NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Forum on Microbial Threats. Global Climate Change and Extreme Weather Events: Understanding the Contributions to Infectious Disease Emergence: Workshop Summary. Washington (DC): National Academies Press (US); 2008.

Cover of Global Climate Change and Extreme Weather Events

Global Climate Change and Extreme Weather Events: Understanding the Contributions to Infectious Disease Emergence: Workshop Summary.

Show details

1Climate Change Challenges


The contributions that comprise this chapter establish the context of workshop discussions and depict the “big picture” within which the papers collected in subsequent chapters are set. The three presenters represented herein—Sir Andrew Haines of the London School of Hygiene and Tropical Medicine; Paul Epstein of Harvard Medical School; and keynote speaker Donald Burke of the University of Pittsburgh—offer varied, and occasionally contrasting, perspectives on what is known, suspected, or unknown regarding the consequences of global climate change for health and, more specifically, for infectious disease emergence. At the same time, each of these contributors observes the accelerating pace of ecological upheaval and emphasizes the inherent complexity of biological responses to climate change and extreme weather events, which frequently involve nonlinear “tipping points.” These characteristics inspire both uncertainty and urgency in the quest to better understand, anticipate, and respond to the potentially wide-ranging health effects of climate change.

In the chapter’s first paper, Haines reviews the Intergovernmental Panel on Climate Change’s (IPCC’s) most recent findings on global climate change to date and the panel’s predictive scenarios for the future. He then describes several approaches that have been taken to identify and model the potential health impacts of climate change, along with the methodological challenges presented by such studies. Several examples illustrate how variations in climate—in the form of floods, droughts, and other extreme weather events—influence the range and transmission dynamics of infectious diseases, and therefore suggest potential effects of climate change. Yet, as Haines observes, “in the case of infectious diseases there is still considerable controversy about the degree to which climate change has been responsible for changes in the incidence and distribution of disease. This is due to the potential contribution of other factors, such as changing land-use patterns, human behavior, and methodological issues including the use and analysis of appropriate climate data.”

Haines also reviews several efforts to date to estimate the future impact of climate change on infectious diseases, which—although individually problematic—support his overall conclusion that “it is likely that the disease burden as a result of climate change will [increase] substantially over time and will be particularly concentrated in the poorer populations.” Thus, his proposed strategies to address the negative health effects of climate change focus on the poor: first, by improving their access to basic public health services (clean water, sanitation, immunization); second, by providing them with cleaner fuels, which offer both immediate health benefits and long-term protection for the atmosphere. “Infectious diseases are one of a number of categories of health outcomes that are likely to be affected adversely by climate change,” Haines concludes. “Public health policies should take into account the need to adapt to a changing climate, as well as the potential for near-term benefits to health from a range of policies to mitigate climate change.”

Haines’s paper is followed by two reprinted pieces, authored (in the first case) and coedited by Epstein (in the second), that illustrate the breadth of biological responses to climate change. The first manuscript is an essay, originally published in the New England Journal of Medicine in October 2005—weeks after Hurricane Katrina devastated the Gulf Coast—that focuses on the wide-ranging health effects of extreme weather. The second, an excerpt from the report Climate Change Futures: Health, Ecological and Economic Dimensions (Center for Health and the Global Environment, 2005), introduces a “multidimensional assessment” incorporating trend analysis, case studies, and scenarios that focus on health, ecological, and economic impacts of climate change. This project, undertaken in collaboration by the Center for Health and the Global Environment at Harvard Medical School, the United Nations Development Programme, and Swiss Re, a global reinsurance company, was designed to assess threats posed by climate change to the institution of insurance, a “time-tested method for adapting to change.” Such threats go far beyond immediate property damage, and indeed even health consequences, to the social and political stability of regions affected by climate disasters (see also Chapter 4).

In his workshop presentation, Epstein emphasized the methodologies that make such threat assessments possible. He identified three phenomena that underlie climate- and weather-related changes in disease distribution:

  1. Since 1950, nighttime and winter warming have occurred twice as fast as overall global warming.
  2. The pace of warming in temperate, boreal, and polar latitudes is occurring faster than warming in the tropics.
  3. Since the first International Geophysical Year in 1957, when many global measurements were initiated, the world’s oceans have accumulated 22 times the amount of heat that the atmosphere has, accelerating the global hydrological cycle.

There are no appropriate, independent controls for the study of global climate change on Earth, and the experiment we are conducting (with an n of 1) cannot be repeated, Epstein observed. Therefore, he explained, a wide range of methodologies must be harnessed to assess changes in biological variables, such as the geographic range and incidence of diseases in relation to changes in temperature and precipitation. Monitoring and mapping produce data that can be integrated into geographic information systems (GISs) to identify and compare physical and biological phenomena. Further, GISs overlay multiple sets of data, providing input for descriptive and mathematical models that project the biological impacts of various climate change scenarios. Models are used for understanding dynamics, for predicting outcomes, and for decision making (Hilborn and Mangel, 1997).

In his presentation, Epstein also described methods for analyzing data gathered across scientific disciplines (e.g., diseases affecting a range of taxonomic groups) in order to reveal patterns and emerging trends associated with climate change, calculate rates of change (i.e., in the geographic range, prevalence, and incidence of infectious diseases), and compare observations with predicted outcomes. Researchers also conduct experiments, called “fingerprint” studies, to compare data with model projections (such as those undertaken by the IPCC to analyze climate models and energy fluxes driven by increases in heat-trapping, greenhouse gases).

Many of the methodologies used to study the effects of climate change yield correlations, rather than proof of causation, Epstein acknowledged. However, he added, such associations and their plausible mechanisms can be then tested via qualitative, schematic, and quantitative models. Bayesian methods of assessing causality based on prior probabilities and prior knowledge via first physical principles can also be used to analyze the effects of global climate change, he said.

Moreover, Epstein asserted, when observational data from multiple sources match model projections (i.e., the findings are internally consistent) and can be explained by plausible biological mechanisms (e.g., changes in observed altitude ranges in tandem with observed temperature changes), the composite pattern warrants further attention. This may take the form of analyzing attribution probabilities for anomalous events; for example, such an assessment indicates that global warming increased the likelihood of the European heat wave of 2003 two- to fourfold (Stott et al., 2004). Extreme conditions that favor infectious disease outbreaks via multiple pathways may be revealed by “cluster analyses” or characterized through the use of “principal component analyses,” which identify spatial and temporal associations among variables.

Despite the existence of such methodologies, “our current understanding of the relationships between climate and weather, and epidemic infectious diseases, is insufficient to make credible predictions about future threats posed by infectious diseases under various global change scenarios,” Burke argues in the chapter’s final paper. To support this contention, he presents detailed analyses of the transmission dynamics of two infectious diseases: influenza and dengue fever. These studies reveal that oscillations in disease incidence occur in the absence of seasonal transmission effects; if these patterns coincide with seasonal variation, small changes in transmissibility may, under some circumstances, produce considerable variability from year to year in epidemic disease occurrence (Dushoff et al., 2004).

Burke notes approvingly that conclusions by “respected scientific bodies” regarding the probable impact of global climate change on epidemic infectious diseases remain measured since the publication of the landmark report Under the Weather: Climate, Ecosystems, and Infectious Diseases (NRC, 2001) by the interdisciplinary committee that he chaired. “This caution honestly reflects the uncertainties involved, which in turn reflect the difficulty of the underlying scientific problems,” he states. Calling readers’ attention to the recommendations for future research and surveillance made in that report (see also Box SA-3), Burke concludes that “it is safe to say that [these recommendations] continue to be relevant.”


Andy Haines, M.B.B.S., M.D.

London School of Hygiene and Tropical Medicine

Greenhouse gases are now accumulating in the atmosphere at unprecedented rates. The annual growth rate of carbon dioxide (CO2) concentration was highest over the last 10 years since the beginning of continuous direct atmospheric measurements (IPCC, 2007a). The atmospheric concentration of CO2 now greatly exceeds the natural range over the last 650,000 years. CO2 is the most important greenhouse gas produced by humankind and accounts for around 77 percent of the total. The concentrations of all three major greenhouse gases: CO2, methane, and nitrous oxide, are at the highest levels for at least 10,000 years and have resulted in clear changes in the Earth’s climate. Of the last 12 years (1995–2006), 11 are among the 12 warmest years since 1850 when instrumental records began. Over past 100 years (1906–2005), global surface temperature has increased by 0.74°C (90 percent, uncertainty interval 0.56–0.92) with warming faster over land than over oceans. Of course climate will still vary (e.g., in February 2008, NOAA [National Oceanic and Atmospheric Administration] predicted La Niña conditions—the cold phase of the El Niño Southern Oscillation [ENSO]—to continue throughout spring 2008; however, by May La Niña began to transition to more ENSO-neutral conditions [NOAA Climate Prediction Center, 2008]). But after accounting for these known climate fluctuations, the long-term warming trend will continue.

Figure 1-1 shows the trends for temperature, global average sea level, and observed decreases in snow and ice. Sea level rise is due to a combination of thermal expansion of the oceans together with increased melting of glaciers and polar ice sheets. Other significant changes in climate include declines in precipitation from 1900 to 2005 in the Sahel, Mediterranean, southern Africa, and parts of southern Asia. More intense and longer droughts have occurred over increasing areas since the 1970s, particularly in the tropics and subtropics. The IPCC concludes that increased drying as a result of temperature increases and reductions in precipitation has contributed to changes in drought.

FIGURE 1-1. Observed changes in (A) global average surface temperature; (B) global average sea level rise from tide gauge (blue) and satellite (red) data; and (C) Northern Hemisphere snow cover for March-April.


Observed changes in (A) global average surface temperature; (B) global average sea level rise from tide gauge (blue) and satellite (red) data; and (C) Northern Hemisphere snow cover for March-April. All changes are relative to corresponding averages for (more...)

In contrast, precipitation has increased significantly in eastern areas of North and South America, as well as northern Europe and northern and central Asia. There is an apparent increase in intense tropical cyclone activity since 1970 in the North Atlantic.

Over the next two decades, warming of about 0.2°C per decade has been projected for a range of emission scenarios according to the IPCC. After that, different scenarios of socioeconomic development and the use of mitigation strategies result in markedly different trajectories for greenhouse gas emissions (see Figure 1-2). The different scenarios result in best estimates for temperature change in 2090–2099 relative to 1980–1999, ranging between 1.8°C for B1 scenario and 4°C for A1F1 scenario, but the likely range of estimates is even wider, extending to 6.4°C for the latter scenario. Regional-scale changes are outlined in Box 1-1.

FIGURE 1-2. Multimodel averages and assessed ranges for surface warming (compared to the 1980–1999 base period) for the SRES scenarios A2 (red), A1B (green), and B1 (blue), shown as continuations of the twentieth-century simulation.


Multimodel averages and assessed ranges for surface warming (compared to the 1980–1999 base period) for the SRES scenarios A2 (red), A1B (green), and B1 (blue), shown as continuations of the twentieth-century simulation. The latter two scenarios (more...)

Box Icon

BOX 1-1

Regional-Scale Changes. Changes include the following: Warming greatest over land and at most high northern latitudes and least over the Southern Ocean and parts of the North Atlantic Ocean, continuing recent observed trends

Climate Variability, Climate Change, and Health

It has been known for thousands of years, at least since the time of Hippocrates, that climatic variations can influence health, particularly through changes in temperature and precipitation, as well as extreme weather events. Growing scientific consensus about the existence of global climate change has rekindled interest in linkages between climate and health. The potential range of impacts is wide and they have been reviewed extensively (IPCC, 2007b).

There are a number of approaches to studying the potential health impacts of climate change. These include studies of the associations between past climate variability and disease; of the associations between trends in climatic variables over recent decades and the epidemiology of diseases; and of the response of vector species to changes in temperature and rainfall. In addition, there have been a number of approaches to modeling the potential future impacts of climate change on health.

The study of potential associations between climate change and health poses a number of methodological challenges including the need to consider confounding factors as possible explanations of apparent associations between climatic variables and health outcomes. Such confounding factors may include changes in resistance to insecticides (in the case of vector-borne diseases); changes in resistance to commonly used drugs for treatment (e.g., in the case of malaria); migration of populations, which may result in the exposure of nonimmune populations to infectious diseases; and changes in the performance of disease surveillance systems over time. Some diseases, such as malaria, exhibit differences in local transmission dynamics that complicate attempts to model the likely effect of climate change. Improvements in public health infrastructure leading to improved adaptation to climate change could, in the future, attenuate relationships between the changing climate and health outcomes. Climate change is likely to be a long-term process that will evolve over decades and centuries while our understanding of the linkages between climate and health is based largely on studies of short-term variability.

