Education And Debate

Global climate change: the potential effects on health

BMJ 1997; 315 doi: (Published 27 September 1997) Cite this as: BMJ 1997;315:805
  1. Anthony J McMichael (t.mcmichael{at},
  2. Andrew Hainesb, professor of primary health care
  1. a Department of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, London WC1E 7HT
  2. b Joint Department of Primary Care and Population Sciences, Royal Free Hospital School of Medicine and University College London Medical School, London NW3 2PF
  1. Correspondence to: Professor McMichael


    Excess carbon dioxide, methane, and other gases which trap heat are accumulating in the troposphere, the earth's lower atmosphere, because of the scale and type of human economic activity. Climate scientists predict that the resultant increase in the troposphere's “radiative forcing” will warm the earth's surface.1 2 3 Indeed, in its recent second assessment report, the Intergovernmental Panel on Climate Change—a multidisciplinary scientific body established by the United Nations in 1988 to advise governments—concluded that on balance an anthropogenic influence upon the global climate was now “discernible.” 1

    The intergovernmental panel forecasts an increase in the average world temperature of 1.0-3.5°C over the coming century.1 This forecast is necessarily uncertain because the sensitivity of climate to atmospheric change is imperfectly understood and because future trends in gaseous emissions and modulating processes (for example, the cooling effects of industrial aerosol emissions) cannot be foreseen accurately. Nevertheless, the expected rate of climate change over the coming century would be far greater than any natural change in world climate since the advent of agriculture 10 000 years ago.

    Anthropogenic climate change signifies that for the first time the aggregate global impact of humankind exceeds the physical and ecological limits of the biosphere.4 The potential consequences of this and other global changes (including stratospheric ozone depletion, loss of biodiversity, worldwide land degradation, and depletion of aquifers) are wide ranging. We can expect that climate change will affect the health and wellbeing of human populations in diverse ways. This greatly extends the temporo-spatial scale of environmental health beyond our usual concern with localised and immediate exposures to toxic or infectious agents.4 A major research task, therefore, is the application of current knowledge to forecasting probable health effects. The primary objective is to provide indicative forecasts of an important consequence that will assist pre-emptive policy making.5

    Evaluating scientific reports on climate change

    Climate change is a new and rapidly developing topic of scientific inquiry and health risk assessment. We were involved in the detailed review of published studies needed for the chapter on health impact in the second assessment report of the Intergovernmental Panel on Climate Change.10 Preparation of the report was subject to wide ranging international peer review with contributions, corrections, and criticisms from over 60 scientists worldwide. That text was then expanded to form a book of 10 chapters.7 This entailed a further intensive review process which included scientists from various United Nations agencies. In preparing this article immediately after those two reviews we have updated information in light of other recent publications.

    Summary points

    Most climatologists believe that global climate change is now occurring as a result of emission of greenhouse gases, especially from fossil fuel combustion. Its expected rate over the coming century will far exceed that of any other climatic change in the 10 000 years since agriculture and human settlement began

    Changes in climate conditions and, possibly, weather variability would affect human health through several processes, many mediated by disturbances of ecological systems

    Changes in the environment to which human biology and culture are adapted or disturbances of ecosystems that set the conditions for health would generally have adverse effects on health

    More readily foreseen effects (resulting from thermal stresses and changed patterns of air pollution) will probably have less impact than will those resulting from complex changes in ecological infrastructure (altered patterns of infectious dis- ease, agricultural production, coastal ecology, etc)

    The types and circumstances of exposure would vary geographically. Populations would vary in their vulnerability because of differences in location, social and technical resources, and concurrent health status

    Assessing potential health impact: issues of context

    Forecasting the potential impact of climate change on health calls for the development of risk assessment methods based on scenarios.5 Details are discussed in the paper to follow (Oct 3),6 but it is necessary to note three important issues of context which make this a challenging task. Firstly, the assessment of health risks based on scenarios must accommodate much unavoidable uncertainty, both in the scenarios themselves and in our knowledge of the response of the various linked systems in the causal chain.5 7 Secondly, many of the health effects would occur as a result of disturbances in complex, non-linear ecological systems which are difficult to model. In addition, the impact of climatic change will depend on interactive effects with coexistent global change processes (see fig 1). Thirdly, the vulnerability of populations is determined by a wide variety of factors (box).

