Content area
Purpose
Understanding how social media use (SMU) embedded in daily routines influences mental health in older adulthood is crucial. We explored whether integration-based SMU relates to depressive symptoms and whether social support mediates that link, considering online network size.
Method
A cross-sectional survey of 371 community-dwelling adults aged ≥55 years measured SMU integration, social support, depressive symptoms, demographics, health, and network size. Mediation analysis with bias-corrected bootstrapping assessed direct and indirect paths.
Results
Greater SMU integration corresponded with slightly higher depressive scores. Higher social support predicted lower symptoms but did not explain the SMU–depression association. Participants with medium-sized networks reported the strongest support; larger networks offered no additional benefit.
Conclusion
Deeply embedding social media in daily routines may modestly intensify depressive feelings, whereas maintaining a moderate circle of online ties seems most supportive. Interventions should foster balanced engagement and manageable networks among older adult users. [Research in Gerontological Nursing, xx(x), xx–xx.]
Full text
Depression among older adults stands as a critical public health concern, with its rising prevalence fueled by prolonged social isolation, declining physical health, and reduced social support (Czaja et al., 2021; Herbolsheimer et al., 2018; Santini et al., 2020). Amid an increasingly digital society, it is vital to understand how older adults engage with social media platforms and whether these technologies can alleviate or exacerbate depression (Chen et al., 2022; Madden, 2010). Some evidence indicates that higher social media use (SMU) is linked to elevated depressive symptoms, possibly due to heightened upward social comparisons (Lewin et al., 2023). Conversely, other studies suggest SMU can reduce depression by enhancing perceived social support and fostering social connectedness (Li et al., 2024; Zhang et al., 2021). These contradictory findings may stem from variations on how SMU is conceptualized and measured, as well as differences in older adults' individual characteristics (Baker & Algorta, 2016; Guzman et al., 2023; Lei et al., 2024; McDougall et al., 2016). As daily interactions and services increasingly shift online, understanding how SMU influences older adults' mental health and social well-being is critical for guiding community programs and policy interventions.
The mechanisms through which SMU influences mental health in older adults remain underexplored (Cotten et al., 2023). Despite growing interest in digital aging, interpretation of current findings is limited by key conceptual and methodological weaknesses. Most studies operationalize social media use with simple metrics, such as frequency or time online, overlooking how deeply the medium is woven into daily routines and emotional life. The Social Media Use Integration Scale (SMUIS) captures this nuance by assessing affective attachment and routine incorporation, yet it has rarely been applied to older populations (Jenkins-Guarnieri et al., 2013; McDougall et al., 2016). Established research on the stress buffering model shows that perceived social support mitigates the impact of stressors on depressive symptoms (Cohen & Wills, 1985; Santini et al., 2020). Whether social media fosters such support in later life remains uncertain. Moreover, the structural features of online networks are largely overlooked (Guzman et al., 2023). Evidence from social network science suggests that medium-sized networks of approximately 100 to 200 ties are optimal for sustaining meaningful interaction without overwhelming cognitive or emotional capacity, a pattern consistent with socioemotional selectivity theory, which argues that older adults prioritize emotionally satisfying relationships (Dunbar, 2016; Hill & Dunbar, 2003). Evidence on how online network size or composition influences the benefits of SMU is limited (Janssen et al., 2025; Kim & Fingerman, 2022). Addressing these gaps is essential for nurses and policymakers who seek evidence-based guidance on digital engagement programs for older adults.
The current cross-sectional study surveyed adults aged ≥55 years in the Los Angeles area from March to April 2021 to clarify how integration-based SMU relates to depressive symptoms and to determine whether social support helps explain that relationship. Using the SMUIS to capture the depth of online engagement, the Medical Outcomes Study Social Support Survey (MOS-SSS) to gauge multidimensional support, and the Patient Health Questionnaire-9 (PHQ-9) to index depressive symptoms, we tested a mediation model that controlled for sociodemographic factors, self-rated physical health, and the size of participants' online networks. We hypothesized that SMU as measured by stronger integration of social media into daily life would be associated with fewer depressive symptoms indirectly through higher social support, and we further posited that this indirect association would be most pronounced among individuals with medium-sized online networks (101 to 200 ties), reflecting an optimal balance between the benefits and costs of maintaining digital relationships.
Method
Study Design and Procedures
This cross-sectional study was approved by the University of California Los Angeles Institutional Review Board. A convenience sample of community-dwelling adults aged ≥55 years who resided in the Los Angeles metropolitan area (zip-code verified) was recruited through a blended offline–online strategy. Paper flyers were posted, with permission, at senior centers, senior housing complexes, community centers, home health agencies, and outpatient clinics. Because in-person outreach was constrained by the coronavirus disease 2019 (COVID-19) pandemic, a paid, geotargeted Facebook advertising campaign was added. Recruitment and data collection occurred over a continuous 4-week period, from March 2, 2021, to April 2, 2021, after which the survey portal was closed. Eligible participants (a) were aged ≥55 years, (b) lived in Los Angeles County or the surrounding counties, (c) maintained their own internet account, and (d) had used at least one social media platform (e.g., Facebook, Instagram, Twitter, YouTube, LinkedIn, Reddit, WhatsApp, Viber, Skype) within the preceding 2 weeks. After electronic informed consent, participants completed an anonymous, self-administered Qualtrics® questionnaire that required approximately 10 to 15 minutes to complete. As an incentive, all respondents were entered into a random drawing for a $250 electronic gift card. A total of 371 older adults completed the survey.
