Climate change and disability-adjusted life years.

Introduction

Climate change was selected as a risk factor in the category of "environmental risks" in the World Health Organization (WHO) global-burden-of-disease (GBD) study, reflecting increasing concern about global warming and its impact on population health (Mathers et al., 2004). By comparison with other factors, such as tobacco smoking, alcohol consumption, sanitation, and hygiene, climate change has not been treated as a major risk factor contributing to the global burden of diseases (Ezzati et al., 2003). This circumstance reflects the limited number of studies conducted in this research field.

[ILLUSTRATION OMITTED]

National burden-of-disease studies have been conducted in some countries, including Australia, New Zealand, Chile, South Africa, Sweden, and the Netherlands, but the disability-adjusted life years (DALYs) lost because of climate change have not been investigated in these countries (Bradshaw et al., 2003; Concha, 2004; Mathers, Vos, & Stevenson, 1999; New Zealand Ministry of Health, 2001; Smith, 2000). The DALY is a summary measure that combines information on mortality and disability to represent population health as a single number (Gold, Stevenson, & Fryback, 2002). Although there is controversy in the calculation of DALYs, this measurement is the most commonly used index of health status in global-burden-of-disease studies. DALYs have not, however, been used as a health outcome in most research examining climate change and its contribution to population health. Using DALYs attributable to climate change would make it possible to compare these risks with others identified in global-burden-of-disease studies to establish health priorities and evaluate the efficiency of environmental policies (Hollander, Melse, Lebret, & Kramers, 1999).

Many studies examining association between climate and human health use extra deaths/mortality as the stated health outcome (Keatinge & Donaldson, 2004). In some cases, however, the death rate for a disease is relatively low (e.g., infectious diseases in a developed country) compared with the total number of affected people. Therefore, it is inappropriate to use mortality figures alone to estimate the impact of climate change on human health. This article will 1) review the association between DALYs and climate change in burden-of-disease studies, 2) assess the methodological issues and challenges in such studies, and 3) discuss further research directions.

Climate Change and DALYs in Global-Burden-of-Disease Studies

The WHO GBD study provided useful results in measuring global health status and attributable risk factors (WHO, 2004). Very few studies have, however, examined the relationship between climate change and disease burden, although the impact may be substantial. According to calculations taken from the 2000 global-burden-of-disease study (WHO, 2004), 5,517,000 DALYs were attributable to climate change worldwide. This contribution to the global burden of disease is relatively small compared with contributions from other risk factors, such as tobacco (59,081,000 DALYs) and unsafe water, sanitation, and hygiene (54,158,000 DALYs) (Ezzati, Lopez, Rodgers, Vander, & Murray, 2002). The geographic distribution of DALYs lost because of climate change is uneven across the world, with almost 50 percent of the DALYs lost in regions with high child and adult mortality in Southeast Asia, followed by the losses in regions in Africa (23 percent) and the Eastern Mediterranean (14 percent) (WHO, 2002a). Given the size of its population, Africa has the largest number of DALYs lost per 100,000 inhabitants as a result of climate change (Table 1). It should be acknowledged that because of the lack of research in developing countries on this topic, actual DALYs may be much higher. According to the limited results from WHO, children under five years of age suffer the most from the consequences of climate change, with 88 percent of lost DALYs attributable to climate change occurring in this age group in both developed and developing countries. This proportion is higher than that observed for other environmental risks, including unsafe water, urban air pollution, indoor smoke, and lead exposure (WHO, 2002b) (Table 2).