There are likely to be interactions between climate change and other environmental changes, such as deforestation, growth in global travel, increased local population mobility, and depletion of aquifers in some regions. For example, deforestation may change the distribution of disease vectors as well as contributing to climate change, and migration of populations into formerly forested areas may result in increased exposure to a number of diseases. A study in the Peruvian Amazon suggested that the abundance of the malaria vector Anopheles darlingi was two hundredfold higher in deforested areas than in pristine forest and this could not be attributed solely to increased population density (Vittor et al., 2006). Deforestation may increase malaria risk in the Americas and Africa, but reduce it in Southeast Asia (Guerra et al., 2006).

This paper focuses on potential relationships between infectious disease and climate change, but there are a range of other health outcomes such as the effects of heat waves and potential reductions in cold-related deaths, particularly in temperate countries; the direct effect of wind, storms, and floods causing deaths and injuries; and the effect of droughts on malnutrition and food security. Floods and droughts, together with less extreme changes in precipitation, can also have important implications for a range of water-related diseases, as well as impacting health through the effects of increases in malnutrition and consequent reductions in immunity to disease. A summary of the pathways through which climate change may affect human health including infectious diseases is shown in Figure 1-3.

FIGURE 1-3. Pathways by which climate change may affect human health, including infectious diseases.


Pathways by which climate change may affect human health, including infectious diseases. SOURCE: Reprinted from Haines and Patz (2004) with permission from the American Medical Association. Copyright 2004. All rights reserved; adapted from Patz et al. (more...)

Rainfall, Temperature, and Disease

Water-related diseases include water-borne diseases due to ingestion of pathogens in contaminated water and water-washed diseases as a result of poor hygiene. A recent global overview of cross-sectional studies based on 36 published reports from low- and middle-income countries between 1954 and 2000 suggests an average of 4 percent increase in diarrhea incidence in children less than 5 years of age per 10 mm decrease in rainfall per month (Lloyd et al., 2007). Currently, more than 2 billion people live in relatively dry regions of the world and are likely to suffer disproportionately from lack of access to clean water. While there have been substantial improvements in the management of diarrheal disease, particularly because of the widespread use of oral rehydration therapy, child mortality remains unacceptably high, particularly in sub-Saharan Africa. Diarrheal diseases claim almost 2 million lives a year in children under 5. However, it is important to recognize that most freshwater is used for irrigation rather than personal consumption; therefore, the relationship between reduced freshwater availability and diarrheal diseases may be indirect. Hand washing with soap has a protective effect against diarrheal disease (Curtis and Cairncross, 2003), and reduced availability and/or increased costs of freshwater may lead to reduced hand washing where rainfall is low.

Heavy precipitation may be associated with outbreaks of water-borne diseases, such as cryptosporidiosis (Curriero et al., 2001). Although the global overview referred to above did not find a relationship with temperature, other studies have found associations between higher temperatures and increased episodes of diarrheal diseases in Peru, the Pacific Islands, and Australia (Checkley et al., 2004; McMichael et al., 2003; Singh et al., 2001). The association between sea surface temperatures and cholera transmission has been most convincingly shown in the Bay of Bengal (Colwell, 1996).

Time-series analysis of weekly cases of salmonellosis using data from 16 sites in industrialized countries suggested an approximately linear increase in reported cases above the threshold of about 6°C. The study focused on the analysis of sporadic cases only. The association appeared to be particularly evident in the case of Salmonella enteritidis, with a lag of around 1 week between the increased temperature and the increase in cases, suggesting that the effect may be on the replication of salmonella after food has been prepared (Kovats et al., 2004).

One of the best documented examples of the way in which climate can drive the onset of disease is the relationship between the dry northern winds (called the Harmattan) and meningococcal meningitis epidemics in West Africa. The mean weeks of the winter maximum wind speed and of the onset of the epidemic are identical and usually occur between February 7 and 15 (Sultan et al., 2005). Although the causal mechanism is not fully understood, the disease may result from the effects of mucosal drying and abrasion as a result of the strong dust-laden winds.

There is some evidence that the geographical distribution of meningococcal meningitis in West Africa has expanded in the recent past, possibly as a result of changes in land use and climate (Molesworth et al., 2003).

The El Niño/Southern Oscillation and Health

El Niño events are large-scale ocean-atmospheric climate phenomena emanating from the central and east-central equatorial Pacific Ocean that have occurred for thousands of years. A major feature is the upwelling of warm water off the coast of Peru and Ecuador. Associated by distant connections (teleconnections) with climatic changes in Australia, Indonesia, the Pacific highlands, and East Africa, as well as parts of Latin and North America, the El Niño cycle is usually between 3 and 7 years. The El Niño (warm event) is frequently followed by a La Niña (opposite, cold) event. The Southern Oscillation is the name given to the seesaw of air pressure differences between the east and west Pacific, which is associated with the El Niño phenomenon and leads to the full name El Niño/Southern Oscillation (ENSO).

Figure 1-4 demonstrates some of the teleconnections associated with El Niño and indicates where health outcomes, such as increased risk of epidemic malaria, are experienced after the onset of an El Niño event (Kovats et al., 2003). ENSO is the most important climatic cycle that contributes to natural disasters. Drought is twice as frequent worldwide in the year after the onset of the El Niño than in other years, particularly in southern Africa and South Asia. In an average El Niño year, around 35 people per 1,000 worldwide are affected by a natural disaster, more that four times that of non-El Niño years based on analysis of data for three decades between 1963 and 1993 (Bouma et al., 1997).

FIGURE 1-4. ENSO teleconnections and risk map for malaria.


ENSO teleconnections and risk map for malaria. SOURCE: Reprinted from Kovats et al. (2003) with permission from Elsevier.

The relationship between El Niño and intense rainfall is also strong in many areas, although unlike drought it is not seen on a global scale because flood-related disasters are relatively localized and the risk is increased during both El Niño and La Niña phases in different parts of the world. During an El Niño event, storm activity in parts of the Pacific is increased and decreased in the Atlantic so that hurricanes in the Caribbean and the Gulf of Mexico tend to be less common during an El Niño but more common during a La Niña, while typhoons are more likely to occur near the Marshall Islands in the Pacific Ocean during an El Niño event compared to other times because of the western shift in storm tracks (Spennemann and Marschner, 1995).

The study of the relationship between ENSO events and health outcomes provides useful evidence for causal relationships between climatic variability and human health, which can help to improve understanding of the potential impacts of climate change. It is not known precisely how climate change will affect the El Niño phenomenon, but such events will occur in the future against a background of climate change, which is likely to increase the health and social impacts.

For a causal relationship between the ENSO cycle and a health outcome to be inferred, it is necessary to have appropriate climatic data for the region to be studied. There also needs to be a plausible biological relationship between the health outcome in question and climate variability, and appropriate statistical analysis is required, taking into account autocorrelation and using, where available, long time series of data over decadal periods. There have also been numerous case reports of associations of disease outbreaks with single ENSO events, although these provide less compelling evidence of association.

The analysis of linkages between health outcomes and El Niño events is complicated by the fact that they are not identical but vary in intensity and regional impacts. For example, in 1997 and 1998 the anticipated drought did not occur in the southern African region and some areas received above average rainfall.

Despite the limitations of available data, a number of robust associations have been described at various sites throughout the world, particularly with epidemic malaria in countries such as Colombia, Venezuela, and Guyana (Kovats et al., 2003). In highland areas, such as northern Pakistan, higher temperatures associated with El Niño may increase the transmission of malaria. In many desert fringe regions, such as the Punjab and the Thar Desert, increased rainfall during the post-El Niño and La Niña years has historically increased epidemic risk. Conversely, El Niño-related droughts have been associated with malaria outbreaks in Sri Lanka, Colombia, and Venezuela, possibly as a result of reduced river flows, which in turn permitted increased breeding of mosquitoes.

The yearly incidence of dengue has also been associated with the El Niño year in Indonesia. Monthly cases of dengue on some but not all Pacific Islands have been associated with the Southern Oscillation index. In Australia, epidemic years for the Ross River virus disease over a 70-year period up to 1998 showed some association with the Southern Oscillation index, but two other studies did not show a relationship between yearly notifications of this disease and the index. Associations of health outcomes with climatic factors may be localized and may not be detectable with aggregated data.

The association between cholera and ENSO has been established in Dhaka, Bangladesh, and the relationship has become apparent in recent decades (Rodo et al., 2002).

The annual incidence of visceral leishmaniasis in Bahia State, northeastern Brazil, was significantly related to sea surface temperature in the Pacific Ocean (Franke et al., 2002) probably as a result of the association between the ENSO cycle and drought in northeast Brazil.

The El Niño event of 1993 resulted in increased rainfall in the southern United States, which was then followed by a drought that facilitated the emergence of hantavirus pulmonary syndrome. There was also a fivefold increase in reported cases associated with the 1997–1998 El Niño. Although there is a plausible mechanism for ways in which climatic factors can lead to substantial increases in the rodent population and thus hantavirus transmission, a consistent association with ENSO has not been established (Engelthaler et al., 1999). Likewise in the case of Rift Valley fever in East Africa, there is good evidence that epidemics in the dry grasslands are triggered by heavy rainfall events, but there is no association between this disease in Kenya and the ENSO index (Linthicum et al., 1999).

Extreme Climate Events and Infectious Diseases

Case reports suggest that extreme climatic events such as floods, droughts, and storms can be associated with outbreaks of infectious disease, but the association is not always seen and is likely to depend on socioeconomic factors, damage to infrastructure, and the extent of population displacement.

There are a number of mechanisms by which the risk of flooding is likely to increase as a result of climate change. These include the melting of glaciers, increased frequency of episodes of heavy precipitation, and sea level rise. However, an extensive review of the effects of floods in Europe (Hajat et al., 2003) showed that infectious disease outbreaks were rarely a major public health problem. A more consistent finding seems to be increased prevalence of common mental disorders (i.e., anxiety, depression) following exposure to floods. This is likely to be due to loss of familiar possessions, forced evacuation, loss of livelihood, and increased poverty.

However, following floods, increases in diarrheal and respiratory diseases have been reported in both high- and low-income countries. Crowding of displaced populations may contribute to transmission. In low-income countries in particular, there may be outbreaks of leptospirosis, hepatitis, and vector-borne diseases, including malaria and Rift Valley fever (Ahern et al., 2005). On occasions, such as in Tanzania in 1997–1998, flooding had the opposite effect and washed away vector breeding sites.

Populations in low-lying areas are vulnerable to the effects of sea level rise, particularly in low-income countries where flood defenses may not be upgraded. For example, up to 57 percent of people in unprotected dry land areas in Bangladesh could be inundated if there was a 4°C temperature increase and, as a consequence, a 100 cm increase in sea level together with increases in monsoon precipitation and discharge into major rivers (BCAS/RA/Approtech, 1994). Deltaic regions are particularly vulnerable to the effects of sea level rise as illustrated in Figure 1-5. Salinization of freshwater aquifers may occur as a result of incursion of seawater. Even in industrialized countries, densely populated urban areas are at risk from sea level rise, as shown by the impact of Hurricane Katrina on New Orleans. After Hurricane Katrina there were increases in diarrheal disease incidence due to fecal contamination of drinking water (Manuel, 2006).

FIGURE 1-5. Relative vulnerability of coastal deltas as indicated by estimates of the population potentially displaced by current sea-level trends to 2050 (extreme >1 million; high 1 million to 50,000; medium 50,000 to 5,000).


Relative vulnerability of coastal deltas as indicated by estimates of the population potentially displaced by current sea-level trends to 2050 (extreme >1 million; high 1 million to 50,000; medium 50,000 to 5,000). Climate change would exacerbate (more...)

The potential health effects of drought in developing countries are likely to be wide ranging (see Figure 1-6). It is important to note, however, that domestic water consumption represents only 2 percent of the global total, and a flow sufficient to meet the domestic water requirements of around 1,000 people would be sufficient to irrigate only 1 hectare of land, which is capable of feeding a couple of families (Shiklomanov, 2000). It has also been pointed out that the weight of evidence from studies in low-income countries suggests that ready access to water that results in increased quantities used for hygiene is likely to be more important than water quality improvements by themselves in determining benefits for health (Cairncross, 2003). This is likely to be due to the fact that most endemic diarrheal disease is transmitted person-to-person by hands, food, and other fomites because of poor hygiene.

FIGURE 1-6. Potential health effects of drought in developing countries.


Potential health effects of drought in developing countries. SOURCE: Adapted from Kovats et al. (2003) with permission from Elsevier.

A study in six sub-Saharan African countries showed the potential for HIV/AIDS to amplify the effects of drought on childhood malnutrition, particularly in periurban and urban populations where HIV prevalence was high (Mason et al., 2005). Malnutrition associated with drought increases susceptibility to diseases such as measles, particularly in countries such as Somalia where immunization rates are low (Shepherd-Johnson, 2006).

