tcamt\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e3,292\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e7.4312\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1.9129\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1.0986\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e13.9995\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003eage\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e7,825\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e61.6157\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e7.5615\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e75\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003egender\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e7,825\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0.4682\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0.499\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003eedu\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e7,825\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e2.0235\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1.0239\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003emarry\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e7,825\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0.8096\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0.3927\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003efamily_size\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e7,825\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e2.822\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1.5504\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e13\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003egrandchildu16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e7,825\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0.2276\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0.912\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e14\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003e4.2Regression Results\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTable 6 presents the regression results examining the relationship between retirement, grandchild caregiving, intergenerational support, and mental health outcomes. The effects of retirement, caregiving intensity, and intergenerational support on mental health are estimated, controlling for various socio-demographic variables.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.2.1OLS Regression Results\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe Ordinary Least Squares (OLS) regression results show a significant negative association between retirement and depressive symptoms (\u0026beta;= - 0.276, p \u0026lt; 0.001), indicating that retirement is linked to a substantial reduction in depressive symptoms. Satisfaction with parent-child relationships is also negatively associated with depressive symptoms (\u0026beta;= -0.210, p \u0026lt; 0.001), suggesting that greater satisfaction with these relationships correlates with fewer depressive symptoms.\u003c/p\u003e\n\u003cp\u003eGrandchild caregiving is associated with improved mental health outcomes, with caregiving linked to a reduction in depressive symptoms (\u0026beta;= -0.113, p = 0.010). However, this effect is weaker compared to retirement and parent-child relationship satisfaction.\u003c/p\u003e\n\u003cp\u003eCaregiving intensity shows a contrasting pattern. Higher caregiving intensity is associated with increased depressive symptoms (\u0026beta; = 0.00094, p= 0.001), suggesting that while caregiving for grandchildren can enhance emotional well-being, more intensive caregiving is linked to greater psychological strain.\u003c/p\u003e\n\u003cp\u003eSelf-rated health (SRH) is strongly associated with depressive symptoms, with poorer self-reported health corresponding to higher levels of depressive symptoms (\u0026beta; = -0.304, p \u0026lt; 0.001). Financial support to children is negatively associated with depressive symptoms (\u0026beta;= -0.032, p= 0.001), indicating that older adults who provide financial support to their children experience fewer depressive symptoms. However, financial support received from children does not significantly affect mental health (\u0026beta;= -0.013, p = 0.306).Detailed OLS regression results, including coefficient estimates, standard errors, and significance levels, are presented in Table 6.\u003c/p\u003e\n\u003cp\u003eTable6: Regression Results - OLS Regression\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eCoefficient\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eStandard Error\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003et-Statistic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 87px;\"\u003e\n \u003cp\u003ep-Value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 110px;\"\u003e\n \u003cp\u003e95% Confidence Interval\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eRetire\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-0.2756\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.0479\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-5.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 87px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 110px;\"\u003e\n \u003cp\u003e[-0.3697,-0.1816]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eSati_Child\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-0.2106\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.0241\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-8.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 87px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 110px;\"\u003e\n \u003cp\u003e[-0.2579, -0.1633]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eWPC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-0.1198\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.0476\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-2.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 87px;\"\u003e\n \u003cp\u003e0.012\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 110px;\"\u003e\n \u003cp\u003e[-0.2132, -0.0264]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eHTSPCH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.0011\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.0004\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e2.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 87px;\"\u003e\n \u003cp\u003e0.004\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 110px;\"\u003e\n \u003cp\u003e[0.0004, 0.0019]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eSRH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-0.304\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.0168\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-18.