Measurements
Depression. The primary outcome, depression, was assessed using the PHQ-9, a widely validated, nine-item self-administered tool that screens, monitors, and quantifies the severity of depressive symptoms (Kroenke et al., 2001). Each item is rated on a scale of 0 to 3, where 0 = not at all, 1 = several days, 2 = more than half the days, and 3 = nearly every day. Summing these scores yields a total score ranging from 0 to 27, with higher scores indicating greater depressive symptom severity. For this study, the PHQ-9 demonstrated good internal reliability, with a Cronbach's alpha of 0.82.
Social Support. The mediating variable, social support, was measured with the MOS-SSS (Sherbourne & Stewart, 1991). This 18-item survey includes four subscales (emotional/informational, tangible, affectionate, and positive social interactions) and provides a total social support score. Each item is rated on a 5-point Likert scale from 1 (none of the time) to 5 (all of the time), with higher scores reflecting greater social support. In our study, the total MOS-SSS score had high internal reliability (Cronbach's alpha = 0.97).
Social Media Use. The SMUIS was used to measure the engaged use of social media activity or SMU (Jenkins-Guarnieri et al., 2012). Although the development of the SMUIS was focused on Facebook, Jenkins-Guarnieri et al. (2012) acknowledged how the instrument is flexible enough to be adapted for use to the multiple manifestations of general online social media use (Berryman et al., 2017; Maree, 2017; McDougall et al., 2016). For the current study, we followed Jenkins et al.'s (2012) recommendations of replacing the word “Facebook” in the item statements with “social media.” The scale has a 6-point Likert response to indicate the level of agreement or disagreement, ranging from 1 (strongly disagree) to 6 (strongly agree). Total possible score on the SMUIS ranges from 25 to 62, with higher values indicating a greater level of engaged SMU. Jenkins-Guarnieri et al. (2012) reported reliability of 0.91. The SMUIS has a good reported internal consistency of 0.83 (Jenkins-Guarnieri et al., 2012). In our sample, the 10-item SMUIS showed good internal consistency (Cronbach's alpha = 0.83).
Covariates. To control for confounding, we included covariates with well-documented links to depressive symptoms in later life. Variables were chosen a priori because they are repeatedly associated with depressive symptoms in later life. Age and gender shape depression risk trajectories (Cacioppo et al., 2010), whereas lower educational attainment and non-married status are linked to higher depressive burden (Fiske et al., 2009; Teo et al., 2013). Poor self-rated physical health and greater multimorbidity independently predict late-life depression (Ambresin et al., 2014). Household income, employment status, and online network size capture socioeconomic resources and social context known to influence mental health (Dunbar, 2016; Santini et al., 2020). Age at survey response was calculated from participants' birth year. Gender was categorized as female or male; education as college graduate or not; ethnicity as White only or non-White; marital status as married or not married; and employment as retired or not retired. Household income was treated as an ordinal variable, reflecting the ordered categories of <$10,000 to $49,000, $50,000 to $99,000, $100,000 to $149,000, and ≥$150,000. Self-reported physical health was rated on a single 5-point scale from excellent (5) to poor (1). Online social network structure was measured by asking respondents to estimate the combined number of family members and friends on their primary social media account, as this item captures a self-estimated headcount of all contacts the participant considers family or friends, regardless of reciprocity or tie strength. Participants were encouraged to open the account while answering to improve accuracy. Due to small sample sizes in certain categories and to enhance the robustness of our analyses, we collapsed the original eight categories of online network size into three groups: small (1 to 100 connections), medium (101 to 200 connections), and large (>200 connections). This categorization aligns with theoretical perspectives on social network size and its impact on social support and mental health outcomes among older adults (Antonucci et al., 2014; Dunbar, 2016). In subsequent analyses, the collapsed online network size variable was treated as categorical.
Data Analysis
Data were analyzed using Intellectus Statistics and verified with SPSS version 30. Descriptive statistics summarized participant characteristics and the key study variables (SMU integration, social support, depressive symptoms). Assumptions of normality, homoscedasticity, and multicollinearity were examined before inferential tests.
The hypothesized mediation model is shown in Figure 1. We used the classic regression-based mediation approach because the study tested a single observed mediator and the sample size was adequate for ordinary least-squares estimation. This method also allows bias-corrected bootstrapping of indirect effects without the distributional assumptions of structural equation modeling (SEM), making it more straightforward than SEM for the current research question (Baron & Kenny, 1986; Preacher & Hayes, 2004). First, social support was regressed on social media integration while controlling for age, gender, educational level, ethnicity, marital status, employment, income, self-rated physical health, and online network size, which yielded path a. Second, depressive symptoms were regressed on both social support and social media integration with the same covariates, generating path b and the direct path c'. Third, the indirect effect was calculated as a × b and its 95% confidence interval (CI) was derived from 5,000 bias-corrected bootstrap resamples following Preacher and Hayes (2004). Mediation was considered significant when that interval excluded zero. Statistical significance was set at a two-tailed alpha level of 0.05.
