Background

J Med Internet Res

jmir

Journal of Medical Internet Research

J Med Internet Res

1438-8871

JMIR Publications

Toronto, Canada

v27i1e67039

10.2196/67039

Original Paper

Internet Usage, Social Participation, and Depression Symptoms Among Middle-Aged and Older Adult Chinese Individuals Before and During the COVID-19 Pandemic: Evidence From the China Health and Retirement Longitudinal Observational Study

Zhang

Siqian

MD123*Dong

Shiju

MD123*Yu

Zexuan

MD4Bi

Shuwen

MD123Wen

Wen

MD123Li

Jiajia

PhD123

Department of Social Medicine and Health Management, Cheeloo College of Medicine, School of Public Health, Shandong University

44 WenHua West Road, Li'xia District

Jinan

ChinaNHC Key Lab of Health Economics and Policy Research, Shandong University

Jinan

ChinaShandong Provincial Key New Think Tank, Center for Health Management and Policy Research, Shandong University

Jinan

ChinaThe Dartmouth Institute for Health Policy and Clinical Practice, Dartmouth Hitchcock Medical Center, Geisel School of Medicine at Dartmouth

Lebanon

United States

Sarvestan

Javad

Sheldrake

Elena

Song

Jie

Phankitnirundorn

Krit

Lee

Yen-Fen

Correspondence to Jiajia Li, PhD, Department of Social Medicine and Health Management, Cheeloo College of Medicine, School of Public Health, Shandong University, 44 WenHua West Road, Li'xia District, Jinan, 250012, China, 86 053188382269; lijiajia@sdu.edu.cn*

these authors contributed equally

2025

2292025

e67039

300920243007202531072025

2025

This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research (ISSN 1438-8871), is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included.

Background

While the relationship between internet usage and depression, and the mediating role of social participation in this association, has garnered significant attention, the COVID-19 pandemic has disrupted traditional forms of social participation. The extent to which this disruption has altered the association remains underexplored.

Objective

This study aims to examine the impact of internet usage on depressive symptoms among middle-aged and older adults and to assess how the mediating role of social participation differs before and during the COVID-19 pandemic.

Methods

Data for this study were sourced from the fourth wave (2018: 15,326 observations) and the fifth wave (2020: 15,758 observations) of the China Health and Retirement Longitudinal Study. A 2-way fixed-effects model incorporating an instrumental variable was employed to investigate the relationship between internet usage and depressive symptoms among middle-aged and older adults. Furthermore, a mediation effect model was applied following propensity score matching to assess changes in the mediating role of social participation on the relationship between internet usage and depressive symptoms before and during the pandemic, as well as to explore changes in 3 internet usage functions. Subgroup analyses based on age and urban-rural residence were performed to assess heterogeneity in this association.

Results

The average Center for Epidemiological Studies Depression Scale (CESD-10) score among middle-aged and older Chinese people increased from 1.94 in 2018 to 1.98 in 2020. Internet usage was negatively associated with depressive symptoms (β=−.041; P<.01), with social participation serving as a partial mediator. The proportion of the mediating effect of social participation on the relationship between distinct internet usage functions and depression shifted during the pandemic: the social function declined from 12.55% in 2018 to 9.30% in 2020, while the informational and recreational function increased from 7.53% and 11.29% in 2018 to 8.85% and 16.37% in 2020, respectively. Both the total and indirect effects were statistically significant across subgroups, with a higher mediation proportion observed among rural residents and women than among urban residents and men.

Conclusions

Internet usage was negatively associated with depressive symptoms among middle-aged and older adults. Despite a noticeable increase in depression scores in 2020 compared with 2018, the mediating effect of social participation diminished, while the direct effect of internet usage became more pronounced during the pandemic. These findings underscore the need to prioritize mental health recovery in this population, promote diverse forms of social participation, and leverage the internet’s potential to enhance social participation and mental well-being.

internet usage functions2-way fixed-effectmediating effectgender differencesurban-rural differences

IntroductionBackground

Depression is now recognized as the predominant psychological issue among older adult populations [1]. From 1994 to 2020, approximately 31.74% of the global older adult population experienced depression, with the incidence rising to 40.78% in low-income countries, significantly higher than the 17.05% observed in high-income nations [1]. The sudden outbreak of the COVID-19 pandemic has rapidly increased the prevalence of depression [2]. The prevalence of depression among American adults surged from 8.5% before the pandemic to 27.8% during its occurrence [3]. Similarly, the prevalence of depression among older adults in Asia and Europe varied from 14.6% to 47.2% during the pandemic [4].

There are several potential factors that may contribute to the increased prevalence of depression during the pandemic, such as difficulty in receiving medical services, particularly mental health services, which has led to reduced social interaction and increased loneliness [5-9]. During this period, the internet has become a vital channel for socialization, kinship supports, and medical consultations, drawing increased attention to its role [10-16]. Previous studies have confirmed the positive impact of the internet on reducing depression in both adults and older individuals [17-19]. However, as a disadvantaged group in internet usage, older adults usually lack basic information literacy [20,21] and are more prone to experiencing digital exclusion [22-25]. For instance, in Australia, while 87% of the population use the internet, only 55% of those aged 65 years and older engage with web-based platforms [26]. Similarly, the internet usage rate of Chinese older adult individuals was merely 13%, far below the 73.8% usage rate among younger groups [27]. The older adults, being a vulnerable population in terms of both health and internet usage, necessitate further research on their mental well-being and the impact of the internet.

Literature ReviewPrepandemic Internet Usage, Social Participation, and DepressionInternet Usage and Depression

The internet plays an increasingly crucial role in different aspects of people’s lives due to advancements in information technology [28]. As an online platform, it offers convenient access to information, provides entertainment through short videos and games, and facilitates communication among family and friends via email, online chat, and video calls [29].

Academic interest is intensifying in the exploration of how internet usage might influence mental health, with a particular focus on symptoms of depression. Four mechanisms could elucidate the link between internet usage and depression [30-32]. First, internet usage influences depression by sustaining and broadening the social networks of older adults, thereby reducing social isolation and improving the quality of their interpersonal relationships [33,34]. Second, communication through the internet has been found to help older adults strengthen connections with family and friends and gain informal social support [35,36], leading to reduced levels of depression and increased life satisfaction [37]. Third, internet usage has also demonstrated its usefulness in helping middle-aged and older individuals obtain health-related information, increasing their health literacy and modifying behaviors, resulting in a decrease in the occurrence of chronic diseases [38,39]. Finally, the internet provides a wide range of entertaining activities that can bring pleasure to middle-aged and older adult individuals, fulfilling their emotional needs, alleviating symptoms of depression, and enhancing cognitive functions [32,40]. Conversely, some studies have reported that increased time spent on the internet correlates with higher levels of depressive symptoms [41,42]. Specifically, excessive internet usage can result in decreased engagement in other activities and a smaller social network, fostering feelings of social isolation, loneliness, and depression [43-45].

Social Participation and Depression

Social participation refers to the process of building connections with others by engaging in social or community activities [46]. Previous studies have demonstrated that active participation in social activities is often associated with improved mental health outcomes [47,48]. In particular, social participation has been found to have a protective effect in relieving depression among older adults through emotional social support and increasing cognitive demands [47,49,50].

However, with aging and the resultant changes in social roles, middle-aged and older adult individuals may face difficulties in maintaining their current interpersonal relationships and social networks [51]. Older adults are more susceptible to feelings of isolation and may experience poorer mental health due to limited social engagement, including factors such as lack of social connections, smaller social networks, or reduced social interaction [52]. Interestingly, individuals with limited social ties often derive significant benefits from formal social engagements, such as participating in volunteer or charity work, or attending educational or training courses, which can lead to reductions in depressive symptoms and improvement in life quality [53].

