Does social capital enhance self-rated health of married women? Findings from a population-based study

Background: Social capital is one of the main concepts that has received attention in the health literature. This study aimed to investigate the association between social capital and self-rated health in married women in four western provinces of Iran. Methods: This cross-sectional study was conducted using a multi-stage sampling method in four main western provinces of Iran. In total, 1531 married women participated in our study. The questionnaire included demographic variables, women's self-rated health, and social capital. Results: The mean (SD) self-rated health was 3.83 (0.77). Pearson correlation test showed that among the social capital subscales, trust had the most (r = 0.116, p<0.001) and networks the least signicant correlations (r = 0.095, p<0.001) with self-rated health among the participants. Social capital shows a weak association with self-rated health (Coecient=0.0030, p=0.001). Conclusion: This study showed that there was a positive correlation between social capital dimensions with the self-rated health among women, however it was relatively weak association. This may be due to the complex nature of health and its related factors such as social capital.


Introduction
Self-rated health (SRH) as a simple and good indicator of health has been used widely in social epidemiological studies [1]. SRH assess perception of people about their general health [2]. According to the literature, people who perceived their health as poor have a greater risk of mortality and worst overall health pro le [3,4]. During the last decades, empirical evidence showed that social capital is one of the important social determinants of health Evidence shows that social capital is associated with a range of health-related behaviors and their consequences including criminal behaviors, mental health, smoking, suicidal ideation, suicide attempts and mortality [5][6][7][8][9]. A study in the Brazilian general population found that people who had fewer friends, and provide less support for neighborhood people, and had lower community-based participation and activities were more likely to rate their health negatively [10].
Evidence also shows that high levels of community participation were associated with higher social cohesion and better mental health, physical functioning, and general health [8].
Social capital de ned by Putnam "as social trust, cross-trade norms, and social network density" [11].
Social capital is a context-based concept that its conceptualization, measurement, and effect may differ from a society to another society with different economic, social and cultural backgrounds. Iran is one of the developing and under transition countries that has experienced unstable political and economic situation and is under great sanctions in the last decade, accompanied by fundamental but gradual social changes. It seems that this unstable political and socioeconomic status could increase or decrease social capital and its in uences on the population's health. Despite the enormous costs spent annually on disease prevention and health promotion programs, many of the key determinants of health such as social relationships, citizens' trust, and social capital especially among women are still neglected [12].
Distribution of different types and levels of social capital are not equal among various groups of a society. The related evidence suggests that there is a difference between social capital of men and women especially married women. Women constitute main part of the Iranian population who can play an important role in sustaining levels of social capital. However, there is little evidence about the level and characters of social capital and its effect on SRH of married women in Iran.
A previous study research demonstrated that the existence of social capital could provide various social support, for people including nancial, information, appraisal practical and emotional supports, among which the provided emotional support more clearly helps a person's health [13]. Giddens believes that emotional support is a kind of protective shell by which people protect themselves when faced with life challenges [14]. Although the association of social capital and health have long be studied [15], the previous studies are mainly from developed and high-income countries, and there is a little evidence on the relationship between social capital and health especially among women from developing countries.
The married women as an important subgroup of population have an important role in family and society's health. Due to gender speci c predictors of health, evaluating of predictors health among married women are necessary. Therefore, this study aimed to investigate the association between social capital and SRH in married women in four western provinces of Iran.

