Chapter 16

Consequences and Options for Human Health

Coordinating Lead Authors: Carlos Corvalan, Simon Hales, Alistair Woodward

Lead Authors: Diarmid Campbell-Lendrum, Kristie Ebi, Fernando De Avila Pires, Colin L. Soskolne
Contributing Authors: Colin Butler, Andrew Githeko, Elisabet Lindgren, Margot Parkes
Review Editors: Paul Epstein, Jorge Rabinovich, Philip Weinstein

Main Messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469

16.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469

16.1.1 Overview of Health in the Context of Ecosystems
16.1.2 Impacts of Ecosystem Goods and Services on Health
16.1.3 Health in Scenarios
16.1.4 Typologies of Response Options and How They Apply to Health

16.2 Response Options and Actions outside the Health Sector . . . . . . . . . 476
16.2.1 Case Study: Climate Change, Land-use Changes, and Tick-borne
Diseases—Illustrative Example from Sweden
16.2.2 Case Study: Responding to the Risk of Water-borne Campylobacteriosis
16.2.3 Case Study: Linking Ecosystems and Social Systems for Health and
Sustainability—River Catchments
16.2.4 Case Study: Ciguatera (Fish Poisoning) and Ecological Change

16.3 Response Options and Actions by the Health Sector . . . . . . . . . . . . . 480

16.3.1 Improved Decision-making in the Health Sector
16.3.2 Methods for Measuring and Prioritizing Environmental Influences on
Health
16.3.3 Methods for Selecting Interventions to Protect Health
16.3.4 Addressing Risk Perception and Communication

16.4 Cross-sectoral Response Options and Actions . . . . . . . . . . . . . . . . . 482

16.4.1 Health, Social Development, and Environmental Protection
16.4.2 Linking Health and Ecosystem Responses

16.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483

REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 484

467
468 Ecosystems and Human Well-being: Policy Responses

BOXES 16.2 Causal Pathway from Driving Forces, through Exposures to

16.1 Disease Burden and Summary Measures of Population Health Outcomes, in the Context of Ecosystem Change
Health 16.3 Potential Ecological Pathways in Fish Poisoning (Ciguatera)
16.2 Global Burden of Disease Attributable to Climate Change TABLES
16.3 Developments in Linking Disease to Environmental Factors 16.1. Relationships between Ecosystem Services and the Major
16.4 Examples of Responses to Improve Human Health Categories of Disease
16.2. Water and Sanitation-related Diseases
FIGURES 16.3. Examples of Current and Future Vulnerability and Adaptation
16.1 Associations between Health, Other Aspects of Human Well- 16.4. Examples of Potential Health Implications of Sectoral
being, and Ecosystem Services Responses
 Consequences and Options for Human Health 469

Main Messages Where ill-health is caused, directly or indirectly, by excessive consump-