Internet Usage and Social Participation

The internet, unconstrained by time and space, enables individuals to establish connections with the external world. Using online platforms such as WeChat and Weibo, older adults can exchange information, share daily life events, and discover clubs, volunteer activities, and community organizations [54]. As a vital way for older adults to engage in offline activities, the internet facilitates the establishment of new social connections, cultivates a sense of belonging, and ultimately contributes to enhancing their mental health [55,56].

Internet Usage, Social Participation, and Depression During the PandemicInternet Usage During the Pandemic

In efforts to minimize the spread of COVID-19 pandemic, individuals were advised to stay at home and maintain social distancing [57]. The decrease in face-to-face contact led to a significant uptick in the amount of time the public spent on internet usage, particularly on social media [15]. A study on Chinese adults revealed a substantial increase in internet usage, with 72% of participants indicating a growing dependence on online connectivity [58]. Social media usage among adults increased significantly, with weekly social media usage increasing from 17.2 hours to 21.4 hours [59]. Similarly, the older adults reported comparable increases in their use of digital platforms. A study of older adults aged 60 years and older in the United States found that they have significantly increased their use of social media and spent more time on their computers and tablets [60].

Internet Usage and Depression During the Pandemic

During the COVID-19 pandemic, older adults turned to online chat as a substitute for face-to-face communication with family members who were not living together [61]. This increased internet usage has led to more frequent contact between older adults and their children, reducing social loneliness and perceived stress levels, thus positively impacting their mental health [62,63]. Additionally, older adults have more time to use the internet for leisure activities related to their hobbies and interests, which can mitigate the negative impacts of the pandemic and enhance their well-being [64].

It is critical to acknowledge that social media can spread misinformation, alarmist content, and exaggerated narratives, potentially fostering fear, anxiety, and depression across the broader population [65]. Individuals who excessively relied on social media for COVID-19 information and continuously focused on pandemic updates may experience higher levels of anxiety and depression symptoms during the pandemic [66-69].

Social Participation and Depression During the Pandemic

To prevent the spread of the COVID-19 pandemic, individuals were urged to minimize outdoor activities. A global survey showed a dramatic decrease in adults’ engagement in social activities with family, friends, neighbors, or through recreation, noting a 71.15% drop in total social participation scores as a result of COVID-19 home confinement [70]. In the United States, older adults experienced significant reductions in social interactions and engagements with their families and friends, with 1 out of every 4 older individuals reporting fewer than 1 in-person interaction per week with family and friends [71,72]. The Japanese older adults reported a decrease in both the extent and intensity of community participation [73]. Pandemic-related social distancing measures reduced older adults’ social participation in community groups and family activities, leading to a decline in physical activity and increased feelings of loneliness, which negatively affected their physical and mental health [4,8,74,75].

Study Objectives

This study aims to explore changes in the impact of internet usage on depression among middle-aged and older adults before and during the pandemic. This study further explores whether the mediating effect of social participation changed during the pandemic. Finally, it examines how social participation plays a role in the relationship between different internet usage functions and depressive symptoms and conducts subgroup analyses by urban or rural residence and gender.

MethodsSample and Data Collection

This study employs data from the China Health and Retirement Longitudinal Study (CHARLS). CHARLS uses a probability proportional to size sampling method, systematically selecting samples from counties, villages, residential areas, households, and individuals. The CHARLS baseline survey covered 150 counties or districts and 450 villages or urban communities across the country, involving 10,257 households and 17,708 individuals, offering comprehensive representation of the middle-aged and older adult population in China. CHARLS respondents are followed every 2 years, using a face-to-face computer-assisted personal interview [76]. At present, CHARLS has conducted 5 waves of surveys (2011, 2013, 2015, 2018, and 2020). For each respondent interviewed in the baseline survey, CHARLS will follow up with them in each subsequent survey. The primary respondents and their spouses from the baseline survey will be tracked throughout the entire life cycle of CHARLS. The response rate for the panel sample has consistently remained above 86% in all rounds of follow-up surveys. Specifically, the repeat participation rates for the 2018 and 2020 surveys were 86.46% and 86.81%, respectively [77]. From late 2019 to early 2020, China experienced the COVID-19 pandemic, which led to a reduction in traditional forms of social participation. The fifth wave of CHARLS was conducted from July to September 2020. During the routine prevention and control phase of the COVID-19 pandemic, it is possible to observe changes in internet usage, social participation, and depressive symptoms among Chinese middle-aged and older adults. Therefore, we used data from the 2018 and 2020 surveys to compare changes in the mediating role of social participation in the relationship between internet usage and depression among middle-aged and older adults before and during the pandemic.

The data-cleaning process proceeded as follows: (1) the data from the 2 waves were matched and merged, resulting in a total of 39,084 observations; (2) cases with missing values for Center for Epidemiological Studies Depression Scale (CESD-10) and internet usage were excluded, totaling 6978 observations; and (3) an additional 1022 cases with missing values in control variables were removed. A total of 15,326 respondents in 2018 and 15,758 respondents in 2020 were selected, with 12,549 respondents participating repeatedly. We used data from the 2018 and 2020 waves to construct the panel dataset, encompassing a total of 31,390 observations.

Ethical Considerations

Prior to the survey, each participant received a written informed consent form outlining the purpose of the study. In accordance with the Declaration of Helsinki, all CHARLS surveys obtained ethical approval from the Institutional Review Board of Peking University (IRB00001052-11015). The survey content is strictly confidential, and all data about respondents are protected by data security and privacy laws. Images and supplementary materials ensure participant anonymity, with no identifiable information included.

VariablesDepression Symptoms

The dependent variable in this study was depression symptoms, measured using the CESD-10 scores. This scale consists of 10 items and has good validity in measuring emotional and physical distress [78]. The Chinese version of the CESD Scale used in CHARLS demonstrated good reliability and validity, with Cronbach α value reaching 0.815 [79]. In this study, the CESD-10 demonstrated good internal consistency, with a Cronbach α value of 0.800, and sampling adequacy, with a Kaiser-Meyer-Olkin value of 0.881. This scale comprises 8 negative questions and 2 positive questions, each with 4 possible responses. For the negative questions, respondents were scored on a scale of 0‐3 based on the frequency of their symptoms: “rarely or none of the time (<1 day), some or a little of the time (1‐2 days), occasionally or a moderate amount of the time (3‐4 days), and most or all of the time (5‐7 days).” The positive questions were scored in the opposite direction. A cumulative score ranging from 0 to 30 was computed, where higher scores signify elevated levels of depressive symptoms.

Internet Usage

In this study, the primary explanatory variable was internet usage. In the CHARLS questionnaire, participants were asked: “Have you used the Internet in the past month,” with responses coded as 1 for “yes” and 0 for “no.” Internet usage functions were measured using the question, “What do you usually do on the Internet?” Referring to the studies by Jiang and Luo [80] and Du et al [81], internet usage was divided into 3 distinct functions: social internet usage (eg, chatting), informational internet usage (eg, watching news and managing finances), and recreational internet usage (eg, watching videos and playing games).

Social Participation

This study used social participation as a mediating variable, which was assessed using 6 types of activities from the CHARLS questionnaire [82,83]. The selected activities included (1) visiting and socializing with friends; (2) playing mahjong, chess, cards, and engaging in community room activities; (3) providing help to noncohabiting relatives, friends, or neighbors; (4) attending social, sports, or other kinds of clubs; (5) involving in community organizations; and (6) volunteering, engaging in charity work, or caring for sick or disabled individuals not living with the respondent. Referring to the study by Du et al [84] and Yang et al [85], each activity was assigned 1 point, resulting in a total score ranging from 0 to 6, with higher scores indicating higher levels of social participation among respondents.