    Fig 1
    Fig 1

    Global environmental change—causes and consequences. Climate change is shown as part of a larger complex of changes

    Factors determining the vulnerability of populations8

    • Poverty and the associated lack of resources and technical infrastructure

    • Nutritional status (for example, those people of sub-Saharan Africa who lack food security9)

    • Isolation

    • Physical location (for example, populations of the coast or small islands, who are exposed to rising seas and salt water intrusion into soil and fresh water)

    • Cultural inflexibility

    • Political rigidity

    Variations in vulnerability would be compounded by regional differences in climate change. The modelling of climate change on a regional basis is improving. Increases in temperature would generally be greater at higher latitude, and changes in precipitation would vary in such a way that droughts would occur in some regions that are already semiarid while rainfall would increase in others.1

    Potential health effects of climatic change

    The various potential health effects of climatic change—direct and indirect, immediate and delayed—are summarised in table 1.7 10 Some would be beneficial. For example, warmer temperatures would mean milder winters in cool-temperate countries, which should reduce the mortality peak among older people in winter.11 An increase in summer temperatures in currently hot zones might reduce the viability of mosquito populations. Most of the anticipated effects, however, would be adverse. Shifts in climate means and variability would perturb important physical and biological systems to which human health is biologically and culturally adjusted.

    Table 1

    Mediating processes and direct and indirect potential effects on health of changes in temperature and weather

    View this table:

    Direct effects of climate change

    The direct health effects would include changes in mortality and morbidity from heatwaves and thermal stress.12 13 14 Climatologists forecast an increase in the frequency of heatwaves (because of both the rising mean temperature and a possible increase in weather variability). Relatively little is known, however, about the capacity of human populations to adapt physiologically and culturally to such changes over decades. Also, since extremes of heat and cold affect elderly and sick people in particular, the average impact on life expectancy may not be large.10 13

    Other direct effects would include those on respiratory health of altered concentrations of aeroallergens (such as spores and moulds),15 of air pollutants such as ozone produced by photochemical reactions that are sensitive to temperature, and the many different health consequences of an altered tempo of extreme weather events, including storms and floods.7

    Indirect effects

    Indirect health effects, many of which would result from the perturbation of complex ecological systems, are of potentially greater consequence. These include changes in the range and activity of infectious diseases borne by vector organisms;altered transmission of infections from person to person (including food poisoning and transmission of pathogens carried by water); the nutritional and health consequences of regional changes in agricultural productivity; and the various consequences of rising sea levels.7 10 Public health would also be affected by population movements and by regional conflicts over food and water in shortages caused by climatic changes.16 17

    Infectious diseases

    Changes in the risks of infectious disease would occur particularly as a result of climatic influences on the vector organisms for diseases such as malaria, dengue fever, trypanosomiasis, the viral encephalitides (including those in temperate zones), and schistosomiasis.10 18 19 Increased temperatures and changed rainfall would affect the range, proliferation, and behaviour of vector organisms and intermediate hosts and the viability and maturation rates of the infectious agents. While richer, developed countries might be able to maintain public health defences against an extension (or reintroduction) of vector borne infections, populations on the margins of endemic areas in tropical and subtropical countries would be at increased risk.

