Background Residents of environmental justice (EJ) communities often suffer significant health inequities due to pollutants and to adverse social conditions. However, methodologies for assessing such communities seldom account for both kinds of factors. While traditional environmental health risk assessments use single-pollutant, single-source measures of chronic risk alone, this project develops a technique for assessing “overall risk burden” in EJ communities. Cumulative risks from aggregate (that is, multi-agent, multi-pathway, multi-source, over time) exposures are combined with an index representing a wide range of social determinants of health. The work is being piloted in West Port Arthur, Texas, an EJ community flanked by petrochemical plants and a seaport, and characterized by poverty and disadvantage. Materials and methods Place-based social determinant indicators relevant to EJ communities were identified from Health Impact Assessment tools, then added to novel measures of cumulative risk developed by the U.S. Environmental Protection Agency and others, forming a composite index for overall risk burden [1,2]. Indicators were mapped to show spatial disparities across the community [3]. A community participatory mapping exercise was conducted to collect local knowledge of overall risk burden and to support local capacity to participate in and understand the project. Results Preliminary findings suggest that the neighborhood’s proximity to petrochemical plants and to the seaport poses health risks to residents not only due to exposure to pollutants but also to the adverse health impact of living in an area with physical isolation due to poorly designed roadways, poor food security, poor child care availability, high unemployment and a depressed economy. Spatial concentrations of poverty and segregation further undermine health in the community. In fact, non-chemical social determinants of health may have a greater combined effect on health than exposure to toxicants in this community. Further study of this differential effect is needed. Conclusions