Drought has variable effects on the incidence and distribution of vector-borne diseases such that, for example, reductions in mosquito activity during droughts may be followed by increases in disease transmission once the drought ends because of the increased number of susceptible hosts (Woodruff et al., 2002). In other cases, stagnation of water in residual pools may cause short-term increases in the transmission of malaria. Long-term drought may result in the contraction of areas suitable for malaria transmission.

Assessment of the likely effect of climate change on malnutrition is complex because the impacts on food production and consumption depend on a range of factors, including agricultural practices, the potential role of carbon dioxide fertilization in improving some crop yields (the effect of increasing concentrations of carbon dioxide in improving the yields of some crops may be reduced when crops are stressed as a result of high temperatures or changes in precipitation), patterns of land ownership, and the ability of disadvantaged populations to purchase food. Concern has also been expressed about the potential competition for land between biofuels and agricultural production for food, although this is not an inevitable consequence and will depend on the policy choices that are made (Haines et al., 2007). Vulnerability to increased malnutrition as a result of climate change is likely to be greatest in regions currently most vulnerable to food insecurity, particularly sub-Saharan Africa (FAO, 2005).

Both fatalities and direct economic losses of national per capita income from natural disasters are higher by orders of magnitude in low- and middle-income countries compared to high-income countries (Linnerooth-Bayer et al., 2005). For example, a study of the impact of Hurricane Mitch on the livelihood of rural poor in Honduras showed that one of every two households surveyed incurred medical, housing, or other costs due to the hurricane. Relief amounted to less than one-tenth of the losses incurred by households (Morris et al., 2002). Such economic losses accentuate poverty and contribute substantially to the adverse effects of climate-related disasters on public health. For example, an estimate of mortality due to floods in Mozambique in 2000 suggested that the increase in infant mortality associated with around a 14 percent drop in gross domestic product in the flooded provinces made a substantial contribution to the overall deaths due to flooding (Cairncross and Alvarinho, 2006).

Is There Evidence That Climate Change Has Begun to Affect Human Health?

There is good evidence that climate change has caused an earlier onset of the spring pollen season in the Northern Hemisphere, with resulting changes in the seasonality of allergic rhinitis (Emberlin et al., 2002). The summer of 2003 was probably the hottest in Europe since 1500, and climate change is thought to have at least doubled the risk of a heat wave such as that experienced in 2003 (Stott et al., 2004).

However, in the case of infectious diseases there is still considerable controversy about the degree to which climate change has been responsible for changes in the incidence and distribution of disease. This is due to the potential contribution of other factors, such as changing land use patterns, human behavior, and methodological issues, including the use and analysis of appropriate climate data. Northern shifts in tick distribution have been observed in Sweden (Lindgren et al., 2000), and the incidence of tick-borne encephalitis (TBE) in Sweden has increased considerably since the mid-1980s. A study of cases of TBE in Stockholm County between 1960 and 1998 showed that increases in disease incidence were significantly related to a combination of two consecutive mild winters, as well as temperatures favoring spring development and extended autumn activity in the prior year, and temperatures allowing activity early in the incidence year (Lindgren and Gustafson, 2001). This suggests that milder winters and the early arrival of spring may have contributed to the increased incidence of TBE, but other factors may be implicated, such as changes in land use and land cover leading to increases in the wildlife hosts of ticks, together with the presence of more people in endemic locations (Randolph, 2001).

There has been considerable interest in the possible role of climate change as a factor in the increases in malaria incidence in the East African highlands over recent decades. Using an analysis of data for four sites, Hay and colleagues (2002) asserted that there were no significant trends in climate variables and therefore concluded that climate change had played no role in malaria resurgence in the region. However, the use of low-resolution data to investigate the relationship was criticized, and it was suggested that the apparent lack of an association could not be interpreted as convincing evidence that climate change had not played a role (Patz, 2002). Subsequently, an updated time series (Pascual et al., 2006) using an additional 5 years of data demonstrated that, using both nonparametric and parametric statistical analyses, there was evidence of significant warming trends of around 0.5°C at all sites. It was suggested that the “observed temperature changes would be significantly amplified by the mosquito population dynamics with a difference in the biological response at least 1 order of magnitude larger than that in the environmental variable” (Pascual et al., 2006).

Other authors have shown that climatic factors play a role in malaria epidemics in the East African highlands. The study of climate variability, seasonality, and the number of monthly malaria outpatients over 10- to 15-year periods in seven highland sites in East Africa showed substantial spatial variation in the sensitivity of malaria outpatient numbers to climate variability, with between 12 and 63 percent of variance attributed to climate variability (Zhou et al., 2004). The study of malaria epidemic risk in Ethiopia showed that epidemics were significantly more often preceded by a month of abnormally high minimum temperature during the previous 3 months than would be expected by chance (Abeku et al., 2003).

The recent United Nations Development Programme (UNDP) Human Development Report (UNDP, 2007) has documented the growing burden of climate disasters, which is greater than can be explained by population growth alone. Weather-related insurance losses are increasing faster than the population, inflation, and coverage, but the greatest impacts are in developing countries where the majority lack insurance. Between 2000 and 2004, more than 250 million people per year were affected by hydrometeorological disaster (with increases in floods, droughts, lightning strikes, and the intensity of tropical cyclones), compared to less than 50 million per year between 1975 and 1979. A recent report has documented the contribution of increased sea surface temperature to increased Atlantic hurricane activity in recent years (Saunders and Lea, 2008).

Estimating the Impact of Climate Change on Infectious Diseases

A growing number of studies that have modeled projected impacts of climate change on health have been reviewed by the IPCC (IPCC, 2007b). A study of the potential effect of climate change on malaria transmission in Africa that assessed the impacts of 3 climate scenarios suggests a modest (5 to 7 percent) increase in the population at risk, largely due to expansion into higher altitudes. It also suggested a prolongation of the transmission season in some areas, leading to a 16 to 28 percent increase in the total number of person-months exposure (Tanser et al., 2003). This analysis, although based on a very large database of historical malaria surveillance data, has been criticized for oversimplifying the situation (1) by underestimating the variability in response of local vector species to climatic change and (2) because extension of the transmission season does not necessarily translate into a proportional increase in mortality or clinical disease (Reiter et al., 2004). Inevitably, however, all models involve some simplification of assumptions, and for example, few studies take adaptive capacity into account.

Exercises to estimate the global burden of disease due to climate change have been undertaken under the auspices of the World Health Organization (WHO). Epidemiological models were used to estimate the impact of climate change on a number of health outcomes (malaria, diarrheal disease, malnutrition, flood deaths, and direct effects from the heat and cold). The analysis suggests that although there are likely to be some benefits, particularly lower cold-related mortality in temperate zones, these benefits will be greatly exceeded by negative impacts on health, particularly in terms of infectious diseases and malnutrition in low-income countries. The methodological approach has been outlined elsewhere (McMichael et al., 2004), and on aggregate, the estimates suggest that compared to baseline, climate change had caused around 150,000 deaths annually by 2000, an equivalent to 0.3 percent of global deaths per year or 0.4 percent of global disability-adjusted life-years (DALYs) lost annually. The estimate may well be conservative because the baseline used was the average climate for 1960–1991 when climate change was probably already under way and the range of health outcomes was limited. Although increasing wealth and some level of adaptation could blunt the adverse effects, it is likely that the disease burden as a result of climate change will increase substantially over time and will be particularly concentrated in the poorer populations. Nevertheless, populations in all regions of the world are likely to experience some adverse effects, particularly if temperature increases exceed 2°C, at which temperature the probability of major adverse events—such as melting of ice caps and disruption of ecosystems—appears to be unacceptably high, as judged for example by policy makers of the European Union (EU) who have undertaken to pursue negotiations with the aim of keeping temperature increases below that level (European Commission, 2007).

Adaptation Strategies

Adaptation to climate change may take a number of forms; physiological and behavioral adaptation may take place without policy changes, but properly designed adaptation strategies can result in near-term benefits to public health, as well as improving the resilience of populations to future climate change. Policies to improve access to clean water and sanitation, promote hygiene behaviors, promote uptake of immunization, and strengthen health systems are needed in any event in order to improve the chances of attaining the UN Millennium Development Goals (Haines and Cassels, 2004). Increased use of insecticide-impregnated nets and appropriate antimalarial drug combinations that take into account prevailing patterns of drug resistance, as well as effective vector control strategies such as indoor residual spraying, can all help to reduce the burden of malaria.

The threat of climate change has resulted in increased interest in climate-based early warning health systems for heat waves and climate-sensitive diseases. Early warning systems must be integrated into local health systems if they are to have an impact. One example is the highland malaria project (HIMAL), which aims to create and test functional systems for malaria early warning and early detection including district-level surveillance and predictive modeling (Abeku et al., 2004). Most routine disease surveillance systems lack the ability to provide accurate and timely indications of increases in the number of cases of malaria. There is a need to improve the routine collection of data on parasitologically confirmed cases of malaria because febrile illnesses other than malaria have to be considered as possible causes of outbreaks (Cox et al., 2007). Seasonal forecasts can also be used to increase preparedness for climate variability and extreme events associated with phenomena such as El Niño; these approaches were used to warn specific governments when a strong El Niño was developing in 1997–1998 (Hamnett, 1998).

Meeting the Energy Needs of the Poor While Reducing Greenhouse Gas Emissions

Meeting the energy needs of the poor will also help to reduce vulnerability to climate change. Currently there are around 1.6 billion people without electricity (Modi et al., 2006), and 2.4 billion use solid fuels (wood, dung, coal) in their households. Meeting the essential energy needs of the poor will take around 1 percent of current world energy use and in addition reduce exposure to high levels of indoor air pollution (with an attributable annual mortality of about 1.6 million) (WHO, 2002). Concerted action is needed to improve access to more efficient cook stoves and to assist poor populations to move to cleaner fuels, such as kerosene, liquefied petroleum gas, or biogas. Electrification, using renewable energy where possible, can improve adaptation by supplying electricity to maintain the cold chain for vaccines, to provide a reliable power source for health facilities, and to make possible the use of information and communication technologies. The need for policies that prevent dangerous anthropogenic interference with the climate while addressing the energy needs of disadvantaged people is an essential challenge for the current era (Haines et al., 2007).

Although most renewable energy technologies can provide near-term benefits for health, for example by reducing exposure to air pollution, as well as mitigating greenhouse gas emissions, it is important that health impact assessments are undertaken. For example, a range of public health problems related to dams (which can be used to generate hydroelectricity and to promote adaptation to climate change through improved irrigation) have been documented, including increases in the prevalence of schistosomiasis, the introduction of Rift Valley fever, and increases in the burden of malaria. A recent systematic review concluded that although it was not possible to quantify the attributable fraction of the malaria burden due to dam building and irrigation, future water resource development projects should include in-depth assessment of potential effects (Keiser et al., 2005).


Climate change is likely to have far-reaching implications for human health and development. The Stern Review (Stern, 2006) has extensively reviewed the economic rationale for mitigation and adaptation polices, and also suggested that there is a strong economic case for action in the near term because the effects of climate change could result in losing around 5 percent of the gross world product (GWP) by the middle of the twenty-first century, perhaps even reaching 20 percent or more if the full range of effects is considered.

Infectious diseases are one of a number of categories of health outcomes that are likely to be affected adversely by climate change. Public health policies should take into account the need to adapt to a changing climate, as well as the potential for near-term benefits to health from a range of policies to mitigate climate change. Research funders should increase resources available to improve our understanding of the linkages between climate change, other environmental changes, and human health.


Paul R. Epstein, M.D., M.P.H.

In 1998, Hurricane Mitch dropped six feet of rain on Central America in three days. In its wake, the incidence of malaria, dengue fever, cholera, and leptospirosis soared. In 2000, rain and three cyclones inundated Mozambique for six weeks, and the incidence of malaria rose fivefold. In 2003, a summer heat wave in Europe killed tens of thousands of people, wilted crops, set forests ablaze, and melted 10 percent of the Alpine glacial mass.

This summer’s blistering heat wave was unprecedented with regard to intensity, duration, and geographic extent. More than 200 U.S. cities registered new record high temperatures. In Phoenix, Arizona, sustained temperatures above 100°F (38°C) for 39 consecutive days, including a week above 110°F (43°C), took a harsh toll on the homeless. Then came Hurricane Katrina, gathering steam from the heated Gulf of Mexico and causing devastation in coastal communities.

These sorts of extreme weather events reflect massive and ongoing changes in our climate to which biologic systems on all continents are reacting. So concluded the United Nations Intergovernmental Panel on Climate Change2, a collaboration of more than 2000 scientists from 100 countries. In 2001, the panel concluded that humans are playing a major role in causing these changes, largely through deforestation and the combustion of fossil fuels that produce heat-trapping gases such as carbon dioxide.

FIGURE 1-7. Hurricane Katrina passing over the Gulf of Mexico.