Figure 1. - Path diagram illustrates the hypothesized mediation model testing whether social support mediates the association between social media use and depression among older adults. Path a denotes the effect of social media use on social support, path b denotes the effect of social support on depressive symptoms, and path c' denotes the direct effect of social media use on depressive symptoms after accounting for social support. The model controls for age, gender, ethnicity, educational level, marital status, employment status, physical health rating, income, and online network size.
Results
Descriptive Statistics
Participants' (N = 371) characteristics are listed in Table 1. Mean age was 70.2 years (SD = 8.12 years, range = 55 to 96 years). Most participants were female, White, college-educated, and retired. The sample was evenly split between married and not married, with most having good physical health. On the PHQ-9, the mean depressive symptoms score was 3.96 (SD = 3.86, range = 0 to 23, median = 3), aligning with the upper end of “no or minimal depression” (scores <5). However, a subset of participants (approximately 25%) scored ≥5, indicating at least mild depressive symptoms. Perceived social support, as measured by the MOS-SSS, averaged 62.5 (SD = 19.48, range = 18 to 90, median = 64), reflecting moderate-to-high support in this sample. Mean SMUIS score was 45.56 (SD = 6.28, range = 25 to 62, median = 46), suggesting moderate integration of social media into daily life.
| Variable | n (%) |
|---|---|
| Gender | |
| Female | 316 (85.2) |
| Male | 53 (14.3) |
| Missing | 2 (0.5) |
| Educational level | |
| Some college | 76 (20.5) |
| Graduated college | 295 (79.5) |
| Ethnicity | |
| White | 315 (85) |
| Non-White | 56 (15) |
| Marital status | |
| Married | 189 (51) |
| Not married | 182 (49) |
| Employment status | |
| Retired | 231 (62.3) |
| Employed | 140 (37.7) |
| Income (USD) | |
| <10,000 to 49,000 | 120 (32.4) |
| 50,000 to 99,000 | 78 (21) |
| 100,000 to 149,000 | 92 (24.8) |
| ≥150,000 | 72 (19.4) |
| Missing | 9 (2.4) |
| Physical health | |
| Poor | 7 (2) |
| Fair | 52 (14) |
| Good | 161 (43.4) |
| Very good | 123 (33.2) |
| Excellent | 28 (7.5) |
| Online network size | |
| Small (1 to 100) | 186 (50.1) |
| Medium (101 to 200) | 102 (27.5) |
| Large (>200) | 82 (22.1) |
| Missing | 1 (0.3) |
| Variable | Mean (SD) (Range) |
|---|---|
| Age (years) | 70.2 (8.12) (55 to 96) |
| Depression (PHQ-9a) | 3.96 (3.86) (0 to 23) |
| Social support (MOS-SSSb) | 62.5 (19.48) (18 to 90) |
| Social media use (SMUISc) | 45.56 (6.28) (25 to 62) |
Regression Results for Depressive Symptoms
In the regression analysis of depressive symptoms (PHQ-9), several variables were significantly associated after controlling for SMU, social support, and online network size and other covariates (Table 2). Social support was significantly negatively related to depressive symptoms (B = −0.03, standard error [SE] = 0.01, t = −2.53, p = 0.012), indicating that higher levels of social support corresponded with lower levels of depressive symptoms. Social media use intensity (SMUIS) now showed a small but significant positive association with depressive symptoms (B = 0.06, SE = 0.03, t = 1.99, p = 0.047). Age showed a significant negative association with depressive symptoms (B = −0.09, SE = 0.03, t = −3.30, p = 0.001), suggesting that older participants reported fewer symptoms. Compared with the poor health reference group, participants rating their health as fair reported markedly fewer symptoms (B = −4.62, SE = 0.97, t = −4.77, p < 0.001), and those rating their health as excellent also reported fewer symptoms (B = −0.88, SE = 0.39, t = −2.28, p = 0.023), whereas the good and very good categories were not significantly different. Ethnicity emerged as a significant factor, with participants identifying as White reporting higher depressive symptoms compared to non-White participants (B = 1.21, SE = 0.53, t = 2.29, p = 0.023). Marital status approached significance (B = 0.86, SE = 0.47, t = 1.84, p = 0.067). Gender, educational level, employment, income, and online network size remained non-significant.
