The Role of Social Capital on Utilization of Maternal and Child Health Services in Low- and Middle- Income Countries: Mixed-Methods Review

Background: Social capital is dened as social relations that may provide individuals and groups with access to resources and supports in community networks. It has become an important issue to be studied in the eld of public health. This study aimed to review pieces of literature about the role of social capital on the utilization of maternal and child health services in low- and middle- income countries. Methods: Mixed-methods research review and synthesis, using three databases (PubMed, Scopus, and Science Direct), was performed. Besides, Google Scholar and Google search engines were used. Both quantitative and qualitative studies conducted in low- and middle- income countries, published in English, in grey and commercial literature were considered. Results: A total of 1,545 studies were identied, of which, 328 records were removed due to duplication. Then, 13 records were selected after reading titles, abstracts, and full texts. Of these eligible studies, six studies were included for quantitative synthesis, and seven of them for qualitative synthesis. Moreover, six quantitative studies and seven qualitative studies were included. Of the six reviewed quantitative studies, ve of them addressed the relationship between social capital and health facility delivery. Likewise, women who lived in communities with higher membership in groups that help form intergroup bridging ties had higher odds of antenatal care use. Synthesized qualitative ndings revealed that women received some form of emotional, informational, and instrumental support from their network members. Receiving health information from trusted people and socio-cultural factors inuenced the use of maternal health services. Conclusions: Social capital has a great contribution to the utilization of maternal and child health services. Countries aiming to reduce maternal mortality should identify context-specic approaches to benet from its social capital. On the other hand, measurement tools for social capital require due attention as there is no uniformity across studies and most of them were conducted using a cross-sectional design. Hence, further longitudinal and interventional studies should be employed to pin down the direction of causality.

Moreover, involvement of religious leaders, health extension workers, women developmental army leaders, and selected community members could enhance use of maternal and child health services. Women who received health information from people they trust are more likely to access and use health services [33]. Social trust has positive association with better health and safety of a community. Neighborhoods with higher levels of social trust experience lower rates of health and health related problems, and have fewer signs of physical disorder, making residents of these neighborhoods feel safer [34][35][36][37].
Despite studies in India [24,38], Tanzania [39] and Cameroon [40] countries indicated the role of SC in improving uptake of maternal and child health services, some other studies have identi ed its negative consequences as exclusion of outsiders, excess claims on group members, restrictions on individual freedoms, and downward leveling norms [41,42]. To date, there is no study that systematically synthesizes the available literature focusing on social capital's role to improve maternal and child health services use in LMICs. Therefore, this review aimed to synthesize the available literature about the role of social capital on the utilization of maternal and child health services including antenatal care (ANC), institutional delivery and postnatal care (PNC) in LMICs. The ndings of the study will inform policy and decision makers to improve maternal and child health services use in LMICs.

Methods
This systematic review was conducted according to the Joanna Briggs Institute (JBI) manual for mixed evidence synthesis [43]. The protocol was registered with the PROSPERO database (registration number: CRD42021226923). PICO: population (women and children), intervention (social capital), context (Lowand middle-income countries), and outcome (utilization of maternal and child health services).

Inclusion criteria
In this systematic review, the inclusion criteria were: Types of studies: both quantitative and qualitative study designs Setting: studies that were conducted in one or more LMICs based on the World Bank criteria Search and search strategy We used Pubmed, Scopus and Science Direct data bases to search articles. In addition, Google Scholar and Google search engines were used for grey literature. To search literature in databases, several searching techniques were used, for example, the presence of medical subject headings (MeSH), text words and key words in the title or abstract. When we search in Google Scholar and Google search engines, the rst 100 hits were included. Moreover, we tried to access literature in websites of international organizations such as world health organization and World Bank.
Our search strategy combined terms related to the four domains: Firstly, the term related to 'social capital' like 'social support' OR 'social trust', 'social network' OR 'community network', OR social cohesion were searched. The second term, 'maternal and child health services', comprised 'antenatal care' OR health facility delivery OR postnatal care. The third term, 'utilization' contained utilization OR access OR use OR uptake. Then, countries in 'LMICs' regions were embraced in the search strategy. All synonym keywords and subject headings were combined with the "OR" Boolean operator. Finally, the four domains including social capital, maternal health services, utilization and LMICs were combined with the Boolean operator "AND". According to the World Bank classi cation [44], articles were chosen for review if they were conducted in LMICs, measured social capital or related terms as an exposure and one or more of the maternal and child health services as an outcome of interest. (Appendix 1)

Study selection
The literature search was conducted by EWM. On completion, titles and abstracts of the papers identi ed were reviewed. All retrieved records were downloaded into EndNote, stored and categorized by database of origin. Potential duplicates were removed. Reviewers (EWM and DAZ) independently applied the inclusion criteria to all titles. All papers that appeared to meet the inclusion criteria or where there was uncertainty were taken through to the next stage of full text review. Any disagreements on article exclusion or inclusion were resolved by consensus.
Data extraction and management Data were extracted from articles or reports using an excel sheet. The main information collected from each study contain last name of author(s), year of publication, study methods (study setting, study participants, study design, the year of data collection, sample size and data analysis), key ndings and limitation acknowledged by the author(s) of the study.