Control Variables

Based on previous studies [84,86-89], we included several control variables in the analysis that have been associated with depressive symptoms in the CHARLS dataset. These variables included gender (man=0, woman=1), age (continuous variable), residence (urban=0, rural=1), education (illiterate=0, primary school or below=1, middle school=2, and high school or above=3), marital status (others=0, married=1), chronic diseases (none=0, 1 type=1, and two or more types=2), sleep duration, nap duration, physical activity, drinking habits (no=0, yes=1), annual per capita consumption expenditure, household size (continuous variable), household composition (living alone=0, living with spouse only=1, and others=2), health care accessibility (no=0, yes=1), and employment status (unemployed=0, employed=1). Marital status was categorized as “others,” which includes those who are separated, divorced, widowed, or never married. Sleep duration was divided into 5 groups:<6 hours, 6‐7 hours, 7‐8 hours, 8‐9 hours, and ≥9 hours per night [90]. Nap duration was categorized into 4 groups: nonnappers (0 minutes), short nappers (<30 minutes), moderate nappers (30‐90 minutes), and excessive nappers (>90 minutes) [91]. Physical activity was measured based on intensity of activity performed each week and was classified into 3 groups: light, moderate, and intensive physical activity [92]. Health care accessibility was defined based on the concept of the “15-minute medical circle,” referring to the ability of residents to access health care services within a 15-minute walk or via convenient transportation.

Data AnalysisTwo-Way Fixed-Effects and Instrumental Variables Model

Two-way fixed-effects (TWFE) model is a panel data method that accounts for both individual and time fixed effects, thereby controlling for unobserved confounders that are constant across time or individuals. Given that the CESD-10 score used in this study was a continuous variable, we employed a TWFE model to examine the impact of internet usage and social participation on depressive symptoms over time. Furthermore, given the potential endogeneity issues, we used smartphone usage as an instrumental variable for internet usage for middle-aged and older adults. The instrumental variable–fixed-effects (IV-FE) model was used to estimate the impact of internet usage on depression symptoms.

In the first stage, the instrumental variable (smartphone usage) was used to predict the endogenous variable, internet usage, yielding the fitted values, Internet_useit→OLSIVit, to obtain the fitted values Internet_useit^. In the second stage, the outcome variable, depressive symptoms, was regressed on the fitted internet usage values obtained from the first stage. The premise of using IV was the existence of endogenous explanatory variables. In addition, the correlation and validity of IV were tested. The basic model was as follows:

(1)Depressionit=β0Internet_useit++β1XXit+σi+γt+μit

(2)Internet_useit=α0Zit+α1Tit+σi+γt+εit

where Depressionit represents the continuous outcome variable and Internet_useit is a binary indicator of internet usage. XXit denotes a vector of control variables. Zit represents smartphone use, and Tit is the exogenous in the equation of Depressionit variable, σi and γt capture individual and time fixed effects, respectively, while μit and εit are the error terms.

Propensity Score Matching–Based Mediating Effect Model

To address potential selection bias arising from the non-random nature of internet usage, we implemented propensity score matching (PSM) prior to mediating effect analysis [93,94]. Propensity scores were estimated using k-nearest neighbor matching, radius matching, and kernel-matching methods. The following covariates were included in the PSM adjustment: gender, age, residence, education, marital status, presence of chronic disease, sleep duration, nap duration, physical activity, drinking behavior, per capita household expenditure, household size and composition, health care accessibility, and employment status. To explore the mechanism underlying the impact of internet usage on depression among middle-aged and older individuals, a mediating effect model was constructed following the approach of Wen et al [95]. The models were set as follows:

(3)Social_participationit=α2+β2Internet_useit+β3Xit+δit

(4)Depressionit=α3+β4Internet_useit+β5Social_participationit+β6Xit+ϕit

where β2 represents the regression coefficient of internet usage on mediating variable, and β3 denotes the regression coefficient of internet usage on CESD-10 score among middle-aged and older individuals after adding mediating variable. Mediation analysis was conducted using the sgmediation package in Stata (version 14.0; StataCorp), with 5000 bootstrap replications to validate the mediating effect of social participation before and during the COVID-19 pandemic [96]. This allowed us to estimate the mediating effect of social participation and obtain bias-corrected 95% CIs for the indirect, direct, and total effects. The root-mean-square error of approximation for the mediation model was 0.000, which is below the cutoff value of 0.08, indicating excellent model fit. Moreover, we examined the proportion of the mediating effect of social participation between different internet usage functions and depression symptoms.

ResultsDescriptive Statistics

Figure 1 shows the flowchart of the study population selection process. Table S1 in Multimedia Appendix 1 shows the participant characteristics at each wave of data collection. The results indicated an increase in depression symptom scores in 2020 compared with 2018. CESD-10 scores rose from 1.94 (SD 0.86) to 1.98 (SD 0.84). Social participation scores dropped from 0.81 (SD 0.99) in 2018 to 0.79 (SD 1.00) in 2020. In 2018, the majority of respondents were women (7820/15,326, 51.02%), with 70.59% residing in rural areas and 87.75% being married. Of the respondents in 2020, the majority lived in rural areas (9967/15,758, 63.25%) and most of them were women (8162/15,758, 51.80%). A large proportion had a primary school education or below (6886/15,758, 43.70%). The majority of respondents were married ( 13,599/15,758, 86.30%).

Figure 1.

Flowchart of the study population selection process. CHARLS: China Health and Retirement Longitudinal Study.

Two-Way Fixed-Effects Regression Results

The results of the TWFE model are shown in Table 1. As shown, both internet usage (β=−.041; P<.01) and social participation (β=−.016; P<.05) have a negative impact on depression symptoms. Additionally, depression symptoms worsened between 2018 and 2020 (β=.060; P<.001).

Table 1.

Results of regressions with 2-way fixed effects (N=31,084).

Variable	β coefficient	SE	P value
Internet usage (reference: no)	−0.041	0.015	.008
Social participation	−0.016	0.007	.02
Residence (reference: urban)	0.012	0.020	.06
Marital status (reference: Others)	−0.171	0.049	.001
Chronic diseases (reference: None)
1 type	0.018	0.013	.16
≥2 types	0.106	0.017	<.001
Sleep duration (hours), reference: <6
6-7	−0.091	0.016	<.001
7-8	−0.123	0.019	<.001
8-9	−0.163	0.020	<.001
≥9	−0.128	0.027	<.001
Nap duration (reference: nonnappers)
Short nappers	−0.020	0.017	.39
Moderate nappers	−0.004	0.016	.81
Excessive nappers	−0.017	0.022	.44
Physical activity (reference: light)
Moderate	−0.008	0.016	.61
Intensive	−0.005	0.016	.77
Drinking (reference: no)	0.022	0.008	.01
Household expenditure per capita	−0.009	0.007	.03
Household size	−0.007	0.005	.26
Household composition (reference: living alone)
Living with spouse only	0.007	0.035	.84
Others	0.025	0.039	.53
Employment status (reference: no)	−0.035	0.019	.07
Year (reference: 2018)	0.048	0.010	<.001

IV-FE Regression Results

Table 2 reports the results of the IV-FE regression. The coefficient of internet usage on depression is −0.044, statistically significant at the 1% level. The Kleibergen-Paap rk LM statistic and the Cragg-Donald Wald F statistic confirm the validity of the instruments [97]. This indicates that the negative impact of internet usage on depression among middle-aged and older adults remains robust after controlling for sources of endogeneity.

Table 2.

Result of the instrumental variable–fixed-effects method (N=25098).