| Variable | B | SE | t | p |
|---|---|---|---|---|
| (Intercept) | 8.34 | 2.62 | 3.19 | 0.002 |
| Social media use | 0.06 | 0.03 | 1.99 | 0.047 |
| Social support | −0.03 | 0.01 | −2.53 | 0.012 |
| Age | −0.09 | 0.03 | −3.30 | 0.001 |
| Gender (female) | 0.50 | 0.53 | 0.94 | 0.348 |
| Education (graduated college) | −0.02 | 0.47 | −0.04 | 0.965 |
| Ethnicity (White) | 1.21 | 0.53 | 2.29 | 0.023 |
| Marital status (married) | 0.86 | 0.47 | 1.84 | 0.067 |
| Employment (retired) | −0.34 | 0.46 | −0.74 | 0.463 |
| Income [ref. <$50,000] | ||||
| $50,000 to $99,000 | −0.64 | 0.42 | −1.52 | 0.129 |
| $100,000 to $149,000 | −0.13 | 0.40 | −0.34 | 0.737 |
| ≥$150,000 | 0.76 | 0.41 | 1.86 | 0.063 |
| Physical health [ref. poor] | ||||
| Fair | −4.62 | 0.97 | −4.77 | <0.001 |
| Good | 0.39 | 0.82 | 0.48 | 0.633 |
| Very good | 0.32 | 0.60 | 0.53 | 0.594 |
| Excellent | −0.88 | 0.39 | −2.28 | 0.023 |
| Network size [ref. small ≤100] | ||||
| Medium (101 to 200) | 0.16 | 0.45 | 0.36 | 0.723 |
| Large (>200) | 0.32 | 0.49 | 0.66 | 0.512 |
Regression Results for Social Support
In the regression analysis of social support (MOS-SSS), marital status, selected physical health rating, and medium online network size were significantly associated with perceived social support after controlling for SMU and other covariates (Table 3). Being married was significantly linked to higher perceived social support (B = 12.94, SE = 2.27, t = 5.69, p < 0.001), indicating that married participants experienced more social support than those who were not married. Physical health ratings were not uniformly significant, as only the very good category differed from the poor reference group and did so in the unexpected direction (B = −6.97, SE = 3.02, t = −2.31, p = 0.022). In addition, participants with medium online network sizes (101 to 200 connections) reported significantly higher social support compared to those with small networks (1 to 100 connections) (B = 5.65, SE = 2.29, t = 2.47, p = 0.014). SMU did not have a significant association with social support (B = −0.06, SE = 0.16, t = −0.40, p = 0.686). Other variables, including age, gender, educational level, ethnicity, employment status, income, and a large online network (>200 connections) did not show significant associations with social support in this model.
| Variable | B | SE | t | p |
|---|---|---|---|---|
| (Intercept) | 46.65 | 13.10 | 3.56 | <0.001 |
| Social media use | −0.06 | 0.16 | −0.40 | 0.686 |
| Age | 0.16 | 0.14 | 1.09 | 0.275 |
| Gender (female) | 0.86 | 2.72 | 0.32 | 0.753 |
| Education (graduated college) | −1.85 | 2.39 | −0.78 | 0.438 |
| Ethnicity (White) | 0.68 | 2.70 | 0.25 | 0.802 |
| Marital status (married) | 12.94 | 2.27 | 5.69 | <0.001 |
| Employment (retired) | −2.48 | 2.35 | −1.06 | 0.292 |
| Income [ref. <$50,000] | ||||
| $50,000 to $99,000 | −0.49 | 2.14 | −0.23 | 0.819 |
| $100,000 to $149,000 | 2.37 | 2.01 | 1.18 | 0.240 |
| ≥$150,000 | 2.44 | 2.07 | 1.18 | 0.240 |
| Physical health [ref. poor] | ||||
| Fair | 6.39 | 4.93 | 1.30 | 0.196 |
| Good | −0.35 | 4.17 | −0.08 | 0.934 |
| Very good | −6.97 | 3.02 | −2.31 | 0.022 |
| Excellent | −2.25 | 1.97 | −1.14 | 0.254 |
| Network size [ref. small ≤100] | ||||
| Medium (101 to 200) | 5.65 | 2.29 | 2.47 | 0.014 |
| Large (>200) | 1.91 | 2.50 | 0.76 | 0.445 |
Mediation Analysis
A mediation analysis was conducted to examine whether social support mediates the relationship between SMU and depressive symptoms among older adults, while controlling for age, gender, educational level, ethnicity, marital status, employment status, income, physical health rating, and online network size. The direct effect of SMU on depressive symptoms was modest but statistically significant (B = 0.06, 95% CI [0.001, 0.12], p = 0.047), indicating that greater SMU was directly associated with higher depressive symptom scores even after accounting for social support and all covariates. In addition, the indirect effect of SMU on depressive symptoms through social support was B = 0.002 (95% CI [−0.007, 0.01]), which includes zero in the CI, demonstrating it was not statistically significant. In other words, because the bootstrapped CI for the product of paths a and b covers zero, these data suggest that social support does not significantly mediate the relationship between SMU and depressive symptoms in this sample. These findings imply that, among older adults in the current study, SMU shows a small direct association with depressive symptoms, but no indirect association through perceived social support.
Discussion
Social Media Use and Depression
In the current study, greater integration of social media into daily life as captured by the SMUIS was modestly but significantly associated with higher depressive symptom scores among older adults (B = 0.06, p = 0.047). Although the effect size is small, it suggests that more psychologically immersive engagement with social media platforms may coincide with slightly increased depressive symptoms in later life. This finding reflects the mixed broader literature. Some studies report no substantial association between SMU and depressive symptoms (Ang & Chen, 2019; McDougall et al., 2016), whereas other work has highlighted potential benefits, such as enhanced social connectedness (Chopik, 2016; Hwang et al., 2021; Teo et al., 2019; Wang et al., 2023). One distinguishing feature of the current study is the use of the SMUIS, which moves beyond frequency-based metrics to assess how strongly social media is embedded in daily routines and emotional connections. Unlike traditional measures focusing on usage frequency or duration, the SMUIS captures psychological aspects of engagement, offering a more layered view of how social media fits into users' lives. This subtle distinction is particularly relevant for older adults, where the manner of engagement may be more impactful than mere participation. In addition, contextual factors likely shaped the observed relationship. Data collection occurred during the COVID-19 pandemic, a period marked by heightened reliance on digital communication and a simultaneous “infodemic” of misinformation. Although social media served as an important conduit for staying connected during physical distancing, it also increased exposure to distressing news and polarizing discourse. These circumstances may have reshaped how older adults engaged with social media, potentially altering its usual patterns and effects on depression. Thus, although our results do not substantiate a direct link between SMU and depression in this population, the unique nature of older adults' engagement in the context of the COVID-19 “infodemic” calls for a more thorough examination of how, when, and why these platforms might influence mental well-being in later life. Future work should look beyond usage frequency to account for the quality, depth, and broader context of online interactions.
