Climate Change 2001: Working Group II: Impacts, Adaptation and Vulnerability

	Other reports in this collection

9.3. Sensitivity, Vulnerability, and Adaptation

There are uncertainties regarding the sensitivity (i.e., rate of change of the outcome variable per unit change in the input/ exposure variable) of many health outcomes to climate or climate-induced environmental changes. Relatively little quantitative research, with estimation of exposure-response relationships, has been done for outcomes other than death rates associated with thermal stress and changes in the transmission potential of several vector-borne infectious diseases. There has been increased effort to map the current distribution of vectors and diseases such as malaria by using climate and other environmental data (including satellite data).

Continuation of recent climatic trends soon may result in some shifts in the geographic range and seasonality of diseases such as malaria and dengue. In reality, however, such shifts also would depend on local topographical and ecological circumstances, other determinants of local population vulnerability, and the existence and level of adaptive public health defenses. There has been some recent debate in the scientific literature about whether there is any evidence of such shifts yet (Epstein et al., 1997; Mouchet et al., 1998; Reiter, 1998a,b). It is not yet clear what criteria are most appropriate for assessment of climatic influences on such changes in infectious disease patterns. A balance is needed between formal, statistically based analysis of changes within a particular local setting and a more synthesizing assessment of the consistency of patterns across diverse settings and across different systems—physical, biotic, social, and public health. As with climate change itself, there is an inherent difficulty in detecting small climate-induced shifts in population health outcomes and in attributing the shift to a change in climate.

Population vulnerability is a function of the extent to which a health outcome in that particular environmental-demographic setting is sensitive to climate change and the capacity of the population to adapt to new climate conditions. Determinants of population vulnerability to climate-related threats to health include level of material resources, effectiveness of governance and civil institutions, quality of public health infrastructure, access to relevant local information on extreme weather threats, and preexisting burden of disease (Woodward et al., 1998). Thus, vulnerability is determined by individual, community, and geographical factors:

• Individual factors include:
  • Disease status (people with preexisting cardiovascular disease, for example, may be more vulnerable to direct effects such as heat waves).
  • Socioeconomic factors (in general, the poor are more vulnerable).
  • Demographic factors (the elderly are more vulnerable to heat waves, for example, and infants are more vulnerable to diarrheal diseases).
• Community factors may include:
  • Integrity of water and sanitation systems and their capacity to resist extreme events.
  • Local food supplies and distribution systems.
  • Access to information, including early warnings of extreme climate events.
  • Local disease vector distribution and control programs.
• Geographical factors may include:
  • The influence of El Niño cycle or the occurrence of extreme weather events that are more common in some parts of the world.
  • Low-lying coastal populations more vulnerable to the effects of sea-level rise.
  • Populations bordering current distributions of vector-borne disease particularly vulnerable to changes in distribution.
  • Rural residents often with less access to adequate health care, and urban residents more vulnerable to air pollution and heat island effects.
  • Environmentally degraded and deforested areas more vulnerable to extreme weather events.

Understanding a population's capacity to adapt to new climate conditions is crucial to realistic assessment of the potential health impacts of climate change (Smithers and Smit, 1997). This issue is addressed more fully in Section 9.11.