Death attributable to heat exposure is a well-documented clinical entity, often occurring in younger people engaged in vigorous physical activity or in older people during heat waves. In the United States, only a limited number of deaths, typically in the hundreds, are directly attributed to heat annually. The effect of heat on mortality has also been assessed using time-series methods that characterize the relationship between daily mortality counts on the same or recent days. Analyses of mortality data for the United States and other developed countries show that mortality counts are positively associated with temperatures above the moderate range, with the strength of association becoming greater at the upper end of the range. When examined across cities, this relationship tends to be stronger in more northern locations and temperature is predictive of mortality, even at ambient temperatures below heat wave levels. Are the clinical and epidemiological patterns parallel? Perhaps, but we lack needed information on actual temperature exposures of susceptible individuals in relevant microenvironments; that is, the relationship between ambient temperatures, measured at central stations, and temperatures indoors need to be characterized. We have recently described data collection approaches that can be used for this purpose. This presentation will synthesize the clinical and epidemiological evidence, using an exposure-based framework.