Variable	β coefficient	SE	P value
Internet usage (reference: no)	−0.044	0.016	.006
Social participation	−0.016	0.007	.02
Residence (reference: rural)	0.012	0.021	.56
Marital status (reference: married)	0.171	0.049	<.001
Chronic diseases (reference: ≥2 types)
None	−0.106	0.015	<.001
One type	−0.089	0.017	<.001
Sleep duration (hours), reference:≥9
<6	0.128	0.028	<.001
6-7	0.037	0.028	.18
7-8	0.005	0.028	.86
8-9	−0.034	0.028	.21
Nap duration (reference: excessive nappers)
Nonnappers	0.017	0.023	.45
Short nappers	−0.003	0.027	.91
Moderate nappers	0.013	0.021	.52
Physical activity (reference: intensive)
Light	0.005	0.016	.77
Moderate	−0.003	0.014	.83
Drinking (reference: yes)	−0.022	0.008	.01
Household expenditure per capita	−0.009	0.007	.19
Household size	−0.007	0.006	.26
Household composition (reference: others)
Living alone	−0.025	0.037	.84
Living with spouse only	−0.017	0.019	.51
Employment status (reference: yes)	0.034	0.018	.06
Year (reference: 2020)	−0.049	0.011	<.001
KP-LM^a	5883.626
CD-Wald^b	3.0e+05

^aKP-LM: Kleibergen-Paap rk LM statistic, testing the relevance of IV.

^bCD-Wald: Cragg-Donald Wald F statistic, testing the weak relevance of IV.

Mediating Effects After PSM Analysis

Figures 2 and 3 and Table 3 illustrate the PSM-based indirect effects examined between internet usage and depression before and during the COVID-19 pandemic. The findings demonstrated that social participation significantly (P<.001) played a role in mediating the relationship between internet usage and depression symptoms both in 2018 and 2020. Particularly, internet usage among middle-aged and older adults led to a notable increase in their level of social participation (2018, β=.435, P<.001; 2020, β=.336, P<.001), indicating a moderate to strong standardized effect size. Significant direct effects (2018, β=−.118, P<.001; 2020, β=−.095, P<.001) as well as indirect effects (2018, β= −.010, P<.01; 2020, β=−.011, P<.001) of internet usage on depression were observed among the middle-aged and older groups. The coefficient reflecting the impact of internet usage on depression decreased when social participation was included as a mediator. With the inclusion of social participation, the coefficient changed from −0.128 (P<.001) to −0.118 (P<.001) in 2018, and from −0.106 (P< .001) to −0.095 (P<.001) in 2020.

The balance test results are shown in Tables S2-S4 in Multimedia Appendix 1. After matching, the standardized differences of covariates were substantially reduced, and no significant differences were observed between the treatment and control groups. These results indicate that systematic differences between groups were largely eliminated.

Table S5 in Multimedia Appendix 1 shows the mediation analysis results using 3 different matching methods: k-nearest neighbor, radius, and kernel matching. The findings consistently support the main results, demonstrating a robust mediating effect of social participation on the relationship between internet usage and depressive symptoms among middle-aged and older Chinese adults.

Figure 2.

Path diagram of the impact of internet usage on depression symptoms in 2018. The model controls for residence, marital status, chronic diseases, sleep duration, nap duration, physical activity, drinking, household expenditure per capita, household size, household composition, and employment status. ^*P<.001.

Figure 3.

Path diagram of the impact of internet usage on depression symptoms in 2020. The model controls for residence, marital status, chronic diseases, sleep duration, nap duration, physical activity, drinking, household expenditure per capita, household size, household composition, and employment status. ^*P<.001.

Table 3.

Mediating effect of social participation between internet usage and depression.

Types	β coefficient (bootstrapped 95% CI)	SE	P value
2018
Indirect effect	−0.010 (−0.018 to −0.003)	0.004	.008
Direct effect	−0.119 (−0.160 to −0.075)	0.022	<.001
Total effect	−0.128 (−0.170 to −0.086)	0.022	<.001
2020
Indirect effect	−0.011 (−0.016 to −0.006)	0.003	<.001
Direct effect	−0.095 (−0.126 to −0.064)	0.015	<.001
Total effect	−0.106 (−0.137 to −0.076)	0.016	<.001

Proportion of Mediating Effects in Social Function, Getting Information, and Recreational Function

Table 4 shows the indirect effects of internet usage functions on depression, both prior to and during the COVID-19 pandemic. It reveals that social participation acted as a significant partial mediator in the relationship between different internet usage functions and depression. To be specific, in 2018 and 2020, the indirect effect of social participation on social function and depression was −0.012 (P<.01), and −0.010 (P<.001), respectively. Similarly, the indirect effects of social participation on getting information and depression were −0.011 (P<.01) and −0.011 (P<.001), while for recreational function and depression, the indirect effects were −0.012 (P<.01) in 2018 and −0.011 (P<.001) in 2020. Figure S1 in Multimedia Appendix 1 shows the proportion of mediating effects of social participation between different internet usage functions and depression. It illustrated that in 2018, the mediating effect of social participation between social function and depressive symptoms was more pronounced than in 2020. Conversely, the proportion of the mediating effect of social participation between informational function and depressive symptoms did not significantly change before and during the pandemic.

Table 4.

The indirect effects of internet usage functions analyzed by pandemic time.

Indirect effects of internet usage analyses by functions	β (bootstrapped 95% CI)	SE	P value
2018
Social function → Social participation → Depression symptoms	−.012 (−0.020 to −0.004)	0.004	.002
Informational function → Social participation → Depression symptoms	−.011 (−0.019 to −0.003)	0.004	.007
Recreational function → Social participation → Depression symptoms	−.012 (−0.019 to −0.004)	0.004	.003
2020
Social function → Social participation → Depression symptoms	−.010 (−0.015 to −0.005)	0.002	<.001
Informational function → Social participation → Depression symptoms	−.011 (−0.017 to −0.006)	0.003	<.001
Recreational function → Social participation → Depression symptoms	−.011 (−0.016 to −0.006)	0.003	<.001

Subgroup Analysis

Table 5 presents the mediating effects of social participation across subgroups defined by residence (urban vs rural) and gender (man vs woman). Social participation significantly mediated the relationship between internet usage and depressive symptoms (P<.001). Specifically, internet usage was associated with reduced depression scores among middle-aged and older adults, with pronounced heterogeneity observed across demographic subgroups.

In 2018, the indirect effect of social participation was not significant among rural residents (β=−.006, 95% CI −0.016 to 0.005) and women (β=−.007, 95% CI −0.017 to 0.004). By 2020, however, these effects became significant in both subgroups—rural: β=−.012 (95% CI −0.019 to −0.005); woman: β=−.011 (95% CI −0.018 to −0.005)—with higher mediation proportions observed compared with urban residents and men. The direct effects of internet usage on depression also showed consistent negative associations across all groups, with rural residents and women exhibiting slightly stronger direct effects in 2018. By 2020, direct effects converged across subgroups, while the emergence of significant indirect effects in rural groups and women highlighted the growing role of social participation as a mediator. The total effect of social participation on the relationship between internet usage and depressive symptoms was greater among rural residents and women, both before and during the pandemic.

Table 5.

Subgroup analysis of the mediating effect of social participation between internet usage and depressive symptoms.

Types	Residence		Gender
Types	Urban, β (bootstrapped 95% CI)	Rural, β (bootstrapped 95% CI)	Man, β (bootstrapped 95% CI)	Woman, β (bootstrapped 95% CI)
2018
Indirect effect	−.014^a(−0.025 to −0.002)	−.006(−0.016 to −0.005)	−.012^a(−0.024 to −0.001)	−.007(−0.017 to 0.004)
Direct effect	−.090^a(−0.151 to −0.028)	−.146^b(−0.205 to −0.087)	−.110^b(−0.169 to −0.052)	−.127^b(−0.189 to −0.066)
Total effect	−.103^b(−0.164 to −0.042)	−.151^b(−0.210 to −0.092)	−.123^b(−0.181 to −0.065)	−.134^b(−0.196 to −0.073)
2020
Indirect effect	−.008^c(−0.017 to 0.001)	−.012^b(−0.019 to −0.005)	−.009^a(−0.017 to −0.002)	−.011^a(−0.018 to −0.005)
Direct effect	−.089^b(−0.142 to −0.036)	−.093^b(−0.132 to −0.054)	−.103^b(−0.153 to −0.053)	−.102^b(−0.146 to −0.058)
Total effect	−.097^b(−0.149 to −0.045)	−.105^b(−0.144 to −0.066)	−.112^b(−0.162 to −0.063)	−.113^b(−0.157 to −0.069)

^aP<.05.