Chronic disease is one of the leading public health issues in the world, with main risk factors including smoking, physical inactivity, obesity, high blood pressure, and poor nutrition (Centers for Disease Control and Prevention, 2005; Mathers et al., 1999). Climate change with increased temperature might bring about more extreme weather events, which may have effects on both infectious and chronic diseases, including chronic respiratory disorders, heat- or cold-related illnesses such as cardiovascular diseases, injuries, and psychological disorders (McMichael, Haines, Slooff, & Kovats, 1996). Climate change may increase the mortality burden of such chronic diseases, particularly for children, elderly, and frail people. Unfortunately, so far, there has been no study examining the DALYs lost because of such chronic diseases attributable to climate change. Obviously, climate change has not been treated as an "important" risk factor within the burden-of-disease studies for chronic diseases. It should be emphasized that the potential risks of extreme weather and catastrophic climate will certainly increase and will result in more damage to human health (Folland & Karl, 2001). The 2003 European heat wave and extra mortalities were a typical example (Stott, Stone, & Allen, 2004; Grynszpan, 2003).

Infectious diseases affect a large number of people around the world, especially in developing countries (WHO, 2002c). It is therefore important to understand the underlying causes of the burden of infectious diseases. The researchers for the 2000 global-burden-of-disease study made great efforts to answer this question by examining many potential risk factors, including climate change. It was estimated that 2 percent of DALYs lost because of diarrheal diseases and 2 percent of DALYs lost because of malaria could be attributed to climate change (WHO, 2004). It was also expected that the risk of diarrhea could be up to 10 percent higher by 2030 in regions experiencing climate change, compared with regions without such change (WHO, 2003). Since only limited studies have explored such exposure-response relationships, these estimates remain uncertain. Furthermore, studies of DALYs from other relevant infectious diseases attributable to climate change, such as dengue, Japanese encephalitis, West Nile virus infections, and schistosomiasis have not been reported.

The WHO study of the global burden of disease is of great value. The limitations of the study should be acknowledged, however. First, only four conditions related to climate change were included in the study: diarrheal diseases, malaria, other unintentional injuries, and protein-energy malnutrition. Climate change has diverse effects on population health by affecting the environment and pathogen and human behaviors (McMichael, Haines, Slooff, & Kovats, 1996). In addition to the four conditions included in the GBD study, other health burdens are related to climate change, including vectorborne, waterborne, and foodborne diseases; chronic respiratory disorders; heat- or cold-related illness and deaths; injuries; and psychological disorders. The selection mainly reflected the availability of data, and it under-represented the biggest climate-related causes of disease. Second, the assessment time frame may be too short. In the global-burden-of-disease assessment, the estimated impacts of climate change have been projected only as averages for a 30-year period centered in the 2020s and 2050s. Further studies should estimate the long-term and persistent impacts of climate change beyond the 2020s-to-2050s range, depending on the accuracy of future climatic and demographic scenarios. Third, estimating the burden of disease at a global level using "global" criteria is necessary, but the results may not be sufficiently precise for government policy makers. Therefore, the burden of disease attributable to climate change should be examined at the national level, the state/provincial level, or both to provide more specific evidence and policy suggestions for local decision makers.

Methodological Issues

Environmental Burden-of-Disease Studies

The calculation of DALYs attributable to climate change is based on the methods used in the environmental-burden-of-disease studies, in which 135 major exposures were evaluated (WHO, 2004). Three categories of data are required in an assessment of environmental burden of disease, namely the distribution of exposure to risk factors among various populations, information on exposure-response relationships, and data on DALYs lost because of the risk factors (Pruss-Ustun, Mathers, Corvalan, & Woodward, 2000). Generally, two approaches have been used in environmental-burden-of-disease studies: exposure-based and scenario-based (Pruss-Ustun et al.). While the exposure-based approach uses current measurements for analysis, the scenario-based approach selects characteristic exposure scenarios to compare current levels of exposure to those that would occur under alternative, hypothetical scenarios. This approach is necessary when it is not possible to specify a direct relationship between the proximal cause of disease and the disease outcome. The burden of disease attributable to climate change, a distal cause of diseases, could be examined by the scenario-based approach.

