Association between individual social capital and depressed mood in older adults in Iran: results from baseline data of Birjand Longitudinal Aging Study

Abstract

Objectives

To examine the association between individual social capital and depression in older adults in Iran and to test the hypothesis that higher levels of social capital are inversely associated with depressive symptoms.

Design

Cross-sectional study using baseline data from a longitudinal cohort.

Setting

Community-based study conducted in primary care settings across urban and rural areas of Birjand County, Eastern Iran.

Participants

A total of 1348 community-dwelling individuals aged 60 years and older were recruited through multistage stratified cluster random sampling. Participants who were bedridden or had end-stage disease (life expectancy<6 months) were excluded.

Primary and secondary outcome measures

The primary outcome was depression status, measured using the Patient Health Questionnaire 9 items, with a score≥10 indicating depression. The main explanatory variable was social capital, assessed using a validated 69-item questionnaire capturing domains such as collective activity, social trust and network structure. Univariable and multivariable logistic regression analyses were conducted to estimate adjusted ORs and 95% CIs for associations between depression and social capital dimensions. Statistical analyses were performed using Stata V.12.0

Results

Of the total participants, 268 (19.94%) were identified as having depressive symptoms, with a significantly higher prevalence among women (27.44%) compared with men (11.88%). Depression was more prevalent among those in the lowest wealth quintile (32.09%) and individuals with low literacy levels (28.10%). Participation in collective activities was inversely associated with depression in the second (OR=0.62, 95% CI (0.42 to 0.93)), third (OR=0.45, 95% CI (0.29 to 0.71)), fourth (OR=0.59, 95% CI (0.37 to 0.93)) and fifth (OR=0.37, 95% CI (0.22 to 0.61)) quintiles. Social trust was also associated with lower odds of depression in the third (OR=0.62, 95% CI (0.39 to 0.99)) and fourth (OR=0.64, 95% CI (0.42 to 0.97)) quintiles. Furthermore, the second (OR=0.63, 95% CI (0.40 to 0.99)) and fifth (OR=0.38, 95% CI (0.23 to 0.63)) quintiles of social network structure were inversely related to depression. These findings suggest that higher levels of social capital, particularly in terms of collective participation, trust and social networks, are associated with a reduced likelihood of depressive symptoms in older adults.

Conclusions

Higher levels of social capital, particularly collective engagement, interpersonal trust and diverse social networks, are associated with lower odds of depression in older adults. These findings support the need for community-based interventions to strengthen social capital as a strategy for mental health promotion among the elderly in low-income and middle-income settings.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The study used a large, community-based sample drawn through a multistage stratified cluster sampling method, improving representativeness and generalisability within the regional context.

• Validated and culturally adapted instruments were employed to measure both depression (Patient Health Questionnaire 9 items) and multidimensional social capital, ensuring measurement reliability.

• A Directed Acyclic Graph was developed to inform confounder selection, enhancing the causal interpretability of the multivariable model.

• The cross-sectional design precludes causal inference and cannot determine the directionality of the observed associations.

• Reliance on self-reported data for both exposure and outcome measures introduces potential recall and social desirability biases, and residual confounding from unmeasured variables cannot be ruled out.

Introduction

The number of older adults (≥60 years) is increasing in a surprising way all over the world.¹ Although Iran is not at the top of the list of countries with the largest population of older adults today, due to the sharp decrease in the population growth rate, it is expected to be among the countries with the largest population of older adults in the not-too-distant future.² Because of this, geriatric depression remains a tremendous public health challenge.³ Recent evidence has suggested that approximately one-third of older adults in Iran have depressive disorders.⁴ Depression can impair functional ability, reduce the quality of life, increase older adults’ mortality and inflict a heavy economic burden on them, society and the healthcare system.⁵