^bP<.001.

^cP<.01.

DiscussionPrincipal Findings

Using data from the CHARLS and applying the IV-FE method, this study found that internet usage was negatively associated with depressive symptoms among middle-aged and older adults. Social participation partially mediated this relationship. Although depression scores increased in 2020 compared with 2018, the mediating effect of social participation weakened, while the direct effect of internet was strengthened during the pandemic. Additionally, the proportion of the mediating effect of social participation on the relationship between internet usage functions and depressive symptoms shifted over time. Specifically, the social function declined from 12.55% in 2018 to 9.30% in 2020, whereas the informational and recreational functions increased from 7.53% and 11.29% in 2018 to 8.85% and 16.37% in 2020, respectively.

Findings from the study indicate an increase in the average CESD-10 score among middle-aged and older Chinese individuals, rising from 1.94 in 2018 to 1.98 in 2020. This upward trend was less pronounced compared with other countries during the same time frame [98-100]. The prevalence of milder or greater depressive symptoms increased by 1.6 times among the older adults with diabetes in the United States. In the United Kingdom, the prevalence of mental health problems in adults rose from 24.3% in 2017‐2019 to 37.8% in April 2020. Additionally, the prevalence of depression among older adults in Ireland surged from 7.2% before the pandemic to 19.8% during the pandemic. Depressive symptoms in China during the COVID-19 pandemic were less severe than in other countries, primarily due to China’s web-based mental health services and community support [101,102].

This study suggests that internet usage had a significant reduction of 4.10% in depressive symptoms, aligning with findings from previous research [103-106]. Meanwhile, we found that the internet played a mitigating role in the worsening of depressive symptoms among middle-aged and older adults following the outbreak. The reason may be that the internet enables the older adults to access health-related information, monitor health indicators, and enhance communication with family and friends [12,39]. Particularly, the internet serves as a platform for middle-aged and older adults to gather information on social activities, leading to greater engagement in such activities [107,108]. Through internet technology, middle-aged and older adult people can explore new social participation opportunities, assume new social roles, and expand their social networks based on existing interpersonal relationships [109,110]. Notably, our study indicates that web-based social participation may act as a complement to offline engagement, aiding individuals in avoiding social isolation [33,111].

In this study, we also explore the relationship between different internet usage functions, social participation, and depressive symptoms before and during the COVID-19 pandemic. First, in terms of social function, the proportion of the mediating effect of social participation decreased from 12.55% in 2018 to 9.30% in 2020. One explanation is that the social functions of the internet increase enhanced interactions among people and facilitate their engagement in various volunteer activities [112], thereby expanding interpersonal networks, improving the quality of interpersonal relationships, and obtaining more emotional support [34]. Another explanation may be a substitution effect between social functions and social participation, wherein certain offline social activities, such as playing mahjong or card games, were replaced by web-based activities. Second, the mediation effect proportion of social participation between information acquisition function and depressive symptoms had slightly increased before and during the pandemic. One possible reason for this is that people pay more attention to health and pandemic-related information (eg, infection and mortality rates) during the pandemic [69,113]. Finally, concerning recreational functions, the mediating effect of social participation increased during the pandemic. The internet provides a wide range of recreational activities, allowing individuals to choose leisure activities based on their interests [84]. Additionally, during the pandemic, people had ample spare time to partake in web-based recreational activities, such as gaming or watching videos, as a means to alleviate anxiety and depression [114].

Given the urban-rural digital divide and gender-based differences in internet usage patterns [85], stratified analyses revealed that the proportion of mediation was higher among rural residents and women than among urban residents and men. This difference may reflect disparities in infrastructure and access across settings [34]. Despite limitations in rural internet infrastructure, older adults who do access the internet may gain stronger social and psychological benefits. For rural populations, internet usage may serve as a critical tool to overcome reduced mobility and limited social opportunities during the pandemic, thereby fostering social inclusion and emotional support that contribute to reduced depressive symptoms [85]. Gender differences in the mediating role of social participation may also stem from distinct usage patterns and social norms. Women tend to use the internet more for social communication and emotional connection, which may enhance the protective effect of internet usage against depression via increased social participation. Men, on the other hand, may use the internet more for information seeking or entertainment, which might have a less direct impact on social engagement and thus a weaker mediating effect.

Limitations

Several limitations should also be aware of in this study. First, the study did not perform a comprehensive analysis of the intensity and types of internet usage, including daily time spent on the web, which could impact mental health. Second, reliance on self-reported data for depression and internet usage may introduce biases, such as recall bias or social desirability. Third, reliance on cross-sectional data for mediation analysis in our study limits our capacity to determine a causal relationship between internet usage and depression. Future studies should delve deeper into the causal association between internet usage duration, device preference, and mental health in middle-aged and older adults, and examine the role of different types of social participation.

Conclusions

Using national survey data, this study examined the mechanisms of internet usage on depression in middle-aged and older adults, across periods both before and during the COVID-19 pandemic. Internet usage was found to mitigate depression levels, with social participation serving as a key mediator between internet usage and depression. Therefore, measures to promote and amplify the positive impact of internet usage among older individuals are vital, such as offering community-based computer training, developing age-friendly web-based apps, and leveraging big data for providing services tailored to the older adults. Considering the influence of digital literacy on the effectiveness of internet usage, targeted training programs for older adults, especially for rural residents and women, should be prioritized. Mental health recovery in aging populations should also be advanced by integrating internet usage facilitation and social engagement strategies into broader public health and social welfare systems.

This work was supported by the Humanities and Social Sciences Foundation of Ministry of Education (number 21YJC630060), Natural Science Foundation of China (number 72274108), and Natural Science Foundation of Shandong province (number ZR2022MG003).

Data Availability

The data used in this study are available at China Health and Retirement Longitudinal Study [115]. Authors are required to apply for permission to use the data.

JL and SZ contributed to the conception or design of the study. SZ contributed to data collection and data analysis. SZ and SD drafted the paper while JL guided the revisions and funding acquisition. ZY, SB, and WW contributed to the writing–editing and supervision. SZ and SD contributed equally to this work and should be considered cofirst authors.

None declared.

Abbreviations

CESD

Center for Epidemiological Studies Depression Scale

CHARLS

China Health and Retirement Longitudinal Study

IV-FE

instrumental variable–fixed effects

PSM

propensity score matching

TWFE

2-way fixed effects

References1

Zenebe

Akele

W/Selassie

Necho

Prevalence and determinants of depression among old age: a systematic review and meta-analysis

Ann Gen Psychiatry2021121820155

10.1186/s12991-021-00375-x

34922595

Shah

SMA

Mohammad

Qureshi

MFH

Abbas

Aleem

Prevalence, psychological responses and associated correlates of depression, anxiety and stress in a global population, during the coronavirus disease (COVID-19) pandemic

Community Ment Health J202101571101110

10.1007/s10597-020-00728-y

Ettman

Abdalla

Cohen

Sampson

Vivier

Galea

Prevalence of depression symptoms in US adults before and during the COVID-19 pandemic

JAMA Netw Open202009139e2019686

10.1001/jamanetworkopen.2020.19686

32876685

Sepúlveda-Loyola

Rodríguez-Sánchez

Pérez-Rodríguez

Impact of social isolation due to COVID-19 on health in older people: mental and physical effects and recommendations