Social Media Use and Social Support
Our findings also revealed no significant association between SMU and social support, contrasting with previous research that suggested positive relationships between SMU and increased social support (Nam, 2021; Zhang et al., 2021). This divergence may be partially explained by our use of the SMUIS, which measures social media integration through two distinct dimensions: emotional connection and integration into social routines. This refined measurement suggests that mere integration of social media into daily routines may not automatically translate into enhanced perceptions of social support among older adults.
Our analysis of online network size revealed a nuance pattern. Participants with medium-sized online networks (101 to 200 connections) reported significantly higher social support compared to those with small networks (1 to 100 connections). However, those with large networks (>200 connections) did not report significantly different levels of social support compared to participants with small networks. This finding further supports a potential threshold or “sweet spot” effect of a network that is large enough to sustain meaningful interaction yet small enough to remain manageable may maximize social support (Dunbar, 2016; Nabi et al., 2013). The pattern echoes social network theory's notion of maintenance costs—the cognitive and emotional effort required to keep ties active, implying that, for older adults, very large networks may dilute interaction quality or become overwhelming (Szabo et al., 2019). Lack of additional benefits from having a large network may reflect older adults' preference for more intimate, meaningful online connections rather than extensive, potentially superficial networks.
The disconnect between SMU and the effects of network size on social support raises important questions about the quality versus quantity of online social interactions. The current study's findings suggest that, for older adults, integration of social media into daily routines does not automatically increase social support. Instead, a moderate circle of approximately 100 to 200 online ties seems to offer the most benefit. Very small networks may leave users with too few active contacts, whereas very large networks can dilute intimacy and impose cognitive maintenance costs, ultimately providing no added support. Older adults' sense of support could hinge on the depth of their interactions rather than on accumulating even more connections, a nuance that must be considered when examining social media engagement's role to mental health outcomes.
Mediation Analysis
The mediation analysis indicates that deeper integration of social media into daily life relates directly to slightly higher depressive symptoms among older adults, even after extensive adjustment for demographic and health factors and online network size. Although the effect is small, it is clinically relevant because late-life depression compounds risks for disability and mortality, so even marginal increases deserve attention (Santini et al., 2020). This direct link may arise from mechanisms such as heightened exposure to distressing content or upward social comparison (Baker & Algorta, 2016; Lewin et al., 2023), or from the displacement of restorative offline activities during prolonged screen time (Czaja et al., 2021). By contrast, the indirect path through perceived social support was nonsignificant, challenging the stress-buffering hypothesis that support transmits the mental health benefits of social engagement (Cohen & Wills, 1985). These results highlight the complex nature of social support in later life and underscore the need for further research into whether and how online interactions can genuinely supplement, rather than merely echo, offline forms of social support for older adults. Further, the quality of online interactions may be more influential than the overall intensity of SMU in shaping an older adult's mental health outcomes.
Although the SMUIS provided useful insights into how social media was integrated into participant's daily lives, it was not able to capture whether older adults' interactions were primarily passive (e.g., scrolling, lurking) or more active (e.g., posting, messaging, video chatting). Future research should consider measuring interaction quality, from emotional valence of exchanges, perceived intimacy, and tie strength, to better explain whether certain types of engagement foster stronger social support and buffer against depressive symptoms. Pandemic-era data collection could also have disrupted usual support exchanges, muting any protective effect. The overabundance of information, including health-related content, misinformation, and even negative social conflict, during this period may have complicated the role of social support. Examining post-pandemic dynamics (e.g., more stable usage patterns, better digital literacy) could yield different findings on the role of social media in older adults' well-being. Continued investigation into these nuanced dynamics, specifically on the quality of interactions post-pandemic, will be essential for determining whether and how social media use can ultimately nurture, rather than erode, mental well-being in later life.
