The goal of this study is to identify the underlying structure of Health Competence and its value as a predictor of access to care among the Latino population in Baltimore, Maryland. Data on a cross-sectional urban probability sample were collected from 330 foreign-born Latino men and women aged 21-75 years residing in Baltimore at the time of the survey. Principal components analysis yielded a two-component solution: the first component comprised "factors enabling" access to care; the second, "perceived barriers". When testing the predictive power of the Health Competence construct using model building and the log likelihood criteria the "enabling factors" added significantly (P < .005) to the power of the socio-demographic variables to predict access to care, making it an important tool for programs aiming to improve US Latinos' health.