J Nutr Health Aging2020249938947

10.1007/s12603-020-1469-2

33155618

Yurumez Korkmaz

Gemci

Cosarderelioglu

Attitudes of a geriatric population towards risks about COVID-19 pandemic: in the context of anxiety and depression

Psychogeriatrics202109215730737

10.1111/psyg.12731

34132456

Silva

Fonseca

Ferreira

Depression in older adults during the COVID-19 pandemic: a systematic review

J Am Geriatr Soc20230771723082325

10.1111/jgs.18363

37029710

Mistry

Ali

ARMM

Hossain

Exploring depressive symptoms and its associates among Bangladeshi older adults amid COVID-19 pandemic: findings from a cross-sectional study

Soc Psychiatry Psychiatr Epidemiol20210856814871497

10.1007/s00127-021-02052-6

33661353

Social isolation and loneliness among older adults in the context of COVID-19: a global challenge

Glob Health Res Policy2020527

10.1186/s41256-020-00154-3

32514427

Coley

Baum

Trends in mental health symptoms, service use, and unmet need for services among US adults through the first 8 months of the COVID-19 pandemic

Transl Behav Med20220216122273283

10.1093/tbm/ibab133

34662427

Fan

Ghani

Machairas

COVID-19 prevention and treatment information on the internet: a systematic analysis and quality assessment

BMJ Open20200910109e040487

10.1136/bmjopen-2020-040487

32912996

Gong

Cai

Chen

Wang

Internet hospitals help prevent and control the epidemic of COVID-19 in China: multicenter user profiling study

J Med Internet Res20200414224e18908

10.2196/18908

32250962

Zhou

Internet use and Chinese older adults’ subjective well-being (SWB): the role of parent-child contact and relationship

Comput Human Behav202106119106725

10.1016/j.chb.2021.106725

Liu

Zhang

Effects of internet hospital consultations on psychological burdens and disease knowledge during the early outbreak of COVID-19 in China: cross-sectional survey study

J Med Internet Res2020084228e19551

10.2196/19551

32687061

Liu

Social media use, eHealth literacy, disease knowledge, and preventive behaviors in the COVID-19 pandemic: cross-sectional study on Chinese netizens

J Med Internet Res20201092210e19684

10.2196/19684

33006940

Saud

Mashud

Ida

Usage of social media during the pandemic: seeking support and awareness about COVID-19 through social media platforms

J Public Affairs2020e02417

10.1002/pa.2417

Xie

Charness

Fingerman

Kaye

Kim

Khurshid

When going digital becomes a necessity: ensuring older adults’ needs for information, services, and social inclusion during COVID-19

J Aging Soc Policy2020324-5460470

10.1080/08959420.2020.1771237

32507061

Guo

Zhu

Does mobile internet use affect the loneliness of older Chinese Adults? An instrumental variable quantile analysis

Int J Environ Res Public Health20220541995575

10.3390/ijerph19095575

35564970

Liu

Peng

Chen

Zhang

Mediating and moderating effects of internet use on urban-rural disparities in health among older adults: nationally representative cross-sectional survey in China

J Med Internet Res202325e45343

10.2196/45343

Guo

Feng

Liu

The temperature of internet: internet use and depression of the elderly in China

Front Public Health2022101076007

10.3389/fpubh.2022.1076007

Jun

A study on cause analysis of digital divide among older people in Korea

Int J Environ Res Public Health2021081418168586

10.3390/ijerph18168586

34444334

Blažič

Overcoming the digital divide with a modern approach to learning digital skills for the elderly adults

Educ Inf Technol (Dordr)2020251259279

10.1007/s10639-019-09961-9

Heponiemi

Kaihlanen

Kouvonen

Leemann

Taipale

Gluschkoff

The role of age and digital competence on the use of online health and social care services: a cross-sectional population-based survey

Digit Health2022820552076221074485

10.1177/20552076221074485

35111333

Hong

Zhou

Fang

Shi

The digital divide and health disparities in China: evidence from a national survey and policy implications

J Med Internet Res20170911199e317

10.2196/jmir.7786

28893724

Morris

Goodman

Brading

Internet use and non-use: views of older users

Univ Access Inf Soc2007057614357

10.1007/s10209-006-0057-5

Cornejo Müller

Wachtler

Lampert

Digital divide-social inequalities in the utilisation of digital healthcare

Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz202002632185191

10.1007/s00103-019-03081-y

31915863

Australian Bureau of Statistics

Household use of information technology

20180328

2025-08-24

https://www.abs.gov.au/statistics/industry/technology-and-innovation/household-use-information-technology/latest-release

China Internet Network Information Center

The 52nd statistical report on internet development in China

20230828

2025-08-24

https://www.cnnic.net.cn/n4/2023/0828/c88-10829.html

McMillan

Morrison

Coming of age with the internet: a qualitative exploration of how the internet has become an integral part of young people’s lives

New Media Soc2006817395

10.1177/1461444806059871

Pan

Jordan-Marsh

Internet use intention and adoption among Chinese older adults: from the expanded technology acceptance model perspective

Comput Human Behav20100926511111119

10.1016/j.chb.2010.03.015

Cotten

Ford

Hale

Internet use and depression among retired older adults in the United States: a longitudinal analysis

J Gerontol B Psychol Sci Soc Sci201409695763771

10.1093/geronb/gbu018

24671896

Zhang

Jiang

Zhang

Xiao

The longitudinal associations between perceived importance of the internet and depressive symptoms among a sample of Chinese adults

Front Public Health2023111167740

10.3389/fpubh.2023.1167740

37361143

Zhang

Wang

Liu

Does internet use promote mental health among middle-aged and older adults in China?

Front Psychol202213999498

10.3389/fpsyg.2022.999498

Zhao

The effect of internet use on depressive symptoms in middle-aged and older adults with functional disability: the mediating role of social isolation

Front Public Health2023111202541

10.3389/fpubh.2023.1202541

37492141

Nan

Xie

Internet use and depression among Chinese older adults: the mediating effect of interpersonal relationship

Front Public Health2023111102773

10.3389/fpubh.2023.1102773

36935716

Yang

Niu

The impact of internet use on health status among older adults in China: the mediating role of social support

Front Public Health2023111108096

10.3389/fpubh.2023.1108096

36908418

Russell

Campbell

Hughes

Ageing, social capital and the Internet: findings from an exploratory study of Australian “silver surfers”

Australas J Ageing2008062727882

10.1111/j.1741-6612.2008.00284.x

18713197

Lam

SSM

Jivraj

Scholes

Exploring the relationship between internet use and mental health among older adults in England: longitudinal observational study

J Med Internet Res20200728227e15683

10.2196/15683

32718913

AlMuammar

Noorsaeed

Alafif

Kamal

Daghistani

The use of internet and social media for health information and its consequences among the population in Saudi Arabia

Cureus202109139e18338

10.7759/cureus.18338

34722089

Zhang

Gao

Association between daily internet use and incidence of chronic diseases among older adults: prospective cohort study

J Med Internet Res2023071725e46298

10.2196/46298

37459155

Zhou

Impact of internet use on cognitive decline in middle-aged and older adults in China: longitudinal observational study

J Med Internet Res20220124241e25760

10.2196/25760

35072642

A U-shaped association between internet use and depression in China

Am J Health Behav20221230466627636

10.5993/AJHB.46.6.5

36721286

Yuan

Liu

Internet use and depressive symptoms among Chinese older adults: two sides of internet use

Front Public Health2023111149872

10.3389/fpubh.2023.1149872

36969621

Kraut

Patterson

Lundmark

Kiesler

Mukopadhyay

Scherlis

Internet paradox. A social technology that reduces social involvement and psychological well-being?