Calculating DALYs Lost Because of Climate Change

A scenario-based calculation of the environmental burden of disease attributable to risk factors would include four steps: defining exposure scenarios, determining different levels of exposure in study populations, calculating the relative risk for each scenario, and calculating the impact fraction and disease burden (Pruss-Ustun et al., 2000). With this approach, Valent and co-authors (2004) studied the burden of disease attributable to inadequate water supply among children and adolescents in Europe. Pruss, Kay, Fewtrell, and Bartram (2002) used a similar approach to estimate the burden of disease from water, sanitation, and hygiene at a global level using nine different scenarios.

On the basis of the approach described above, the comparative risk assessment (CRA) module has been used to evaluate burden of disease due to climate change. Using the CRA module, McMichael, Campbell-Lendrum, and Kovats (2000) included four procedures in their calculation: determining the scenarios of greenhouse gas emissions, predicting future climate change by the general circulation model, estimating the health impact of each climate scenario, and then converting to global-burden-of-disease indicators. A more detailed account of these procedures is given in Table 3. The findings of this work were published in a report by WHO (2004) and are summarized in Table 1 and Table 2.

Challenges in the Study of Burden of Disease Due to Climate Change

Difficulties are inherent in the measurement of the burden of disease due to climate change. First, exposure scenarios associated with climate change are difficult to measure because there is no clear minimum exposure group to be used as a reference. In the global-burden-of-disease study, averagelevels of greenhouse gases in 1961-1990 were chosen as the theoretical minimum scenarios, with alternative scenarios for comparison based on the density of carbon dioxide during this period (Kay, Pruss, & Corvalan, 2000) Second, it is difficult to identify appropriate health indicators for the assessment of the consequences of climate change. The commonly selected indicators of health status in this area are death/mortality (El-Zein, Tewtel-Salem, & Nehme, 2004; O'Neill, Zanobetti, & Schwartz, 2003; Simon, Lopez-Abente, Ballester, & Martinez, 2005). Summary measurements such as DALYs have not been calculated except in a recent WHO report (WHO, 2004). Third, it is very difficult to demonstrate a direct exposure-response relationship between climate variability/climate change and disease outcome, as many intermediate steps may be involved.

The impact of climate on human health has certainly been underestimated. Many studies have estimated the extra death/mortality attributable to extreme weather since the 1970s (Lye & Kamal, 1977; Macfarlane & Waller, 1976; Watson & Team, 2001). Calculating extra deaths, however, is not enough to assess the impacts of climate change, given the large morbidity burden from infectious diseases and injuries caused by climate change (Kovats, Campbell-Lendrum, & Matthies, 2005). So far, there are no studies reporting morbidity or DALYs lost because of chronic disease--for example, cardiovascular diseases and respiratory diseases, which account for most deaths in the world--attributable to climate change. Moreover, because chronic diseases affect people over long periods of time, it is unclear how many effects could be attributable to climate change during the whole course of the disease process.

Climate change and variation have affected and will continue to affect the transmission of infectious diseases (Anyamba, Chretien, Small, Tucker, & Linthicum, 2006). For infectious diseases, however, including both vectorborne and foodborne diseases, it is even more difficult to estimate DALYs attributable to climate change, which is just one distal cause of such diseases. Many factors--including host characteristics, vectors, environment, population density, socioeconomic status, and vaccination--contribute to the transmission of infectious diseases. For example, in the epidemiology of malaria transmission, ambient temperature, precipitation, and humidity may be chosen as climatic indicators for the estimation of the risk of malaria rather than changes in the mosquitos population, for which routine data is usually unavailable on a long time-scale (Martens, Niessen, Rotmans, Jetten, & McMichael, 1995). Therefore, the situation presents a need to investigate the burden of infectious diseases attributable to climate change.

There are still other challenges for climate change researchers, such as what to do about uncertainty regarding interaction between socioeconomic status, adaptation measurements, and climate change; how to evaluate the impacts of climate change that may occur in the more distant future; and how to deal with nonlinear changes in global climate systems and great data problems (i.e., lack of necessary information for the most vulnerable parts of the world).

