What are the socioeconomic risk factors of depression among the elderly in China: a systematic review

Abstract

Background

Numerous studies have explored the socioeconomic risk factors of depression in late life, including within China, however, there has been no systematic review and synthesis of these findings. Our systematic review aims to identify, evaluate, and synthesize current evidence on potential socioeconomic risk factors associated with depression among the elderly in China.

Methods

Four electronic academic databases (MEDLINE [via OVID], Social Sciences Citation Index (SSCI) [via Web of Science], APA PsycArticles [via EBSCOhost], and Wanfang [a Chinese database]) were searched from 2005 to March 2022 for national studies reporting evidence on potential associations between socioeconomic risk factors and depression among the elderly in China. Elder is defined as a person who is aged 60 and over. Forward and backward citation searches were also conducted. Data was extracted using a standardized form. Study methodological quality was assessed using the Liverpool University Quality Assessment Tools (LQAT). A narrative synthesis was performed.

Results

Thirty studies that met the review inclusion criteria were included. Across the studies, the evidence indicates that higher educational status, better financial status, better housing conditions, active social engagement, more social support and better neighborhood conditions reduced the risk of depression among the elderly in China. Evidence on associations between working status was equivocal, two studies found evidence suggesting that working after 60 increased the likelihood of depression, while another study suggested working after 60 decreased the likelihood of depression.

Conclusions

The findings highlight the importance of considering socioeconomic factors in the prevention and treatment of depression among the elderly in China. To support elderly mental health, policies should focus on expanding educational opportunities, strengthening financial security, improving housing conditions, and supporting flexible employment options for the elderly. Additionally, enhancing social engagement, reinforcing social support networks, and developing community-based initiatives can help create a more inclusive and supportive environment for healthy aging. Further research is needed on the relationships between working status and depression among the elderly in China.

Trial registration

The protocol was registered on PROSPERO: CRD42023483862.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-026-26199-6.

Introduction

Depression is a significant public health concern among the elderly in China. Since 2010, it has emerged as the second leading cause of years lived with disability in China [1]. The negative effects of widespread aging and elderly depression have heightened concerns about its prevalence among this population. A meta-analysis reported a 23.6% prevalence of depression among elderly individuals in China. Moreover, it revealed that a lower prevalence of depression is associated with higher educational levels [2]. Further studies have demonstrated that depression can lead to declining physical health and increases in morbidity, mortality, and suicide among the elderly [3–6].

Given the high prevalence and severe impact of depression among the elderly in China, it is crucial to identify and understand its risk factors. Socioeconomic status (SES) is one such determinant, with individuals of lower SES potentially having a higher risk of depression [7–9]. Health inequalities commonly refer to the distribution of health by socioeconomic position in the United Kingdom [10]. While variations in health occur naturally, social inequalities in health are systematic, socially produced, unfair, unjust, and unnecessary [11, 12]. Socioeconomic inequalities in health occur when there are systemic health differences between groups with unequal social status [13]. Link and Phelan [14] argued that SES is a fundamental cause of health inequalities, which is linked with resources such as money, knowledge, power, prestige and various social connections. They also highlighted that the more advantaged members of society, i.e., the individuals with higher SES, had greater access to flexible resources, enhancing their adaptive capacity. The privileged position of these individuals affords them easier access to crucial resources to protect themselves from health risks, for example, accessing new health information, adopting health technologies, and changing behavior in response to health risks [14]. China is entering into an aging society due to factors such as population control policies including the former one-child policy, increased life expectancy, and pension reforms. Additionally, increased inequality has become a serious issue among the elderly population. As shown in Zhan’s [15] study, migration may increase population aging in rural areas and widen the income inequality in rural areas. Zhan’s study indicated that the effect of migration on income inequality is larger than the effect of birth, death, and natural aging in China [15]. Hanewald [16] identified that the primary causes of elderly income inequality in China were urban–rural gap and educational inequality, with the former partially mediated by the latter. This inequality tended to accumulate over lifetime and further reinforced among older age due to the Chinese public pension system [16].