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 87px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 110px;\"\u003e\n \u003cp\u003e[-0.3369, -0.2711]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003ePension\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-0.0151\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.0491\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 87px;\"\u003e\n \u003cp\u003e0.758\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 110px;\"\u003e\n \u003cp\u003e[-0.1114, 0.0812]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eLog_Fcamt\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-0.0125\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.0128\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 87px;\"\u003e\n \u003cp\u003e0.329\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 110px;\"\u003e\n \u003cp\u003e[-0.0376, 0.0126]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eLog_Tcamt\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-0.0318\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.0098\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-3.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 87px;\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 110px;\"\u003e\n \u003cp\u003e[-0.0510, -0.0126]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eAge^2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-0.0001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.00003\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-2.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 87px;\"\u003e\n \u003cp\u003e0.005\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 110px;\"\u003e\n \u003cp\u003e[-0.0001, -0.00002]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eMarried\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-0.2156\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.0511\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-4.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 87px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 110px;\"\u003e\n \u003cp\u003e[-0.3158, -0.1154]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eFamily Size\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-0.0396\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.0135\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e-2.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 87px;\"\u003e\n \u003cp\u003e0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 110px;\"\u003e\n \u003cp\u003e[-0.0661, -0.0131]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eGrandchildu16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.0364\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.0234\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e1.55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 87px;\"\u003e\n \u003cp\u003e0.121\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 110px;\"\u003e\n \u003cp\u003e[-0.0096, 0.0824]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003eConstant\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e2.615\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e0.177\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 98px;\"\u003e\n \u003cp\u003e14.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 87px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 110px;\"\u003e\n \u003cp\u003e[2.2679, 2.9620]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNote: *** p \u0026lt; 0.01, ** p \u0026lt; 0.05, * p \u0026lt; 0.10.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.2.2Instrumental variable regression results\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo address endogeneity concerns, we conducted Instrumental Variable (IV) regression using urban-rural registration and mandatory retirement age as instruments for retirement status. The IV regression results largely confirm the findings from the OLS model.Financial support to children continues to show a significant negative association with depressive symptoms (\u0026beta; = -0.023, p = 0.031). In contrast, financial support received from children does not significantly affect mental health outcomes (\u0026beta;= -0.014, p = 0.268).\u003c/p\u003e\n\u003cp\u003eThe Instrumental Variable (IV) regression results largely confirm the findings from the OLS model. Retirement remains significantly associated with improved mental health (\u0026beta; = -0.435, p \u0026lt; 0.001), consistent with the OLS results. Parent-child satisfaction continues to have a protective effect on mental health (\u0026beta;= -0.209, p \u0026lt; 0.001), and grandchild caregiving is linked to fewer depressive symptoms (\u0026beta;= -0.109, p = 0.012). However, caregiving intensity remains a significant risk factor for mental health deterioration (\u0026beta; = 0.00094, p = 0.001). Financial support to children continues to show a significant negative association with depressive symptoms (\u0026beta; = -0.023, p= 0.031), whereas financial support received from children does not significantly affect mental health outcomes (\u0026beta; = -0.014, p = 0.268).The full results of the IV regression, including coefficient estimates, standard errors, and confidence intervals, are presented in Table 7.\u003c/p\u003e\n\u003cp\u003eTable 7: Regression Results - IV Regression\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"590\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 112px;\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003eCoefficient\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003eStandard Error\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003eZ-Statistic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003ep-Value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e95% Confidence Interval\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 112px;\"\u003e\n \u003cp\u003eRetire\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e-0.4352\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.0886\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e-4.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e[-0.6088, -0.2616]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 112px;\"\u003e\n \u003cp\u003eSati_Child\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e-0.2093\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.0271\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e-7.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e[-0.2624, -0.1561]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 112px;\"\u003e\n \u003cp\u003eWPC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e-0.1149\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.0465\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e-2.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.013\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e[-0.2060, -0.0238]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 112px;\"\u003e\n \u003cp\u003eHTSPCH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e0.0011\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.0004\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e3.