Measures
The rst part of the questionnaire included demographic variables such as age, spouse's age, number of family members, education, spouse's education, history of illness, and hospitalization, occupation status, and spouse's occupation. The second part of the questionnaire was a question on women's SRH. We asked participants "How would you rate your general health status?". This question measures SRH on a 5-point Likert scale and has been proposed by WHO [16]. The third part of the questionnaire was a questionnaire for measuring the social capital that was developed based on the Nahapiet and Goshal social capital questionnaire [17]. This questionnaire has three structural, relational, and cognitive dimensions that examine seven subscales of networks, trust, collaboration, mutual understanding, relationships, values, and commitment. It consists of 28 items based on a ve-point Likert scale (strongly agree = 5, agree = 4, disagree = 3, disagree = 2, and strongly disagree = 1). A higher score indicates more social capital. In the present study, according to the target group and the research objectives, the questionnaire was translated into Persian and tailored. The face and content validity of the questionnaires was assessed by an expert's panel including three sociologists, three psychologists, and two health promotion specialists. According to the experts' comments, Content Validity Index (CVI), and Content Validity Ratio (CVR) were 0.92 and 0.84, respectively. To assess the reliability, the questionnaire was distributed among 150 individuals who were similar to the statistical sample. Cronbach's alpha coe cient for the whole questionnaire was 0.93. Cronbach's alpha coe cient for the structural, cognitive and relational, dimensions was 0.87, 0.95, and 0.90, respectively Data Analysis Data were analyzed by STATA version 12. Descriptive analyses are reported as numbers (percentages), means (standard deviations). To determine the relationships between the variables, Pearson correlation.
Multiple linear regression analysis was conducted to determine the adjusted association between social capital and self-rated health. To correct for the population distribution, poststrati cation adjustments were applied using sampling weights. P-value < .05 was considered statistically signi cant.

Results
Out of 1650 distributed questionnaire, a total of 1531married women completed the questionnaire, with a response rate of 94%. The mean (SD) age of the respondents was 33.67 (11). The lowest percentage of age groups belonged to 60 years or older (3.3%). Nearly 30% of the respondents had a bachelor's degree.
Seventy-six percent of the respondents were housewives and only a small percentage (1.1%) were retired. The majority of the respondents (90.4%) reported that they have no history of chronic disease (Table 1). Nearly 6% of participants were either unsatis ed or very unsatis ed with their state of health, while 74% were either satis ed or very satis ed. The mean (SD) SRH and overall social capital were 3.83 (0.77) and 97.66 (23.50), respectively. Table 2 present the mean (SD) subscales of social capital dimensions and SRH.
Pearson correlation test showed that among the social capital subscales, trust had the most signi cant correlation (r = 0.116, p < 0.001) and networks had the least correlation (r = 0.095, p < 0.001) with SRH.
Among the dimensions of social capital, the total Relational dimension had the strongest correlation (r = 0.120, p < 0.001) and the total cognitive dimension had the weakest correlation (r = 0.103, p < 0.001) with SRH. Also, results of the partial correlation showed that the relationship between social capital and SRH was positive and signi cant (r: 0.092, p = 0.001) after controlling of age, education, chronic disease, and occupation (Table 3).
Multiple linear regression test was performed to investigate the predictors of women's SRH. The variables of education, occupation, age, chronic disease, and social capital were entered into the model. The results showed that the level of education is associated with the SRH. Only the diploma level of education compared to the reference group (below diploma) was not signi cant in the model (Coe cient = 0.022, p = 0.709). As shown in Table 3, the presence or absence of chronic disease had signi cant adjusted association with women's SRH (Coe cient = 0.750, p = 0.001). Where social capital had the least impact (Coe cient = 0.0030, p = 0.001) ( Table 4).