FIGURE 1-7Hurricane Katrina passing over the Gulf of Mexico

The map shows the three-day average of sea-surface temperatures from August 25, 2005, through August 27, 2005, and Hurricane Katrina growing in strength and breadth as it passes over the unusually warm Gulf of Mexico. Yellow, orange, and red areas are at or above 82°F (27.8°C, the temperature required for hurricanes to strengthen). Since the 1970s, the number of category 4 and 5 hurricanes has increased as sea temperatures have risen.

SOURCE: Scientific Visualization Studio at NASA.

Since 2001, we’ve learned substantially more. The pace of atmospheric warming and the accumulation of carbon dioxide are quickening; polar and alpine ice is melting at rates not thought possible several years ago3; the deep ocean is heating up, and circumpolar winds are accelerating; and warming in the lower atmosphere is retarding the repair of the protective “ozone shield” in the stratosphere. Moreover, ice cores that are drilled in Greenland indicate that the climate can change abruptly. Given the current rate of carbon dioxide buildup and the projected degree of global warming, we are entering uncharted seas.

As we survey these seas, we can see some of the health effects that may lie ahead if the increase in very extreme weather events continues4. Heat waves like the one that hit Chicago in 1995, killing some 750 people and hospitalizing thousands, have become more common2. Hot, humid nights, which have become more frequent with global warming, magnify the effects. The 2003 European heat wave—involving temperatures that were 18°F (10°C) above the 30-year average, with no relief at night—killed 21,000 to 35,000 people in five countries.

But even more subtle, gradual climatic changes can damage human health. During the past two decades, the prevalence of asthma in the United States has quadrupled, in part because of climate-related factors. For Caribbean islanders, respiratory irritants come in dust clouds that emanate from Africa’s expanding deserts and are then swept across the Atlantic by trade winds accelerated by the widening pressure gradients over warming oceans. Increased levels of plant pollen and soil fungi may also be involved. When ragweed is grown in conditions with twice the ambient level of carbon dioxide, the stalks sprout 10 percent taller than controls but produce 60 percent more pollen. Elevated carbon dioxide levels also promote the growth and sporulation of some soil fungi, and diesel particles help to deliver these aeroallergens deep into our alveoli and present them to immune cells along the way.

The melting of the earth’s ice cover has already become a source of physical trauma. In Alaska, Inuits report an increase in accidents caused by walking on thin ice3. Ocean warming and Arctic thawing are also spawning severe winter storms and hazardous travel conditions in the continental United States. Although tropical sea surfaces are warming and becoming saltier, parts of the North Atlantic are freshening from melting polar ice and increased amounts of rain falling at high latitudes. Contrasting barometric pressures over changing oceans increase winds and propel storms.

Meanwhile, in the past three decades, widening social inequities and changes in biodiversity—which alter the balance among predators, competitors, and prey that help keep pests and pathogens in check—have apparently contributed to the resurgence of infectious diseases. Global warming and wider fluctuations in weather help to spread these diseases: temperature constrains the range of microbes and vectors, and weather affects the timing and intensity of disease outbreaks5. Disease-bearing ticks in Sweden are moving northward as winters become warmer, and models project a similar shift in the United States and Canada. The encroachment of human housing on wilderness and reductions in the populations of predators of deer and competitors of mice are largely responsible for the current spread of Lyme disease.

Mosquitoes, which can carry many diseases, are very sensitive to temperature changes. Warming of their environment—within their viable range—boosts their rates of reproduction and the number of blood meals they take, prolongs their breeding season, and shortens the maturation period for the microbes they disperse. In highland regions, as permafrost thaws and glaciers retreat, mosquitoes and plant communities are migrating to higher ground6.

The increased weather variability that accompanies climate instability contributed to the emergence of both the hantavirus pulmonary syndrome and West Nile virus in the United States. Six years of drought in the Southwest apparently reduced the populations of predators, and early heavy rainfall in 1993 produced a bounty of piñon nuts and grasshoppers for rodents to eat. The resulting legions of white-footed mice heralded the appearance of hantavirus in the Americas. The origin of the 1999 outbreak of West Nile virus in New York City remains a mystery, but city-dwelling, bird-biting Culex pipiens mosquitoes thrive in shallow pools of foul water that remain in drains during droughts. When dry springs yield to sweltering summers, viral development accelerates and, with it, the cycle of mosquito-to-bird transmission. During the hot, arid summer of 2002, West Nile virus traveled across the country, infecting 230 species of animals, including 138 species of birds, along the way. Many of the affected birds of prey normally help to rein in rodent populations that can spread hantaviruses, arenaviruses, and yersinia and leptospira bacteria, as well as ticks infected with Borrelia burgdorferi.

Extremely wet weather may bring its own share of ills. Floods are frequently followed by disease clusters: downpours can drive rodents from burrows, deposit mosquito-breeding sites, foster fungus growth in houses, and flush pathogens, nutrients, and chemicals into waterways. Milwaukee’s cryptosporidium outbreak, for instance, accompanied the 1993 floods of the Mississippi River, and norovirus and toxins spread in Katrina’s wake. Major coastal storms can also trigger harmful algal blooms (“red tides”), which can be toxic, help to create hypoxic “dead zones” in gulfs and bays, and harbor pathogens.

Prolonged droughts, for their part, can weaken trees’ defenses against infestations and promote wildfires, which can cause injuries, burns, respiratory illness, and deaths. Shifting weather patterns are jeopardizing water quality and quantity in many countries, where groundwater systems are already being overdrawn and underfed. Most montane ice fields are predicted to disappear during this century—removing a primary source of water for humans, livestock, and agriculture in some parts of the world.

A still greater threat to human health comes from illnesses affecting wildlife, livestock, crops, forests, and marine organisms. The Millennium Ecosystem Assessment of 2005 revealed that 60 percent of the resources and life-support systems examined—from fisheries to fresh water—are already in decline or are being used in unsustainable ways. The resulting biologic impoverishment may have important consequences for our air, food, and water.

FIGURE 1-8. Increase from 1992 (left) to 2002 (right) in the amount of the Greenland ice sheet melted in the summer.

FIGURE 1-8Increase from 1992 (left) to 2002 (right) in the amount of the Greenland ice sheet melted in the summer

The extent of seasonal melting on the Greenland ice sheet has been observed by satellite since 1979. The melt zone (dark gray), where summer warmth turns snow and ice around the edges of the ice sheet into slush and water, has been expanding inland and to record-high elevations in recent years. When the meltwater seeps through cracks in the ice sheet, it may accelerate melting and allow ice to slide more easily over bedrock, speeding its movement to the sea. In addition to contributing to a rising sea level, this process adds freshwater to the ocean, with potential effects on ocean circulation and regional climate.

SOURCE: Map by C. Grabhorn. Reprinted from the Arctic Climate Impact Assessment (2004) with permission from Cambridge University Press and C. Grabhorn.

Crops are being confronted with more volatile weather, vanishing pollinators, and the proliferation of pests and pathogens. One fungal disease, soybean rust, is thought to have been ushered into the United States by Hurricane Ivan last fall. Warmth and moisture will favor its propagation.

And many habitats are not faring well. Our coastal zones, for example, are in trouble: coral reefs are suffering from warming-induced “bleaching,” excess waste, physical damage, overfishing, and fungal and bacterial diseases. Reefs provide a buffer against storms and groundwater salinization and offer protection for fish, the primary protein source for many inhabitants of island nations. One reef resident, the cone snail, produces a peptide that is 1000 times as potent as morphine and that is not addictive. We may never know what other potential treatments will be lost as reefs deteriorate.

All in all, it would appear that we may be underestimating the breadth of biologic responses to changes in climate. Treating climate-related ills will require preparation, and early-warning systems forecasting extreme weather can help to reduce casualties and curtail the spread of disease. But primary prevention would require halting the extraction, mining, transport, refining, and combustion of fossil fuels—a transformation that many experts believe would have innumerable health and environmental benefits and would help to stabilize the climate.

The good news is that we may also be underestimating the economic benefits of the clean-energy transition. When the financial incentives are adequate, renewable energy, energy-efficient and hybrid technologies, “green buildings,” and expanded public transportation systems can constitute an engine of growth for the 21st century.


The Center for Health and the Global Environment Harvard Medical School


“Imagining the unmanageable” was to be the subtitle for the Climate Change Futures report. But the devastating series of intense, immense fall hurricanes besetting the United States displaced it. What were once extreme scenarios for the US have occurred, and the consequences have cascaded across the physical landscape, overwhelming the capacities of health, ecological and economic systems to absorb, adapt to and manage the change.

Hurricane Katrina killed over 1,000 people, displaced over a million people, and spread oil, toxins, microorganisms and deep losses throughout the US Gulf Coast. It revealed deep-seated inequities and vulnerabilities, and the shock waves have reverberated through all sectors of society. The need for prevention has become embedded into our future political landscape.

While no one event is diagnostic of climate change, the relentless pace of unusually severe weather since 2001—prolonged droughts, heat waves of extraordinary intensity, violent windstorms and more frequent “100-year” floods—is descriptive of a changing climate.

The reasons for the changed weather patterns are well understood. Five years ago, Levitus and colleagues at the US Department of Commerce’s National Oceanic and Atmospheric Administration reported that the world’s oceans had warmed to a depth of two miles in five decades. This year Barnett and colleagues at the Scripps Institution of Oceanography reported that the oceans had absorbed 84% of the globe’s warming and that the warming pattern is unmistakably attributable to human activities.

Because of the natural cycles on which global warming is superimposed, the overall frequency of hurricanes ebbs and flows. But, since the 1970s, tropical storm destructiveness (peak winds and duration) (Emanuel 2005) and the frequency of category 4 and 5 storms (Webster et al. 2005) have essentially doubled. These observations are correlated with warming tropical seas, and the scientists project that continued warming will likely enhance the frequency of large storms still further.

Warm sea surfaces evaporate quickly and, with the deep ocean warming, the water replenishing that which evaporates is also warm and fuels subsequent storms. A warmer atmosphere also holds more water vapor, and the accelerated water cycle generates more droughts and more floods (see Figure 1-9; Trenberth, 2005).

FIGURE 1-9. Warm ocean waters fuel hurricanes.


Warm ocean waters fuel hurricanes. This image depicts the three-day average of sea surface temperatures (SSTs) from August 25–27, 2005, and the growing breadth of Hurricane Katrina as it passed over the warm Gulf of Mexico. Yellow, orange and (more...)

This fall’s succession of megastorms is, at the very least, a harbinger of what we can expect more of in a changing climate (Kerr 2005). The series may also mark a turning point in our understanding of how an energized climate system is exaggerating natural phenomena and of just how rapidly climate has changed.

This multidimensional assessment of climate change includes trends, case studies and scenarios—with a focus on health, ecological and economic dimensions. One surprise is the vulnerability of the energy sector—the primary source of increased heat in Earth systems. The risks to oil production compound the threats to the electricity grid from heat waves and the instabilities of pipelines grounded in thawing tundra.

At the same time, recovery, adaptation and prevention open the door to enormous opportunities. Developing a diversified portfolio of safe, well-distributed and nonpolluting energy sources, with hybrids and complementing technologies, can fortify energy security, bolster public health, promote economic activity and help stabilize the climate. Bold initiatives and innovative measures spearheading a well-funded, well-insured clean energy transition may be just the components needed to build a sustainable engine of growth for the 21st century.

Paul Epstein and Evan Mills

Executive Summary

Climate is the context for life on earth. Global climate change and the ripples of that change will affect every aspect of life, from municipal budgets for snow-plowing to the spread of disease. Climate is already changing, and quite rapidly. With rare unanimity, the scientific community warns of more abrupt and greater change in the future.

Many in the business community have begun to understand the risks that lie ahead. Insurers and reinsurers find themselves on the front lines of this challenge since the very viability of their industry rests on the proper appreciation of risk. In the case of climate, however, the bewildering complexity of the changes and feedbacks set in motion by a changing climate defy a narrow focus on sectors. For example, the effects of hurricanes can extend far beyond coastal properties to the heartland through their impact on offshore drilling and oil prices. Imagining the cascade of effects of climate change calls for a new approach to assessing risk.

The worst-case scenarios would portray events so disruptive to human enterprise as to be meaningless if viewed in simple economic terms. On the other hand, some scenarios are far more positive (depending on how society reacts to the threat of change). In addition to examining current trends in events and costs, and exploring case studies of some of the crucial health problems facing society and the natural systems around us, “Climate Change Futures: Health, Ecological and Economic Dimensions” uses scenarios to organize the vast, fluid possibilities of a planetary-scale threat in a manner intended to be useful to policymakers, business leaders and individuals.