    Recent empirical (analogue) studies in Zimbabwe, Rwanda, and Ethiopia, which have used natural climate variability to foreshadow aspects of future climate change, indicate that such change would cause malaria to move to higher altitudes and thereby affect highland populations who are currently protected.18 20 21 Other recent analogue studies have confirmed the sensitivity of both malaria and dengue fever to climatic variations between years.22 23 24 Interestingly, there have been reports of a recent upward movement of malaria, dengue fever, or their mosquito vectors in several continents in conjunction with retreating glaciers and ascent of alpine plants.25

    Simulation of scenarios of standard climatic change using integrated mathematical modelling shows that the potential geographical range for transmitting malaria would expand.26 27 28 29 Approximately 45% of the world's population currently live in zones that are defined climatically as areas of potential malaria transmission, and assuming that other factors remain constant this would increase to around 60% towards the end of the next century.10 26 27 Figure 2 shows the potential future transmissibility of malaria under likely conditions of climate change as modelled by the widely used ECHAM-1 general circulation model.26 Similar mathematically modelled projections exist for dengue fever and schistosomiasis.30

    Fig 2
    Fig 2

    Estimated changes in the potential transmission of malaria as a result of climate change brought about by doubling the atmospheric carbon dioxide concentration (from Martens27). Top map shows current distribution of climatic zones, colour coded accorded to the estimated capacity to transmit malaria; middle map shows estimated distribution of equivalent potential transmission zones under climate change later next century; lower map shows difference between these two—that is, the estimated change in potential transmissibility of malaria. Transmission potential is scaled arithmetically from zero to 1 (maximum) in the upper and middle maps and as a multiplication factor (2=doubling) in the lower map

    The highly aggregated models used to make global forecasts necessarily gloss over many regional differences. With sufficient localised information it is possible to scale down and retune the mathematical model to match local conditions. The application of such a model to forecast the changes in potential malaria transmission in Zimbabwe is shown in figure3

    Fig 3
    Fig 3

    Changes in the potential transmission of Plasmodium falciparum malaria in Zimbabwe modelled under three scenarios of climate change (from Martens27). Note that the highland region runs centrally from the north east to the south west. For the baseline climate, malaria prevalence is shown on a percentage scale; for the three other maps the change in transmission potential is shown as a multiplier of that baseline prevalence

    Climate change in temperate regions would also affect some infectious diseases carried by vector organisms10

    • Viral encephalitis borne by ticks: occurs in parts of western Europe and Scandinavia and is sensitive to the climate

    • Leishmaniasis: currently endemic in the rural Mediterranean region of Europe and in the eastern Mediterranean. Climate change may extend the habitat of the sandfly vector northwards

    • Lyme disease: caused by the spirochaete Borrelia burgdorferi, transmitted in Europe and the northeastern United States by the tick Ixodes ricinus. Temperature influences the tick's life cycle of three stages (larvae, nymphs, adults) and, hence, the probability of transmission of the spirochaete

    • Malaria: endemic malaria was widely present in Europe earlier this century.31 It is appropriate to consider how bioenvironmental stresses induced by climate might eventually breach existing protective public health measures.

    Climatic change would also influence various infections which are directly transmitted, especially those caused by contamination of drinking water and food.32 Recent evidence suggests that the spread of cholera is aided by warmer coastal and estuarine waters and their associated algal blooms, since phytoplanktonic and zooplanktonic organisms act as a natural reservoir for Vibrio cholerae.71733

    Food production, hunger, and malnutrition

    Climate change would affect the production of food, especially that of cereal crops. Regional gains and losses would reflect the local balance of changes in temperature, soil moisture, carbon dioxide “fertilisation,” and pest and pathogen activity. Studies using mathematical modelling consistently show that tropical and subtropical countries would be most affected. Furthermore, many communities that rely on traditional agriculture may lack the resources and adaptability to switch to alternative crops and production methods. Although there would be some mid-continental drying in summer months in North America and Europe,73435 agricultural production in high latitude, temperate zones (for example, Canada and much of Siberia) might increase. Long term climatic changes, however, might have a net negative impact on world agriculture.1034