A previous systematic review by Chen et al. [17] examined the prevalence and risk factors of depression among older adults in China, identifying poor physical health, cognitive impairment, and lower functional ability as key contributors. Additionally, demographic and socioeconomic factors such as financial status, living arrangements, marital status, education level, and gender were found to influence depression risk, with older women, as well as those who were unmarried, living alone, financially disadvantaged, or had lower education levels, being particularly vulnerable. Conversely, social support was found to alleviate depressive symptoms [17]. While this previous review provides important background information, its studies included range from 1997 to 2010, highlighting the need for an updated synthesis of recent findings. A recent meta-analysis by Barrass et al. [18] explored the association between socioeconomic position and depression across Southeast Asian low- and middle-income countries, finding that lower education, financial difficulties, subjective economic status, and unemployment were consistently linked to higher depression risk [18]. While this meta-analysis offers a regional perspective, its findings are not specific to China, underscoring the need for a more focused examination of the socioeconomic risk factors affecting elderly depression in China. The concept of social rank has also been proposed as a key factor in understanding depression. A systematic review by Wetherall et al. [19] explored the Social Rank Theory of Depression, suggesting that individuals with lower perceived social status are more likely to experience depressive symptoms and suicidal ideation [19]. This previous systematic review further indicated that social rank may serve as a psychosocial mechanism linking SES to depressive symptoms, providing an additional psychological explanation for how socioeconomic disparities contribute to mental health inequalities.

Close examination of potential socio-economic risk factors for depression among the elderly is important, as factors such as social engagement and social support, along with other factors including education, income, employment status and housing, may be modifiable. Changes to these determinants may, through carefully designed targeted interventions and policies, improve public health and eliminate or ameliorate health inequalities. Oakes & Rossi [20] defined SES as 1) material endowments, 2) skills, abilities and knowledge, and 3) social network and the status, power, trustworthiness, and abilities of its members [20]. Following Oakes and Rossi, we conceptualize SES as including material endowments, skills, abilities and knowledge, social network, status, power and trustworthiness. We utilize 7 categories of SES in reviewing the evidence: 1) educational level, 2) financial status, 3) housing conditions, 4) working status, 5) social engagement, 6) social support and neighborhood conditions, and 7) composite measures of SES—to capture studies that constructed SES indicators by combining multiple dimensions. Variables that do not directly reflect socioeconomic position or access to material and social resources, such as age, gender, and marital status, were treated as potential confounding variables rather than primary SES indicators.

A large section of the elderly population in China is affected by these adverse conditions. According to the fourth survey of the National Survey on Elderly People in Urban and Rural China in 2015, 71% of the elderly population had an educational level of elementary or below, while 29% had secondary education or above [21]. This survey showed that in 2014, the average annual per capita income for the elderly in urban areas was CNY 23,930 (around £2393 in 2014) and CNY 7,621 (around £762) for the elderly in rural areas. The average annual per capita consumption expenditure in 2014 for the elderly in both urban and rural areas was CNY 14,764 (around £1476) [21]. According to the same survey in 2015, 45.6% of the elderly population engaged in social activities [21]. Based on China’s multidimensional poverty index in 2014, the average deprivation score among people living in multidimensional poverty was 41% [22]. The 2015 China Health and Retirement Longitudinal Study (CHARLS) showed that the rates of housing quality poverty, housing quantity poverty and the combined housing quality-quantity poverty among the elderly population were 48.8%, 39.7% and 21.4% respectively [23]. The survey researchers defined housing quality poverty as a respondent’s house with no potable water, no toilet for sole use, or no kitchen, whilst housing quantity poverty referred to number of rooms per household member being fewer than one [23]. These statistics demonstrate the pervasive nature of these issues, underscoring the importance of understanding how these common risk factors influence depression susceptibility.

Therefore, we conducted a systematic review to answer the following research question: What are the socio-economic risk factors for depression among the elderly in China? By identifying these key factors, our review makes important contributions across multiple domains. First, it offers valuable insights for policymaking and clinical practice through its national-level analysis of socioeconomic risk factors for late-life depression. Second, it advances a macro-level conceptual framework for understanding how socioeconomic factors influence depression in aging populations. Finally, it identifies critical directions for future research, including investigations into the relationship between working status and depression among older adults.

Methods

Our systematic review follows the Centre for Reviews and Dissemination (CRD)’s guide to undertaking systematic reviews [24], the guidance on the Conduct of Narrative Synthesis in Systematic Reviews [25], and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [26]. As our study does not include a meta-analysis, we followed the Synthesis Without Meta-analysis (SWiM) guidelines to ensure transparency in our narrative synthesis. A completed SWiM checklist is provided in Additional file 1. These guidelines were selected as they help ensure transparency and completeness in conducting and reporting systematic reviews, particularly in the context of examining complex social determinants of health. Our study has been registered at PROSPERO (no CRD42023483862). The original research question was: 'What are the risk factors for elderly depression in China?' However, during the full-text screening stage, we observed that a large number of studies met the broad inclusion criteria, making the analysis too extensive to provide focused insights. To ensure the specificity and policy relevance of our findings, we restricted inclusion to studies that: (1) examined socioeconomic risk factors (given their modifiability through policy interventions), and (2) employed nationally representative samples (to enhance generalizability across China compared to local-level studies). Repeating the screening with this more focused question from the beginning would not have altered the articles/studies included. Instead, it would have simply shifted some (but not all) exclusions from the full text screening stage to the title and abstract screening stage. This adjustment did not affect the findings of our review.