Further Directions

It is vital to quantify the burden of disease attributable to climate change at national and local levels because of increased concerns about the impact of climate change, especially as the Kyoto Protocol has just come into force (United Nations Framework Convention on Climate Change, 2005), restricting the global emission of greenhouse gases. Unfortunately, research on DALYs and climate change is very limited. Future studies, at both national and local levels, should assess the relationship between climate change and diseases using different climate scenario models and appropriate analyses. In addition, studies linking DALYs with climate change should be conducted in various populations in different ecological regions, particularly in developing countries. Such studies could include both exposure-based and scenario-based analyses. Historical-data analysis may be required for a better understanding of DALYs attributable to climate change. Projections for future risks under different scenarios are necessary and more significant for local decision makers. Closer attention should be paid in these studies to children, who are most affected by climate change, especially in developing countries (WHO, 2004).

In conclusion, quantitative studies on the burden of disease attributable to climate change are very scarce. DALYs lost because of climate change in the world could be very high, especially in developing countries. The challenge lies in modeling complex causal relationships and in multidisciplinary cooperation. There is a long way to go in the investigation of the DALYs attributable to climate change, an undertaking that is needed to protect the health of the most disadvantaged of the world.

Corresponding Author: Ying Zhang, Department of Public Health, University of Adelaide, Level 9,10 Pulteney Street, Adelaide, South Australia. E-mail: [e-mail].

REFERENCES

Anyamba, A., Chretien, J.P., Small, J., Tucker, C.J., & Linthicum, K.J. (2006) Developing global climate anomalies suggest potential disease risks for 2006-2007. International Journal of Health Geographies, 28, 60-68.

Bradshaw, D., Groenewald, P., Laubscher, R., Nannan, N., Nojilana, B., Norman, R., Pieterse, D., & Schneider, M. (2003). Initial burden of diseases estimates from South Africa, 2000. Cape Town: South African Medical Research Council.

Centers for Disease Control and Prevention. (2004). The burden of chronic diseases and their risk factors. Retrieved December 4, 2005, from www.cdc.gov/nccdphp/burdenbook2004/toc.htm.

Concha, M. (2004). Burden of disease in Chile--Global forum for health research. Retrieved November 15, 2004, from www.globalforumhealth.org/Non_compliant_pages/forum3/forum3doc823.htm.

El-Zein, A., Tewtel-Salem, M., & Nehme, G. (2004). A time-series analysis of mortality and air temperature in Greater Beirut. Science of the Total Environment, 330, 71-80.

Ezzati, M., Hoorn, S.V., Rodgers, A., Lopez, A.D., Mathers, C.D., & Murray, C.J.L. (2003). Estimates of global and regional potential health gains from reducing multiple major risk factors. Lancet, 362, 271-275.

Ezzati, M., Lopez, A.D., Rodgers, A., Vander, H., Murray, C.J.L. (2002). Selected major risk factors and global and regional burden of disease. Lancet, 360, 1347-1350.

Folland, C., & Karl, T. (2001). Observed climate variability and change. In: Climate change 2001: The scientific basis. Cambridge, U.K.: Cambridge University Press.

Grynszpan, D. (2003). Lessons from the French heatwave. Lancet, 362, 1169-1172.

Gold, M.R., Stevenson, D., & Fryback, D.G. (2002). HALYs and QALYs and DALYs, oh my: Similarities and differences in summary measures of population health. Annual Review of Public Health, 23, 115-121.

Hollander, A.E.Md., Melse, J.M, Lebret, E., & Kramers, P.G.N. (1999). An aggregate public health indicator to represent the impact of multiple environmental exposures. Epidemiology, 10, 606-617.

Pruss-Ustun, A., Mathers, C., Corvalan, C., & Woodward, A. (2000). Introduction and methods: Assessing the environmental burden of disease at national and local levels. (Environmental burden of disease series, No. 1). Retrieved August 7, 2006, from http://www.who.int/quantifying_ehimpacts/publications/9241546204/en/.

Kay, D., Pruss, A., & Corvalan, C. (2000). Methodology for assessment of environmental burden of disease. Geneva: ISEE Session on Environmental Burden of Disease, WHO.