tion of ecosystem services, substantial reductions in overconsumption
Human health is both a product and a determinant of well-being. Mea- would have major health benefits while simultaneously reducing pres-
sures to ensure ecological sustainability would safeguard ecosystem ser- sure on life-support systems (high confidence). Both human health and the
vices and therefore benefit health in the long term (high confidence). In environment would benefit from a reduction in overconsumption. This would
this chapter, health is the central concern, while noting the reciprocal relation- improve health in the short term as well as contribute to short-term ecological
ships with other determinants of well-being. Negative health effects of ecosys- sustainability. Implementing better transportation practices and systems could
tem disruption are already evident in many parts of the world. In the long term, lead to decreased injuries, increased physical activity in sedentary populations,
some ecosystem change is inevitable. To limit the damage to human health as well as reduction in local air pollution and greenhouse gas emissions. Inte-
caused by these changes, mitigation strategies that reduce the driving forces grating national agriculture and food security policies with the economic, social,
of consumption, population increase, and inappropriate technology use are and environmental goals of sustainable development could be achieved, in
needed. part, through ensuring that the environmental and social costs of production
and consumption are more fully reflected in the price of food and water. Re-
Ecosystem disruption damages health through complex pathways. Local duced consumption of animal products in rich countries would have benefits
conditions exert a very strong influence on the nature, extent, and timing for human health and for ecosystems.
of the effects on health. Social adaptations may minimize, displace, or
postpone effects of ecosystem disruption on human health, but there are Society needs to balance technological and institutional development. To
limits to what can be achieved. Human societies have developed methods achieve this balance, governments must incorporate environmental, social, and
(such as agricultural systems or water supplies) that allow natural resources to health costs (both gains and losses) into measurements of progress.
be appropriated for social benefit. Piped water supplies and other man-made
resource appropriation systems provide human populations with a buffer in Approaches that perpetuate or worsen social inequalities may protect the
times of environmental change. These social adaptations are usually designed health of privileged populations, but are likely to result in the worst
to minimize local impacts. Many effects of ecosystem disruption on health are global health outcome overall. Globalization processes increasingly link the
displaced, either geographically (such as the costs of rich countries’ overcon- health and well-being of privileged with poor populations. A selective approach
sumption) or into the future (for example, long-term consequences of climate whereby the health and well-being of a small fraction of the global population is
change or desertification). promoted at the expense of the majority entails a high risk for both populations.

To understand the potential negative health impacts of ecosystem

change, two aspects need to be considered: the current vulnerability of 16.1 Introduction
the population affected and their future adaptive capacity. These two con-
There are well-defined relationships between health and the other
siderations are closely related, since vulnerable populations are less able to
components of well-being as defined in the MA framework. Ma-
plan and implement adaptive responses. Vulnerability and adaptive capacity terial lack, for example, is a strong determinant of health (and
are also tied to other aspects of well-being (material minimum, freedom and indeed of other aspects of well-being). Both at the country level
choice, social relations and security). and within countries, poorer communities have a worse health
profile than richer ones. Among poor communities, women and
Decisions about health and ecosystems must consider how one is re- children often bear the largest burdens of disease (WHO 2002).
lated to the other. Choices that are made about the management of eco- At the global level, poorer countries are still battling traditional
systems can have important consequences for health, and vice versa. hazards such as lack of clean water and sanitation, which contrib-
Consideration of ecosystem change enlarges the scope of health responses ute considerably to the burden of disease in these countries. For
by highlighting ‘‘upstream’’ causes of disease, injury, and premature death. example, the African region with 11% of the world’s population
The health sector can make an important contribution to reducing the damage has over 50% of the world’s burden of disease resulting from in-
caused by environmental disruptions, but the greatest gains will be made by fectious and parasitic diseases; in contrast, in the European region,
interventions that are partly or wholly placed in other sectors. The health sector with 14% of the world’s population, the burden of disease in this
bears responsibility for revealing the links and indicating which interventions category is less than 2% of the world’s total (based on WHO
are needed. Decision-makers need to consider the connections between health regions, measured in ‘‘disability-adjusted life-years’’ (WHO
and other sectors. Where there are trade-offs, it is important for politicians, 2002). (See Box 16.1.)
regulators, and the public to understand the consequences of taking one path On the other hand, the lack of good health is a major determi-
in preference to another. nant of poverty. For example, it is estimated that Africa’s GDP
could have been $100 billion larger if malaria had been eliminated
some 35 years ago (WHO 2000). GDP declines by about 1%
Where a population is weighed down by disease related to poverty and
when more than 20% of the population is infected with HIV
lack of entitlement to essential resources such as shelter, nutritious food,
(WHO 2002); see also reports of the Commission on Macroeco-
or clean water, the provision of these resources should be the priority nomics and Health (WHO 2001).
for healthy public policy (high confidence). The links between ecosystems Several studies have shown an important connection between
and human health are seen most clearly among deprived communities, which social relations and health. People with good social networks live
lack the ‘‘buffers’’ that the rich can afford. Within poor communities, poverty- longer and are generally healthier (Skrabski et al. 2003). Similarly,
related diseases are more prevalent among women and children, often due to people who fall ill recover faster when good social networks are
culturally related resource distribution. Poor communities are the most directly in place. Lack of security (or vulnerability) is also associated with
dependent upon productive ecosystems. This means that the poorest and most morbidity or mortality, although the relationship is often con-
disadvantaged communities can be among the first to benefit from ecosystem founded with material lack (that is, vulnerable communities are
protection. There are economic considerations also: a healthy community is often poor). Communities and individuals can be vulnerable for
more capable of sustaining local ecosystems than an unhealthy one. other than economic reasons. For instance, all inhabitants of low-
470 Ecosystems and Human Well-being: Policy Responses