Limitations and Strengths
Although our study contributes valuable insights into the relationship among SMU, social support, and depression in older adults, several limitations must be acknowledged. First, there is lack of diversity in our sample, with an overrepresentation of women, White individuals, and those who are college-educated. Similar studies on social media and depression have also reported an overrepresentation of female participants (Lin et al., 2021; McDougall et al., 2016). Due to COVID-19 restrictions, our recruitment was primarily conducted online, which tends to attract more female and college-educated participants (Gao et al., 2020; Meshi et al., 2020). Although our study adds to the literature by highlighting women's greater inclination to engage in online recruitment and participation, future research should strive for a more diverse sample to enhance the applicability of results across different demographic groups. Second, the cross-sectional design of the current study precludes any causal inferences. Observed associations among SMU, social support, and depression do not imply causation. Longitudinal and intervention studies are needed to explore the directionality and potential causative relationships among these variables over time. Third, we modeled SMU as a mediator, assuming it boosts social support, which then lowers depression; however, it may instead moderate the relationship, altering how strongly existing support offsets depression without creating that support itself. Future studies should test this moderating role to see whether SMU amplifies, weakens, or otherwise reshapes the SMU–support–depression link in later life. Fourth, reliance on self-reported data introduces possible response biases, as participants may underreport or over-report health-related information due to recall or social desirability bias. Fifth, our online network size measure captured only the breadth of a contact list and did not distinguish close from weak ties or assess qualities, such as reciprocity, interaction frequency, or emotional tone. Subsequent research should combine structural counts with tie-level descriptors to determine which network attributes truly foster social support and buffer depression. Finally, we focused on SMU integration and did not consider platform type, synchronous versus asynchronous features, usage frequency, exposure to negative content, or passive versus active engagement, all factors that may shape the social media–mental health link in older adults. Incorporating these variables will provide a more comprehensive understanding of digital engagement and well-being.
Despite these limitations, our study has several notable strengths. Using a relatively large sample of 371 older adults bolsters the statistical power of the findings, particularly given the limited representation of older adults in studies exploring the integration of SMU into daily life and its effects on depression. The current study used established instruments, including the PHQ-9 for evaluating depressive symptoms and MOS-SSS for measuring social support, thereby providing reliable and precise assessment of essential variables. The application of the SMUIS facilitated a deeper comprehension of SMU beyond mere usage metrics (e.g., frequency of use, binary responses), addressing a deficiency in current research. The advantages of medium-sized online social networks for social support were emphasized through the analysis of their structural characteristics, a topic frequently neglected in current literature. Conducting the study during the COVID-19 pandemic adds important context, offering unique insights into the dual role of digital connectivity as a resource and challenge amid widespread social isolation. These strengths collectively offer substantial evidence for understanding the complex interaction among SMU, social support, and depression in older adults.
Implications
Findings from the current study have several important implications for health care practitioners, policymakers, and researchers interested in older adults' digital engagement and mental health. The modest but significant direct association between social media integration and depressive symptoms suggests that health care providers should consider inquiring about the nature and extent of older adults' SMU during mental health assessments, particularly for those showing signs of depression. The absence of a mediating effect through social support challenges assumptions that social media automatically enhances perceived support for older adults, highlighting the need for more nuanced approaches to digital literacy programs that emphasize quality of interactions over mere platform adoption. The “sweet spot” finding regarding mediumsized online networks (101 to 200 connections) provides concrete guidance for social technology interventions, suggesting that efforts to expand older adults' online social circles should aim for moderation rather than maximization. The COVID-19 pandemic context of our data collection underscores the importance of considering temporal and societal factors when interpreting the relationship between SMU and mental health outcomes, as crisis periods may alter typical patterns of digital engagement and support exchange. These results contribute to the growing literature indicating that the psychological aspects of SMU specific to its integration in daily life and emotional connections may be more consequential for older adults' mental health than traditional frequency-based metrics alone. Our findings highlight the importance of developing age-specific best practices for SMU that balance the potential benefits of digital connection with strategies to mitigate negative psychological impacts, particularly as digital platforms become increasingly central to service delivery and social participation for aging populations.
Conclusion
The current study examined how the depth of SMU integration, size of online networks, and perceived social support relate to depressive symptoms in community-dwelling adults aged ≥55 years. Deeper integration of social media into daily routines showed a small but significant direct association with higher depressive symptom scores after adjustment for demographic, online network size, and health factors. Although higher social support was independently linked to fewer depressive symptoms, social support did not explain the connection between SMU and depression. Participants whose online networks contained approximately 101 to 200 contacts reported the greatest social support, whereas larger networks offered no additional benefit. These results highlight the nuanced nature of digital engagement in later life. Integration of SMU in their daily life alone is unlikely to relieve depressive symptoms and, in some cases, may worsen them if engagement exposes older adults to distressing content, fuels unfavorable social comparisons, or displaces restorative offline activities. At the same time, maintaining a moderatesized circle of meaningful online ties appears to foster the strongest sense of support. Programs that aim to improve mental health through technology should help older adults curate networks of manageable size; learn active and supportive communication styles, such as private messaging and video chatting; and develop habits for filtering misinformation and negative content. Clinicians might routinely ask about the nature of a patient's online activities, not just whether the patient uses social media at all. Community organizations can integrate digital skill workshops with offline group events so that online ties reinforce, rather than replace face-to-face contact. Future longitudinal and experimental studies should disentangle how different platforms, engagement patterns, and tie strengths influence well-being beyond pandemic conditions, thereby guiding evidence-based policies for healthy aging in a connected world.