One of the critical steps to prevent depression in older adults is to identify the risk factors associated with depression.⁶ Common risk factors of depression include advanced age, being woman and unfavourable economic levels.⁷ As a person ages, they often experience significant life changes that can increase their risk of depression.^{6 7} Various factors substantially contribute to the development of depression in the elderly population. Physical health issues such as illness and disability, chronic or severe pain, decreased cognitive ability and damage to body image due to surgery or illness can all contribute to depression.⁸ Social factors also play a significant role, including living alone, limited and declining social support due to death or relocation, reduced mobility due to illness and loss of driving privileges. Additionally, lifestyle changes such as retirement can lead to a loss of identity, self-confidence and financial security, thereby increasing the risk of depression. Physical limitations in previously enjoyable activities can affect an individual’s sense of purpose. Psychological and emotional challenges, including fear of death and anxiety due to financial problems, health issues, abuse or neglect, are also contributing factors. Finally, loss and grief from ending relationships with friends, family members and pets, or losing a spouse or partner, significantly cause depression in older adults.⁵,⁷

With the development of social determinants of health, the role of social capital in the mental health of humans is increasingly recognised.⁸ Social capital is a multifaceted concept that includes multiple dimensions, each used to describe a phenomenon in social relations at the individual and societal levels.⁹,¹¹ Studies in many countries,⁹¹²,¹⁵ including Iran, have shown that social capital is associated with depression in older people. Increasing attention has been focused on the impact of social capital on the mental health of older adults.¹⁶ Although several risk factors have been investigated in previous research,^{5 9 10 14} it is still unclear whether there is a synergistic or antagonistic effect between social capital and other important factors influencing depression.¹⁷ In practice, older people are exposed to two or more risk factors, such as low economic level and low trust and reciprocity, which make them more vulnerable to depression.⁹ Knowledge about these findings will provide more appropriate and accurate measures to protect older adults from depressive disorders.

Therefore, in the present study, we aimed to investigate the relationship between social capital and depression in older adults in Iran. Specifically, we will state the different dimensions of social capital and then examine their relationship with the category of depression in older adults. We investigated whether this study’s dimensions of social capital were associated with depression and then further explored the combined effect of social capital and other common risk factors on depression.

Method

Setting and sampling

The present study involves a cross-sectional analysis of the baseline data collected from the Birjand Longitudinal Aging Study (BLAS). This cohort study commenced in October 2018 in Birjand, the provincial capital city in the eastern region of Iran. The BLAS aims to investigate the impact of ageing on various health outcomes in a representative cohort of Iranian older adults. This analysis provides a preliminary glimpse into the characteristics of the study participants at baseline and sets the stage for future longitudinal analyses. The BLAS is an ongoing cohort study, and the current analysis used the total available baseline cohort at the time of data extraction

In this study, the elderly population is considered to be individuals aged 60 years and above. The BLAS sampling methodology was implemented through a multistage stratified cluster random sampling technique. The city was initially demarcated into 70 blocks based on the geographic information system and the aged population. Applying this approach ensured that the sample was representative of the target population and minimised the risk of bias. Specifically, the sampling method was based on the weight of three distinct age groups: individuals between 60–69, 70–79 and 80+ years old. A simple random selection process was used in rural regions based on the health registry code. The sample consisted of 400 eligible subjects, with 40 subjects chosen from each of the 10 villages in Birjand County. Overall, this sampling strategy ensured a diverse and representative sample for the study.

The criteria for participant inclusion in this study consisted of individuals aged 60 years and above who had expressed an interest in participating. In contrast, the exclusion criteria comprised individuals who were bedridden or had an end-stage disease with a life expectancy of fewer than 6 months. Further discussion about the sampling method can be found in previous related publications.¹⁸

Patient and public involvement

Patients or the public were not involved in the design, conduct, reporting or dissemination plans of our research.

Data collection

Demographic information, including age, sex, marital status, educational status, job and living arrangement, was collected by trained researchers through interviews conducted with participants and their close informants. Additionally, the medical history of participants, including the history of medically approved diseases and duration of involvement with health problems, was gathered from the participants and their informants. These data were acquired to provide a comprehensive record of the participants’ personal and medical backgrounds.