tem integrity and human health do not always go hand in hand,

BOX 16.1 in the short term at least. Populations may flourish in degraded
Disease Burden and Summary Measures of Population environments, but only where it is possible to import resources
Health and services from elsewhere.
The disease burden encompasses the total amount of disease or pre-
mature death within the population. Comparing burden fractions attrib-
16.1.1 Overview of Health in the Context of
utable to several different risk factors requires, first, knowledge of the Ecosystems
severity/disability and duration of the health deficit, and second, the Definitions of health vary across cultures. Some cultures focus on
use of standard units of health deficit. The widely used Disability- physical evidence of bodily structure and function; others have a
Adjusted Life Year (DALY) is the sum of: much broader conception. For instance, for the indigenous peo-
ple of New Zealand, the Maori, dimensions of health include
• years of life lost from premature death (YLL) access to heritage and a sense of communion with the environ-
• years of life lived with disability (YLD). ment (Durie 2001). In its constitution, the World Health Organi-
zation defines health as ‘‘a state of complete physical, mental and
YLL takes into account the age at death. YLD takes into account social well-being and not merely the absence of disease or infir-
disease duration, age of onset, and a disability weight that reflects the mity.’’ However it is framed, health serves both as a cause of well-
severity of the disease. being and as a consequence. In the absence of good health, it is
To compare the attributable burdens for disparate risk factors we difficult to claim a state of well-being, but many of the compo-
need to know (1) the baseline burden of disease, in the absence of the nents of well-being (such as shelter, sustenance, and social rela-
particular risk factor, (2) the estimated increase in the risk of disease/ tions) are themselves important determinants of health status.
death per unit increase in risk factor exposure (the ‘‘relative risk’’), and Each of the categories of services provided by ecosystems is
(3) the current or estimated future population distribution of exposure. relevant to health. For instance, provisioning services include the
The avoidable burden is estimated by comparing projected burdens production of food and fiber. In their capacity as regulators of the
under alternative exposure scenarios.
environment, ecosystems influence the quality and flow of water
and the local climate. If these services are impaired, the impacts
An example of an application of this method in the field of ecosys-
on health tend to be direct, and relatively acute. More difficult to
tem change and health is the assessment of the burden of disease
demonstrate are the effects of cultural services provided by ecosys-
attributable to climate change.
tems—these are nonmaterial benefits such as spiritual, recre-
ational, and aesthetic outcomes. But even if these aspects of
human experience are not encompassed in the definition of
lying islands are vulnerable to the effects of sea level rise, although health, they do have an influence on physical and mental func-
their individual responses will vary depending on economic and tioning. The history of colonization, for example, shows clear
social conditions (Nurse and Sem 2001). Lack of control is an- links between loss of spiritual and cultural identity and rates of
other important cause of vulnerability. Many indigenous popula- disease and premature mortality (Kunitz 1994).
tions, for example, face ecosystem changes introduced by forces The links between ecosystem change and cultural services are
outside their control (such as economic interests), and these described in detail in the next chapter. The effects of ecosystem
threats can have an impact on their overall well-being, including disturbance on human health may be relatively direct, or occur at
their mental health. the end of long causal chains, dependent on many intermediate
For most diseases, the burden of disease is not borne evenly by events, and subject at many points to modifying influences. There
all members of a community. For example, children and pregnant are also relatively direct links between human health and the
women are at much greater risk for morbidity and mortality from health of animal populations that share a common environment
malaria, particularly if malnourished, whereas morbidity and mor- (Epstein et al. 2003). (See also MA Current State and Trends, Chap-
tality due to heat waves is highest among the elderly (Kilbourne ter 14.)
1997; Greenwood and Mutabingwa 2002). In general, the vul- As an example of long causal chains, environmental changes
nerability of a population to a health risk depends on the level of affecting river flows might lead to disputes over water rights, so-
material resources, effectiveness of governance and civil institu- cial unrest, forced migrations of large populations, conflict, and,
tions, quality of public health infrastructure, access to relevant indirectly, increased rates of disease and injury (WCD 2000). The
information, and existing burden of disease (Woodward et al. connections between ecosystem functioning and human health
1998). These factors are not uniform across a region or nation; may be bi-directional. Thus where there is environmental disrup-
rather, there are geographic, demographic, and socioeconomic tion leading to poor health, there may be compounding effects
differences. Failure to understand the reasons why particular pop- and ‘‘vicious cycles’’ established (Woodward et al. 2000). For ex-
ulation subgroups are vulnerable to a health outcome can reduce ample, land degradation and soil loss leads to crop failure, hunger,
the effectiveness of response options. and health problems. These health problems will more likely be
Exposure to hazards is often a function of resources, with the experienced by women and children, and, as such, can affect not
resource-poor most likely to be in harm’s way. For example, in only current but future health through poor growth, increased
impoverished communities, economically tied to large urban cen- disease burdens in later life, less productivity, etc. But there are
ters, but unable to afford safe housing, mudslides cause hundreds effects in the other direction also: populations with high levels of
of deaths each year (IFRCRCS 2002). Even where there is no chronic health problems can put less energy and time into grow-
economic or social differentiation in the exposure to hazards, the ing crops, preventing erosion, and managing agricultural re-
impacts on health may vary considerably from place to place or sources.
group to group. The difference between direct and indirect effects applies to
Figure 16.1 describes the links between ecosystem services the temporal and spatial scales over which these effects occur.
and well-being with a focus on human health. Note that ecosys- Owing to the many intermediate factors that may be involved,
 Consequences and Options for Human Health 471

