Since the SAR, much of the additional research on health impacts has examined natural climate variability in relation to interannual variations in infectious diseases—particularly vector-borne diseases—and the relationship between daily weather and mortality in various urban populations. Predictive modeling of the impact of climate scenarios on vector-borne disease transmissibility has undergone further development. Meanwhile, however, data sets that allow study of the effects of the health impacts of observed longer term trends in climate remain sparse.

The research task of assessing the actual and potential health impacts of climate change has several distinctive characteristics and poses four major challenges to scientists:

1. Anticipated anthropogenic climate change will be a gradual and long-term process. This projected change in mean climate conditions is likely to be accompanied by regional changes in the frequency of extreme events. Changes in particular health outcomes already may be occurring or soon may begin to occur, in response to recent and ongoing changes in world climate. Identification of such health effects will require carefully planned epidemiological studies.
2. In epidemiological studies (in which associations are observed with or without knowledge of likely causal mechanisms), there often are difficulties in estimating the role of climate per se as a cause of change in health status. Changes in climate typically are accompanied by various other environmental changes. Because most diseases have multiple contributory causes, it often is difficult to attribute causation between climatic factors and other coexistent factors. For example, in a particular place, clearing of forest for agriculture and extension of irrigation may coincide with a rise in regional temperature. Because all three factors could affect mosquito abundance, it is difficult to apportion between them the causation of any observed subsequent increase in mosquito-borne infection. This difficulty is well recognized by epidemiologists as the "confounding" of effects.
3. It is equally important to recognize that certain factors can modify the vulnerability of a particular population to the health impacts of climate change or variability. This type of effect-modification (or "interaction") can be induced by endogenous characteristics of the population (such as nutritional or immune status) or contextual circumstances that influence the "sensitivity" of the population's response to the climate change (such as unplanned urbanization, crowding, or access to air conditioning during heat waves). Deliberate social, technological, or behavioral adaptations to reduce the health impacts of climate change are an important category of effect-modifying factor.
4. Simulation of scenario-based health risks with predictive models entails three challenges. These challenges relate to validity, uncertainty, and contextual realism:
   • Valid representation of the main environmental and biological relationships and the interacting ecological and social processes that influence the impact of those relationships on health is difficult. A balance must be attained between complexity and simplicity.
   • There are various sources of (largely unavoidable) uncertainty. There is uncertainty attached to the input scenarios of climate change (and of associated social, demographic, and economic trends). Subsequently, there are three main types of uncertainties in the modeling process itself: "normal" statistical variation (reflecting stochastic processes of the real world); uncertainty about the correct or appropriate values of key parameters in the model; and incomplete knowledge about the structural relationships represented in the model.
   • Climate change is not the sole global environmental change that affects human health. Various large-scale environmental changes now impinge on human population health simultaneously, and often interactively (Watson et al., 1998). An obvious example is vector-borne infectious diseases, which are affected by climatic conditions, population movement, forest clearance and land-use patterns, freshwater surface configurations, human population density, and the population density of insectivorous predators (Gubler, 1998b). In accordance with point 2 above, each change in health outcome must be appropriately apportioned between climate and other influences.
     

IPCC Homepage