The Patient Health Questionnaire 9 items (PHQ-9) is a widely used tool to assess the level of depressed mood. The questionnaire comprises nine items, and the total score ranges from 0 to 27, with higher scores indicating higher levels of depression. Several studies have demonstrated the validity and reliability of PHQ-9, making it a reliable tool for evaluating depressed mood. The psychometric properties of this instrument in the Persian language have been previously validated in studies, further enhancing its reliability and validity.¹⁹,²³

The social capital questionnaire used in this study consisted of 69 items, developed and validated in the Urban HEART study in Tehran by a multidisciplinary team. We have included a blank copy of the original Farsi version of this Questionnaire as online supplemental file 1. This questionnaire evaluates social capital components at various levels, including close family, relatives, friends, neighbours, colleagues, people of the same ethnicity and religion, and the general public. The domains include voluntary participation, collective activity, trust, social cohesion, social support, participation in association activities and social reciprocity. Psychometric evaluations of the instrument have demonstrated acceptable internal consistency, with Cronbach’s alpha values ranging from 0.70 to 0.83 across key domains (eg, trust, reciprocity, social participation). Construct validity was supported through exploratory and confirmatory factor analyses, and discriminant validity was shown by its ability to distinguish between different socioeconomic groups and neighbourhood types. This tool has been used in multiple population-based studies in Iran and is suitable for assessing individual-level social capital in both urban and semiurban settings.^{24 25}

Statistical analysis

Continuous data were presented as mean and SD, while categorised data were presented as frequency and percent. All statistical analyses were conducted using Stata software (V.12.0, StataCorp, College Station, Texas, USA). A threshold of p<0.20 was used for variable selection during model building, while p<0.05 was considered the criterion for statistical significance in final interpretation. The association between dependent dummy variables and independent variables was assessed using univariable and multiple logistic regression models. In contrast, univariable and multiple linear regression models evaluated the association between continuous dependent and independent variables. In the logistic regression analyses, the primary dependent variable was a dummy variable representing depression status. This binary variable was constructed based on the PHQ-9 Score: participants with a score of 10 or greater were coded as 1 (indicating presence of depression), and those with a score below 10 were coded as 0 (indicating absence of depression). This dummy outcome variable was used to assess associations with various independent variables in both univariable and multivariable logistic regression models.

We also developed a Directed Acyclic Graph (DAG) to map the hypothesised causal structure between social capital (exposure) and depression (outcome) among older adults. The DAG was constructed based on domain-specific knowledge and visually represents the relationships between key sociodemographic variables. Variables such as age, sex, education, marital status, job status, living arrangement and wealth index (WI) were identified as potential confounders. This conceptual model validated the inclusion of these variables in our multivariable regression analyses and supported the overall modelling strategy. The structure of the DAG is presented in online supplemental file 2.

The WI is a composite index used as a proxy indicator of household-level wealth. First, we chose variables that could distinguish between relatively rich and poor populations. To construct the WI, we initially considered a total of 12 asset variables representing household possessions. After conducting a preliminary principal component analysis (PCA), two variables were excluded due to their low factor loadings and limited contribution to the overall variance explained. As a result, PCA was performed using the remaining 10 asset variables. Then, PCA was carried out on these 10 variables: refrigerator ownership, vacuum cleaner ownership, microwave oven ownership, washing machine ownership, dish machine ownership, water purifier ownership, LED television ownership, smartphone ownership, automobile ownership and house ownership. The first factor from PCA was taken to construct the WI to represent the household’s wealth. Information is collected on household assets that are indicative of wealth. Each asset is assigned a factor score based on the analysis of principal components. The asset scores are standardised using a standard normal distribution with a mean of zero and a SD of one. Breakpoints that define wealth quintiles as lowest, second, middle, fourth and highest are created using the standardised scores. Each household is given a standardised score for each asset, which varies depending on whether or not the household possesses that asset. Household scores are summed, and individuals are ranked accordingly. The sample is divided into five equal groups, each with the same number of individuals, known as population quintiles.