    Poor and economically underdeveloped populations who would be unable to offset reduced agricultural yields by trade would be adversely affected. Hunger and malnutrition increase the risk of infant and child mortality and cause physical and intellectual stunting. Energy levels, work capacity, and health status are reduced in adults. One widely quoted study has forecast the additional number of hungry people attributable to climate change by the year 2060, allowing for plausible future trajectories of demographic, economic, and trade liberalisation processes.34 The estimate varied between an additional 40 million and 300 million relative to a future background total of around 600 million hungry people

    Water supplies, essential for agriculture, animal husbandry, and personal hygiene, may also be adversely affected by climatic change in many regions. Tensions over shortages of fresh water, especially in low to middle latitude locations where adjoining countries share river basins,16 would be exacerbated by changes in rainfall related to the climate.710

    Sea level rise: effects on coastal regions and populations

    The Intergovernmental Panel on Climate Change has forecast that sea level will rise approximately 40 cm by 2100—a rate of increase several times faster than that over the past century. The importance of this forecast lies in the fact that over half the world's population lives within 60 km of the sea. An increase would affect various processes and structures on which food production, economies, and public health depend. The displacement of communities living on the coast or on small islands would be a particular health hazard, especially in poorer, populous countries with few material resources.7936 At today's population level an increase of half a metre would approximately double the number of people who experience flooding annually—currently around 46 million.37 The disposal of sewage and waste water, the physical safety of coastal structures, the viability of coral reefs and wetlands (fish nurseries), and the ecology of some infectious diseases (for example, malaria and cholera) would also be affected.710 A rise in sea level, ocean warming, and changes in currents and nutrient flows would all contribute to alterations in marine ecosystems.10 In relation to health, the consequences include algal blooms (implicated in cholera transmission) and the production of toxins in edible fish and shellfish.10

    International and national developments

    Recognition that climate change may adversely affect human health has prompted a rapid expansion of research and evaluation into the topic. The box shows noteworthy developments during 1996-7.

    Recent developments, 1996-7

    • The Intergovernmental Panel on Climate Change has included a full chapter on potential health impacts in its second assessment report.1 The first report, in 1990, gave scant attention to health

    • The World Health Organisation, United Nations Environment Programme, and World Meteorological Organisation have published Climate Change and Human Health, a comprehensive volume which considers the implications for science and policy7

    • n Britain, the Climate Change Impacts Review Group of the Department of Environment has published a report with a major section on potential health consequences for the British population.38 The Global Environmental Research Office has published a report advocating interdisciplinary, jointly funded, research initiatives.39 The Medical Research Council and Department of Health were represented in that process

    • Various national governments and research agencies have funded research into climate change and human health. These include the United States, Canada, the Netherlands, and Australia—but there has been no equivalent initiative in Britain

    Research needs and ethical considerations

    The main research strategies for estimating the health effects of future climate change (analogue studies, integrated mathematical modelling, the use of expert judgment in relation to diffuse or complex consequences, and experimental studies of specific components of the climate change process), the role of monitoring, and the implications for the formulation of social policy are the focus of the accompanying paper.6 It is important to note here, however, that it is misleading to view future and present environmental health problems as competing for attention in research and policymaking. The economic structures, social inequities, industrial practices, and consumer behaviour that underlie many of today's public health problems also underlie the growing pressures on the world's natural, life supporting systems. The advent of global environmental health hazards further underlines the continuing vulnerability of poorer, economically underdeveloped populations. We should therefore be seeking far reaching solutions through social policy—solutions that reduce both today's public health problems and the likelihood of future damage to the global biophysical systems that support life.


    We received assistance and helpful comments from Sari Kovats.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    16. 16.
    17. 17.
    18. 18.
    19. 19.
    20. 20.
    21. 21.
    22. 22.
    23. 23.
    24. 24.
    25. 25.
    26. 26.
    27. 27.
    28. 28.
    29. 29.
    30. 30.
    31. 31.
    32. 32.
    33. 33.
    34. 34.
    35. 35.
    36. 36.
    37. 37.
    38. 38.
    39. 39.