Am Psychol19980953910171031

10.1037//0003-066x.53.9.1017

9841579

Meshi

Cotten

Bender

Problematic social media use and perceived social isolation in older adults: a cross-sectional study

Gerontology2020662160168

10.1159/000502577

31522178

Morahan-Martin

Schumacher

Loneliness and social uses of the internet

Comput Human Behav200311196659671

10.1016/S0747-5632(03)00040-2

Levasseur

Richard

Gauvin

Raymond

Inventory and analysis of definitions of social participation found in the aging literature: proposed taxonomy of social activities

Soc Sci Med201012711221412149

10.1016/j.socscimed.2010.09.041

21044812

Cohn-Schwartz

Pathways from social activities to cognitive functioning: the role of physical activity and mental health

Innov Aging202043igaa015

10.1093/geroni/igaa015

32665981

Nielsen

Hinrichsen

Madsen

Participation in social leisure activities may benefit mental health particularly among individuals that lack social connectedness at work or school

Mental Health Soc Inclusion20211116254341351

10.1108/MHSI-06-2021-0026

Bourassa

Memel

Woolverton

Sbarra

Social participation predicts cognitive functioning in aging adults over time: comparisons with physical health, depression, and physical activity

Aging Ment Health201702212133146

10.1080/13607863.2015.1081152

Choi

Han

Chang

Social participation and depressive symptoms in community-dwelling older adults: emotional social support as a mediator

J Psychiatr Res202105137589596

10.1016/j.jpsychires.2020.10.043

33168196

Huang

Chinese middle-aged and older adults’ internet use and happiness: the mediating roles of loneliness and social engagement

J Appl Gerontol202112401218461855

10.1177/0733464820959168

32954905

Cornwell

Waite

Social disconnectedness, perceived isolation, and health among older adults

J Health Soc Behav2009035013148

10.1177/002214650905000103

19413133

Santini

Jose

Koyanagi

The moderating role of social network size in the temporal association between formal social participation and mental health: a longitudinal analysis using two consecutive waves of the Survey of Health, Ageing and Retirement in Europe (SHARE)

Soc Psychiatry Psychiatr Epidemiol202103563417428

10.1007/s00127-020-01961-2

33037448

Filsinger

Freitag

Internet use and volunteering: relationships and differences across age and applications

Int J Volunt Nonprofit Organ2019023018797

10.1007/s11266-018-0045-4

Cornejo

Tentori

Favela

Enriching in-person encounters through social media: a study on family connectedness for the elderly

Int J Hum Comput Stud201309719889899

10.1016/j.ijhcs.2013.04.001

Inoue

Wann

Lock

Sato

Moore

Funk

Enhancing older adults’ sense of belonging and subjective well-being through sport game attendance, team identification, and emotional support

J Aging Health2020327-8530542

10.1177/0898264319835654

30873892

Sjödin

Wilder-Smith

Osman

Farooq

Rocklöv

Only strict quarantine measures can curb the coronavirus disease (COVID-19) outbreak in Italy, 2020

Euro Surveill20200425132000280

10.2807/1560-7917.ES.2020.25.13.2000280

32265005

Huang

Chen

Huang

Substance and internet use during the COVID-19 pandemic in China

Transl Psychiatry20210923111491

10.1038/s41398-021-01614-1

34556627

Luo

Chen

Liao

Social media use in China before and during COVID-19: preliminary results from an online retrospective survey

J Psychiatr Res2021081403538

10.1016/j.jpsychires.2021.05.057

34090101

Emerson

Coping with being cooped up: social distancing during COVID-19 among 60+ in the United States

Rev Panam Salud Publica202044e81

10.26633/RPSP.2020.81

32612645

Noguchi

Hayashi

Kubo

Tomiyama

Ochi

Hayashi

Living alone and depressive symptoms among older adults in the COVID-19 pandemic: role of non-face-to-face social interactions

J Am Med Dir Assoc2023012411721

10.1016/j.jamda.2022.10.014

36435273

Özdemir

Çelen

Social loneliness and perceived stress among middle-aged and older adults during the COVID-19 pandemic

Curr Psychol20230130110

10.1007/s12144-023-04301-0

36742062

Zhou

Lonely

Feeling lonely, engaging online during the outbreak of COVID-19: a longitudinal study of Chinese older adults

Innov Aging202377igad090

10.1093/geroni/igad090

37841215

Nimrod

Changes in internet use when coping with stress: older adults during the COVID-19 pandemic

Am J Geriatr Psychiatry202010281010201024

10.1016/j.jagp.2020.07.010

32771312

González-Padilla

Tortolero-Blanco

Social media influence in the COVID-19 pandemic

Int Braz J Urol20200746Suppl.1120124

10.1590/S1677-5538.IBJU.2020.S121

32550706

Gao

Zheng

Jia

Mental health problems and social media exposure during COVID-19 outbreak

PLoS One2020154e0231924

10.1371/journal.pone.0231924

32298385

Nekliudov

Blyuss

Cheung

Excessive media consumption about COVID-19 is associated with increased state anxiety: outcomes of a large online survey in Russia

J Med Internet Res20200911229e20955

10.2196/20955

32788143

Wong

FHC

Liu

Leung

DKY

Consuming information related to COVID-19 on social media among older adults and its association with anxiety, social trust in information, and COVID-safe behaviors: cross-sectional telephone survey

J Med Internet Res20210211232e26570

10.2196/26570

33523825

Zou

Zhang

Sznajder

Factors influencing anxiety among WeChat users during the early stages of the COVID-19 pandemic in Mainland China: cross-sectional survey study

J Med Internet Res20210517235e24412

10.2196/24412

33878025

Ammar

Chtourou

Boukhris

COVID-19 home confinement negatively impacts social participation and life satisfaction: a worldwide multicenter study

Int J Environ Res Public Health2020082717176237

10.3390/ijerph17176237

32867287

Freedman

Kasper

Changes in older adults’ social contact during the COVID-19 pandemic

J Gerontol B Psychol Sci Soc Sci2022075777e160e166

10.1093/geronb/gbab166

Zhang

Luo

Hou

Crisis response during the COVID-19 pandemic: changes in social contact and social participation of older Americans

PLoS One2023187e0284935

10.1371/journal.pone.0284935

37498870

Sato

Kondo

Pre-pandemic individual- and community-level social capital and depressive symptoms during COVID-19: a longitudinal study of Japanese older adults in 2019-21

Health Place20220374102772

10.1016/j.healthplace.2022.102772

35228058

Egeljić-Mihailović

Brkić-Jovanović

Krstić

Simin

Milutinović

Social participation and depressive symptoms among older adults during the Covid-19 pandemic in Serbia: a cross-sectional study

Geriatr Nurs202244814

10.1016/j.gerinurse.2021.12.014

34999368

Noguchi

Hayashi

Kubo

Tomiyama

Ochi

Hayashi

Association between decreased social participation and depressive symptom onset among community-dwelling older adults: a longitudinal study during the COVID-19 pandemic

J Nutr Health Aging202125910701075

10.1007/s12603-021-1674-7

34725663

Zhao

Smith

Strauss

Yang

Cohort profile: the China Health and Retirement Longitudinal Study (CHARLS)

Int J Epidemiol2014024316168

10.1093/ije/dys203

23243115

Yaohui

Xinxin

Yafeng

The china health and retirement longitudinal study wave 5

2020

2025-05-18

National School of Development, Peking University

https://charls.charlsdata.com/Public/ashelf/public/uploads/document/2020-charls-wave5/application/CHARLS_2020_User_Guide_Chinese.pdf

Andresen

Malmgren

Carter

Patrick

Screening for depression in well older adults: evaluation of a short form of the CES-D

Am J Prev Med1994031027784

10.1016/S0749-3797(18)30622-6

8037935

Lei

Sun

Strauss

Zhang

Zhao

Depressive symptoms and SES among the mid-aged and elderly in China: evidence from the China Health and Retirement Longitudinal Study national baseline