The study participants were divided into five groups based on their PHQ-9 scores. A PHQ-9 Score of less than 5 was categorised as minimal or no depression, while a score of 5–9 was considered as mild depression. A score of 10–14 was classified as moderate depression, a score of 15–19 was moderately severe and a score of 20–27 was classified as severe depression. A PHQ-9 Score of ≥10 was used as the cut-off to classify participants as having depression.²¹,²³

Results

Population characteristics

A total of 1348 participants were included in the study, with 697 (51.71%) women. The average age of all participants was 69.73 years, with a SD of 7.53 years. Of all the participants, 245 (18.20%) were single, divorced or widowed. On the other hand, 605 (45.01%) were found to be individuals with low literacy levels. It should be mentioned that only 106 (7.88%) had academic degrees. Out of the total population, 268 individuals, who account for 19.94%, were identified as suffering from depression. From the job status point of view, 501 individuals (37.27%) were retired, while 170 participants (12.64%) were still working and employed. Table 1 demonstrates the baseline phase data of participants based on depression status and PHQ-9 Score.

Table 1. Baseline phase data of participants based on depression status.

	Not depressed (n=1076)	Depressed (n=268)	P value
Age (years), mean (SD)	69.52 (7.39)	70.66 (8.01)	0.027
Gender (female), %	505 (72.56)	191 (27.44)	<0.001
Gender (male), %	571 (88.12)	77 (11.88)
Marital status (n=1345), %			<0.001
Married	902 (82.15)	196 (17.85)
Single, divorced or widowed	173 (70.61)	72 (29.39)
Educational status, %			<0.001
Low literacy levels	435 (71.90)	170 (28.10)
Primary school	360 (84.51)	66 (15.49)
High school	66 (89.19)	8 (10.81)
Diploma	119 (89.47)	14 (10.53)
Academic degree	96 (90.57)	10 (9.43)
Job status, %			<0.001
Retired	456 (91.02)	45 (8.98)
Employed	142 (83.53)	28 (16.47)
Housekeeper	427 (70.46)	179 (29.54)
Unemployed	51 (76.12)	16 (23.88)
Wealth quintile, %			<0.001
First quintile	182 (67.91)	86 (32.09)
Second quintile	204 (76.69)	62 (23.31)
Third quintile	221 (82.16)	48 (17.84)
Fourth quintile	230 (84.87)	41 (15.13)
Fifth quintile	239 (88.52)	31 (11.48)
Total social capital score (n=1249), %			<0.001
First quintile	173 (69.76)	75 (30.24)
Second quintile	198 (78.57)	54 (21.43)
Third quintile	198 (79.84)	50 (20.16)
Fourth quintile	206 (82.07)	45 (17.93)
Fifth quintile	221 (88.40)	29 (11.60)
Collective activity domain of social capital (n=1313), %			<0.001
First quintile	152 (69.41)	67 (30.59)
Second quintile	209 (77.70)	60 (22.30)
Third quintile	244 (83.56)	48 (16.44)
Fourth quintile	218 (80.15)	54 (19.85)
Fifth quintile	227 (86.97)	34 (13.03)
Voluntary help domain of social capital (n=1316), %			<0.001
First quintile	204 (74.73)	69 (25.27)
Second quintile	144 (77.01)	43 (22.99)
Third quintile	201 (74.44)	69 (25.56)
Fourth quintile	293 (88.79)	37 (11.21)
Fifth quintile	209 (81.64)	47 (18.36)
Social trust domain of social capital (n=1325), %			0.001
First quintile	176 (78.92)	47 (21.08)
Second quintile	166 (77.57)	48 (22.43)
Third quintile	193 (84.65)	35 (15.35)
Fourth quintile	225 (72.82)	84 (27.18)
Fifth quintile	299 (85.19)	52 (14.81)
Social coherence domain of social capital (n=1322), %			<0.001
First quintile	158 (71.49)	63 (28.51)
Second quintile	134 (77.01)	40 (22.99)
Third quintile	205 (76.49)	63 (23.51)
Fourth quintile	312 (85.01)	55 (14.99)
Fifth quintile	246 (84.25)	46 (15.75)
Social network domain of social capital (n=1344), %			<0.001
First quintile	191 (72.08)	74 (27.92)
Second quintile	183 (80.97)	43 (19.03)
Third quintile	96 (75.00)	32 (25.00)
Fourth quintile	344 (78.54)	94 (21.46)
Fifth quintile	262 (91.29)	25 (8.71)
Social support (n=1329), %			0.538
No	673 (79.55)	173 (20.45)
Reciprocity (n=1335), %			0.353
First quintile	186 (76.54)	57 (23.46)
Second quintile	196 (78.71)	53 (21.29)
Third quintile	223 (80.51)	54 (19.49)
Fourth quintile	197 (80.93)	50 (19.07)
Fifth quintile	266 (81.80)	53 (18.20)