Keatinge, W.R., & Donaldson, G.C. (2004). The impact of global warming on health and mortality. Southern Medical Journal, 97(11), 1093-1099.

Kovats R.S., Campbell-Lendrum, D., & Matthies, F. (2005) Climate change and human health: Estimating avoidable deaths and disease. Risk Analysis, 25, 1409-1418.

Lye, M., & Kamal, A. (1977). Effects of a heat wave on mortality rates in elderly in-patients. Lancet, 1, 529-531.

Macfarlane, A., & Waller, R. (1976). Short-term increases in mortality during heat waves. Nature, 264, 434-436.

Martens, W, Niessen, L., Rotmans, J., Jetten, T., & McMichael, A. (1995). Potential impact of global climate change on malaria risk. Environmental Health Perspectives, 103, 458-464

Mathers, C., Vos, T., & Stevenson, C. (1999). The burden of disease and injury in Australia. Canberra: Australian Institute of Health and Welfare.

Mathers, C.D., Bernard, C., Iburg, K.M, Inoue, M., Fat, D.M., Shibuya, K., Stein, C., Tomijima, N., & Hongyi, X. (2004). Global burden of disease in 2002: Data sources, methods and results (Global Programme on Evidence for Health Policy Discussion, Paper No. 54). Retrieved August 7, 2007, from http://www.who.int/healthinfo/paper54.pdf.

McCarthy, J.J., Canziani, O.F., Leary, N.A., Dokken, D.J., White, K.S., eds. (2001). Climate change 2001: impacts, adaptation, and vulnerability. Cambridge: Cambridge University Press.

McMichael, A., Campbell-Lendrum, D., & Kovats, S. (2000). Annex 6.7: Comparative risk assessment of the health effects of climate change. In: 12th annual meeting of the international society for environmental epidemiology (ISEE). Retrieved August 7, 2007, from http://www.who.int/quantifying_ehimpacts/methods/en/wsh0007an6.pdf.

McMichael A.J., HainesA., Slooff, R., & Kovats,S. (Eds.). (1996). Climate change and human health. Geneva: World Health Organization.

New Zealand Ministry of Health. (2001). The burden of disease and injury in New Zealand public health intelligence (Occasional Bulletin No. 1). Wellington, New Zealand: The Ministry of Health.

O'Neill, M., Zanobetti, A., & Schwartz, J. (2003). Modifiers of the temperature and mortality association in seven U.S. cities. American Journal of Epidemiology, 15, 1074-1082.

Pruss, A., Kay, D., Fewtrell, L., & Bartram, J. (2002). Estimating the burden of disease from water, sanitation, hygiene at a global level. Environmental Health Perspectives, 110, 537-542.

Stott, PA., Stone, D.A., & Allen, M.R. (2004). Human contribution to the European heatwave of 2003. Nature, 432, 610-614.

Simon, F., Lopez-Abente, G., Ballester, E., & Martinez, E (2005). Mortality in Spain during the heat waves of summer 2004. Eurosurveillance, 10, 156-161.

Smith, K.R. (2000). National burden of disease in India from indoor air pollution. Proceedings of the National Academy of Sciences of the United States of America (PNAS), 97, 13286-13293.

United Nations Framework Convention on Climate Change. (2005). Retrieved March 29, 2005, from http://unfccc.int/2860.php.

Valent, F., Little, D., Bertollini, R., Nemer, L., Barbone, F., & Tamburlini, G. (2004). Burden of disease attributable to selected environmental factors and injury among children and adolescents in Europe. Lancet, 363, 2032-2039.

Watson, R.T., & Team, C.W. (Eds.). (2001). Climate change 2001: Synthesis report. Cambridge, UK: Cambridge University Press.

World Health Organization. (2003). Climate change and human health--Risks and responses [Summary]. Geneva: Author.

WHO. (2002a). Attributable DALYs by risk factor and WHO subregion. Retrieved December 11, 2004, from http://www.who.int/quantifying_ehimpacts/global/en/.