Search strategy

Our study systematically searched both English and Chinese academic search engines and bibliographic databases: MEDLINE [via OVID], Social Sciences Citation Index (SSCI) [via Web of Science], APA PsycArticles [via EBSCOhost], and Wanfang [a Chinese database]. Full search strategies are included in Additional file 2. Supplementary searches were conducted in accordance with the CRD 2009 guidance. These searches were especially valuable for identifying evidence on complex social determinants of health, which encompass diverse concepts and measures that are often challenging to capture [24]. For backward citation tracking, we reviewed the reference lists of the studies included to identify any relevant articles having been cited. For forward citation tracking, we used search engines Google Scholar and Web of Science to search for articles that cited the studies included. We consulted experts working in mental health for recommendations of additional references. Four groups of search terms were used as follows:

1. China OR Chinese OR Hong Kong OR Taiwan OR Macau.

2. AND Late Adulthood OR Aging OR Ageing OR Retired OR Aged OR Senior citizen OR Old OR Elderly OR Geriatric OR Senescent.

3. AND Risk factors OR Risk factor OR Related factors OR Linked factors OR Associated factors OR Predictable factors OR Determinant factors OR Determinants OR Intervention OR Strategy OR Training.

4. AND Depression OR Mental health OR Well-being.

For the first round of literature searches, the following four criteria were specified:

1. Country: Studies conducted in China, focusing on Chinese participants.

2. Target group: Individuals aged 60 years or older.

3. Determinant factors: Risk factors influencing the outcomes.

4. Primary outcome: Depression, typically assessed using standardized questionnaires.

Inclusion and exclusion criteria

Based on the four criteria established above during the first round of literature searches, we further specified the five inclusion criteria: 1) Study population: participants in the studies should have or should be more than 60 years old, of any gender; participants in the studies should be The People's Republic of China residents who are residing in China. 2) Study setting: studies should have been conducted in China, in other words, the data should be collected in China. Only studies with broadly representative (not selective) samples conducted at the national level were included in this systematic review. Findings from nationally representative studies/samples are more likely to be transferable or generalizable to wider populations and interventions across China than studies based on selective or local population samples. Studies should have been either using nationally recognized datasets or independently collected primary data, provided they sampled a diverse elderly population. While we prioritized studies that covered both urban and rural areas, rural-only studies were also included, as rural populations constitute a significant proportion of the elderly population in China and can capture socioeconomic disparities in different settings. Participants in the studies had to be drawn from community settings since our review aimed at the general elderly population or elderly individuals at risk for depression. 3) Study characteristics: studies had to be original research reported in Chinese or English, and published between 2005 and 2022, as this timeframe builds on the work of Chen et al. [17], whose review focused on depression prevalence and general risk factors rather than specifically on socioeconomic risk factors, highlighting the need for an updated synthesis that captures recent demographic trends, socioeconomic policies, and methodological advancements. 4) Potential determinants/exposures of the study: The studies needed to examine at least one socio-economic risk factor for depression among the elderly. 5) Study outcomes: The studies needed to use established, validated questionnaires for measuring depression. There were no restrictions on the methods of measuring risk factors.

The two exclusion criteria included: 1) Study population: elderly who have depression at baseline and elderly diagnosed with any kind of psychiatric disorder will be excluded, including bipolar disorder and schizophrenia. Studies focusing on selective populations of the elderly, such as those residing in nursing institutions or with specific comorbidities (e.g., hypertension, diabetes), were excluded. These selective populations may not provide a representative sample for the broader investigation of socio-economic risk factors associated with elderly depression. 2) Study characteristics: studies that were not specific to depression were excluded. Review-level studies, such as systematic reviews, and literature reviews, were not included, but the reference list of such reviews were searched for relevant studies as part of the backward citation searches. Non-empirical publications, such as opinion and discussion pieces, were not included. Please see the list of excluded studies and their reasons for exclusion in the Additional file 3.