Comparing social capital components among subjects with and without depressed mood

Based on the findings derived from the PHQ-9 and social capital questionnaire, it was observed that out of a total of 1348 individuals surveyed, 268 individuals (19.88%) reported symptoms indicative of depression. Notably, the data highlighted a significant discrepancy in depression rates between older women and men, with 27.44% of older women experiencing depression compared with 11.88% of older men (table 1).

Individuals in the lowest wealth quintile (first quintile) exhibited the highest prevalence of depression, with 86 depressed older adults (32.09%), whereas those in the highest wealth quintile (fifth quintile) demonstrated the lowest proportion, with 31 cases (11.48%) (table 1).

Furthermore, married participants displayed a lower incidence of depression, with 196 cases (17.85%) compared with individuals who were single, divorced or widowed, among whom the prevalence was notably higher at 29.39% and 72 depressed people (table 1)

Educational attainment was also found to be significantly associated with depression rates (table 1). Notably, individuals classified as individuals with low literacy levels exhibited the highest prevalence of depression, with 170 older people (28.10%), while those with academic degrees reported the lowest rates at 9.43% and only 10 cases.

The study delved into the impact of occupational status on depression and revealed that housekeepers registered the highest depression prevalence at 29.54% and 179 cases. Moreover, unemployment emerged as a significant contributor to depression, with a prevalence rate of 23.88% and 16 cases (table 1). Conversely, retirees displayed the lowest depression rates, with a prevalence of 8.89% and 45 older people.

Association of depressed mood and social capital in univariable and multivariable logistic regression

Multivariable logistic regression on the association of social capital and depression status, with a threshold p value less than 0.2, was used to assess the impact of the total social capital score on depression (table 2). The selection of covariates included in the final model was guided by a DAG, as described in the Methods section. The crude model revealed a significant association between all quintiles of social capital and the lowest level. The results were statistically significant after controlling for potential covariates/confounders, such as demographic characteristics, additional education level, marital status, job status, living arrangement and wealth indicators. The final model also demonstrated a substantial association across all quintiles compared with the lowest quintile (second quintile: OR: 0.74, 95% CI: 0.49 to 1.13, p=0.164; third quintile: OR: 0.75, 95% CI: 0.48 to 1.15, p=0.182; fourth quintile: OR: 0.65, 95% CI: 0.42 to 1.01, p≤0.055; fifth quintile: OR: 0.49, 95% CI: 0.30 to 0.80, p≤0.005).

Table 2. Multivariable logistic regression on the association of social capital and depression status.

	Quintiles of social capital	OR (95% CI)	P value
Crude	First (Ref.)	1
	Second	0.63 (0.42 to 0.94)	0.025
	Third	0.58 (0.39 to 0.88)	0.010
	Fourth	0.50 (0.33 to 0.77)	0.001
	Fifth	0.30 (0.19 to 0.49)	<0.001
Model 1	First (Ref.)	1
	Second	0.68 (0.45 to 1.03)	0.065
	Third	0.65 (0.43 to 0.99)	0.045
	Fourth	0.62 (0.40 to 0.96)	0.033
	Fifth	0.43 (0.26 to 0.70)	0.001
Model 2	First (Ref.)	1
	Second	0.72 (0.47 to 1.10)	0.124
	Third	0.73 (0.47 to 1.13)	0.155
	Fourth	0.63 (0.41 to 0.99)	0.044
	Fifth	0.48 (0.29 to 0.79)	0.004
Final model	First (Ref.)	1
	Second	0.74 (0.49 to 1.13)	0.164
	Third	0.75 (0.48 to 1.15)	0.182
	Fourth	0.65 (0.42 to 1.01)	0.055
	Fifth	0.49 (0.30 to 0.80)	0.005

Model 1: adjusted for age and sex; model 2: adjusted for additional education level, marital status, job status, living arrangement and wealth indicators. The final model was calculated using the backward elimination method with a threshold of p<0.20.