Discussion
Our study aimed to investigate the association between social capital and SRH among the married women in four western provinces of Iran. Finding of our study could ll gaps in our knowledge of social capital and SRH among married women in the context of Iran as a developing country. Our nding showed that mean of SRH among the married women was relatively high. It was inconsistent to those reported in Khawaja and Mowaf study [18] and was similar to Brown et al. [19]. Also, our nding indicated that considering the attainable range of social capital score, the mean of the social capital in the married women was relatively high.
According to our ndings, higher social capital was associated with increase in SRH in the married women. This nding was consistent with the results of previous studies [20][21][22]. Kim et al. (2007) in the United States found that people with higher levels of social capital experienced few days with poor physical and mental health (boredom and ill-health) and activity restriction [23]. Result of a study in Japan showed that women who actively participated in the social group were more socially empowered compared to women with lower social participation, which is conceptually consistent with the present study [24]. Herzong et al.'s (2002) also showed that lower social participation and social support (as indicators of social capital) have negative effects on people's health [25]. However, a study conducted in Iran showed that there was no signi cant association between participation in social or local activities, being active in the social context, and quality of neighborhood communications with perceived health [26]. Also, our nding showed a correlation between the relational dimension of social capital and SRH among the participants. This dimension was more strongly correlated with SRH than others. Also, among the subscales of communication, trust had the highest correlation with SRH. Most theories of social capital emphasize two key elements including trust among social members and participation in social organizations [27]. Social trust is a positive ethical orientation among individuals in a community with essence of community-based moral expectations [28]. Girdano et al. (2010) in a study showed that an inability to trust others was signi cantly associated with poor health, while social participation was signi cantly associated with improved health over time [29]. Meng et al. (2014) also assessed the association between social capital and health among urban and rural population in China and found that among the components of social capital, only trust had a signi cant effect on health [30]. The ndings of our study showed that cognitive social capital had a positive and signi cant association with participants' SRH. This nding is consistent with the past study by Rajabi et al. (2013) among Iranian teachers [12]. Cognitive social capital addresses qualitative dimensions of social capital and encompasses variables such as values, reciprocal norms, and collaboration. It can affect overall health by creating health-related behavioral norms, controlling risky behaviors, social support, and creating uno cial information exchange tools. While the cognitive dimension of social capital is considered as a key factor in moving individuals to health action, the structural dimension is also effective as facilitator. Social support as a component of cognitive social capital promotes health through highlighting the sense of belonging and solidarity in individuals as well as enhancing the performance and self-esteem [12].
Our ndings also showed that the structural dimension of social capital was correlated with the women's SRH. The growth of voluntary service associations and participation of women in those as a structural dimension of social capital can enhance the vertical and horizontal relationships between people and improve their health. The associations can build trust and partnership at the macro, middle and micro levels of community through warm and symmetrical relationships and creating material and emotional opportunities, and subsequently strengthening the social relations network [31].
According to our nding, education and employment were associated with SRH among the participants. These ndings are consistent with those in Youse et al. study (2010) [32]. Previous studies showed that the mental health status of employees is better than non-employed, which is consistent with the results of the present study [33,34]. According to Keyes and Shapiro (2004), attention to the effect of social context in health is an important issue which should be considered in healthcare and medical interventions [35]. Although Field (2006) has emphasized that the psychological approach to trust and control are apart from social environment and government policies [11], however, the general pattern of evidence such as the ndings of our study and similar studies illustrate the positive association between social capital and health. Putnam discussed that social capital through social networks is related to health in four ways: rst; social networks reduce anxiety through tangible bene ts, second; networks reinforce the health measures, third; networks can better address the demand for health services, and fourth, social interaction and activation make the body's immune system more active [36].
Our study has two limitations that have to be considered when interpreting the ndings. First; using a large sample size from four provinces resulting in more precise estimates. Second; a very high response rate (94%) along with a randomly selected sample decreased the possibility of selection bias. However, there would be a possibility of measurement bias, because of the self-reported nature of data gathering.

Conclusion
This study showed that there was a positive association between social capital and its dimensions with the SRH among women, although it was a relatively weak association. This might be due to the complex nature of health and its associated factors especially social capital. promote group activities through community-based voluntary associations can facilitate social participation, develop personal relationship networks and trust, strengthen emotional and social support, and ultimately reduce physical and mental stress and improve health among married women.

Abbreviations
Page 7/11 SRH: self-rated health; CVI: Content Validity Index; CVR: Content Validity Ratio Declarations Ethics approval and consent to participate At the start of each interview, the main study objectives were elucidated and informed consent was obtained. The study was approved by the ethics committee of Kermanshah University of Medical Sciences (approval number: IR.KUMS.REC.1398.014).

Consent for publication
Not applicable.