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e[0.0004, 0.0019]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 112px;\"\u003e\n \u003cp\u003eSRH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e-0.3017\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.0165\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e-18.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e[-0.3342, -0.2693]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 112px;\"\u003e\n \u003cp\u003ePension\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e0.0235\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.0533\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e[-0.0810, 0.1280]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 112px;\"\u003e\n \u003cp\u003eLog_Fcamt\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e-0.0136\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.0128\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e-1.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.288\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e[-0.0387, 0.0115]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 112px;\"\u003e\n \u003cp\u003eLog_Tcamt\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e-0.023\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.0105\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e-2.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.029\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e[-0.0436, -0.0023]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 112px;\"\u003e\n \u003cp\u003eAge^2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e-0.0001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.00003\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e-2.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.013\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e[-0.0001, -0.00002]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 112px;\"\u003e\n \u003cp\u003eMarried\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e-0.2159\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.0548\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e-3.94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e[-0.3234, -0.1085]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 112px;\"\u003e\n \u003cp\u003eFamily Size\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e-0.0411\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.013\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e-3.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e[-0.0665, -0.0157]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 112px;\"\u003e\n \u003cp\u003eGrandchildu16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e0.0302\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.0262\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e1.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e0.249\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e[-0.0211, 0.0816]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 112px;\"\u003e\n \u003cp\u003eConstant\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e2.5244\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e0.1887\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 97px;\"\u003e\n \u003cp\u003e13.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 75px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e[2.1546, 2.8943]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eNote: *** p \u0026lt; 0.01, ** p \u0026lt; 0.05, * p \u0026lt; 0.10.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.3Robustness Checks\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo test the robustness of our findings, we added additional control variables, such as gender and education level, to the regression model. The results remained consistent with the main model, showing no significant changes. Specifically, retirement continued to be significantly negatively associated with depressive symptoms (\u0026beta; = -0.355, p \u0026lt; 0.001). Similarly, parent-child relationship satisfaction (\u0026beta; = -0.219, p \u0026lt; 0.001) and grandchild caregiving (\u0026beta; = -0.103, p = 0.015) maintained their significant protective effects. In contrast, caregiving intensity remained significantly associated with increased depressive symptoms (\u0026beta; = 0.00100, p \u0026lt; 0.001), indicating that excessive caregiving exacerbates mental health challenges.\u003c/p\u003e\n\u003cp\u003eAdditionally, both gender and education level were found to influence mental health outcomes. Female respondents reported fewer depressive symptoms (\u0026beta; = -0.267, p \u0026lt; 0.001), while higher education attainment was associated with better mental health (\u0026beta; = -0.058, p = 0.005). This suggests that higher education may improve mental health by increasing social participation and access to resources.\u003c/p\u003e\n\u003cp\u003eTo further examine the consistency of our results across different demographic groups, we conducted heterogeneity analyses based on gender and urban-rural residence. While some background differences were observed, the regression results showed no significant variations, indicating that the relationships between retirement, caregiving, and mental health are generally consistent across these groups. This suggests that our conclusions are broadly applicable to different subgroups.\u003c/p\u003e\n\u003cp\u003eNonetheless, future research could explore the impact of these variables within specific groups using larger subgroup samples, which would provide further insights into the mechanisms through which retirement and caregiving influence mental health.The detailed results of the robustness checks, including all additional control variables, are presented in Table 8.\u003c/p\u003e\n\u003cp\u003eTable 8: Robustness Check Results\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"594\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003eCoefficient\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003eStd. Error\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003eZ-Statistic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003ep-value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e95% Confidence Interval\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eRetirement Status (Instrumented)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e-0.355***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.101\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e-3.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[-0.552, -0.158]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eParent-Child Satisfaction\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e-0.219***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.027\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e-8.