Most discussions of climate impacts and scenarios stay close to the natural sciences, with scant notice of the potential economic consequences. In addition, the technical literature often “stovepipes” issues, zeroing in on specific types of events in isolation from the real-world mosaic of interrelated vulnerabilities, events and impacts. The impacts of climate change cross national borders and disciplinary lines, and can cascade through many sectors. For this reason we all have a stake in adapting to and slowing the rate of climate change. Thus, sound policymaking demands the attention and commitment of all.

While stipulating the ubiquity of the threat of climate change, understanding the problem still requires a lens through which the problem might be approached. “Climate Change Futures” focuses on health. The underlying premise of this report is that climate change will affect the health of humans as well as the ecosystems and species on which we depend, and that these health impacts will have economic consequences. The insurance industry will be at the center of this nexus, both absorbing risk and, through its pricing and recommendations, helping business and society adapt to and reduce these new risks. Our hope is that Climate Change Futures (CCF) will not only help businesses avoid risks, but also identify opportunities and solutions. An integrated assessment of how climate change is now adversely affecting and will continue to affect health and economies can help mobilize the attention of ordinary citizens around the world and help generate the development of climate-friendly products, projects and policies. With early action and innovative policies, business can enhance the world’s ability to adapt to change and help restabilize the climate.

Why Scenarios?

CCF is not the first report on climate change to use scenarios. The Intergovernmental Panel on Climate Change (IPCC) employs six of the very long-term and very broad scenarios representative of the many scenarios considered. Other organizations have explored scenarios of climate trajectories, impacts for some sectors and the mix of energy sources, to explore the potential consequences of trends and actions taken today. Scenarios are not simple projections, but are stories that present alternative images of how the future might unfold. Handled carefully, scenarios can help explore potential consequences of the interplay of multiple variables and thereby help us to make considered and comprehensive decisions.

The IPCC scenarios, contained in The Special Report on Emissions Scenarios (SRES), make projections into the next century and beyond and assume that climate change will be linear and involve gradual warming. But events of the last five years have overtaken the initial SRES scenarios. Climate has changed faster and more unpredictably than the scenarios outlined. Many of the phenomena assumed to lie decades in the future are already well underway. This faster pace of change on many fronts indicates that more sector-specific, short-term scenarios are needed.

With this in mind, the CCF scenarios are designed to complement the far-reaching IPCC framework. Drawing upon the full-blown, long-term scenarios offered by the IPCC, we have developed two scenarios that highlight possibilities inadequately considered in past assessments of climate change impacts.

Both CCF scenarios envision a climate context of gradual warming with growing variability and more weather extremes. Both scenarios are based on “business-as usual,” which, if unabated, would lead to doubling of atmospheric CO2 from pre-industrial values by midcentury. Both are based on the current climate trends for steady warming along with an increase in extremes, with greater and costlier impacts. The compilation of extreme weather events of all types shows a clear increase over the past decade in the number of extremes occurring in both hemispheres (see Figure 1-10, below).

FIGURE 1-10. These data are taken from EMDAT (Emergency Events Database) from 1975 to 2002.


These data are taken from EMDAT (Emergency Events Database) from 1975 to 2002. Compiled by the Center for Research on the Epidemiology of Disasters (CRED) at the Universite-Catholique de Louvain in Brussels, Belgium, this data set draws from multiple (more...)

Overall costs from catastrophic weather-related events rose from an average of US $4 billion per year during the 1950s, to US $46 billion per year in the 1990s, and almost double that in 2004. In 2004, the combined weather-related losses from catastrophic and small events were US $107 billion, setting a new record. (Total losses in 2004, including non-weather-related losses, were US $123 billion; Swiss Re 2005a). With Hurricanes Katrina and Rita, 2005 had, by September, broken all-time records yet again.

Meanwhile, the insured percentage of catastrophic losses nearly tripled from 11% in the 1960s to 26% in the 1990s and reached 42% (US $44.6 billion) in 2004 (all values inflation-corrected to 2004 dollars, Munich Re NatCatSERVICE).

As an insurer of last resort, the US Federal Emergency Management Agency has experienced escalating costs for natural disasters since 1990. Moreover, in the past decade, an increasing proportion of extreme weather events have been occurring in developed nations (Europe, Japan and the US) (see chart below).

The first impact scenario, or CCF-I, portrays a world with an increased correlation and geographical simultaneity of extreme events, generating an overwhelming strain for some stakeholders. CCF-I envisions a growing frequency and intensity of weather extremes accompanied by disease outbreaks and infestations that harm humans, wildlife, forests, crops and coastal marine systems. The events and their aftermaths would strain coping capacities in developing and developed nations and threaten resources and industries, such as timber, tourism, travel and the energy sector. The ripples from the damage to the energy sector would be felt throughout the economy.

BOX 1-2Key Points

  1. Warming favors the spread of disease.
  2. Extreme weather events create conditions conducive to disease outbreaks.
  3. Climate change and infectious diseases threaten wildlife, livestock, agriculture, forests and marine life, which provide us with essential resources and constitute our life-support systems.
  4. Climate instability and the spread of diseases are not good for business.
  5. The impacts of climate change could increase incrementally over decades.
  6. Some impacts of warming and greater weather volatility could occur suddenly and become widespread.
  7. Coastal human communities, coral reefs and forests are particularly vulnerable to warming and disease, especially as the return time between extremes shortens.
  8. An increasingly unstable climate could shift and settle into a new equilibrium, allowing a measure of adaptation and the opportunity to rapidly reduce the global environmental influence of human activities, namely fossil fuel combustion and deforestation.
  9. A well-funded, well-insured program to develop and distribute a diverse suite of means to generate energy cleanly, efficiently and safely offers enormous business opportunities and may present the most secure means of restabilizing the climate.
  10. Solutions to the emerging energy crisis must be thoroughly scrutinized as to their life cycle impacts on health and safety, environmental integrity, global security and the international economy.

In CCF-I, an accelerated water cycle and retreat of most glaciers undermine water supplies in some regions and land integrity in others. Melting of permafrost (permanently frozen land) in the Arctic becomes more pronounced, threatening native peoples and northern ecosystems. And gradually rising seas, compounded by more destructive storms cascading over deteriorating barrier reefs, threaten all low-lying regions.

Taken in aggregate, these and other effects of a warming and more variable climate could threaten economies worldwide. In CCF-I, some parts of the developed world may be capable of responding to the disruptions, but the events would be particularly punishing for developing countries. For the world over, historical weather patterns would diminish in value as guides to forecasting the future.

The second impact scenario, CCF-II, envisions a world in which the warming and enhanced variability produce surprisingly destructive consequences. It explores a future rife with the potential for sudden, wide-scale health, environmental and economic impacts as climate change pushes ecosystems past tipping points. As such, it is a future inherently more chaotic and unpredictable than CCF-I.

BOX 1-3Vulnerabilities in the Energy Sector

  • Heat waves generate blackouts.
  • Sequential storms disrupt offshore oil rigs, pipelines, refineries and distribution systems.
  • Diminished river flows reduce hydroelectric capacity and impede barge transport.
  • Melting tundra undermines pipelines and power transmission lines.
  • Warmed inland waters shut down power plant cooling systems.
  • Lightning claims rise with warming.

Each stage in the life cycle of oil, including exploration, extraction, transport, refining and combustion, carries hazards for human health and the environment. More intense storms, thawing permafrost and dried riverbeds, make every stage more precarious.

Some of the impacts envisioned by the second scenario are very severe and would involve catastrophic, widespread damages, with a world economy beset by increased costs and chronic, unmanageable risks. Climate-related disruptions would no longer be contained or confined.

Threshold-crossing events in both terrestrial and marine systems would severely compromise resources and ecological functions, with multiple consequences for the species that depend upon them. For example:

  • Repeated heat waves on the order of the 2003 and 2005 summers could severely harm populations, kill livestock, wilt crops, melt glaciers and spread wildfires.
  • The probability of such extreme heat has already increased between two and four times over the past century and, based on an IPCC climate scenario, more than half the years by the 2040s will have summers warmer than that of 2003.
  • Chronic water shortages would become more prevalent, especially in semi-arid regions, such as the US West.
  • With current usage levels, more environmentally displaced persons and a changing water cycle, the number of people suffering water stress and scarcity today will triple in two decades.

Other non-linear impact scenarios include:

  • Widespread diebacks of temperate and northern forests from drought and pests.
  • Coral reefs, already multiply stressed, collapse from the effects of warming and diseases.
  • Large spikes occur in property damages from a rise of major rivers. (A 10% increase in flood peak would produce 100 times the damage of previous floods, as waters breach dams and levees.)
  • Severe storms and extreme events occurring sequentially and concurrently across the globe overwhelm the adaptive capacities of even developed nations; large areas and sectors become uninsurable; major investments collapse; and markets crash.

CCF-II would involve blows to the world economy sufficiently severe to cripple the resilience that enables affluent countries to respond to catastrophes. In effect, parts of developed countries would experience developing nation conditions for prolonged periods as a result of natural catastrophes and increasing vulnerability due to the abbreviated return times of extreme events.

Still, CCF-II is not a worst-case scenario.

A worst-case scenario would include large-scale, nonlinear disruptions in the climate system itself—slippage of ice sheets from Antarctica or Greenland, raising sea levels inches to feet; accelerated thawing of permafrost, with release of large quantities of methane; and shifts in ocean thermohaline circulation (the stabilizing ocean “conveyor belt”).

Finally, there are scenarios of climate stabilization. Restabilizing the climate will depend on the global-scale implementation of measures to reduce greenhouse gas emissions. Aggressively embarking on the path of non-fossil fuel energy systems will take planning and substantive financial incentives—not merely a handful of temporizing, corrective measures.

This assessment examines signs and symptoms suggesting growing climate instability and explores some of the expanding opportunities presented by this historic challenge.

Applying the Scenarios

In choosing how to apply the two impact scenarios, we have focused on case studies of specific health and ecological consequences that extend beyond the more widely studied issue of property damages stemming from warming and natural catastrophes. In each case study, we identify current trends under way and envision the future consequences for economies, social stability and public health.

Infectious diseases have resurged in humans and in many other species in the past three decades. Many factors, including land-use changes and growing poverty, have contributed to the increase. Our examination of malaria, West Nile virus and Lyme disease explores the role of warming and weather extremes in expanding the range and intensity of these diseases and both linear and non-linear projections for humans and wildlife.

The rising rate of asthma (two to threefold increase in the past two decades; fourfold in the US) receives special attention, as air quality is affected by many aspects of a changing climate (wildfires, transported dust and heat waves), and by the inexorable rise of atmospheric CO2 in and of itself, which boosts ragweed pollen and some soil molds.

We also examine the public health consequences of natural catastrophes themselves, including heat waves and floods. An integrated approach exploring linkages is particularly useful in these instances, since the stovepipe perspective tends to play down the very real health consequences and the manifold social and economic ripples stemming from catastrophic events.

Another broad approach of the CCF scenarios is to study climate change impacts on ecological systems, both managed and natural. We examine projections for agricultural productivity that, to date, largely omit the potentially devastating effects of more weather extremes and the spread of pests and pathogens. Crop losses from pests, pathogens and weeds could rise from the current 42% to 50% within the coming decade.

BOX 1-4Case Studies in Brief

Infectious and Respiratory Diseases

  • Malaria: Malaria is the deadliest, most disabling and most economically damaging mosquito-borne disease worldwide. Warming affects its range, and extreme weather events can precipitate large outbreaks. This study documents the fivefold increase in illness following a 6-week flood in Mozambique, explores the surprising role of drought in northeast Brazil, and projects changes for malaria in the highlands of Zimbabwe.
  • West Nile Virus: West Nile virus (WNV) is an urban-based, mosquito-borne infection, afflicting humans, horses and more than 138 species of birds. Present in the United States, Europe, the Middle East and Africa, warm winters and spring droughts play roles in amplifying this disease. To date, there have been over 17,000 human cases and over 650 deaths from WNV in North America.
  • Lyme Disease: Lyme disease is the most widespread vector-borne disease in the US and can cause long-term disability. Lyme disease is spreading in North America and Europe as winters warm, and models project that warming will continue to shift the suitable range for the deer ticks that carry this infection.
  • Asthma: Asthma prevalence has quadrupled in the US since 1980, and this condition is increasing in developed and underdeveloped nations. New drivers include rising CO2, which increases the allergenic plant pollens and some soil fungi, and dust clouds containing particles and microbes coming from expanding deserts, compounding the effects of air pollutants and smog from the burning of fossil fuels.

Extreme Weather Events

  • Temperature: Heat waves are becoming more common and more intense throughout the world. This study explores the multiple impacts of the highly anomalous 2003 summer heat wave in Europe and the potential impact of such “outlier” events elsewhere for human health, forests, agricultural yields, mountain glaciers and utility grids.
  • Flooding: Floods inundated large parts of Central Europe in 2002 and had consequences for human health and infrastructure. Serious floods occurred again in Central Europe in 2005. The return times for such inundations are projected to decrease in developed and developing nations, and climate change is expected to result in more heavy rainfall events.