Study selection and data extraction

All the articles from the database searches were exported into Endnote X9 reference management software to remove duplicates (through a three-stage process: one automatic [matching name, date, and title]; and two manual stages). All screening was performed using EPPI Reviewer 4 systematic review management software. The selection process for inclusion consisted two stages (title and abstract screening, and full text screening). In the first selection round, two reviewers independently screened a random selection of 20% of all the titles and abstracts to identify potentially eligible studies based on the inclusion/exclusion criteria. This was followed by a screening “calibration” exercise to ensure consistency in the screening of the remaining 80% of papers. The main reviewer then screened the remaining 80% of titles and abstracts. Studies meeting the inclusion criteria during the title and abstract screening stage were retained as potentially eligible studies. In the second selection round, two reviewers independently screened all the full-texted studies. Among the 5,812 studies identified in the original search, 383 studies were included for full-text screening, with 30 studies included in the final analyses (see Fig. 1). Any queries or disagreements were resolved through discussion or consultation with three senior reviewers. The main reviewer performed all data extractions, while other reviewers double-checked the extracted data for accuracy and completeness. The data extraction included: authors, year of publication, data source and year, sample size for this analysis, participants’ age, sex (percentage of women), study design, depression measurement, and the main results on socioeconomic risk factors.

Fig. 1.

PRISMA flow diagram of progression of studies in the systematic review

Quality assessment

Risk of bias and quality assessments were conducted independently by two reviewers. Discrepancies between reviewers in the quality assessment were discussed, re-examined and resolved through consensus, using the Liverpool University Quality Assessment Tool (LQAT) [27]. LQAT incorporates a star rating system to evaluate and quantify the absence of bias, misclassification, and confounding across various study designs. Based on results of the LQAT methodological quality assessment, all findings from the studies included were reported in order of methodological quality (from highest to lowest), in line with Popay et al.’s (2016) guidance on narrative synthesis [25].

Results

From an initial 5812 studies, 30 studies on socioeconomic risk factors of depression among the elderly in China were included and synthesized. Figure 1 presents the study selection process through each stage of the review.

Study characteristics and risk of bias

The 30 included studies were published between 2016 and 2022, all of which were national-level studies conducted in China. Among them, three studies focused specifically on rural elderly populations. All studies included used quantitative methods: five were longitudinal studies, and 25 were cross-sectional. Table 1 presents their quality ratings based on the LQAT, while Tables 2, 3, 4, 5, 6, 7 and 8 provide the complete detailed results on different socioeconomic factors (see Additional file 4). Additional file 5 includes information on the characteristics of each study. Based on the LQAT criteria, eleven studies were graded as high methodological quality, and nineteen as medium; no studies were graded as low methodological quality. Table 2 summarizes the association between each socioeconomic risk factor and depression. To ensure consistency, all results and coefficients reported were reframed/harmonized in the same direction to facilitate direct comparisons across studies.

Table 1.

Quality assessment of the included studies on socioeconomic risk factors of depression among Chinese elderly

Authors (Years of publication)	Quality assessment	Authors (Years of publication)	Quality assessment
Ni (2017) [28]	High	Li (2017) [29]	Medium
Chen (2019) [30]	High	Chen (2021) [31]	Medium
Cui (2022) [32]	High	Fang, M (2019) [33]	Medium
Fang, M (2019) [34]	High	Feng (2019) [35]	Medium
Gao (2020) [36]	High	He (2021) [37]	Medium
Guo (2017) [38]	High	Hu (2018) [39]	Medium
Li (2016) [40]	High	Liu (2022) [41]	Medium
Li (2021) [42]	High	Yu, Y (2021) [43]	Medium
Liang, Y (2020) [44]	High	Wang (2018) [45]	Medium
Liu (2021) [46]	High	Xu (2016) [47]	Medium
Wang (2019) [48]	High	Yu, X (2021) [49]	Medium
		Li, M (2019) [50]	Medium
		Li, C (2019) [51]	Medium
		Xiao (2021) [52]	Medium
		Xie (2020) [53]	Medium
		Xie (2021) [54]	Medium
		Liang, K (2020) [55]	Medium
		Tao (2018) [56]	Medium
		Yang (2017) [57]	Medium

Table 2.

