Several recent articles have pointed to the effect of social context on heart disease mortality after adjusting for individual level indicators. This study investigates the contributions of individual socioeconomic factors (sex, race, and education) and social context at the neighborhood level (wealth, education, social capital, and racial/ethnic composition), and the county level (social inequality, human and social capital, economic and demographic characteristics) on premature cardiovascular mortality. Death certificate information was obtained for all those who died of heart disease in Texas, USA, in 1991. Deaths were geocoded to obtain block-group, census tract, and county social context from the census. Multilevel hierarchical models quantified the contributions of individual characteristics and block-group, tract, and county social context on years of potential life lost to heart disease. Cross-level analyses investigated the interaction between individual and contextual factors. Being female, having more education, and residing in areas with higher median house value were associated with less premature mortality. Although blacks and Hispanics lost more years of life to heart disease than whites, blacks and Hispanics living in tracts with higher own racial/ethnic group density lost fewer years of life than their peers living in less homogenous tracts. At the county level, premature mortality was negatively associated with social capital. The tract and county level variances were statistically significant indicating the importance of social context to premature heart disease mortality. Plausible mechanisms through which these effects operate are explored. Social context at the block-group, tract, and county level played an important role, though a smaller role than individual factors, in explaining years of life lost to heart disease.