Ambresin, G., Chondros, P., Dowrick, C., Herrman, H., & Gunn, J. M. (2014). Self-rated health and long-term prognosis of depression. Annals of Family Medicine, 12(1), 57–65. 10.1370/afm.1562 PMID: 24445104
Ang, S., & Chen, T.-Y. (2019). Going online to stay connected: Online social participation buffers the relationship between pain and depression. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 74(6), 1020–1031. 10.1093/geronb/gby109 PMID: 30260444
Antonucci, T. C., Ajrouch, K. J., & Birditt, K. S. (2014). The convoy model: Explaining social relations fromamultidisciplinary perspective. The Gerontologist, 54(1), 82–92. 10.1093/geront/gnt118 PMID: 24142914
Baker, D. A., & Algorta, G. P. (2016). The relationship between online social networking and depression:Asystematic review of quantitative studies. Cyberpsychology, Behavior and Social Networking, 19(11), 638–648. 10.1089/cyber.2016.0206 PMID: 27732062
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173–1182. 10.1037/0022-3514.51.6.1173 PMID: 3806354
Berryman, C., Ferguson, C. J., & Negy, C. (2017). Social media use and mental health among young adults. Psychiatric Quarterly,1– 8. 10.1007/s11126-017-9535-6
Cacioppo, J. T., Hawkley, L. C., & Thisted, R. A. (2010). Perceived social isolation makes me sad: 5-year cross-lagged analyses of loneliness and depressive symptomatology in the Chicago Health, Aging, and Social Relations Study. Psychology and Aging, 25(2), 453–463. 10.1037/a0017216 PMID: 20545429
Chen, E., Wood, D., & Ysseldyk, R. (2022). Online social networking and mental health among older adults:Ascoping review. Canadian Journal on Aging, 41(1), 26–39. 10.1017/S0714980821000040 PMID: 33736725
Chopik, W. J. (2016). The benefits of social technology use among older adults are mediated by reduced loneliness. Cyberpsychology, Behavior and Social Networking, 19(9), 551–556. 10.1089/cyber.2016.0151 PMID: 27541746
Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98(2), 310–357. 10.1037/0033-2909.98.2.310 PMID: 3901065
Cotten, S. R., Ghaiumy Anaraky, R., & Schuster, A. M. (2023). Social media use may not be as bad as some suggest: Implication for older adults. Innovation in Aging, 7(3), igad022. 10.1093/geroni/igad022 PMID: 37179658
Czaja, S. J., Moxley, J. H., & Rogers, W. A. (2021). Social support, isolation, loneliness, and health among older adults in the PRISM randomized controlled trial. Frontiers in Psychology, 12, 728658. 10.3389/fpsyg.2021.728658 PMID: 34675843
Dunbar, R. I. M. (2016). Do online social media cut through the constraints that limit the size of offline social networks? Royal Society Open Science, 3(1), 150292. 10.1098/rsos.150292 PMID: 26909163
Fiske, A., Wetherell, J. L., & Gatz, M. (2009). Depression in older adults. Annual Review of Clinical Psychology, 5(1), 363–389. 10.1146/annurev.clinpsy.032408.153621 PMID: 19327033
Gao, J., Zheng, P., Jia, Y., Chen, H., Mao, Y., Chen, S., Wang, Y., Fu, H., & Dai, J. (2020). Mental health problems and social media exposure during COVID-19 outbreak. PLoS One, 15(4), e0231924. 10.1371/journal.pone.0231924 PMID: 32298385
Guzman, A. A., Brecht, M.-L., Doering, L. V., Macey, P. M., & Mentes, J. C. (2023). Social media use and depression in older adults:Asystematic review. Research in Gerontological Nursing, 16(2), 97–104. 10.3928/19404921-20230220-05 PMID: 36944173
Herbolsheimer, F., Ungar, N., & Peter, R. (2018). Why is social isolation among older adults associated with depressive symptoms? The mediating role of out-of-home physical activity. International Journal of Behavioral Medicine, 25(6), 649–657. 10.1007/s12529-018-9752-x PMID: 30350258
Hill, R. A., & Dunbar, R. I. M. (2003). Social network size in humans. Human Nature: An Interdisciplinary Biosocial Perspective, 14(1), 53–72. 10.1007/s12110-003-1016-y PMID: 26189988
Hwang, J., Toma, C. L., Chen, J., Shah, D. V., Gustafson, D., & Mares, M.-L. (2021). Effects of web-based social connectedness on older adults' depressive symptoms:Atwo-wave cross-lagged panel study. Journal of Medical Internet Research, 23(1), e21275. 10.2196/21275 PMID: 33439143
Janssen, J. H. M., van Tilburg, T. G., van Ingen, E. J., Corten, R., & Peeters, G. M. E. E. G. (2025). Social connectedness in older adults: The potential of social internet use to maintainastrong and stable personal network. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 80(4), 14. 10.1093/geronb/gbaf014 PMID: 39862222
Jenkins-Guarnieri, M. A., Wright, S. L., & Johnson, B. (2013). Development and validation ofasocial media use integration scale. Psychology of Popular Media Culture, 2(1), 38–50. 10.1037/a0030277
Kim, Y. K., & Fingerman, K. L. (2022). Daily social media use, social ties, and emotional well-being in later life. Journal of Social and Personal Relationships, 39(6), 1794–1813. 10.1177/02654075211067254 PMID: 37727534
Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2001). The PHQ-9. Journal of General Internal Medicine, 16(9), 606–613. 10.1046/j.1525-1497.2001.016009606.x