Association between socioeconomic factors and depression

Authors (Years of publication)	Composite measures of high SES	High educational level	High financial status	Better housing conditions	Still working	High social engagement	High social support and neighborhood conditions
Ni (2017) [28]						Decreased
Chen (2019) [30]							Decreased
Cui (2022) [32]		Decreased	Decreased			Decreased
Fang (2019) [34]		No association	Decreased	Decreased
Gao (2020) [36]			Decreased
Guo (2017) [38]		Decreased	Decreased
Li (2016) [40]		Decreased	Decreased	Decreased		Decreased	Decreased
Li (2021) [42]		Decreased	Decreased
Liang (2020) [44]		Decreased	Decreased	Decreased	Increased	Decreased
Liu (2021) [46]		Decreased	Decreased			Decreased
Wang (2019) [48]		Decreased	Decreased
Li (2017) [29]		Decreased	Decreased
Chen (2021) [31]		Decreased	Decreased				Decreased
Fang (2019) [33]		No association	Decreased	Decreased
Feng (2019) [35]		Decreased	Decreased			No association	Decreased
He (2021) [37]			Decreased				Decreased
Hu (2018) [39]		Decreased	Decreased				Decreased
Liu (2022) [41]						Increased
Yu (2021) [43]			Decreased				Decreased
Wang (2018) [45]		Decreased		Decreased			Decreased
Xu (2016) [47]		Decreased	Decreased		Decreased	Decreased
Yu (2021) [49]		Decreased	Decreased			Decreased	Decreased
Li (2019) [50]		Decreased	Decreased		Increased	Decreased
Li (2019) [51]		Decreased					Decreased
Xiao (2021) [52]		Decreased				Decreased
Xie (2020) [53]		Decreased
Xie (2021) [54]		Decreased	No association
Liang (2020) [55]	Decreased
Tao (2018) [56]			Decreased			Decreased
Yang (2017) [57]		Decreased	Decreased			Decreased

Note: Decreased means a negative association with a p value of < 0.05, e.g. higher educational level was associated with decreased depression; increased means a positive association with a p value of < 0.05, e.g. still working was associated with the increasement of depression; and no association with depression means it’s not significant with a p value > 0.05

All 30 studies included used secondary data from various sources/databases and employed different measurements to assess depression. Specifically, twenty-two studies used the China Health and Retirement Longitudinal Study (CHARLS), six studies used the China Longitudinal Aging Social Survey (CLASS), one study used the Sample Survey on the Aged Population in Urban/Rural China (SSAPUR), and one study used the Chinese Longitudinal Healthy Longevity Survey (CLHLS). The measurement for depression included: 10-item of Center for Epidemiologic Studies Depression Scale (CES-D) [58], 9-item CES-D [59], 12-item CES-D [60], and Geriatric Depression Scale-Short Form (GDS-15) [61].

Socioeconomic risk factors of depression among the elderly in China

Among the thirty studies examining the relationship between SES and depression among the elderly in China, twenty-nine studies found that better SES was associated with a lower risk of depression based on one or more measure of SES.

Educational level

There were twenty-two studies that investigated the relationship between educational level and depression. Of these, eight studies were rated as high methodological quality, and fourteen as medium quality. Twenty studies reported a negative association with depression, indicating that higher education level was associated with a lower risk of depressive symptoms. Two studies found no significant association between education’s level and depression, indicating variability in educational impact across different contexts [33, 34]. These two studies, authored by the same researcher, were rated high and medium quality, respectively, and are further discussed in Sect. 3.4.1.

Financial status

Twenty-two studies investigated the relationship between financial status and depression. Financial status had varied measurements across studies, including consumptions or expenditures, family or personal income, and welfare provisions. Of these, nine studies were rated high methodological quality, and fifteen were medium quality. Twenty-one studies reported a negative association with depression, indicating that elderly individuals with higher financial status had a lower risk of depression. One study (rated medium quality) found no association between financial status and depression in their univariate analysis [54], where the researchers measured financial status using household per capita consumption expenditure, with 57% of participants reporting expenditures within CNY 2,000.

Housing conditions

Five studies examined housing conditions and found a negative association with depression, meaning better housing conditions were associated with a lower risk of depression. Of these, three studies were rated high methodological quality, and two were medium quality.

Working status

The working status of elderly individuals was examined for its associations with depression only in three studies, likely due to most elderly transitioning into retirement. Their findings were inconsistent: two studies (one rated high and one medium quality) found a positive association with depression, meaning elderly individuals who were still working were more likely to experience depression [44, 50]. One study (rated medium quality) reported a negative association with depression, indicating that the elderly individuals who were still working had a lower risk of depression [47]. These contradictory findings highlight the complexity of the relationship between working after 60 and depression.