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[-0.271, -0.167]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eGrandchild Caregiving\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e-0.103*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.042\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e-2.46\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e0.015\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[-0.186, -0.020]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eCaregiving Intensity (hrs/week)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e0.00100***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.00027\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e3.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[0.00047, 0.00153]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eSelf-Rated Health\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e-0.292***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.016\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e-18.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[-0.323, -0.261]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eFinancial Support to Children\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e-0.018\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e-1.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e0.089\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[-0.037, 0.001]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eFinancial Support from Children\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e-0.013\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.013\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e-0.99\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e0.301\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[-0.038, 0.013]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003ePension Status\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e-0.007\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.053\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e-0.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e0.887\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[-0.111, 0.096]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eAge Squared\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e-0.000042\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.000028\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e-1.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e0.132\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[-0.000097, 0.000012]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eGender\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e-0.267***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.036\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e-7.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[-0.338, -0.197]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eEducation Level\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e-0.058**\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.021\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e-2.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e0.005\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[-0.099, -0.018]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eMarital Status\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e-0.135*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.055\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e-2.46\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e0.015\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[-0.244, -0.027]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eFamily Size\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e-0.039**\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.013\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e-2.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[-0.064, -0.014]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eNumber of Grandchildren \u0026lt;16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e0.011\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.025\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e0.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e0.651\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[-0.037, 0.059]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 159px;\"\u003e\n \u003cp\u003eConstant\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e2.571***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.194\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 82px;\"\u003e\n \u003cp\u003e13.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 59px;\"\u003e\n \u003cp\u003e\u0026lt; 0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 124px;\"\u003e\n \u003cp\u003e[2.185, 2.948]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNote: *** p \u0026lt; 0.01, ** p \u0026lt; 0.05, * p \u0026lt; 0.10.\u003c/p\u003e"},{"header":"5 Discussion","content":"\u003cp\u003e\u003cstrong\u003e5.1Overview of Main Findings\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe key findings of this study indicate that retirement, satisfaction with parent-child relationships, and caregiving intensity significantly impact the mental health of older adults. Specifically, retirement is negatively associated with mental health, suggesting that it may alleviate mental health burdens among older adults. In contrast, higher caregiving intensity is negatively correlated with mental health, indicating that excessive caregiving can exacerbate psychological stress. Additionally, satisfaction with parent-child relationships has a protective effect on mental health, highlighting the importance of family support for well-being.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.2Discussion of Regression Model Comparisons\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo validate the robustness of our findings, we employed three different regression models:Ordinary Least Squares (OLS) Regression, Instrumental Variables (IV) Regression, and Multiple Imputation Regression. These models allowed us to examine the relationship between retirement, caregiving, and mental health from multiple perspectives, addressing concerns such as endogeneity and missing data.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.2.1Comparison between OLS and IV Regression\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the OLS Regression, we observed a significant negative relationship between retirement and mental health (\u0026beta; = -0.372, p \u0026lt; 0.001), suggesting that retirement improves mental health in older adults. However, this result may be subject to endogeneity issues, such as reverse causality or omitted variable bias (e.g., health status or social support). To overcome this issue, we applied an Instrumental Variables (IV) approach, utilizing urban-rural household registration status and mandatory retirement age policies as instrumental variables.The IV results confirmed a stronger negative effect of retirement on mental health (\u0026beta; = -0.572, p \u0026lt; 0.001), consistent with the OLS results but with a larger magnitude. This indicates that the IV approach yields a more cautious and dependable estimate of the causal link between retirement and mental health.