The challenges in identifying, quantifying and predicting the health impacts of climate change entail issues of scale, “exposure” specification, and the elaboration of often complex and indirect causal pathways. First, the geographic scale of climate-related health impacts and the typically wide timespans are unfamiliar to most researchers. Epidemiologists usually study problems that are geographically localised, have relatively rapid onset, and directly affect health. The individual is usually the natural unit of observation.
Second, the “exposure” variable – comprising weather, climate variability and climate trends – poses difficulties. There is no obvious "unexposed" group to act as baseline for comparison. Indeed, because there is little difference in weather/climate exposures between individuals in the same geographic locale, comparing sets of persons with different “exposures” is usually precluded. Rather, whole communities or populations must be compared – and, in so doing, attention must be paid to intercommunity differences in vulnerability. For example, the excess death rate during the severe 1995 Chicago heatwave varied greatly between neighbourhoods because of differences in factors such as housing quality and community cohesion.
Third, some health impacts occur via indirect and complex pathways. For example, the effects of temperature extremes on health are direct. In contrast, complex changes in ecosystem composition and functioning help mediate the impact of climatic change on transmission of vector-borne infectious diseases and on agricultural productivity.
A final challenge is the need to estimate health risks in relation to future climatic-environmental scenarios. Unlike most recognized environmental health hazards, much of the anticipated risk from global climate change lies years to decades into the future.