Social engagement

A total of thirteen studies examined the relationship between social engagement and depression: five studies were rated high quality, and eight were medium quality. Eleven studies found a negative association with depression, indicating that elderly individuals who engaged in social activities had a lower prevalence of depression compared to those who did not. One study (rated medium quality) found a positive association with depression, where the researchers examined the moderating effect of the availability of community recreation facilities (CRF) in the association between loneliness and depression. Their results suggested that having access to CRF was associated with an increase in the predictive power of loneliness on depression. It does not necessarily mean that CRF increased depression directly, but the relationship between loneliness and depression was strong when CRF were available [41]. Another study (rated medium quality) found no significant association between social engagement and depression, or more precisely exclusion from social activities insignificantly associated with depression [35]. However, a lower level of exclusion from social activities was significantly associated with being less likely to report poor self-reported health [35].

Social support and neighborhood conditions

All the ten studies that included social support and neighborhood conditions showed a negative association with depression. This indicated that older people with more social support were less likely to experience depression, and those who had better neighborhood conditions (e.g., neighborhoods with better infrastructure) experienced fewer depressive symptoms. Among these ten studies, two studies were rated as high methodological quality, and eight were medium quality.

Composite SES Measure

One study (rated medium quality) used a composite measure of SES—created by standardizing education and income variables to compute their mean—and found that higher SES was associated with lower rates of depression.

Discussion

How does socio-economic risk factors influence depression

Our systematic review identified 30 studies that explored wide-ranging aspects of the socio-economic risk factors for depression among elderly individuals in China. The findings indicate that better SES (including higher educational status, better financial status, better housing conditions, active social engagement, social support and neighborhood conditions) were generally associated with a lower risk of depression.

Higher educational levels were significantly associated with a lower risk of depressive symptoms, aligning with prior research emphasizing the protective role of education in mental health outcomes [62, 63]. This observation could be attributed to several factors. For example, a higher educational level often leads to better financial status, thus having a greater awareness of health and well-being and better access to healthcare. However, two studies by the same first author suggested there was no association between educational level and depression [33, 34]. Notably, these two studies focused on rural participants and emphasized that people with higher personal annual incomes were less likely to have depression. Among their participants, 94.5% of their participants had an educational level of primary school and below, and only 5.5% of the participants attended high school and above. In these two studies, such homogeneity in educational backgrounds may explain why there were no effects of educational status on depression, since the majority of participants fall into a singular category. Additionally, in these rural settings, financial status such as personal annual income might overshadow the influences of education. Given these findings, future policies and interventions aiming to reduce the prevalence of depression in the elderly might benefit from targeting education and its associated socio-economic disparities. Expanding lifelong learning opportunities, community-based education programs, and health literacy initiatives may help mitigate the mental health burden among lower-educated elderly populations, particularly in rural areas.

The financial status of elderly individuals is often negatively correlated with their depressive symptoms. Financial stability plays a critical role in elderly mental health. Being financially secure alleviates the stress related to financial concerns, gives the confidence to plan for the future, and enhances one’s control over their life decision. These may contribute to mental health by providing a sense of security. Furthermore, having better financial status allows the elderly to engage in leisure activities, maintain a healthy lifestyle, and access better healthcare, which can substantially reduce the risk of depression. Pension is an important indicator of financial status among the elderly, as it is associated with the change of lifestyle, increased health investment and reduced economic uncertainty [64]. Given that financial insecurity is a key socioeconomic risk factor, policymakers should strengthen pension systems, expand social welfare programs, and improve financial assistance for low-income elderly populations. Better housing conditions were also associated with lower depressive symptoms. This potentially reflects the broader social and economic implications where better housing conditions often reflect better overall living conditions and better quality of life [65]. In response, housing policies that ensure elderly-friendly infrastructure, promote affordable senior housing, and enhance community-based support systems should be prioritized to improve elderly mental well-being.

The three studies included suggest mixed results regarding the potential risk of depressive symptoms associated with working status in the elderly. The retirement age in China is typically 55 for women and 60 for men. It's crucial to consider the broader context, including working conditions and personal motivations for continued working after 60, as most elderly population transition into retirement at this age. A study found that the elderly individuals who were retired with a pension had a lower depression rate compared to those who were still working [66]. Another study also found that working after retirement increased the risk of depression for the elderly in China, especially for the elderly after 60 years old, women, those with lower educational levels, urban household registration, and those with higher pensions and higher social status [67]. This emphasizes the need for understanding their working conditions and personal motivations in the link with depression. Future research may consider social class and could use qualitative methods to explore further. Policymakers should consider designing flexible and supportive employment policies that allow older adults to work under less stressful conditions if they choose to remain in the workforce. Additionally, financial assistance programs targeting vulnerable elderly groups could further alleviate socioeconomic disparities linked to depression risk.