Natural and Managed Systems

  • Drought: Forests are experiencing numerous pest infestations. Warming increases the range, reproductive rates and activity of pests, such as spruce bark beetles, while drought makes trees more susceptible to the pests. This study examines the synergies of drought and pests, and the dangers of wildfire. Large-scale forest diebacks are possible, and they would have severe consequences for human health, property, wildlife, timber and Earth’s carbon cycle.
  • Agriculture: Agriculture faces warming, more extremes and more diseases. More drought and flooding under the new climate, and accompanying outbreaks of crop pests and diseases, can affect yields, nutrition, food prices and political stability. Chemical measures to limit infestations are costly and unhealthy.
  • Marine Environments: Marine ecosystems are under increasing pressure from overfishing, excess wastes, loss of wetlands, and diseases of bivalves that normally filter and clean bays and estuaries. Even slightly elevated ocean temperatures can destroy the symbiotic relationship between algae and animal polyps that make up coral reefs, which buffer shores, harbor fish and contain organisms with powerful chemicals useful to medicine. Warming seas and diseases may cause coral reefs to collapse.
  • Fresh Water Availability: Water, life’s essential ingredient, faces enormous threats. Underground stores are being overdrawn and underfed. As weather patterns shift and mountain ice fields disappear, changes in water quality and availability will pose growth limitations on human settlements, agriculture and hydropower. Flooding can lead to water contamination with toxic chemicals and microbes, and natural disasters routinely damage water-delivery infrastructure.


Donald S. Burke, M.D.8

University of Pittsburgh

Relationships between weather and health have been appreciated since the dawn of civilization. Hippocrates, in 400 B.C.E. (On Airs, Waters, and Places), observed that

[w]hoever would study medicine aright must learn of the following subjects. First he must consider the effect of the seasons of the year and the differences between them. Secondly he must study the warm and cold winds, both those which are in common to every country and those peculiar to a particular locality. Lastly, the effect of water on the health must not be forgotten.

Can measurements of climate and weather be used to predict epidemics? The purpose of this paper is to review exactly what is and what is not known about the causal relationships between changes in climate and weather and infectious diseases. I do not attempt to present an exhaustive review, but I touch on key issues and guide the interested reader to more comprehensive sources. I focus especially on some computational approaches that can be used to strengthen the state of the art.

Recent Expert Summary Reports

In 2001, I chaired a National Research Council (NRC) committee that issued a report entitled Under the Weather: Climate, Ecosystems, and Infectious Diseases (NRC, 2001). This expert committee concluded that although it is clear that weather fluctuations and seasonal-to-interannual climate variability influence many infectious diseases, published observational and modeling studies must be interpreted cautiously, and consequently the exact future impacts of global climate change remain uncertain. On a brighter note, the Under the Weather committee did go on to say that recent technological and modeling advances are improving our abilities to identify patterns in infectious disease epidemiology.

More recently, two Nobel Prize-winning documents have appeared that illuminate the current knowledge about the relationships of climate change and human health. Last year the IPCC released its Fourth Assessment Report including the “Impacts, Adaptation and Vulnerability” Working Group section (Chapter 8) on human health (IPCC, 2007). The IPCC meticulously explained the potential range of direct, indirect, and socially mediated effects of climate change on human health, including careful caveats about how these effects may be modulated by future environmental, societal, and health system changes. The IPCC then ranked its findings under the headings very high confidence, high confidence, and medium confidence. High confidence negative impacts of climate change on human health included more malnutrition; more deaths, injuries, and diseases from extreme weather events; and more cardiorespiratory diseases from the deterioration of air quality. However, findings related to infectious diseases were careful and circumscribed. The IPCC expressed very high confidence that there would be changes in the distribution of malaria, but noted that in various locales the direction of change could be either an increase or a decrease in malaria, depending on predicted redistributions of rainfall, without a clear conclusion as to whether the net global impact would be positive or negative. Similarly, the group predicted with high confidence that there would be changes in the distribution of many insect vectors of other diseases, in most instances with a negative impact on human health but in some locales a favorable impact. The group expressed medium confidence that there would be a net global increase in the burden of diarrheal diseases. Overall, I interpret the IPCC conclusions to be remarkably measured and quite consistent with our Under the Weather findings.

Former U.S. Vice President Al Gore’s documentary An Inconvenient Truth also treads carefully around possible relationships between global warming and epidemic diseases (Gore, 2006). Here is what he says about infectious diseases in the Oscar-winning feature film:

There are cities that were founded because they were just above the mosquito line. Nairobi is one. Harare is another. There are plenty of others. Now the mosquitoes with warming are climbing to the higher altitudes. There are a lot of vectors for infectious diseases that are worrisome to us that are also expanding their ranges, not only mosquitoes but all of these others as well. And we’ve had 30 so-called new diseases that have emerged just in the last quarter-century. And a lot of them like SARS [severe acute respiratory syndrome] have caused tremendous problems. The resistant forms of tuberculosis. There are others. And there’s been a reemergence of some diseases that were once under control. The avian flu, of course, [is] quite a serious matter as you know. West Nile virus. It came to the eastern shore of Maryland in 1999. Two years later it was across the Mississippi. And two years after that it had spread across the continent. But these are very troubling signs.

Although Vice President Gore cites changes in vector distributions and the threat of emerging diseases, he stops short of claiming a strong causal linkage of particular diseases to climate change.

In sum, assessments by respected scientific bodies have stayed clear of apocalyptic pronouncements. Conclusions about the probable impact of global climate change on epidemic infectious diseases have been cautious and measured.

Uncertainties and Complexities

Why so cautious? My own view is that this caution honestly reflects the uncertainties involved, which in turn reflect the difficulty of the underlying scientific problems. While the idea that infectious diseases are driven by weather and climate is inescapable, our predictive powers remain limited. Foremost among the underlying scientific problems is the fact that epidemics are inherently dynamic, nonlinear processes, composed of many interacting subsystems, only one of which is weather and climate. Other subsystems of the dynamic epidemic system include human social interactions, waxing and waning of immunity at the population level, various health system interventions, and other factors. Ability to detect a strong cause-and-effect relationship between any single weather factor—say ambient temperature—and disease incidence requires a good understanding of the entire system and its dynamics. It is now painfully clear that any robust assessment of the impact of climate change on epidemic infectious diseases must be evaluated using dynamical models; the word arguments and static arrow diagrams (“St. Sebastian diagrams”) so often used in the past are simply not persuasive.

Various infectious diseases are affected differently by environmental conditions (NRC, 2001). As may be seen in Table 1-1, for example, warm temperatures favor insect vector-borne diseases such as malaria and dengue, as evidenced by their tropical distributions and warm season peak incidences; cold temperatures favor influenza, as evidenced by its typical winter appearance; dryness favors meningocci (meningitis) and coccidioidomycosis, as evidenced by their association with arid conditions and dust storms; and wet conditions favor cryptosporidiosis and Rift Valley fever, as evidenced by their association with flooding. Obviously, while climate change in any given direction may increase the incidence of one disease, it may reduce the incidence of other diseases. For any given geographic locale, the net effects on human health may be negative, neutral, or positive, depending on local infectious disease circumstances. Predictions about the net effects of global warming will require a better understanding of its effects on each particular infectious disease.

TABLE 1-1. Examples of Environmental Factors Known to Be Strongly Associated with Certain Specific Infectious Diseases.


Examples of Environmental Factors Known to Be Strongly Associated with Certain Specific Infectious Diseases.

Influenza and dengue are both globally important infectious diseases, capable of serious epidemic spread. Although both are caused by small enveloped RNA viruses, one (influenza) is a cool season disease, and one (dengue) is a warm season disease. A comparison is instructive.

Seasonality of Influenza

A comparison of the incidence curves in the United States, Mexico, Colombia, Brazil, and Argentina shows that the timing of annual influenza incidence varies closely with latitude. Epidemic peaks occur in the winter season in high-latitude countries (alternating with the Northern and Southern Hemisphere winters), but no clear annual peak is seen in countries closer to the equator. In addition to seasonal changes in ambient temperature, a variety of explanations have been proposed to account for the strong wintertime seasonality of influenza (for some recent reviews, see Dowell, 2001; Finkelman et al., 2007; Fisman, 2007; Lofgren et al., 2007). One set of explanations focuses on altered host immunity, possibly due to changes in the photoperiod with shortened days in winter or to low vitamin D caused by decreased exposure to sunshine. Another set of explanations focuses on increased human-to-human exposure in winter—for example, increased social contact among children during the typical school season or breathing shared air in confined spaces. Indeed it is possible that more than one factor contributes to the seasonality of influenza.

Experimental studies in laboratory animals have been valuable in understanding the effects of altered environmental conditions on influenza transmission from animal to animal. More than 40 years ago, Kilbourne and colleagues, studying influenza transmission in caged mice, showed that low temperature and low relative humidity can increase influenza transmission (Schulman and Kilbourne, 1963). Recent studies by Palase and colleagues have provided even more compelling evidence and are summarized here (Lowen et al., 2007). They studied guinea pigs in a climate-controlled setup in which air flow direction, temperature, and relative humidity could be carefully controlled. Pairs of guinea pigs were placed in adjacent cages, one intentionally inoculated with the H3N2 A/Panama/99 strain influenza virus (a typical strain) and placed in the cage upwind of the other noninfected animal. Four pairs of animals were studied at a time under controlled conditions of temperature (5 to 30°C) and relative humidity (20 to 80 percent). Results are shown in Figure 1-11. Under conditions of high temperature or high relative humidity, none of the four downwind guinea pigs breathing infectious air became infected. In contrast, under conditions of low temperature and/or low relative humidity, four of four downwind animals invariably became infected. Increased viral shedding into respiratory secretions was found in animals held at lower temperatures. The authors concluded that both low temperature and low relative humidity increased transmission, the temperature effect mediated by increased virus shedding and the relative humidity effect probably by bio-aerosol formation of droplet nuclei. Although these fine experiments do provide strong evidence of the mechanism relating winter to flu, they provide little help in formulating predictive models on which to base future influenza transmission scenarios. How much of a seasonal change in transmissibility (e.g., shedding, droplet nuclei formation) is needed to see the typical seasonal oscillations of influenza? The answer is that it depends on how the magnitude and timing of these seasonal forces interact with other oscillations inherent in the influenza epidemic processes.

FIGURE 1-11. Transmission of influenza from infected guinea pigs to uninfected exposed guinea pigs under different experimental conditions in which ambient temperature and relative humidity were varied.


Transmission of influenza from infected guinea pigs to uninfected exposed guinea pigs under different experimental conditions in which ambient temperature and relative humidity were varied. Each box shows the results of one experimental study in which (more...)

Dushoff and colleagues recently demonstrated that epidemic systems such as influenza can display intrinsic oscillations without requiring any seasonal effects on transmissibility (Dushoff et al., 2004). Instead they showed that seasonal effects on transmissibility can either resonate with or dampen the intrinsic oscillations of the system, depending on their respective relative frequencies. They formulated a relatively simple model system in which individuals progress from susceptible to infected to recovered or immune and back to susceptible again (S-I-R-S). The incidence of new cases in the system = β × I × S or beta (a measure of transmission) × the number of infected × the number of susceptibles. Even when the transmission variable beta is held constant, the system shows oscillations, and the intrinsic period of the oscillation of incidence is a function of several standard epidemic parameters, including the basic reproductive ratio (R0), the length of infection (L), and the duration of immunity (D). Decreasing R0, prolonging the duration of infection, or lengthening the duration of immunity can all serve to slow the intrinsic periodicity. Importantly, this intrinsic periodicity need not be annual.

Dushoff et al. (2004) then asked what happens if seasonality is imposed. That is, instead of being a fixed constant, β is varied sinusoidally (seasonally). How does this exogenous forcing oscillation of transmissibility interact with the intrinsic S-I-R-S oscillation? They found that a very small seasonal oscillation in β (the transmission parameter) could markedly increase the peak-to-trough amplitude of the system. Figure 1-12 illustrates the effects of a small 4 percent change in the amplitude of the seasonal transmission parameter. Random combinations of were seeded to generate intrinsically oscillating systems. The graph R0, D, and L shows the oscillation amplitudes without and with a tiny 4 percent seasonally varying change in transmission (β). There is a strong nonlinear resonance of the effects of the seasonal forcing on oscillation amplitude when the intrinsic oscillation period is also seasonal (i.e., equal to 1). If natural influenza systems follow the same behaviors as this model system, then it may prove difficult indeed to measure seasonal changes in disease incidence as a function of seasonal changes in the transmission parameter β.

FIGURE 1-12. Graph showing the amplitude of oscillations (y axis, peak-trough ratio) as a function of the endogenous oscillation period (x axis) in a stochastic forced S-I-R-S epidemic model for 2,000 sets of randomly chosen parameters.