Data Synthesis
We performed textual narrative synthesis due to lack of uniformity on the de nition and measurement of social capital and dependent variables. The characteristics, key signi cant ndings, and limitation of the individual studies were presented in tables. The qualitative ndings were synthesized using metaaggregation. The pooled ndings were rst grouped into categories de ned by their similarity of meaning and then combined into one or more synthesized nding(s) that captured their meaning.
Quantitative and qualitative components of mixed methods studies selected for retrieval were assessed by two independent reviewers (EWM and DAZ) for methodological validity prior to inclusion in the review using standardized critical appraisal instrument.
Authors of papers were contacted to request additional data for clari cation, where required. Any disagreements that arise between the reviewers will be resolved through discussion. The results of critical appraisal were reported in narrative form and in a table.

Results
Characteristics of the studies A total of 1,545 studies were identi ed by searching databases and other search engines to synthesize evidence about the role of social capital on utilization of maternal health services in LMICs, of which, 328 records were removed due to duplication. Then, 13 records were selected after reading titles, abstracts and full texts. Of these eligible studies, six studies [24, 38-40, 45, 46] were included for quantitative synthesis and seven [33,[47][48][49][50] of them for qualitative synthesis. This systematic review followed the four-phase ow diagram, Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement [51]. (Fig. 1) Of the six quantitative cross-sectional studies, ve [24,38,40,45,46] were rated as moderate quality and one [39] was rated high quality. Four out of six studies undertook systematic random sampling in selecting their study participants. The samples taken for the study were representative and outcomes were measured using reliable methods. All of the six studies assessed their outcome using objective measures through proxy questions; controlled confounding factors using multivariate and multivariable regression analysis and did not describe those participants who withdraw or refused to participant in the study.
Of the seven qualitative studies, six [33, 48-50, 52, 53] were assessed as high quality, and one [47] was assessed as moderate quality. All seven of these studies lacked description of the congruency between the philosophical perspectives and research methodology used. Failure to describe how the researchers' perspectives may have in uenced the analysis and interpretation of ndings was identi ed as the main weakness in these qualitative studies, potentially undermining credibility. (Appendix 2) In this review, studies showed that the concept of social capital had multiple dimensions. Some studies assessed both structural and cognitive social capital.
Other studies also examined bonding and bridging dimensions. Related to measurement tools for SC, studies used different types of measurement tools. The variations among tools in its content indicated that instruments for measuring social capital are at the developmental stage.

Statistical analysis
According to the nature of variables and objectives to be measured, several researchers employed various analysis techniques for both quantitative and qualitative studies. Exploratory factor analysis (EFA) was used to develop and validate a tool for measuring SC and investigate the in uence of sociocontextual variables [24,[54][55][56][57][58]]. In addition, con rmative factor analysis (CFA) indicated the reliability and validity of social capital scales [55,59]. Role of social capital on utilization of maternal and child health services Of six reviewed quantitative studies, ve of them addressed the relationship between social capital and health facility delivery or professional delivery care. The studies in India examined the association of women's participation in community networks known as self-help group/SHG/ and utilization of maternal health services. One of the studies showed that women from villages with a SHG were more likely to give birth in a health facility [45]. In contrast, other study reported that SHG was not statistically signi cant; however, the presence of SHGs in the community was appreciated for delivering health care messages within and beyond the group [66]. (Table 1) Three of the six studies focused on the association between social capital and the utilization of antenatal care. Women who lived in communities with higher membership in groups that help form intergroup bridging ties had higher odds of antenatal care use, whereas women who lived in communities with higher intragroup bonding ties and collective e cacy had lower odds of antenatal care use [24]. Besides, male partners' involvement in maternal health care during pregnancy has bene ts on maternal health care services access and utilization. The odds of having at least one ANC was higher in women whose male partners' involvement scores were higher [46]. In general, utilization of maternal and child health services which includes antenatal services, institutional delivery and postnatal services were positively associated with number of contacts with health workers during pregnancy [66] and social capital [40].
Qualitative studies conducted in Ethiopia [33], India [47,48], Ghana [49] and Kenya [50] revealed that all women received some form of emotional, informational and instrumental support from their network members during pregnancy and child birth [49]. Besides, receiving health information from trusted people and socio-cultural factors in uenced use of maternal and child health services [33,50]. Mothers-in-law play a crucial role in the utilization of ANC in Nepal. Most of them are illiterate, had not used ANC previously and often discouraged their daughters-in-law from attending ANC services. In most cases, mothers-in-law seemed not to be in favor of ANC and played a negative role. "My mother-in-law doesn't help me. It might be due to her past experiences. She used to do all the work by herself during her time of pregnancy. So she wants me to do the same. I have lots of work here at home, so I don't go for ANC checkups" (Non-user Woman 4) [52]. (Table 3)