The studies included consistently reported the protection effects of engaging in social activities towards depression. Not only does engaging in social activities reduce the risk of depression, the engagement frequency and specific types of social activities also play important roles [68–70]. As China is entering into an aging society, recognizing the importance of social engagement, and creating specific programs to support elderly social engagement is crucial to promote their mental well-being and healthy aging. Initiatives such as subsidized recreational programs, senior community centers, and volunteer opportunities can significantly contribute to improving mental well-being and healthy aging.

Social support and neighborhood conditions consistently emerge as protective factors against depression. The importance of family, community and social networks reinforces the idea that social support is crucial for psychological well-being among the elderly population [71, 72]. Neighborhood conditions act as a part of public social support, and are negatively correlated with depression. Neighborhood-based services aiming to encourage social interaction among the elderly were significantly associated with a decrease in depression, whereas these beneficial effects were greater among low socioeconomic neighborhoods [73]. Policymakers and community leaders should consider implementing measures such as regular community activity programs, facilitating intergenerational activities, and building public activity centers for the elderly population. Expanding neighborhood-based social programs can help build stronger community support networks and improve mental health outcomes for older adults.

Strengths and limitations in the systematic review

To the best of our knowledge, this is the first study to review the links between socio-economic risk factors and depression among the elderly in China. By using the following best practice guidance on conducting systematic reviews, it ensured a robust and structured approach in the identification and selection of studies. By focusing on national-level studies, our review is more generalizable and representative of China’s diverse demographic, social and economic contexts. Additionally, the double screening procedure helped mitigate potential selection bias, reinforcing the credibility of our review.

Our review has some limitations. First, there is an over-reliance on CHARLS data among the studies included, with twenty-two studies out of thirty using the CHARLS database. The issue of over-reliance may introduce a uniform methodological bias and limit the diversity of data sources. Second, all the studies included employed self-reported depression scales to measure depression rather than clinical diagnosis, for example, CESD. The self-reported depression scales are more likely to capture transient psychological distress that may not be the same as clinical disease. Furthermore, it cannot access clinical change [74]. Third, an overarching limitation is the lack of causal approaches, leading to the inability to distinguish between the direction of causal effects. For example, does social engagement reduce the risk of depression or does depression reduce the elderly’s ability to engage socially? Among the thirteen studies that included social engagement as a measure, only two studies used longitudinal data with latent class analysis. Moreover, unmeasured confounding factors, such as genetic risk and early-life adversity, may still influence these relationships, limiting causal interpretations. Fourth, the influence of differing assessment methods for risk factors may also have influenced the results. Meta-analysis was not appropriate due to the differences in measurement methods and high levels of heterogeneity of the coefficients [75, 76]. Given this heterogeneity, we identified a narrative synthesis as the most suitable approach to systematically summarize and interpret the evidence on socioeconomic risk factors for depression among the elderly in China. This method allows for a structured comparison of findings while accounting for differences in analytical frameworks and socioeconomic indicators. Fifth, none of the studies employed qualitative research. Qualitative studies should be included in the future as it can provide deeper insights into lived experiences, cultural context and individual perceptions of the elderly regarding depression.

Gaps and limitations in the body of evidence

While our systematic review provides a large body of evidence on socioeconomic risk factors and depression among the elderly in China, there are several limitations in the body of evidence. First, although one of the inclusion criteria for our review was articles published from 2005, the studies included in the actual results were limited to those published between 2016–2022, which may have missed earlier insights. Second, the national studies included in our review were predominantly secondary data. Third, some studies did not specify the exact questionnaires used for measuring depression, despite using the same database as others (2014 CLASS). It is essential for future research to clearly identify the questionnaires used for depression assessment to ensure the validity and reliability of the findings. Fourth, there were relatively few studies focusing on relationships between housing conditions (n = 5), working status (n = 3) and depression among the elderly in this review.

Implications

Given the overwhelming evidence this study found for socioeconomic inequalities in depression in China, this study has major implications for public health strategies. The National Health Commission of the People's Republic of China released the Healthy China Initiative (2019–2030) in 2019, aiming to achieve health equity by 2030 [77]. Where health equity means that everyone should have an equal chance to attain their full health potential regardless of their socioeconomic positions [78, 79]. The indicators in the Healthy China Initiative were related to measures of aggregate health status, and did not pay sufficient attention to the distribution of health indicators across subgroup populations of different SES [78]. The findings of this study may provide implications for policies to promote health equity.