Graph showing the amplitude of oscillations (y axis, peak-trough ratio) as a function of the endogenous oscillation period (x axis) in a stochastic forced S-I-R-S epidemic model for 2,000 sets of randomly chosen parameters. The imposed 1-year seasonal (more...)

Regrettably, real-world influenza dynamics are more complex than the simple model of Dushoff et al. One aspect of this complexity is that “seasonal influenza” is in fact not one disease process but three: influenza A/H3, influenza A/H1, and influenza B. All three viruses co-circulate worldwide and contribute to the reported total cases of influenza. Data on influenza isolates in the United States were obtained from the WHO “FluNet” website9 and graphed as a continuous 11-year time series, from 1997 through 2007, as shown in Figure 1-13. It can be seen that A/H1 and A/H3 influenza strains dominate in different years; it is possible that they may reciprocally interfere with each other. Another factor complicating any analysis of the relationships between weather and influenza epidemiology is the fact that the three influenza viruses may well have somewhat different values of transmission parameters (βs); however, this has not been directly measured.

FIGURE 1-13. Influenza virus types isolated in the United States between 1997 and 2007.


Influenza virus types isolated in the United States between 1997 and 2007. Influenza H3N2 is shown in red, H1N1 in blue, and B in green. Numbers shown are total isolates typed and reported. SOURCE: WHO (2008).

Epidemics as Partially Decomposable Systems

Because dynamical systems, such as epidemics, are often composed of two or more semi-independent but partially interacting dynamical subsets (e.g., environmental conditions, immunity, health systems), it is essential to isolate and analyze these component dynamical subsystems so as to be able to understand the effects of any particular forcing factor (e.g., ambient temperature). This field is still in its infancy, but there are a number of techniques available. My colleague Derek Cummings and I have experimented with various time-series decomposition techniques borrowed from physics to tease apart the component subsystems from long-term records of the waxing and waning of epidemic diseases (Cummings et al., 2004). Working with our colleagues at the Thailand Ministry of Public Health, we reviewed epidemiological records over many years, entered them into digital format, and applied decomposition methods. Some of these time-series decomposition methods include the well-known Fourier decomposition methods, but we also examined various wavelet decomposition methods, and the Empirical Mode Decomposition. Figure 1-14 is just one example of an analysis of longitudinal epidemic time-series data, showing that this is a partially decomposable system. We applied the Empirical Mode Decomposition (Huang et al., 1998) to analyze 15 consecutive years of the incidence of dengue hemorrhagic fever in Bangkok. A major feature of the Empirical Mode Decomposition is that the method identifies component “modes” of differing frequency, from fast to slow, that together contribute to the full tracing of the epidemic time series. Note that in the Empirical Mode Decomposition, the identified modes are not single standing frequencies, but instead are patterns whose frequencies may vary. For dengue in Bangkok, we identified several major component frequency modes, including a slow 3- to 4-year oscillation that we believe is due to changes in host immunity, a clear 1-year annual oscillation that is probably driven by seasonal changes in weather, and a spiky, irregular, rapid (faster-than-annual) oscillation that may represent local and chance events. Our expectation is that decomposition methods such as these will make it possible to eliminate—as “noise”—the changes in incidence contributed by other modes and to focus on a single mode to understand its intrinsic behaviors and responsiveness to external forces.

FIGURE 1-14. Multiyear time series of incidence of dengue hemorrhagic fever cases in Bangkok decomposed using the Empirical Mode Decomposition method into three modes of different approximate frequencies.


Multiyear time series of incidence of dengue hemorrhagic fever cases in Bangkok decomposed using the Empirical Mode Decomposition method into three modes of different approximate frequencies. When summed, the three modes add up to the original series. (more...)


In this paper, I have argued that our current understanding of the relationships between climate and weather and epidemic infectious diseases is insufficient to make quantitative predictions about future threats posed by infectious diseases under various global climate change scenarios. Nonetheless, I am confident that new state-of-the-art methods, including computational tools, are now available to apply to these difficult scientific problems. I am hopeful that within a few years it may be possible to robustly predict such risks and take steps to intervene to avert potential crises.

It is safe to say that the main “Recommendations for Future Research and Surveillance” of our Under the Weather group continue to be relevant:

  • Research on the linkages between climate and infectious diseases must be strengthened.
  • Further development of disease transmission models is needed to assess the risks posed by climatic and ecological changes.
  • Epidemiological surveillance programs should be strengthened.
  • Observational, experimental, and modeling activities are all highly interdependent and must progress in a coordinated fashion.
  • Research on climate and infectious disease linkages inherently requires interdisciplinary collaborations.


    Overview References

    1. Center for Health and the Global Environment. Climate change futures: health, ecological and economic dimensions. Cambridge, MA: Harvard Medical School; 2005.
    2. Dushoff J, Plotkin JB, Levin SA, Earn DJD. Dynamical resonance can account for seasonality of influenza epidemics. Proceedings of the National Academy of Sciences. 2004;101(48):16915–16916. [PMC free article: PMC534740] [PubMed: 15557003]
    3. Hilborn R, Mangel M. The ecological detective. Princeton, NJ: Princeton University Press; 1997.
    4. NRC (National Research Council). Under the weather: climate, ecosystems, and infectious diseases. Washington, DC: National Academy Press; 2001. [PubMed: 25057537]
    5. Stott PA, Stone DA, Allen MR. Human contribution to the European heatwave of 2003. Nature. 2004;432(7017):610–614. [PubMed: 15577907]

