More and more literatures suggest that sociological factors, instead of health care system, are keys to improve health situations in the United States. This study aims at contributing to these literatures by focusing on how social capital affects health outcomes in Iowa's rural communities. Social capital in this study is measured at the individual and community levels. The first hypothesis of the study is that individual-level social capital has a direct positive effect on health outcomes after controlling for demographic and socioeconomic factors. The second is that community-level social capital positively predicts health outcomes after controlling for individual-level variables in rural communities in Iowa. Data used in the study come from a house quality and health status survey conducted in 2001 by the Rural Development Initiative (RDI) research group in the sociology Department at Iowa State University. Six rural communities were purposively selected from the RDI 99 rural-community list. Approximately 200 households were randomly selected from each of the six communities. Totally 1,244 questionnaires were mailed out, and 646 households filled out and returned the questionnaires. All 646 questionnaires are valid. A variety of statistical methods were carried out. OLS regressions were used to test the individual-level hypothesis. Hierarchical linear models and two-sample t-tests were carried out to test the community level hypothesis. The analyses provide strong evidence that individual-level social capital has a positive significant effect on health outcomes after controlling for age, gender, and household income in the six rural cities. No solid evidence has been found that community-level social capital predicts health outcomes significantly after controlling for individual-level variables. However, the findings don't necessarily suggest that there is no relationship between community-level social capital and health outcomes after individual-level variables are controlled for because the data have a small community-level sample size. It is hard to get significant findings if the sample size is too small. The study suggests for future research directions. A large number of heterogeneous rural communities will help to find evidences of significant effects of community-level social capital on health outcomes in rural communities in Iowa.