To effectively promote health equity, it is crucial to allocate more resources for mental health prevention and care to more disadvantaged groups and places. Policies should adopt the proportionate universalism approach, where policies and actions should be universally applicable but should be strengthened according to the degree of disadvantages [80]. Policies should address the risk factors associated with depression in the early years such as ensuring equitable access to education, as childhood experiences could influence people’s health, educational attainment, and economic participation in later years [80]. Policies should ensure fair wages and good pension coverage to reduce financial stress, since income inequalities could cause negative health effects and increase the psychosocial stress level across the population, especially among those in lower social classes [79]. Policies should create places and spaces that encourage social interaction and community engagement regardless of physical or financial barriers, since enhancing community infrastructure could boost social relations and significantly improve well-being [81]. Implementing these strategies may reduce the socioeconomic inequalities in elderly depression, so as to achieve healthy equity and healthy aging.

Future research directions

In light of the findings and limitations identified in our systematic review, several future research directions should be considered for better understanding socioeconomic factors and depression among the elderly in China. The predominance of secondary data in our systematic review underscores the importance of incorporating primary data in future research. Future research should incorporate more studies on housing conditions and working status as these factors are crucial in understanding the socioeconomic context of depression among the elderly. Furthermore, some of the studies included explored urban/rural and gender differences, which were not the focus of this research. Future research should consider it as subgroups and explore how socioeconomic factors and depression vary across different demographic contexts.

Methodologically, future studies should aim for higher quality studies. The use of consistent measures to enable meta-analysis is also crucial. Both qualitative and quantitative studies should be included in future research. Replicating existing studies in different contexts or with different population groups would strengthen the evidence.

Conclusion

With China's aging population growing and depression among the elderly emerging as a serious public health concern, it is important to understand how socioeconomic risk factors influence depression in China in order to better target policies and interventions aimed at preventing late-life depression and promoting healthy aging. Our systematic review of the relevant literature reveals consistent evidence that several protective factors are associated with reduced depression risk in older adults, including higher educational attainment, greater financial security, improved housing conditions, active social participation, stronger social support networks, and better neighborhood environments. The evidence for a relationship with working status however was equivocal. These socioeconomic factors are all potentially modifiable, and policies targeted at the high risks sub-groups we identify are likely to reduce inequities in mental health. To support elderly mental health, policymakers should consider implementing: (1) lifelong learning and health literacy initiatives; (2) social protection in old age through pension systems and social welfare programs; (3) age-friendly infrastructure and financial assistance; and (4) community-based social engagement programs. Whilst universal support is needed to reduce risk of depression in the elderly in China, the higher risk we identify in more disadvantaged socioeconomic groups (i.e. those with low education, struggling financially, living in poor housing and disadvantaged neighborhoods), indicate that prevention efforts should be pursued with greater intensity in those groups – and approach known as proportional universalism. Our findings underscore the need to address socioeconomic disparities, which could alleviate the burden of late-life depression and guide more targeted, equitable policy interventions.

Abbreviations

CES-D

Center for Epidemiologic Studies Depression Scale

CHARLS

China Health and Retirement Longitudinal Study

CLASS

China Longitudinal Aging Social Survey

CLHLS

Chinese Longitudinal Healthy Longevity Survey

CRD

Centre for Reviews and Dissemination

CRF

Community recreation facilities

GDS-15

Geriatric Depression Scale-Short Form

LQAT

Liverpool University Quality Assessment Tool

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

SES

Socioeconomic status

SSAPUR

Sample Survey on the Aged Population in Urban/Rural Chi

SSCI

Social Sciences Citation Index

Authors’ contributions

JL, BB, and AP conceptualized the study design. AP contributed to the data search and setup for the EPPI reviewer. JL acted as the first reviewer, conducting the screening and data extraction. ZD acted as the second reviewer for screening and data extraction. XZ, AP and BB acted as senior reviewers for resolving disagreements during the screening and data extraction processes. JL drafted the initial manuscript, AP, BB and XZ provided substantive intellectual and editorial revisions. All authors read and approved the final manuscript.

Funding

The research results of this article are sponsored by the Shanxi Federation of Social Sciences SSKLZDKT2025065 and Shanxi Provincial Philosophy and Social Science Planning Program 2025QN076.

Data availability

All data generated or analyzed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.