Soc Sci Med201411120224232

10.1016/j.socscimed.2014.09.028

25261616

Jiang

Luo

Does internet use improve the physical and mental health of older adults in China? A study based on the digital divide theory

Geriatr Nurs202459526534

10.1016/j.gerinurse.2024.08.009

39146643

Qifeng

Jin

Kexin

Research on the impact of internet use on mental health of the elderly: analysis on the moderating role of education

N W Popul J2023442113

10.15884/j.cnki.issn.1007-0672.2023.02.001

Lin

A study on the factors influencing the community participation of older adults in China: based on the CHARLS2011 data set

Health Soc Care Community20170525311601168

10.1111/hsc.12415

28178751

Wang

Chen

Zhou

Shen

Zhang

Melancholy or mahjong? Diversity, frequency, type, and rural-urban divide of social participation and depression in middle- and old-aged Chinese: a fixed-effects analysis

Soc Sci Med201910238112518

10.1016/j.socscimed.2019.112518

31473574

Liao

Multidimensional internet use, social participation, and depression among middle-aged and elderly Chinese individuals: nationwide cross-sectional study

J Med Internet Res2023083025e44514

10.2196/44514

37647119

Yang

Zhang

Cheng

A study on the impact of Internet use on depression among Chinese older people under the perspective of social participation

BMC Geriatr20220824221701

10.1186/s12877-022-03359-y

35999498

Gao

The relationship between internet use and mental health among Chinese residents during the (COVID)-19 pandemic: a cross-sectional study

Am J Health Behav20230630473533538

10.5993/AJHB.47.3.10

37596757

Liu

Fan

Luo

Comparison of depressive symptoms and its influencing factors among the elderly in urban and rural areas: evidence from the China Health and Retirement Longitudinal Study (CHARLS)

Int J Environ Res Public Health20210471883886

10.3390/ijerph18083886

33917216

Xiang

Yan

Liao

Wang

Relationship between chronic diseases and depression: the mediating effect of pain

BMC Psychiatry2021096211436

10.1186/s12888-021-03428-3

34488696

Wang

Bishwajit

Zhou

Intensity, frequency, duration, and volume of physical activity and its association with risk of depression in middle- and older-aged Chinese: evidence from the China Health and Retirement Longitudinal Study, 2015

PLoS One2019148e0221430

10.1371/journal.pone.0221430

Guo

Fei

Zhang

Luo

Chen

Gender- and age-specific associations between sleep duration and prevalent hypertension in middle-aged and elderly Chinese: a cross-sectional study from CHARLS 2011-2012

BMJ Open201609669e011770

10.1136/bmjopen-2016-011770

27601494

Cacchione

Hodgson

Afternoon napping and cognition in Chinese older adults: findings from the China Health and Retirement Longitudinal Study Baseline Assessment

J Am Geriatr Soc201702652373380

10.1111/jgs.14368

27995615

Fan

Lyu

Chinese guidelines for data processing and analysis concerning the International Physical Activity Questionnaire

Zhonghua Liu Xing Bing Xue Za Zhi201408358961964

25376692

Houser

Zhang

Wen

Zhang

Association between using social media WeChat and depressive symptoms among middle-aged and older people: findings from a national survey

BMC Geriatr202212221

10.1186/s12877-022-03054-y

Wang

Duan

Kan

Zhou

Cheng

Tang

The correlation analysis of WeChat usage and depression among the middle-aged and elderly in China: the mediating role of social participation

BMC Public Health2023231

10.1186/s12889-023-15349-9

Wen

Chang

Hau

Liu

Testing and application of the mediating effects

Acta Psychologica Sinica2004

2025-09-02

365614620

https://journal.psych.ac.cn/acps/EN/Y2004/V36/I05/614

Hayes

Scharkow

The relative trustworthiness of inferential tests of the indirect effect in statistical mediation analysis

Psychol Sci201310241019181927

10.1177/0956797613480187

Guo

Cheng

Zhao

Yin

Liu

Impact of internet usage on depression among older adults: comprehensive study

J Med Internet Res2025013127e65399

10.2196/65399

39890098

Briggs

McDowell

De Looze

Kenny

Ward

Depressive symptoms among older adults pre– and post–COVID-19 pandemic

J Am Med Dir Assoc202111221122512257

10.1016/j.jamda.2021.09.003

Chao

Wadden

Clark

Changes in the prevalence of symptoms of depression, loneliness, and insomnia in U.S. older adults with type 2 diabetes during the COVID-19 pandemic: the Look AHEAD Study

Diabetes Care20220114517482

10.2337/dc21-1179

100

Daly

Sutin

Robinson

Longitudinal changes in mental health and the COVID-19 pandemic: evidence from the UK Household Longitudinal Study

Psychol Med202210521325492558

10.1017/S0033291720004432

101

Liu

Yang

Zhang

Online mental health services in China during the COVID-19 outbreak

Lancet Psychiatry20200474e17e18

10.1016/S2215-0366(20)30077-8

32085841

102

Jia

Zhang

Association between mental health and community support in lockdown communities during the COVID-19 pandemic: evidence from rural China

J Rural Stud202102828797

10.1016/j.jrurstud.2021.01.015

33437114

103

Jin

Guo

Zhang

Jing

Internet access, support, usage divides, and depressive symptoms among older adults in China: a nationally representative cross-sectional study

J Affect Disord202302323514523

10.1016/j.jad.2022.12.001

104

Wang

Zhang

Feng

Wang

Does internet use affect levels of depression among older adults in China? A propensity score matching approach

BMC Public Health20191171911474

10.1186/s12889-019-7832-8

31699057

105

Wen

Zhang

Shi

Association between internet use and physical health, mental health, and subjective health in middle-aged and older adults: nationally representative cross-sectional survey in China

J Med Internet Res2023032125e40956

10.2196/40956

36943368

106

Zhang

Shan

Xia

Zhao

Internet accessibility and incident depressive symptoms in middle aged and older adults in China: a national longitudinal cohort study

J Affect Disord2023111340355361

10.1016/j.jad.2023.08.050

37572699

107

Räsänen

Kouvo

LINKED OR DIVIDED BY THE WEB?: internet use and sociability in four European countries

iCO200704102219241

10.1080/13691180701307461

108

Chen

Ding

Wang

The contribution of the internet to promoting mental health for older adults: cross-sectional survey in China

J Med Internet Res2023121925e40172

10.2196/40172

38113096

109

Aggarwal

Xiong

Schroeder-Butterfill

Impact of the use of the internet on quality of life in older adults: review of literature

Prim Health Care Res Dev202012221e55

10.1017/S1463423620000584

33263273

110

Simons

Reijnders

Janssens

Lataster

Jacobs

Staying connected in old age: associations between bonding social capital, loneliness and well-being and the value of digital media

Aging Ment Health202301271147155

10.1080/13607863.2022.2036947

35156860

111

Ang

Chen

Going online to stay connected: online social participation buffers the relationship between pain and depression

J Gerontol B Psychol Sci Soc Sci2019082174610201031

10.1093/geronb/gby109

30260444

112

Szabo

Allen

Stephens

Alpass

Longitudinal analysis of the relationship between purposes of internet use and well-being among older adults

Gerontologist20190195915868

10.1093/geront/gny036

29688332

113

Bessière

Pressman

Kiesler

Kraut

Effects of internet use on health and depression: a longitudinal study

J Med Internet Res20100228121e6

10.2196/jmir.1149

20228047

114

Teng

The impact of mobile Internet use on mental distress among Chinese adults during the COVID-19 pandemic

Front Public Health202210966606

10.3389/fpubh.2022.966606

36339198

115

China Health and Retirement Longitudinal Study [Article in Chinese]2025-08-24

https://charls.pku.edu.cn/

Multimedia Appendix 1

Basic descriptive statistics, balance test results for different matching methods, propensity score matching–based mediating effect, and comparison of proportions of mediating effects on the internet usage function.