Figure 16.1. Associations between Health, Other Aspects of Human Well-being, and Ecosystem Services

there are frequently considerable time-lags between ecosystem and described in detail in previous chapters) and the major cate-
change and health outcomes. For example, loss of biodiversity gories of disease. The table indicates the likely strength of the
may lead to higher mortality and morbidity via diminishing sup- associations, based on knowledge of ‘‘downstream’’ causal path-
plies of bio-pharmaceuticals, but this would be apparent only after ways leading to disease. There is a high level of uncertainty about
some years. In terms of spatial scales, we are most familiar with these judgments, because few studies provide quantitative evi-
local effects (such as flooding and mudslides on steep denuded dence of associations between ecosystem change and disease.
hillsides). More difficult to identify, but perhaps even more im-
portant for human health in the long term, are regional and global [Link] Biodiversity
changes such as acid rain, stratospheric ozone depletion, and the
accumulation of greenhouse gases. Biodiversity underpins the resilience of the ecosystems on which
Environmental and health policies are often determined with- humanity depends. Loss of biodiversity is occurring at an unprec-
out regard for one another, but there are important instances in edented rate, driven by overexploitation of productive ecosys-
which decisions have been swayed by health considerations. Re- tems, other land use changes, climate change, pollution events
moving lead from vehicle fuels is one case—this resulted from the such as oil spills, the transboundary migration of pollutants and
accumulating evidence of risks to child health and has had far hazardous substances, introduced species, and biotechnology. (See
reaching consequences for ecosystems worldwide (Reuer and MA Current State and Trends, Chapter 4.) This depletion of bio-
Weiss 2002). diversity threatens vital ecosystem services, including food, fuel
and fiber, fresh water, nutrient cycling, waste processing, flood
16.1.2 Impacts of Ecosystem Goods and Services on and storm protection, and climate stability. One obvious direct
impact of the loss of biodiversity is a reduction in sources of po-
Health
tential therapeutic chemicals. In general, the links between bio-
There are established links between the state of ecosystems and diversity loss and human health are difficult to demonstrate
the condition of populations that depend on these ecosystems. scientifically, due to the many factors that may confound such an
Some of these links are shown in Table 16.1, which attempts to association, difficulties in modeling nonlinear relationships, and
summarize the complex relationships between ecosystem goods lack of suitable data at appropriate scales (Sieswerda et al. 2001;
and services (as defined in the Millennium Ecosystem Assessment, Huynen et al. 2004). The clearest evidence of an association
472 Ecosystems and Human Well-being: Policy Responses