Discussion
This study aimed to synthesize the available literature about the role of social capital on the utilization of maternal and child health services in LMICs. Women's social capital had great contribution on the utilization of maternal health service including ANC, health facility delivery and PNC. Women from villages with a SHG were more likely to give birth in a health facility [45]. Similarly, the qualitative component of this review revealed that it is socially normative for in-laws, particularly mother in laws, mothers and grandmothers to advice women and provides suggestions on how to experience safe pregnancy and delivery. Pregnant women received some form of emotional, informational and instrumental support from their network members [49].
On the other hand, socio-cultural factors might hinder utilization of maternal health services. In some communities, women have to stay indoors for a month to 40 days; some members of the community cannot go to the hospital for health care services for whatever problem without rst going to herbalists; only opting for a facility delivery if complications arise during the birthing process and TBA play a critical role in the decision-making pathway for choice of place of delivery [50]. In line with these ndings, a previous systematic review reported that in uence of traditional beliefs and sociocultural norms was high during childbirth. Women interpreted their expectations through the lens of family birth stories and social norms [68].
There is no single universally accepted measurement tool for social capital that indicates the tools are in the early stages of development. However, it can be measured at both individual and community levels [30] using quantitative tools [69]. Most reviewed studies assessed both structural and cognitive SC.
Structural SC measured using multiple components including individuals' or groups' participation in neighborhood, public civic activities, networks, organizations, associations and institutions, support from individuals and groups, frequency and breadth of community participation [55-57,

Implication for public policy makers and researchers
The evidence in this review showed that SC has an impact on utilization of maternal and child health services. So, it will be helpful for policy makers to design strategies on strengthening SC of the community that enhances uptake of ANC, health facility delivery and PNC. However, the pathways for SC and which socio-cultural contexts affect it were not as such investigated. Moreover, the measurement of SC is still in its infancy stage. Hence, further research is required to ll these gaps.

Conclusions
Social capital has great contribution on utilization of maternal and child health services including ANC, health facility delivery and PNC even though sociocultural factors in uence it. Measurement tools for SC have no uniformity across studies and most of them were conducted using a cross-sectional design. Hence, further longitudinal and interventional studies should be employed to pin down direction of causality.

Availability of data and materials
Not applicable (systematic review using data published in primary studies).

Competing interests
The authors declare that they have no competing interests.

Funding
No founding used for this study.
EWM conducted the searches of the literature, screened titles, abstracts and full text of articles, extracted and interpreted data and wrote the rst draft of the manuscript. DAZ conduced secondary screening of titles, abstracts and full text of articles, checked data extracted by EWM, and assisted with writing of the manuscript. GDA, YA and GAT assisted with writing the manuscript. All authors read and approved the nal manuscript.   Most women received advice to use facility-based pregnancy and delivery care Informational support It is socially normative for in-laws, particularly Mother in laws[MIL], mothers and grandmothers to advice women and provides suggestions on how to experience safe pregnancy and delivery Network members lived in close proximity, and generally had good relationships, with the women. Woman interacted with her husband and in-laws daily because they lived in the same house When her husband is not there.. .you [MIL] would then talk to any family member available at that time, for that person to look for a motorbike, fuel it and take her to hospital Instrumental support Network members able to help her access and utilize facility delivery Some network members tended to rst seek the involvement of a traditional birth attendant (TBA) during women's labor and did not make timely arrangements to transport women to a facility Some women had numerous sources of support, their own mothers, female relatives and friends.
WDA leaders are good in passing different knowledge to mothers and members of the community during community meetings, women's association meetings, antenatal outreach sessions, and coffee ceremony Promotion of health care services Receiving health information from trusted people enhanced use of maternal health services.
HEWs, WDA and religious leaders are also participating on community mobilization activities including use of full ANC services, health facility delivery and PNC Assistance with community fund raising, facilitating ambulance services or traditional ambulances to get women to the health center for delivery, providing training for model family/WDA

Provision of continuous support
Integrating activities between community leaders, including WDA leaders, religious leaders and HEWs, are all considered to be bridges and enhance strong relationship and communication between HEWs, primary health care units and community members A link between communities and health system There are some members of the community who cannot go to the hospital for health care services for whatever problem without rst going to herbalists In uence of socio-cultural norms Socio-cultural factors in uenced utilization of maternal health services Ill health is as a result of evil spirits and traditional systems of health care were bestplaced to deal with them Only opting for a facility delivery if complications arise during the birthing process In my community, a woman has to stay indoors [for a month to 40 days] Maternal gures would typically provide TBA services and play a critical role in the decision-making pathway for choice of place of delivery Role of signi cant matriarchal gure Islam might forbid women from being seen by other men except their husbands, this was not in line with realities in the health system In uence of religious norms Women would avoid seeking health services like delivering in a health facility if no female provider was available The role of a woman in this community was mainly to give birth and have many children Role of gender stereotypes Gender-related power imbalances were also reported