Association of depressed mood and components of social capital in univariable and multivariable logistic regression

Participation in collective social activities shows a significant association with depression status (table 3). As the level of engagement in collective activities increases from the second to the fifth quintile, the likelihood of depression decreases. The ORs for the second to fifth quintiles are 0.62 (95% CI (0.42 to 0.93)), 0.45 (95% CI (0.29 to 0.71)), 0.59 (95% CI (0.37 to 0.93)) and 0.37 (95% CI (0.22 to 0.61)), respectively, with all p values statistically significant: 0.025, <0.001 and 0.018.

Table 3. Multivariable logistic regression on the impacts of different domains of social capital on depression status.

Different domains of social capital	Quintiles of social capital	OR (95% CI)	P value
Collective activity	First (Ref.)	1
	Second	0.62 (0.40 to 0.94)	0.025
	Third	0.45 (0.29 to 0.70)	<0.001
	Fourth	0.59 (0.38 to 0.91)	0.018
	Fifth	0.37 (0.23 to 0.60)	<0.001
Voluntary help	First (Ref.)	1
	Second	1.01 (0.64 to 1.59)	0.974
	Third	1.04 (0.70 to 1.56)	0.833
	Fourth	0.69 (0.43 to 1.11)	0.123
	Fifth	1.00 (0.64 to 1.55)	0.986
Social trust	First (Ref.)	1
	Second	0.67 (0.43 to 1.06)	0.092
	Third	0.62 (0.40 to 0.99)	0.045
	Fourth	0.64 (0.42 to 0.98)	0.038
	Fifth	0.87 (0.58 to 1.31)	0.517
Social coherence	First (Ref.)	1
	Second	0.77 (0.48 to 1.24)	0.280
	Third	0.72 (0.47 to 1.11)	0.134
	Fourth	0.62 (0.41 to 0.96)	0.032
	Fifth	0.57 (0.37 to 0.90)	0.015
Social support	Having social support	1.03 (0.77 to 1.39)	0.841
Reciprocity	First (Ref.)	1
	Second	1.27 (0.87 to 1.26)	0.216
	Third	1.32 (0.86 to 2.03)	0.203
	Fourth	1.36 (0.87 to 1.42)	0.174
	Fifth	1.40 (0.90 to 1.15)	0.137
Social network	First (Ref.)	1
	Second	0.63 (0.40 to 0.99)	0.046
	Third	0.98 (0.59 to 1.64)	0.947
	Fourth	0.79 (0.54 to 1.14)	0.209
	Fifth	0.38 (0.22 to 0.63)	<0.001
Social capital	First (Ref.)	1
	Second	0.74 (0.48 to 1.13)	0.164
	Third	0.74 (0.48 to 1.14)	0.182
	Fourth	0.64 (0.41 to 1.00)	0.055
	Fifth	0.48 (0.29 to 0.80)	0.005

Results were obtained from the final model using the backward elimination method with a threshold of p<0.2.

Individuals reporting higher levels of social trust, particularly in the third and fourth quintiles, exhibit reduced likelihoods of depression, with ORs of 0.62 (95% CI (0.39 to 0.99)) and 0.64 (95% CI (0.42 to 0.97)) and p values of 0.045 and 0.038.

The second and fifth quintiles indicate that individuals with more dispersed social networks are less likely to experience depression. The ORs for these quintiles are 0.63 (95% CI (0.40 to 0.99)) and 0.38 (95% CI (0.23 to 0.63)), respectively, both statistically significant with p values of 0.046 and <0.001.

Voluntary help, social support, reciprocity and social capital do not exhibit statistically significant effects on depression status. The p values associated with these variables were above the threshold of 0.05, indicating a lack of statistical significance.

Discussion

The present study aimed to investigate the relationship between individual social capital and depression in older adults in Iran, using data from the baseline phase of the BLAS. The findings shed light on the complex interplay between social factors and mental well-being in this population, providing valuable insights for public health interventions and policy development.

