Measurement of social capital in relation to health in low and middle income countries (LMIC): a systematic review.

Social capital is a neglected determinant of health in low and middle income countries. To date, majority of evidence syntheses on social capital and health are based upon high income countries. We conducted this systematic review to identify the methods used to measure social capital in low and middle-income countries and to evaluate their relative strengths and weaknesses. An electronic search was conducted using Pubmed, Science citation index expanded, Social science citation index expanded, Web of knowledge, Cochrane, Trip, Google scholar and selected grey literature sources. We aimed to include all studies conducted in low and middle-income countries, published in English that have measured any aspect of social capital in relation to health in the study, from 1980 to January 2013. We extracted data using a data extraction form and performed narrative synthesis as the measures were heterogeneous. Of the 472 articles retrieved, 46 articles were selected for the review. The review included 32 studies from middle income countries and seven studies from low income countries. Seven were cross national studies. Most studies were descriptive cross sectional in design (n = 39). Only two randomized controlled trials were included. Among the studies conducted using primary data (n = 32), we identified18 purposely built tools that measured various dimensions of social capital. Validity (n = 11) and reliability (n = 8) of the tools were assessed only in very few studies. Cognitive constructs of social capital, namely trust, social cohesion and sense of belonging had a positive association towards measured health outcome in majority of the studies. While most studies measured social capital at individual/micro level (n = 32), group level measurements were obtained by aggregation of individual measures. As many tools originate in high income contexts, cultural adaptation, validation and reliability assessment is mandatory in adapting the tool to the study setting. Evidence on causality and assessing predictive validity is a problem due to the scarcity of prospective study designs. We recommend Harpham et al. s' Adapted Social Capital Assessment Tool (A-SCAT), Hurtado et al. s' six item tool and Elgar et al. s' World Value Survey Social Capital Scale for assessment of social capital in low and middle income countries.