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.2.2Impact of Multiple Imputation Regression\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBecause some variables contained missing values, we applied multiple imputation techniques to address the data gaps, resulting in five completed datasets. Analysis using these imputed datasets confirmed that retirement continued to have a significant negative association with mental health (\u0026beta; = -0.357, p \u0026lt; 0.001).Although there were slight variations in the coefficients compared to the OLS regression, the imputation process had minimal impact on the results, further enhancing the robustness of the estimates.Table 9 presents a side-by-side comparison of the key regression results across OLS, IV, and multiple imputation models, demonstrating the consistency and robustness of our main findings.\u003c/p\u003e\n\u003cp\u003eTable 9: Comparison of Regression Results\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eOLS Regression (Complete Case)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eIV Regression\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eMultiple Imputation Regression\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eRetirement (retire)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.372 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.572 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.357 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eSatisfaction with Child Relationships (sati_child)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.196 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.198 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.200 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eCaregiving Intensity (wpc)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.099 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.090 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.106 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eSelf-Rated Health (srh)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.363 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.359 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.349 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003ePension (pension)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.049 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.009 (p = 0.543)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.049 (p = 0.082)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eFinancial Support from Parents (log_fcamt)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.041 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.041 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.034 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eFinancial Support from Children (log_tcamt)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.012 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.005 (p = 0.168)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e-0.016 (p = 0.080)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003eConstant (_cons)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e2.416 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e2.349 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 148px;\"\u003e\n \u003cp\u003e2.340 (p \u0026lt; 0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNote: *** p \u0026lt; 0.01, ** p \u0026lt; 0.05, * p \u0026lt; 0.10.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.4Rationale for Model Selection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDifferent regression models offer unique advantages in addressing various data issues:\u003c/p\u003e\n\u003cp\u003eOLS Regression provides basic estimates of relationships between variables. However, it assumes no endogeneity, which may introduce bias when analyzing causal relationships, particularly between retirement and mental health.\u003c/p\u003e\n\u003cp\u003eIV Regression addresses endogeneity by using valid instruments\u0026mdash;urban-rural differences and the retirement age policy\u0026mdash;to generate more reliable causal estimates. These instruments influence retirement decisions but do not directly affect mental health, avoiding biases inherent in OLS regression.\u003c/p\u003e\n\u003cp\u003eMultiple Imputation Regression deals with missing data by imputing missing values, reducing bias and improving sample completeness. While the coefficients from multiple imputation are similar to those from OLS, the imputation process strengthens the reliability of the analysis.\u003c/p\u003e\n\u003cp\u003eAlthough the coefficients slightly vary across models, all consistently show that retirement negatively affects mental health. We consider IV regression the most appropriate model, as it effectively addresses endogeneity and provides more accurate causal inferences, while Multiple Imputation Regression ensures the robustness and completeness of the analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.5Robustness of Results\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBy comparing different regression models, we have validated the robustness of our findings. Despite slight variations in the coefficients, the main result\u0026mdash;that retirement significantly negatively affects mental health\u0026mdash;remains consistent across all models. This consistency reinforces the reliability of our results, suggesting that retirement plays a crucial role in improving mental health among older adults in China.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.6Policy Recommendations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.6.1Specific and Actionable Recommendations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo better support older adults who shoulder the demands of caregiving\u0026mdash;particularly for their grandchildren\u0026mdash;local governments should consider introducing targeted financial and community-based interventions. One practical step would be to offer a monthly subsidy of \u0026yen;500 per child to help cover the costs of daycare services. This could provide much-needed relief for families where grandparents spend over 40 hours each week on childcare responsibilities. In fact, pilot programs in cities like Shanghai have already shown that such financial support, coupled with accessible, high-quality childcare, can ease the strain on aging caregivers.\u003c/p\u003e\n\u003cp\u003eBeyond financial aid, municipal authorities could develop neighborhood-level caregiver support hubs. These might include practical training programs, temporary respite care options, and psychological counseling services. Dedicated funding for these initiatives could help create spaces where caregivers receive both emotional and logistical support. Additionally, tax breaks or direct subsidies for family caregivers may help them better manage the competing pressures of work and caregiving\u0026mdash;a balance that is increasingly difficult to maintain without institutional support.