The predictive validity of a social capital index in routine breast and cervical cancer screening.

Because community-based interventions to increase breast and cervical cancer screening have had limited success, concepts that move away from individual-level screening predictors and toward ecological-level predictors may be useful. Social capital encompasses both interpersonal and collective relations with others, and may be appropriate in the investigation of current cancer screening disparities. 

This study used data from the California Health Interview Survey (CHIS) for psychometric analysis and prediction of cancer screening behavior. Psychometric analysis included an expert Delphi panel, as well as validity and reliability analysis. Logistic regression tested the ability of social capital to predict breast and cervical cancer screening behavior against other known screening predictors. This study validated a six-item social capital scale (alpha=0.75) among 20,667 adult women. Multivariate modeling showed that social capital was a significant predictor of breast (OR=1.08, 95%CI=1.03-1.13) and cervical (OR=1.05, 95%CI=1.01-1.10) cancer screening behavior. However, commonly known predictors of cancer screening, such as insurance status and age, were stronger predictors of screening adherence than social capital. This study is the first step in exploring the link between social capital and cancer screening behavior, and may be a promising area of research in cancer prevention and control.