    Haines References

    1. Abeku TA, van Oortmarssen GJ, Borsboom G, de Vlas SJ, Habbema JD. Spatial and temporal variations of malaria epidemic risk in Ethiopia: factors involved and implications. Acta Tropica. 2003;87(3):331–340. [PubMed: 12875926]
    2. Abeku TA, Hay SI, Ochola S, Langi P, Beard B, de Vlas SJ, Cox J. Malaria epidemic early warning and detection in African highlands. Trends in Parasitology. 2004;20(9):400–405. [PMC free article: PMC3173853] [PubMed: 15324728]
    3. Ahern M, Kovats RS, Wilkinson P, Few R, Matthies F. Global health impacts of floods: epidemiologic evidence. Epidemiologic Reviews. 2005;27(1):36–46. [PubMed: 15958425]
    4. BCAS/RA/Approtech. Vulnerability of Bangladesh to climate change and sea level rise: concepts and tools for calculating risk in integrated coastal zone management. Dhaka, Bangladesh: Bangladesh Centre for Advanced Studies (BCAS); 1994.
    5. Bouma MJ, Kovats RS, Goubet SA, Cox JS, Haines A. Global assessment of El Niño’s disaster burden. Lancet. 1997;350(9089):1435–1438. [PubMed: 9371167]
    6. Cairncross S. Water supply and sanitation: some misconceptions. Tropical Medicine and International Health. 2003;8(3):193–195. [PubMed: 12631306]
    7. Cairncross S, Alvarinho M. Floods, hazards and health: responding to present and future risks. Few R, Matthies F, editors. London, UK: Earthscan Publications Ltd; 2006.
    8. Checkley W, Gilman RH, Black RE, Epstein LD, Cabrera L, Sterling CR, Moulton LH. Effect of water and sanitation on childhood health in a poor Peruvian periurban community. Lancet. 2004;363(9403):112–118. [PubMed: 14726164]
    9. Colwell RR. Global climate and infectious disease: the cholera paradigm. Science. 1996;274(5295):2025–2031. [PubMed: 8953025]
    10. Cox J, Abeku T, Beard J, Turyeimuka J, Tumwesigye E, Okia M, Rwakimari J. Detecting epidemic malaria, Uganda. Emerging Infectious Diseases. 2007;13(5):779–780. [PMC free article: PMC2738469] [PubMed: 18044039]
    11. Curriero FC, Patz JA, Rose JB, Lele S. The association between extreme precipitation and waterborne disease outbreaks in the United States, 1948–1994. American Journal of Public Health. 2001;91(8):1194–1199. [PMC free article: PMC1446745] [PubMed: 11499103]
    12. Curtis V, Cairncross S. Effect of washing hands with soap on diarrhoea risk in the community: a systematic review. Lancet Infectious Diseases. 2003;3(5):275–281. [PubMed: 12726975]
    13. Emberlin J, Detandt M, Gehrig R, Jaeger S, Nolard N, Rantio-Lehtimaki A. Responses in the start of betula (birch) pollen seasons to recent changes in spring temperatures across Europe. International Journal of Biometeorology. 2002;46(4):159–170. [PubMed: 12242471]
    14. Engelthaler DM, Mosley DG, Cheek JE, Levy CE, Komatsu KK, Ettestad P, Davis T, Tanda DT, Miller L, Frampton JW, Porter R, Bryan RT. Climatic and environmental patterns associated with hantavirus pulmonary syndrome, Four Corners region, United States. Emerging Infectious Diseases. 1999;5(1):87–94. [PMC free article: PMC2627709] [PubMed: 10081675]
    15. European Commission. EU action against climate change: leading global action to 2020 and beyond. Luxembourg: European Communities; 2007.
    16. FAO (Food and Agriculture Organization). The state of food insecurity around the world: eradicating hunger—key to achieving the millennium development goals. Rome: FAO; 2005.
    17. Franke CR, Ziller M, Staubach C, Latif M. Impact of the El Niño/Southern Oscillation on visceral leishmaniasis, Brazil. Emerging Infectious Diseases. 2002;8(9):914–917. [PMC free article: PMC2732547] [PubMed: 12194766]
    18. Guerra CA, Snow RW, Hay SI. A global assessment of closed forests, deforestation and malaria risk. Annals of Tropical Medicine and Parasitology. 2006;100(3):189–204. [PMC free article: PMC3204444] [PubMed: 16630376]
    19. Haines A, Cassels A. Can the millennium development goals be attained? British Medical Journal. 2004;329(7462):394–397. [PMC free article: PMC509351] [PubMed: 15310610]
    20. Haines A, Patz J. Health effects of climate change. Journal of the American Medical Association. 2004;291(1):99–103. [PubMed: 14709582]
    21. Haines A, Smith KR, Anderson D, Epstein PR, McMichael AJ, Roberts I, Wilkinson P, Woodcock J, Woods J. Policies for accelerating access to clean energy, improving health, advancing development, and mitigating climate change. Lancet. 2007;370(9594):1264–1281. [PubMed: 17868819]
    22. Hajat S, Ebi KL, Kovats S, Menne B, Edwards S, Haines A. The human health consequences of flooding in Europe and the implications for public health: a review of the evidence. Applied Environmental Science and Public Health. 2003;1:13–21.
    23. Hamnett MP. The Pacific ENSO Applications Centre and the 1997–98 El Niño. Online. Pacific ENSO update. 1998. [accessed May 28, 2008]. http://www​.soest.hawaii​.edu/MET/Enso/peu/update.html.
    24. Hay SI, Cox J, Rogers DJ, Randolph SE, Stern DI, Shanks GD, Myers MF, Snow RW. Climate change and the resurgence of malaria in the east African highlands. Nature. 2002;415(6874):905–909. [PMC free article: PMC3164800] [PubMed: 11859368]
    25. IPCC (Intergovernmental Panel on Climate Change). Climate change 2007: the physical science basis. Summary for policymakers. In: Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB, Tignor M, Miller HL, editors. Contribution of Working Group I to the fourth assessment report of the Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge University Press; 2007a.
    26. IPCC (Intergovernmental Panel on Climate Change). Climate change 2007: impacts, adaptation and vulnerability. Technical summary. In: Parry M, Canziani O, Palutikof J, van der Linden P, Hanson C, editors. Contribution of Working Group II to the fourth assessment report of the Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge University Press; 2007b.
    27. Keiser J, De Castro MC, Maltese MF, Bos R, Tanner M, Singer BH, Utzinger J. Effect of irrigation and large dams on the burden of malaria on a global and regional scale. American Journal of Tropical Medicine and Hygiene. 2005;72(4):392–406. [PubMed: 15827275]
    28. Kovats RS, Bouma MJ, Hajat S, Worrall E, Haines A. El Niño and health. Lancet. 2003;362(9394):1481–1489. [PubMed: 14602445]
    29. Kovats RS, Edwards SJ, Hajat S, Armstrong BG, Ebi KL, Menne B. The effect of temperature on food poisoning: a time-series analysis of salmonellosis in ten European countries. Epidemiology and Infection. 2004;132(3):443–453. [PMC free article: PMC2870124] [PubMed: 15188714]
    30. Lindgren E, Gustafson R. Tick-borne encephalitis in Sweden and climate change. Lancet. 2001;358(9275):16–18. [PubMed: 11454371]
    31. Lindgren E, Talleklint L, Polfeldt T. Impact of climatic change on the northern latitude limit and population density of the disease-transmitting European tick Ixodes ricinus. Environmental Health Perspectives. 2000;108(2):119–123. [PMC free article: PMC1637900] [PubMed: 10656851]
    32. Linnerooth-Bayer J, Mechler R, Pflug G. Refocusing disaster aid. Science. 2005;309(5737):1044–1046. [PubMed: 16099976]
    33. Linthicum KJ, Anyamba A, Tucker CJ, Kelley PW, Myers MF, Peters CJ. Climate and satellite indicators to forecast Rift Valley fever epidemics in Kenya. Science. 1999;285(5426):397–400. [PubMed: 10411500]
    34. Lloyd SJ, Kovats RS, Armstrong BG. Global diarrhoea morbidity, weather and climate. Climate Research. 2007;34(2):119–127.
    35. Manuel J. In Katrina’s wake. Environmental Health Perspectives. 2006;114(1):A32–A39. [PMC free article: PMC1332683] [PubMed: 16393645]
    36. Mason JB, Bailes A, Mason KE, Yambi O, Jonsson U, Hudspeth C, Hailey P, Kendle A, Brunet D, Martel P. AIDS, drought, and child malnutrition in southern Africa. Public Health Nutrition. 2005;8(6):551–563. [PubMed: 16236184]
    37. McMichael AJ, Woodruff R, Whetton P, Hennessy K, Nicholls N, Hales S, Woodward A, Kjellstrom T. Human health and climate change in Oceania: risk assessment 2002. Canberra, Australia: Department of Health and Ageing; 2003.
    38. McMichael A, Campbell-Lendrum D, Kovats S, Edwards S, Wilkinson P, Wilson T, Nicholls R, Hales S, Tanser F, Le Sueur D, Schlesinger M, Andronova N. Global climate change. Geneva, Switzerland: World Health Organization; 2004.
    39. Modi V, McDade S, Lallement D, Saghir J. Energy services for millennium development goals. New York: Energy Sector Management Assistance Programme, United Nations Development Programme, UN Millennium Project, and World Bank; 2006.
    40. Molesworth AM, Cuevas LE, Connor SJ, Morse AP, Thomson MC. Environmental risk and meningitis epidemics in Africa. Emerging Infectious Diseases. 2003;9(10):1287–1293. [PMC free article: PMC3033099] [PubMed: 14609465]
    41. Morris SS, Neidecker-Gonzales O, Carletto C, Munguia M, Medina JM, Wodon Q. Hurricane Mitch and the livelihoods of the rural poor in Honduras. World Development. 2002;30(1):49–60.
    42. NOAA (National Oceanic and Atmospheric Administration) Climate Prediction Center. El Niño/Southern Oscillation (ENSO) diagnostic discussion. 2008. [accessed June 16, 2008]. http://www​.cpc.ncep.noaa​.gov/products/analysis_monitoring​/enso_advisory​/ensodisc.html.
    43. Pascual M, Ahumada JA, Chaves LF, Rodo X, Bouma M. Malaria resurgence in the east African highlands: temperature trends revisited. Proceedings of the National Academy of Sciences. 2006;103(15):5829–5834. [PMC free article: PMC1416896] [PubMed: 16571662]
    44. Patz JA. A human disease indicator for the effects of recent global climate change. Proceedings of the National Academy of Sciences. 2002;99(20):12506–12508. [PMC free article: PMC130486] [PubMed: 12271143]
    45. Patz JA, McGeehin MA, Bernard SM, Ebi KL, Epstein PR, Grambsch A, Gubler DJ, Reither P, Romieu I, Rose JB, Samet JM, Trtanj J. The potential health impacts of climate variability and change for the United States: executive summary of the report of the health sector of the U.S. National Assessment. Environmental Health Perspectives. 2000;108(4):367–376. [PMC free article: PMC1638004] [PubMed: 10753097]
    46. Randolph SE. The shifting landscape of tick-borne zoonoses: tick-borne encephalitis and Lyme borreliosis in Europe. Philosophical Transactions of the Royal Society of London. Series B Biological Sciences. 2001;356(1411):1045–1056. [PMC free article: PMC1088499] [PubMed: 11516382]
    47. Reiter P, Thomas CJ, Atkinson PM, Hay SI, Randolph SE, Rogers DJ, Shanks GD, Snow RW, Spielman A. Global warming and malaria: a call for accuracy. Lancet Infectious Diseases. 2004;4(6):323–324. [PMC free article: PMC3130128] [PubMed: 15172336]
    48. Rodo X, Pascual M, Fuchs G, Faruque AS. ENSO and cholera: a nonstationary link related to climate change? Proceedings of the National Academy of Sciences. 2002;99(20):12901–12906. [PMC free article: PMC130557] [PubMed: 12228724]
    49. Saunders MA, Lea AS. Large contribution of sea surface warming to recent increase in Atlantic hurricane activity. Nature. 2008;451(7178):557–560. [PubMed: 18235498]
    50. Shepherd-Johnson D. Drought 2006—boy’s long journey to treatment. 2006. [accessed May 12, 2008]. http://www​​/somalia/reallives_2557.html.
    51. Shiklomanov IA. Appraisal and assessment of world water resources. Water International. 2000;25:11–32.
    52. Singh RB, Hales S, de Wet N, Raj R, Hearnden M, Weinstein P. The influence of climate variation and change on diarrheal disease in the Pacific islands. Environmental Health Perspectives. 2001;109(2):155–159. [PMC free article: PMC1240636] [PubMed: 11266326]
    53. Spennemann DH, Marschner IC. The association between El Niño/Southern Oscillation events and typhoons in the Marshall Islands. Disasters. 1995;19(3):194–197. [PubMed: 7552109]
    54. Stern N. The Stern review: the economics of climate change. Cambridge, UK: Cambridge University Press; 2006.
    55. Stott PA, Stone DA, Allen MR. Human contribution to the European heatwave of 2003. Nature. 2004;432(7017):610–614. [PubMed: 15577907]
    56. Sultan B, Labadi K, Guegan JF, Janicot S. Climate drives the meningitis epidemics onset in west Africa. PLoS Medicine. 2005;2(1):e6. [PMC free article: PMC545199] [PubMed: 15696216]
    57. Tanser FC, Sharp B, le Sueur D. Potential effect of climate change on malaria transmission in Africa. Lancet. 2003;362(9398):1792–1798. [PubMed: 14654317]
    58. UNDP (United Nations Development Programme). Human development report 2007/2008. Fighting climate change: human solidarity in a divided world. New York: United Nations Development Programme; 2007.
    59. Vittor AY, Gilman RH, Tielsch J, Glass G, Shields T, Lozano WS, Pinedo-Cancino V, Patz JA. The effect of deforestation on the human-biting rate of Anopheles darlingi, the primary vector of falciparum malaria in the Peruvian Amazon. American Journal of Tropical Medicine and Hygiene. 2006;74(1):3–11. [PubMed: 16407338]
    60. WHO (World Health Organization). The world health report 2002: reducing risks, promoting healthy life. Geneva, Switzerland: World Health Organization; 2002. [PubMed: 14741909]
    61. Woodruff RE, Guest CS, Garner MG, Becker N, Lindesay J, Carvan T, Ebi K. Predicting Ross River virus epidemics from regional weather data. Epidemiology. 2002;13(4):384–393. [PubMed: 12094092]
    62. Zhou G, Minakawa N, Githeko AK, Yan G. Association between climate variability and malaria epidemics in the east African highlands. Proceedings of the National Academy of Sciences. 2004;101(8):2375–2380. [PMC free article: PMC356958] [PubMed: 14983017]

    The Center for Health and the Global Environment References

    1. Emanuel K. Increasing destructiveness of tropical cyclones over the past 30 years. Nature. 2005;436:686–688. [PubMed: 16056221]
    2. Kerr Richard A. Is Katrina a harbinger of still more powerful hurricanes? Science. 2005;309:1807. [PubMed: 16166493]
    3. Swiss Re. sigma 1/2005. Swiss Reinsurance Company; Zurich: 2005a. Natural Catastrophes and Man-Made Disasters in 2004.
    4. Trenberth K. Uncertainty in hurricanes and global warming. Science. 2005;308:1753–1754. [PubMed: 15961661]
    5. Webster PJ, Holland GJ, Curry JA, Chang H-R. Changes in tropical cyclone number, duration, and intensity in a warming environment. Science. 2005;309:1844–1846. [PubMed: 16166514]

    Burke References

    1. Cummings DAT, Irizarry RA, Huang NE, Endy TP, Nisalak A, Ungchusak K, Burke DS. Traveling waves in the occurrence of dengue haemorrhagic fever in Thailand. Nature. 2004;427(6972):344–347. [PubMed: 14737166]
    2. Dowell SF. Seasonal variation in host susceptibility and cycles of certain infectious diseases. Emerging Infectious Diseases. 2001;7(3):369–374. [PMC free article: PMC2631809] [PubMed: 11384511]
    3. Dushoff J, Plotkin JB, Levin SA, Earn DJD. Dynamical resonance can account for seasonality of influenza epidemics. Proceedings of the National Academy of Sciences. 2004;101(48):16915–16916. [PMC free article: PMC534740] [PubMed: 15557003]
    4. Finkelman BS, Viboud C, Koelle K, Ferrari MJ, Bharti N, Grenfell BT. Global patterns in seasonal activity of influenza A/H3N2, A/H1N1, and B from 1997 to 2005: viral coexistence and latitudinal gradients. PLoS One. 2007;2(12):e1296. [PMC free article: PMC2117904] [PubMed: 18074020]
    5. Fisman DN. Seasonality of infectious diseases. Annual Review of Public Health. 2007;28:127–143. [PubMed: 17222079]
    6. Gore AA Jr. An inconvenient truth. Guggenheim D, editor. Hollywood, CA: Paramount; 2006.
    7. Huang NE, Shen Z, Long SR, Wu MC, Shih HH, Zheng Q, Yen N-C, Tung CC, Liu HH. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proceedings of the Royal Society of London A. 1998;454(1971):903–995.
    8. IPCC (Intergovernmental Panel on Climate Change). Contribution of Working Group II to the fourth assessment report of the Intergovernmental Panel on Climate Change. Chapter 8. Cambridge, UK: Cambridge University Press; 2007. Climate change 2007: impacts, adaptation, and vulnerability.
    9. Lofgren E, Fefferman NH, Naumov YN, Gorski J, Naumova EN. Influenza seasonality: underlying causes and modeling theories. Journal of Virology. 2007;81(11):5429–5436. [PMC free article: PMC1900246] [PubMed: 17182688]
    10. Lowen AC, Mubareka S, Palese P. Influenza virus transmission is dependent on relative humidity and temperature. PLoS Pathogens. 2007;3(10):1470–1476. [PMC free article: PMC2034399] [PubMed: 17953482]
    11. NRC (National Research Council). Under the weather: climate, ecosystems, and infectious diseases. Washington, DC: National Academy Press; 2001. [PubMed: 25057537]
    12. Schulman JL, Kilbourne ED. Transmission of influenza virus infection in mice. II. Some factors affecting the incidence of transmitted infection. Journal of Experimental Medicine. 1963;118:267–275. [PMC free article: PMC2137714] [PubMed: 14074390]
    13. WHO (World Health Organization). Global atlas of infectious diseases. 2008. [accessed March 24, 2008]. http://gamapserver​.who​.int/GlobalAtlas/home.asp.



This article was originally published in the New England Journal of Medicine 353(14):1433–1436 (2005). Reprinted with permission from the Massachusetts Medical Society. Copyright 2005. All rights reserved. An interview with Dr. Epstein can be heard at www​ Dr. Epstein is the associate director of the Center for Health and the Global Environment, Harvard Medical School, Boston.


Houghton JT, Ding Y, Griggs DJ, et al., eds. Climate change 2001: the scientific basis: contribution of the Working Group I to the third assessment report of the Intergovernmental Panel on Climate Change. Cambridge, England: Cambridge University Press, 2001.


Hassol SJ. ACIA, Impacts of a warming Arctic: arctic climate impact assessment. Cambridge, England: Cambridge University Press, 2004.


Leaf A. Potential health effects of global climatic and environmental changes. N Engl J Med 1989; 321:1577–83.


McMichael AJ, Campbell-Lendrum DH, Corvalán CF, et al., eds. Climate change and human health: risks and responses. Geneva: World Health Organization, 2003:250.


Epstein PR, Diaz HF, Elias S, et al. Biological and physical signs of climate change: focus on mosquito-borne diseases. Bull Am Meteorol Soc 1998;78:409–17.


Reprinted with permission from the Center for Health and the Global Environment. 2005. Climate change futures: health, ecological and economic dimensions. Cambridge, MA: Harvard Medical School. Sponsored by Swiss Re and the United Nations Development Programme.


Dean, Graduate School of Public Health.

Copyright © 2008, National Academy of Sciences.
Bookshelf ID: NBK45742


Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...