Table 16.1. Relationships between Ecosystem Services and the Major Categories of Disease. Strength of evidence: ‘‘’’ High,
‘‘’’ Medium, ‘‘’’ Low, ‘‘?’’ Uncertain, ‘‘ ’’ None or not known.
Climate Regulation,
Fresh Flood and Storm Cultural
Biodiversity Food Water Wood Nutrients Waste Control Services

Infectious parasitic Diarrhea     

diseases Malaria    
Other vector-borne
disease    
Acute respiratory
infection    ? ?
Other infectious
diseases    ? ? ?
Noncommunicable Chronic diseases     ?  
diseases Malnutrition    ?   
Mental conditions  ?  ? ?   
Injuries Poisonings  ?  
Drowning 

probably comes from studies showing that high species diversity soil erosion, silting of rivers, and drought, leading to agro-ecosystem
can be an important influence on reduced transmission of zoo- failure (UNEP 2002; Haug et al. 2003). See also MA Current State
notic diseases such as Lyme disease (Ostfeld and Keesing 2000). and Trends, Chapter 5; MA Multiscale Assessment, Chapter 2.
(See MA Current State and Trends, Chapter 14.) Few studies have attempted to quantify the links between
Diversity and health are linked also in agriculture, where food-producing ecosystems and human health. From first princi-
mono-cropping has been associated with increased vulnerability ples, such links might be seen most readily among vulnerable pop-
to acute food shortages and longer-term nutrient deficiencies ulations that live on marginal lands. Childhood stunting was
(Waltner-Toews 2001). There is limited evidence of an associa- associated with local land degradation in one such study (GRID/
tion between experience of the natural world and reduced sick- Arendal 1997). Birth weight was associated with land environ-
ness rates and improved healing (Frumkin 2001). ment classification in Papua New Guinea (Allen 2002).
Human societies have flourished by developing methods (such Undernutrition remains a major health problem in poor
as settled agriculture and water storage) that enhance productive countries, where poverty is a consistently strong underlying de-
ecosystem services for social benefit. Especially in countries domi- terminant (WHO 2002; FAO 2003). Global burden of disease
nated by market economies, these adaptations are often designed estimates indicate that in the year 2000, among the poorest coun-
to minimize short-term, local ecological disturbances, while max- tries, about a quarter of the burden of disease was attributable to
imizing profits. There is a mismatch of scale between social and childhood and maternal undernutrition. Among the rich coun-
ecological systems (Berkes and Folke 1998). tries, diet-related risks (mainly overnutrition) in combination
One result of this is that effects of ecosystem disruption on with physical inactivity accounted for a third of the burden of
health are often displaced geographically (such as the costs of rich disease. Worldwide, undernutrition accounted for nearly 10% of
countries’ overconsumption—climate change being a good ex- global DALYs (WHO 2002).
ample, in which many of the adverse health effects are likely to Aggregate food production is currently sufficient to meet the
appear first in low carbon-emitting countries) or postponed into needs of all, yet of the present world population of just over 6
the future (for example, long-term consequences of climate billion, about 800 million are underfed (FAO 2003), while hun-
change or desertification). But in general, the links between eco- dreds of millions are overfed (WHO 2003a).
system change and human health are seen most clearly among This imbalance has been driven primarily by social factors,
impoverished communities, who lack the ‘‘buffers’’ that the rich though ecological factors may play an increasingly important role
can afford. in the future. In poor countries, the number of people per hectare
of arable land increased from three in 1961–63 to five in 1997–99
[Link] Food (WEHAB 2002a). Poverty and hunger have tended to force peo-
The health of human populations is entirely dependent upon the ple onto marginal drought-prone lands with poor soil fertility.
services of productive ecosystems for food. This is most obvious Where the conditions of poor communities are overshadowed by
in poor countries—especially in rural areas—where food is de- the need to earn foreign exchange for debt repayments, this can
rived almost exclusively from local sources. Human dependence lead to the displacement of subsistence farming by cash crops
on ecosystems for nourishment is less apparent, but ultimately no grown for global corporations (Graber et al. 1995; McMichael
less fundamental, in richer urban communities. Historically, loss 2001).
of productive ecosystem services has led to the collapse of whole Agricultural production tripled in the last four decades, mainly
civilizations. For example, it has been suggested that the Mayan through growth in yield. However, food production has not kept
empire was lost near the end of the first millennium as a result of pace with population increase in many countries and improve-
 Consequences and Options for Human Health 473