The study’s results demonstrate a significant association between social capital and depressive mood among older adults. Specifically, higher levels of social capital, as measured by various components such as participation in collective activities, social trust and social network structure, were consistently associated with decreased odds of depression. These findings align with existing research highlighting the protective effect of social capital on mental health outcomes, especially among the elderly population.¹¹,¹⁷²⁶

From a methodological standpoint, our study employed a quantitative approach using standardised surveys to assess the mediating role of various domains of social capital in the association between social capital and depressive symptoms among older adults in Iran. Precisely, depressive symptoms were measured using PHQ-9, not through interviews. This method lets us quantitatively capture the participants’ depressive symptoms based on their PHQ-9 scores. We gathered data from multiple sources to capture the participants’ subjective experiences and perceptions, enhancing our analysis. In contrast, Sun et al focused solely on quantitative analysis, investigating the direct link between trust and depressive symptoms among older adults in China. Their approach, although valuable, may overlook the underlying mechanisms and subjective interpretations that qualitative methods can elucidate.³¹ Moreover, Bassett and Moore used regression models to explore the relationship between various dimensions of social capital and depressive symptoms among adults in Montreal, Canada. By quantifying neighbourhood social capital and its impact on mental health outcomes, their study contributed to understanding the contextual factors influencing depressive symptoms in urban settings.³²

Regarding the target population, our study specifically targeted older adults in Iran, acknowledging the unique cultural and societal contexts shaping their experiences of social capital and depressive symptoms. By focusing on this demographic group, we aimed to provide tailored insights and interventions to address mental health challenges in rapidly ageing societies. Similarly, Haseda et al investigated the association between community-level social capital and depressive symptoms among older Japanese adults. Their multilevel analysis revealed how contextual factors, such as community cohesion and civic participation, can moderate the impact of income inequality on mental health outcomes.²¹

In comparison with the findings of our study, Han et al and Tengku Mohd et al found direct associations between social support and depressive symptoms among older adults in South Korea and Asia, respectively.^{9 28} While our study focused on social capital, their findings underscored the protective role of social support networks in directly buffering against depressive symptoms. This difference in focus highlights the diverse pathways through which social capital can impact mental health outcomes across different cultural contexts and populations. Similarly, Bassett and Moore’s study in Montreal, Canada, quantified neighbourhood social capital and its impact on depressive symptoms among adults.³² Their findings provided valuable insights into the contextual factors influencing mental health outcomes in urban settings. Although their study did not specifically explore the mediating role of life satisfaction, it complemented our findings by highlighting the importance of community-level social capital in shaping depressive symptoms. Furthermore, Haseda et al’s multilevel analysis among older Japanese adults revealed how community-level social capital, such as community cohesion and civic participation, can moderate the impact of income inequality on mental health outcomes.²¹ While their study focused on income inequality as a contextual factor, their findings resonate with ours in emphasising the contextual influences on mental health and the importance of social capital in mitigating depressive symptoms.

The observed association between participation in collective activities and depression status underscores the importance of social engagement and community involvement in promoting mental well-being among older adults. Active participation in social activities provides opportunities for social interaction, emotional support and a sense of belonging, all crucial for maintaining psychological resilience and reducing the risk of depression.²² When people participate in social activities as a group, they feel like they belong and are supported, and this can help reduce feelings of loneliness and isolation, which are common contributors to depression.^{23 33} Being part of a group can also help distract individuals from negative thoughts and feelings associated with depression and bring moments of joy, pleasure and laughter.^{33 34} Participating in group activities gives people a sense of purpose and meaning.^{33 34} Whether they volunteer for a cause they believe in, play a team sport or contribute to a community project, this sense of accomplishment and fulfilment is essential for mental well-being.³⁵ Many social group activities also involve physical exercise, which is known to positively affect mental health, including reducing symptoms of depression by releasing endorphins and improving overall mood.^{36 37} Social interactions and group activities provide cognitive stimulation, which can help alleviate cognitive symptoms of depression, such as difficulties with concentration and memory.^{38 39} Group activities also foster connections with others with similar interests, values or experiences.⁴⁰ This sense of connection can provide validation, understanding and empathy, which are crucial for improving self-esteem and resilience in individuals struggling with depression.⁴¹ Similarly, the significant protective effect of social trust on depression status suggests that fostering trust and cohesion within social networks may contribute to better mental health outcomes.^{32 42} Trusting relationships provide individuals with a sense of security, social support and a buffer against stressors, reducing vulnerability to depression.⁴³