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.6.2International Experiences as Reference\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLessons from other countries can offer valuable guidance for designing effective caregiving support systems. In Japan, the Long-Term Care Insurance program provides extensive resources to elderly individuals and their families who are involved in intensive caregiving arrangements. This approach has demonstrated measurable improvements in the well-being of both caregivers and care recipients. Given China\u0026rsquo;s aging demographics, drawing inspiration from such structured systems may be a promising direction.\u003c/p\u003e\n\u003cp\u003eThe Nordic countries present another compelling model. Their integrated approach combines public funding, community services, and in-home care, ensuring that older adults receive appropriate support while caregivers benefit from workplace flexibility and strong institutional backing. One of the most notable features of this model is its emphasis on universal accessibility. Adapting elements of this system could help Chinese local governments build a more inclusive, sustainable, and family-centered framework for elder care and caregiver assistance.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.6.3Policy Implementation and Evaluation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBudget Allocation:Local authorities should allocate resources to support community daycare subsidies, caregiver support programs, and long-term care initiatives, with funding distributed based on regional demand to ensure fairness and accessibility. Additional allocations should cover caregiver training activities and the development of caregiver resource centers in community settings.\u003c/p\u003e\n\u003cp\u003eMonitoring and Evaluation:Continuous evaluation is essential to gauge policy effectiveness. Local governments should perform annual assessments to measure caregiver satisfaction, monitor reductions in caregiving-related financial stress, and evaluate improvements in older adults\u0026apos; overall well-being. Regular feedback will allow policymakers to refine their strategies to effectively address caregiving challenges and enhance quality of life.\u003c/p\u003e\n\u003cp\u003ePartnerships with NGOs:Building partnerships with non-governmental organizations and community-based groups can effectively expand service accessibility, particularly for underserved rural areas. Such collaborative efforts can enhance caregiver and elder support networks, boost public awareness of available caregiving resources, and help families better navigate existing support systems.\u003c/p\u003e\n\u003cp\u003eWorkplace Flexibility:Workplace flexibility initiatives should be implemented to better support family caregivers.For example, caregivers could be granted time off or flexible working hours to manage caregiving duties. Companies could be incentivized to offer such benefits, ensuring caregivers are supported both in the workplace and at home.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.6.4Limitations and Future Directions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study has several limitations that should be considered when interpreting the results.First, although the study uses a large, nationally representative sample from the CHARLS dataset, it relies on cross-sectional data, which limits the ability to draw causal conclusions about the relationship between retirement and mental health. Despite using instrumental variables to address potential endogeneity, the results may still be influenced by unobserved confounders affecting both retirement and mental health.\u003c/p\u003e\n\u003cp\u003eSecond, while multiple imputation was employed to handle missing data, residual bias may remain due to sample attrition. Specifically, excluded participants were older and more likely to report poorer health. Although the imputation process helped reduce bias and ensured a balanced final sample with respect to key covariates, unobserved differences\u0026mdash;especially related to age and health\u0026mdash;may still affect the results.\u003c/p\u003e\n\u003cp\u003eFinally, while the study focuses on key variables such as self-rated health and depressive symptoms, it does not consider other potential factors influencing mental health, such as social support, economic status, and personal coping mechanisms. Future research could explore these additional factors and examine the long-term effects of retirement on mental health using longitudinal data, providing a deeper understanding of causal pathways and long-term impacts.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eData Availability Statement\u003c/p\u003e\n\u003cp\u003eThe CHARLS dataset analysed during the current study is publicly available from the official CHARLS repository (http://charls.pku.edu.cn/en).\u003c/p\u003e\n\u003cp\u003eConflict of Interest\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eThe authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest\u003c/em\u003e.\u003c/p\u003e\n\u003cp\u003eAuthor Contributions\u003c/p\u003e\n\u003cp\u003eZT and HN contributed equally to this work and share first authorship. YM served as the corresponding author.\u003c/p\u003e\n\u003cp\u003eZT and YM were responsible for the conceptualization, research design, and methodological framework.\u003c/p\u003e\n\u003cp\u003eYM performed the formal analysis, data curation, and supervised the overall research process.\u003c/p\u003e\n\u003cp\u003eZT and HN collected the data and drafted the initial manuscript.\u003c/p\u003e\n\u003cp\u003eHN also contributed to the translation, language refinement, and final quality control of the article.\u003c/p\u003e\n\u003cp\u003eAll authors contributed to manuscript review and editing and approved the final version for submission.\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eThe authors received no specific funding for this work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study utilized data from the China Health and Retirement Longitudinal Study (CHARLS), which obtained ethical approval from the Institutional Review Board at Peking University. The IRB approval number for the main household survey, including anthropometrics, is IRB00001052-11015, and the IRB approval number for biomarker collection is IRB00001052-11014. All participants provided informed consent prior to data collection.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eFleischmann, M., Xue, B., \u0026amp; Head, J. (2020). Mental Health Before and After Retirement\u0026mdash;Assessing the Relevance of Psychosocial Working Conditions: The Whitehall II Prospective Study of British Civil Servants. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 75(2), 403-413. https://doi.org/10.1093/geronb/gbz042.\u003c/li\u003e\n\u003cli\u003eVo, T.T., \u0026amp; Phu-Duyen, T.T. (2023). Mental health around retirement: evidence of Ashenfelter\u0026apos;s dip. Global Health Research and Policy, 8(1), 35. https://doi: 10.1186/s41256-023-00320-3.\u003c/li\u003e\n\u003cli\u003eKolodziej, I.WK, \u0026amp; Garc\u0026iacute;a-G\u0026oacute;mez, P. (2019). Saved by retirement: Beyond the mean effect on mental health. Social Science \u0026amp; Medicine (1982), 225, 85-97. https://doi: 10.1016/j.socscimed.2019.02.003. Epub 2019 Feb 4.\u003c/li\u003e\n\u003cli\u003eOlesen, K., Rod, N.H., Madsen, I.E., Bonde, J.P., \u0026amp; Rugulies, R. (2015). Does retirement reduce the risk of mental disorders? A national registry-linkage study of treatment for mental disorders before and after retirement of 245,082 Danish residents. Occupational and Environmental Medicine, 72(5), 366-72. https://doi: 10.1136/oemed-2014-102228. Epub 2015 Mar 26.\u003c/li\u003e\n\u003cli\u003eMosconi, G., Vigezzi, G.P., Bertuccio, P., Amerio, A., \u0026amp; Odone, A. (2023). Transition to retirement impact on risk of depression and suicidality: results from a longitudinal analysis of the Survey of Health, Ageing and Retirement in Europe (SHARE). Epidemiology and Psychiatric Sciences, 32, e34. doi: 10.1017/S2045796023000239.\u003c/li\u003e\n\u003cli\u003eOltmans, E., Confectioner, K., Jonkers, R., Kerkhoffs, G. MMJ, Moen, M., Verhagen, E., Wylleman, P., \u0026amp; Gouttebarge, V. (2022). A 12-month prospective cohort study on symptoms of mental health disorders among Dutch former elite athletes. The Physician and Sportsmedicine, 50(2), 123-131.doi: 10.1080/00913847.2020.1868276.\u003c/li\u003e\n\u003cli\u003e\u0026Aring;hlin, J.K., Peristera, P., Westerlund, H., \u0026amp; Magnusson Hanson, L.L. (2020). Psychosocial working characteristics before retirement and depressive symptoms across the retirement transition: a longitudinal latent class analysis. Scandinavian Journal of Work, Environment \u0026amp; Health, 46(5), 488-497. https:// doi: 10.5271/sjweh.3889.\u003c/li\u003e\n\u003cli\u003eOdone, A., Gianfredi, V., Vigezzi, G.P., Amerio, A., Ardito, C., d\u0026apos;Errico, A., Stuckler, D., Costa, G., \u0026amp; (2021). Does retirement trigger depressive symptoms? A systematic review and meta-analysis. Epidemiology and Psychiatric Sciences, 30, 77. https://10.1017/S2045796021000627\u003c/li\u003e\n\u003cli\u003e.Noh, J.W., Kwon, Y.D., Lee, L.J., Oh, I.H., \u0026amp; Kim, J. (2019). Gender differences in the impact of retirement on depressive symptoms among middle-aged and older adults: A propensity score matching approach. PLoS One, 14(3), e0212607. https://10.1371/journal.pone.0212607\u003c/li\u003e\n\u003cli\u003eHuijs, J.JJM, Houtman, I.LD, \u0026amp; Blonk, R.WB. (2019). Development of a new measure for mental retirement; testing of a three-factor structure of mental retirement in different subgroups. BMC Public Health, 19(1), 1604. https://10.1186/s12889-019-7649-5.\u003c/li\u003e\n\u003cli\u003eDenier, N., Clouston, S. A. P., Richards, M., \u0026amp; Hofer, S. M. (2017). Retirement and Cognition: A Life Course View. Advances in Life Course Research, 31, 11-21. doi: 10.1016/j.alcr.2016.10.004.\u003c/li\u003e\n\u003cli\u003eBarman, P., \u0026amp; Sahoo, H. (2024). Through the eyes of grandparents: an in-depth exploration of the nexus between grandchild caring and the psychological well-being of older grandparents. BMC Geriatrics, 24(1), 399. doi: 10.1186/s12877-024-04998-z.\u003c/li\u003e\n\u003cli\u003eHong, Y., \u0026amp; Xu, W. (2023). Continuity and changes in grandchild care and the risk of depression for Chinese grandparents: new evidence from CHARLS. Frontiers in Public Health, 11, 1217998. https://doi: 10.3389/fpubh.2023.1217998.\u003c/li\u003e\n\u003cli\u003eBangerter, L.R., Kim, K., Zarit, S.H., Birditt, K.S., \u0026amp; Fingerman, K.L. (2015). Perceptions of Giving Support and Depressive Symptoms in Late Life. The Gerontologist, 55(5), 770-779. https://doi.org/10.1093/geront/gnt210.\u003c/li\u003e\n\u003cli\u003eKim, K.M., Lee, J.J., \u0026amp; Chung, U.S. (2020). Perceived Health Status of and Moderating Factors in Elderly People Caring for Their Grandchildren. Psychiatry Investigation, 17(4), 275-282. https://DOI: https://doi.org/10.30773/pi.2019.0115.\u003c/li\u003e\n\u003cli\u003eTang, F., Xu, L., Chi, I., \u0026amp; Dong, X. (2016). Psychological Well-Being of Older Chinese-American Grandparents Caring for Grandchildren. Journal of the American Geriatrics Society, 64(11), 2356-2361. doi: 10.1111/jgs.14455.\u003c/li\u003e\n\u003cli\u003eXia, C., Wei, T., Tang, Q., Zheng, H., Chen, G., \u0026amp; Lu, J. (2023). Depression and Associated Factors among Family Caregivers of Children with Disabilities: Analysis of Intergenerational Differences. Healthcare (Basel, Switzerland), 11(19), 2693. https://doi.org/10.3390/healthcare11192693.\u003c/li\u003e\n\u003cli\u003eLee, J., Lee, A., Lee, D., Jung, H.Y., Kim, S.G., \u0026amp; Lee, S.I. (2019). Suicidal Ideation of the Elderly According to Their Involvement in Grandchild Care. Psychiatry Investigation, 16(8), 625-628. https://oi: 10.30773/pi.2019.06.06.\u003c/li\u003e\n\u003cli\u003eWang, H., \u0026amp; Huang, J. (2023). Impacts of grandparenting on older Chinese adults\u0026apos; mental health: a cross-sectional study. BMC Geriatrics, 23(1), 660. https://doi: 10.1186/s12877-023-04396-x.\u003c/li\u003e\n\u003cli\u003eNogueira, A.J., \u0026amp; Ribeiro, M.T. (2023). \u0026quot;I\u0026apos;m Afraid If This Goes Wrong\u0026hellip; What Will Become of Me?\u0026quot;: The Psychological Experience of Grandparents in Pediatric Palliative Care. Healthcare (Basel, Switzerland), 11(17). https://doi.org/10.3390/healthcare11172391.\u003c/li\u003e\n\u003cli\u003eDamman, M., Henkens, K., \u0026amp; Kalmijn, M. (2015). Missing Work After Retirement: The Role of Life Histories in the Retirement Adjustment Process. The Gerontologist, 55(5), 802-813. https://doi: 10.1093/geront/gnt169.\u003c/li\u003e\n\u003cli\u003eMannes, Z. L., Waxenberg, L. B., Cottler, L. B., Perlstein, W. M., Burrell, L. E., Ferguson, E. G., Edwards, M. E., \u0026amp; Ennis, N. (2019). Prevalence and Correlates of Psychological Distress among Retired Elite Athletes: A Systematic Review. International Review of Sport and Exercise Psychology, 12(1), 265-294. https:// doi: 10.1080/1750984X.2018.1469162.\u003c/li\u003e\n\u003cli\u003eAabo, M. G., M\u0026oslash;lgaard, K., \u0026amp; Lassen, A. J. (2023). The worn-out syndrome: Uncertainties in late working life triggering retirement decisions. PLoS One, 18(3), e0282905.doi: 10.1371/journal.pone.0282905.\u003c/li\u003e\n\u003cli\u003eCohen, L., Shiovitz-Ezra, S., \u0026amp; Erlich, B. (2023). Support for Older Parents in Need in Europe: The Role of the Social Network and of Individual and Relational Characteristics. Innovation in Aging, 7(4), igad032. https://doi: 10.1093/geroni/igad032.