WHO. (2002b). Distribution of attributable mortality and DALYs by risk factor, age and sex. Retrieved December 11, 2004, from http://www.who.int/quantifying_ehimpacts/global/en/.

WHO. (2002c) World Health Report. Retrieved December 3, 2004, from http://www3.who.int/whosis/menu.cfm?path=whosis,burden&language=english.

WHO. (2004). Global Burden of Disease results for theyears 2000, 2001 and 2002, vol. 2004. Retrieved December 12, 2004, from: http://www3.who.int/whosis/menu.cfm?path=whosis,burden,burden_estimates&language=english.

Although most of the information presented in the Journal refers to situations within the United States, environmental health and protection know no boundaries. The Journal periodically runs International Perspectives to ensure that issues relevant to our international constituency, representing over 60 countries worldwide, are addressed. Our goal is to raise diverse issues of interest to all our readers, irrespective of origin.

Ying Zhang, M.B.B.S., M.Med.Sci.

Peng Bi, M.B.B.S., Ph.D.

Janet E. Hiller, Ph.D.

TABLE 1 Morbidity Burdens (DALYs) Attributable to Climate Change in the
WHO Subregions (a) in 2000

                      Africa  Africa  America  America
               World  D       E       A        B

DALYs          5,517  1,267   626     3        71
  (X,000)
DALYs             90    358   207     1        16
  per 100,000

                        Eastern        Eastern
               America  Mediterranean  Mediterranean  Europe  Europe
               D        B              D              A       B

DALYs          23       20             748            3       10
  (X,000)
DALYs          32       14             213            1        5
  per 100,000

                       Southeast  Southeast  Western  Western
               Europe  Asia       Asia       Pacific  Pacific
               C       B          D          A        B

DALYs          4       34         2,538      1        169
  (X,000)
DALYs          2       11           201      1         11
  per 100,000

(a) Definitions of WHO subregions: A = regions having very low child and
very low adult mortality, B = regions having low child and low adult
mortality, C = regions having low child and high adult mortality, D =
regions having high child and high adult mortality, and E = regions
having high child and very high adult mortality.
Data retrieved from Attributable DALYs by Risk Factor and WHO Subregion
(WHO, 2002a).

TABLE 2 Age and Sex Distribution of Global Morbidity Burdens (DALYs)
Attributable to Selected Environmental Risk Factors, 2000

                           Distribution of Attributable DALYs
Environmental                       Age               Sex
Risk Factor                0-4  5-14  15-59  60+  Female  Male

Unsafe water, sanitation,  77%  8%    13%     3%  51%     49%
  and hygiene
Urban air pollution        12%  0%    40%    49%  56%     44%
Indoor smoke from          83%  0%     8%     9%  49%     51%
  solid fuels
Lead exposure              75%  0%    16%     8%  55%     45%
Climate change             88%  5%     6%     1%  49%     51%

Data retrieved from Distribution of Attributable Mortality and DALYs by
Risk Factor, Age and Sex (WHO, 2002b).

TABLE 3 Suggested Four Procedures for the Calculation of DALYs
Attributable to Climate Change

Step                     Explanation

1. Greenhouse gas        Emission scenarios are a plausible
   emission scenarios    representation of future emissions of
                         greenhouse gases, which include low, medium,
                         and high emissions (Watson & Team, 2001).
2. General circulation   General circulation models, a complex model
   models                based on chemistry and biological properties,
                         are used to predict future climate change
                         (Watson & Team, 2001).
3. Health impact models  Health impact of each climate scenario on
                         specific health outcomes could be estimated
                         using various abiotic and biotic models, such
                         as integrated assessment modeling (McCarthy,
                         Canziani, Leary, Dokken, & White, 2001).
4. Convert to GBD (a)    Convert to indicators commonly used in GBD
   indicators            studies, such as DALYs, on the basis of GBD
                         study methods (WHO, 2004).

(a) GBD = global burden of disease.