A growing body of research is interested in the roles of health literacy in affecting health outcomes. Improving health literacy among Americans is one of the health goals specified in Health People 2010 by the U.S. Department of Health and Human Services. However, there are some important gaps in the existing literature. Most research has primarily focused on very specific groups of people within specific health care settings using measures of health literacy that are designed only for medical setting. Moreover, the potential variation by age and gender in the association between health literacy and health outcomes has not been systematically examined. In this study, I empirically test two hypotheses using data from a nationally representative sample of 18,102 American adults (2003 National Assessment of Adult Literacy; NAAL): 1) health literacy is positively associated with self-rated health for both men and women, net of such covariates as race/ethnicity, immigrant status, parental education, educational attainment, poverty, welfare receipts, household income and health insurance coverage; 2) the effect of health literacy on self-rated health increases by age for both men and women. I measure the effect of health literacy as the gaps in the likelihood of having higher levels of self-rated health status among three groups of people with different levels of health literacy. I use ordered regression models that predict the likelihood of having higher levels of self-rated health by health literacy and other covariates using the whole sample of men and women, separately, to test Hypothesis 1. Then, I examine the association between health literacy and self-rated health across separate age groups to see how the association changes by age (Hypothesis 2).