\u003c/li\u003e\n\u003cli\u003eHoffman, Y., \u0026amp; Shrira, A. (2019). Variables Connecting Parental PTSD to Offspring Successful Aging: Parent-Child Role Reversal, Secondary Traumatization, and Depressive Symptoms. Frontiers in Psychiatry, 10, 718. https://doi.org/10.3389/fpsyt.2019.\u003c/li\u003e\n\u003cli\u003eEarl, E.J., \u0026amp; Marais, D. (2023). The experience of intergenerational interactions and their influence on the mental health of older people living in residential care. PLoS One, 18(7), e0287369. https://doi: 10.1371/journal.\u003c/li\u003e\n\u003cli\u003eHan, S., Guo, J., \u0026amp; Xiang, J. (2024). Is intergenerational care associated with depression in older adults?. Frontiers in Public Health, 12, 1325049. https://doi.org/10.3389.\u003c/li\u003e\n\u003cli\u003eZhang, J., Chandola, T., \u0026amp; Zhang, N. (2022). Understanding the longitudinal dynamics of rural\u0026ndash;urban mental health disparities in later life in China. Aging \u0026amp; Mental Health, 27(7), 1419\u0026ndash;1428. https://doi.org/10.1080/13607863.2022.2098912.\u003c/li\u003e\n\u003cli\u003eHuang, W. (2024). Economic impact of retirement on the elderly population: A literature review. China Economic Journal, 38(2), 317\u0026ndash;337.10.1080/17538963.2024.2425893.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-public-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pubh","sideBox":"Learn more about [BMC Public Health](http://bmcpublichealth.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pubh/default.aspx","title":"BMC Public Health","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Grandchild Caregiving, Mental Health, Intergenerational Support, Retirement Policies, China, Older Adults","lastPublishedDoi":"10.21203/rs.3.rs-6720729/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6720729/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eThis study explores the impact of retirement, grandchild caregiving, and intergenerational support on the mental health of older adults in China, focusing on how these factors interact.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eUsing data from the 2018 China Health and Retirement Longitudinal Study (CHARLS), we employed Ordinary Least Squares (OLS) regression, Instrumental Variables (IV) regression, and Multiple Imputation regression to address potential endogeneity and missing data.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eRetirement is significantly associated with improved mental health, particularly among older adults with lower caregiving burdens (β = -0.435, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). In contrast, higher caregiving intensity is linked to poorer mental health (β\u0026thinsp;=\u0026thinsp;0.00094, p\u0026thinsp;=\u0026thinsp;0.001). Satisfaction with parent-child relationships plays a protective role in mental health (β = -0.209, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and financial support to children is negatively associated with depressive symptoms (β= -0.023, p\u0026thinsp;=\u0026thinsp;0.031). These findings underscore the interconnectedness of retirement, caregiving, and intergenerational support.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eExtending the working age, without offering meaningful support, can lead to growing mental health issues among older adults\u0026mdash;especially those already managing the demands of grandchild care. Any policy aimed at postponing retirement should be accompanied by tangible resources: local respite programs, training workshops tailored for elder caregivers, and even modest subsidies for childcare. These efforts not only provide relief\u0026mdash;they also enable grandparents to emotionally recover while continuing to support their families without compromising their own well-being. Specifically, interventions such as local respite programs, targeted caregiver workshops, and even modest childcare subsidies could provide meaningful support.\u003c/p\u003e","manuscriptTitle":"The Impact of Retirement and Grandchild Caregiving on Mental Health in China: The Role of Intergenerational Support","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-02 09:48:18","doi":"10.21203/rs.3.rs-6720729/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-07-22T18:00:45+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-20T15:38:21+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-19T02:55:13+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-18T02:57:49+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-16T21:19:26+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-14T06:09:50+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"211278883745069923938320599871834360751","date":"2025-07-12T14:04:46+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"194960336857048757272664162184344344680","date":"2025-07-10T09:35:24+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"207092927242275804296147717106276167032","date":"2025-07-09T14:07:53+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"211101732691724677683548123201747092478","date":"2025-07-09T02:06:14+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"18988038330372940686351701751758338085","date":"2025-07-08T03:14:37+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"20425878829654395887547842818216524953","date":"2025-07-07T20:41:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"73479879403131834276878309907716180761","date":"2025-07-07T15:45:52+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-07-07T14:00:50+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-06-24T06:07:17+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-05-22T11:21:06+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-05-22T11:14:49+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Public Health","date":"2025-05-22T03:28:34+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-public-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pubh","sideBox":"Learn more about [BMC Public Health](http://bmcpublichealth.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pubh/default.aspx","title":"BMC Public Health","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"bf92e63a-9c49-4bbf-930d-74a1735b759f","owner":[],"postedDate":"June 2nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-10-13T16:00:11+00:00","versionOfRecord":{"articleIdentity":"rs-6720729","link":"https://doi.org/10.1186/s12889-025-24774-x","journal":{"identity":"bmc-public-health","isVorOnly":false,"title":"BMC Public Health"},"publishedOn":"2025-10-08 15:57:23","publishedOnDateReadable":"October 8th, 2025"},"versionCreatedAt":"2025-06-02 09:48:18","video":"","vorDoi":"10.1186/s12889-025-24774-x","vorDoiUrl":"https://doi.org/10.1186/s12889-025-24774-x","workflowStages":[]},"version":"v1","identity":"rs-6720729","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6720729","identity":"rs-6720729","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}