Federal law requires the Environmental Protection Agency (EPA) to assess drinking water risks and to set standards to insure a safe drinking water supply. Given limited data, risk assessors rely on high dose experimental data, results from occupational health studies, and other data sets (e.g., in vitro tests) to make predictions about health outcomes. A variety of assumptions and extrapolations must be used to estimate from the results of these studies the likelihood of adverse health effects associated with a lifetime of water consumption. In recognition that the uncertainties in risk assessment may result in economic burdens (i.e., from over regulation) or inadequate public safety (i.e., from under regulation), EPA has established the Research to Improve Health Risk Assessments (RIHRA) program to resolve some of the major health risk assessment issues. The paper presents an overview of the RIHRA program and several examples of health research applicable to drinking water issues.