Home | Business News | Browse by Publication | E | Environmental Health Perspectives

Comparative risk assessment of the burden of disease from climate change.

Article Excerpt
The World Health Organization has developed standardized comparative risk assessment methods for estimating aggregate disease burdens attributable to different risk factors. These have been applied to existing and new models for a range of climate-sensitive diseases in order to estimate the effect of global climate change on current disease burdens and likely proportional changes in the future. The comparative risk assessment approach has been used to assess the health consequences of climate change worldwide, to inform decisions on mitigating greenhouse gas emissions, and in a regional assessment of the Oceania region in the Pacific Ocean to provide more location-specific information relevant to local mitigation and adaptation decisions. The approach places climate change within the same criteria for epidemiologic assessment as other health risks and accounts for the size of the burden of climate-sensitive diseases rather than just proportional change, which highlights the importance of small proportional changes in diseases such as diarrhea and malnutrition that cause a large burden. These exercises help clarify important knowledge gaps such as a relatively poor understanding of the role of nonclimatic factors (socioeconomic and other) that may modify future climatic influences and a lack of empiric evidence and methods for quantifying more complex climate-health relationships, which consequently are often excluded from consideration. These exercises highlight the need for risk assessment frameworks that make the best use of traditional epidemiologic methods and that also fully consider the specific characteristics of climate change. These include the long-term and uncertain nature of the exposure and the effects on multiple physical and biotic systems that have the potential for diverse and widespread effects, including high-impact events. Key words: burden of disease, climate change, national, quantitative comparative risk assessment, regional. Environ Health Perspect 114:1935-1941 (2006). doi:10.1289/ehp.8432 available via http://dx.doi.org/ [Online 11 July 2006]

**********

The process of climate change, including both increases in global average temperatures ("global warming") and changes in other climate characteristics such as the spatial and temporal distribution of precipitation, has important implications for human health. It is important to describe, measure, and predict the health effects of climate change for two reasons. First, this provides a fuller picture of the consequences of mitigating, or failing to mitigate, emissions of greenhouse gases that are the main anthropogenic contribution to climate change. The long persistence of these gases in the atmosphere means that current mitigation activities (or lack of them) will have consequences for all natural and human systems over coming decades and centuries. They should ideally be informed by measures of the overall size and global distribution of likely health effects of climate change throughout suitably long periods to be considered alongside other impacts such as on biodiversity (Parmesan and Yohe 2003; Thomas et al. 2004a). Even imperfect estimates of the full range of global impacts can provide useful information, provided they are accompanied by clear descriptions of the associated assumptions and uncertainties. Second, quantitative studies can help inform policies to adapt to climate changes that are now inevitable because of both natural variability and past greenhouse gas emissions. Such actions typically affect the national or subnational level and require information on the likelihood and expected magnitude of specific health impacts in the local context, allowing for the more appropriate allocation of resources to prevent harm from effects such as extreme weather-related events and changes in disease distributions.

Recent comparisons of natural and anthropogenic influences on regional climate (Stott et al. 2004) have demonstrated that human activity increased the probability of a specific past climate event, with severe health consequences (> 44,000 deaths in the European heat wave of summer 2003) (Kosatsky 2005). However, estimating the full range of effects of climate change on health over long time scales presents additional challenges to epidemiologic methods. These include the absence of an appropriate comparison group, the long period over which human actions affect the climate, the large number of health outcomes potentially affected by climatic change, and the numerous nonclimatic influences on each of these outcomes. For these reasons it is misleading simply to observe long-term trends in climate-related diseases and to attribute these changes directly to anthropogenic climate change (e.g., Kovats et al. 2001; McMichael and Githeko 2001; Reiter 2001). The most plausible estimates of future climate change impacts are instead based on empirically observed relationships between weather or climate conditions and health effects, either in space and/or in time, or, for infectious diseases, on models that capture a detailed understanding of the effects of climate on the biologic processes that determine disease transmission (Rogers and Randolph 2000; Small et al. 2003). Projections of global climate models can be linked to these relationships to indicate how future climate change may influence the level of health outcomes--such as changes in the population living in areas with climates suitable for the transmission of malaria parasites or dengue virus (e.g., Hales et al. 2002; Martens 1998; Rogers and Randolph 2000) or the numbers of people exposed to coastal flooding (Nicholls et al. 1999). These models already provide useful quantitative measures of future risk. However, the results of these models are difficult to relate directly to inform decisions on mitigation (e.g., greenhouse gas emission reduction strategies) because a) many do not attempt to account for changes in nonclimatic influences such as economic development (and hence the ability to protect against disease risk), and b) the model outcomes are often indirectly related to health, and then only to specific diseases. It is therefore difficult to judge the overall magnitude of the likely health impacts of climate change, either globally or in a specific country (e.g., the combined health effect of a projected 10% increase of the population exposed to coastal flooding, a 20% increase in population living in areas suitable for dengue transmission, and a 5% drop in deaths in cold extremes) and compare these with other threats.

These concerns can be addressed partly by using a standard framework for comparison across risk factors and diseases. The World Health Organization (WHO) has recently developed an approach for comparative risk assessment that has been applied to estimate the current and future disease burden from 25 diverse risk factors, including climate change, in a comparable and transparent manner (Ezzati et al. 2002, 2004; Murray and Lopez 1997; WHO 2002). The assessment generated estimates of the numbers of deaths and disability-adjusted life years (DALYs) attributable to each risk factor in the year 2000, along with expected changes in exposures and associated relative risks of disease outcomes, for several time points between 2000 and 2030. A similar approach has been applied to estimate the effects of climate change on health within the Oceania region in the Pacific Ocean for 2020 and 2050. This assessment principally focused on...