Furthermore, the association between social network structure and depression status highlights the importance of diverse and expansive social connections in mitigating depressive symptoms.^{44 45} Individuals with broader social networks may have access to a more comprehensive range of resources, including emotional support, practical assistance and opportunities for social interaction, which can enhance psychological well-being and resilience.⁴⁶,⁴⁸ Regular social interactions can mitigate the adverse effects of social isolation, which is a significant risk factor for depression among older adults.^{49 50} Social interactions provide cognitive stimulation and opportunities for learning, growth and engagement.⁵¹ Participating in conversations, activities and shared experiences can help keep the mind active and challenged, reducing the risk of cognitive decline and depression associated with boredom or inactivity.^{45 50}

The study findings have important implications for the development of interventions and policies aimed at promoting mental health and preventing depression in older adults.⁵²,⁵⁴ Strategies that enhance social capital, such as community-based programmes, social support networks and intergenerational activities, may be effective in reducing the burden of depression among older adults. Interventions focused on strengthening social connections, fostering trust and promoting collective activities can help build resilient communities and empower older adults to maintain their mental well-being. Additionally, efforts to address socioeconomic disparities and improve access to resources and support services may further enhance the protective effects of social capital on depression. Furthermore, the study underscores the need for a holistic approach to mental health promotion that recognises the interconnectedness of individual, social and environmental factors. By addressing social determinants of mental health and promoting social cohesion and support, policymakers and healthcare providers can create environments that facilitate healthy ageing and enhance the quality of life for older adults.

This study has several limitations that should be considered when interpreting the findings. First, the cross-sectional design does not allow for conclusions about causality. Although we identified associations between social capital and depressive symptoms, we cannot determine whether low social capital leads to depression or whether depression contributes to reduced social engagement. Second, while the PHQ-9 is a validated and widely used tool for assessing depressive symptoms, it is a screening instrument rather than a clinical diagnostic measure. Therefore, the classification of participants as having depression based on a cut-off score may include individuals with varying levels of symptom severity, which could affect the precision of our estimates. Third, both exposure and outcome measures were based on self-report. This reliance on participant-reported data introduces the possibility of recall bias and social desirability bias, which may have influenced how individuals reported their social relationships and mental health status. Fourth, although we used a validated, multidimensional tool to measure social capital, the instrument may not fully capture all cultural and contextual aspects of social capital in different subgroups of the Iranian older population, particularly those in rural or underserved settings. Fifth, while we adjusted for several key demographic and socioeconomic variables, the possibility of residual confounding from unmeasured factors, such as comorbid physical illnesses, use of mental health services or prior psychiatric history, remains. Sixth, the study population was limited to a single geographic area (Birjand County), which may affect the generalisability of the results to other regions of Iran or to countries with different social structures and health systems. Finally, the study did not report statistical indicators of model performance, such as pseudo-R², which could help assess how well the regression models explain the variability in depression outcomes. Moreover, because the analysis was based on existing baseline data from an ongoing cohort, no prior sample size calculation was conducted for the specific objectives of this study. These factors may limit the internal validity and statistical power of the results.

Conclusion

This study investigated the association between social capital and depression in older adults in Iran, using data from the baseline phase of the BLAS. The findings provide valuable insights into the complex relationship between social factors and depression in this population. The study revealed a significant association between social capital and depression status among older adults, with higher levels of social capital associated with decreased odds of depression. Specifically, participation in collective activities, social trust and social network structure emerged as significant protective factors against depression, highlighting the importance of social engagement, cohesion and support in promoting depression. These findings have important implications for public health interventions and policy development aimed at preventing depression and promoting mental well-being among older adults. Strategies focused on enhancing social capital, fostering trust and promoting social connections may be effective in reducing the burden of depression and improving the quality of life for older adults.

Data availability statement

Data are available upon reasonable request.