Lei, X., Matovic, D., Leung, W.-Y., Viju, A., & Wuthrich, V. M. (2024). The relationship between social media use and psychosocial outcomes in older adults:Asystematic review. International Psychogeriatrics, 36(9), 714–746. 10.1017/S1041610223004519 PMID: 38287785
Lewin, K. M., Meshi, D., Schuster, A. M., & Cotten, S. R. (2023). Active and passive social media use are differentially related to depressive symptoms in older adults. Aging and Mental Health, 27(1). 10.1080/13607863.2022.2068133
Li, Y., Zhang, X., Chen, Z., Ma, L., Wang, Z., & Shao, J. (2024). Social media use and life satisfaction among Chinese older adults:Amoderated mediation model. Journal of Applied Gerontology. Advance online publication. 10.1177/07334648241309483 PMID: 39700393
Lin, C. Y., Namdar, P., Griffiths, M. D., & Pakpour, A. H. (2021). Mediated roles of generalized trust and perceived social support in the effects of problematic social media use on mental health:Across-sectional study. Health Expectations: An International Journal of Public Participation in Health Care and Health Policy, 24(1), 165–173. 10.1111/hex.13169 PMID: 33253480
Madden, M. (2010). Older adults and social media. https://www.pewresearch.org/2010/08/27/older-adults-and-social-media/
Maree, T. (2017). The Social Media Use Integration Scale: Toward reliability and validity. International Journal of Human-Computer Interaction, 33(12), 963–972. 10.1080/10447318.2017.1301041
McDougall, M. A., Walsh, M., Wattier, K., Knigge, R., Miller, L., Stevermer, M., & Fogas, B. S. (2016). The effect of social networking sites on the relationship between perceived social support and depression. Psychiatry Research, 246, 223–229. 10.1016/j.psychres.2016.09.018 PMID: 27721061
Meshi, D., Cotten, S. R., & Bender, A. R. (2020). Problematic social media use and perceived social isolation in older adults:Across-sectional study. Gerontology, 66(2), 160–168. 10.1159/000502577 PMID: 31522178
Nabi, R. L., Prestin, A., & So, J. (2013). Facebook friends with(health) benefits? Exploring social network site use and perceptions of social support, stress, and well-being. Cyberpsychology, Behavior and Social Networking, 16(10), 721–727. 10.1089/cyber.2012.0521 PMID: 23790356
Nam, S.-J. (2021). Mediating effect of social support on the relationship between older adults' use of social media and their quality-of-life. Current Psychology, 40(9), 4590–4598. 10.1007/s12144-019-00399-3
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36(4), 717–731. 10.3758/BF03206553 PMID: 15641418
Santini, Z. I., Jose, P. E., York Cornwell, E., Koyanagi, A., Nielsen, L., Hinrichsen, C., Meilstrup, C., Madsen, K. R., & Koushede, V. (2020). Social disconnectedness, perceived isolation, and symptoms of depression and anxiety among older Americans(NSHAP):Alongitudinal mediation analysis. The Lancet. Public Health, 5(1), e62–e70. 10.1016/S2468-2667(19)30230-0 PMID: 31910981
Sherbourne, C. D., & Stewart, A. L. (1991). The MOS social support survey. Social Science and Medicine, 32(6), 705–714. 10.1016/0277-9536(91)90150-B
Szabo, A., Allen, J., Stephens, C., & Alpass, F. (2019). Longitudinal analysis of the relationship between purposes of internet use and well-being among older adults. The Gerontologist, 59(1), 58–68. 10.1093/geront/gny036 PMID: 29688332
Teo, A. R., Choi, H., & Valenstein, M. (2013). Social relationships and depression: Ten-year follow-up fromanationally representative study. PLoS One, 8(4), e62396. 10.1371/journal.pone.0062396 PMID: 23646128
Teo, A. R., Markwardt, S., & Hinton, L. (2019). Using Skype to beat the blues: Longitudinal data fromanational representative sample. The American Journal of Geriatric Psychiatry, 27(3), 254–262. 10.1016/j.jagp.2018.10.014 PMID: 30442532
Wang, C., Zhu, Y., Ma, J., & Chu, J. (2023). The association between internet use and depression among older adults in China: The mediating role of social networks. Digital Health, 9, 20552076231207587. 10.1177/20552076231207587 PMID: 37868154
Zhang, K., Kim, K., Silverstein, N. M., Song, Q., & Burr, J. A. (2021). Social media communication and loneliness among older adults: The mediating roles of social support and social contact. The Gerontologist, 61(6), 888–896. 10.1093/geront/gnaa197 PMID: 33284972
This manuscript is derived from the primary author's (AAG) dissertation research. ChatGPT's GPT o1 pro and o3 model assisted in refining the manuscript's language and structural organization, and helped meet journal word count requirements. These tools were employed exclusively for editorial enhancement rather than content generation.
From Cedars Sinai Health System, Nursing Research & Quality Improvement (AAG, BC, HA), and University of California, Los Angeles, School of Nursing (M-LB, PMM, JCM), Los Angeles, California.
Disclosure: The authors have disclosed no potential conflicts of interest, financial or otherwise.
This work is published under https://creativecommons.org/licenses/by-nc/4.0 (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.