ments in yield appear to have slowed (UNEP 2002; WEHAB microorganisms and decreasing dissolved oxygen. Water-related
2002a). It has been estimated that today, nearly a quarter of usable disasters—droughts and floods—also have important health im-
land has reduced productivity and about a billion people are af- pacts. The frequency of heavy rainfall events is likely to increase,
fected by land degradation either through soil erosion, water log- leading to an increase in flood magnitude and frequency and a
ging, or salinity of irrigated land (DFID/EC/UNDP/World Bank reduction in low river flows (IPCC 2001). Heavy rainfall would
2002; UNEP 2002). tend to adversely affect water quality by increasing chemical and
Providing sufficient food for an expected human population biological pollutants flushed into rivers and by overloading sewers
of 8–9 billion people will require major investments in poverty and waste storage facilities. In some parts of the world, climate
alleviation (Mellor 2002). There are also important trade-offs that change also may increase requirements for irrigation water be-
have to be made between various possible uses of productive land. cause of increased evaporation. (See also MA Current State and
Including population health considerations in this weighing of Trends, Chapter 7.)
choices could have important policy implications. The issue of
overconsumption of food is relevant here, for several reasons. [Link] Wood Fuel
First, from economic first principles, overconsumption of food is Most of the world’s population has no access, or limited access, to
encouraged by economic and trade practices, which prioritize electricity supplies, and about two billion people must rely on
short-term profit while externalizing longer-term environmental wood, dung, and agricultural residues for heating and cooking,
and social costs. Second, reductions in animal-based food con- while rich countries typically consume 25 times as much energy
sumption in rich countries could have important ecological bene- per capita as do poor countries (WEHAB 2002b).
fits (WHO 2003a). Intensive meat production, in particular, has Lack of clean, safe power causes a range of health impacts.
major adverse impacts on ecosystems (Leitzmann 2003, Reijnders About half of the world’s population still uses solid fuels for cook-
and Soret 2003). (See also MA Current State and Trends, Chapter ing and 0.5% of DALYs worldwide have been attributed to indoor
8.) air pollution from this source, particularly among women and
children. Urban air pollution, resulting from the combustion of
[Link] Fresh Water fossil fuels for transport, power generation, and industry, ac-
Fresh water is a key resource for human health; it is used for counted for a further 0.5% of DALYs (WHO 2002). Outdoor air
growing food, drinking, washing, cooking, and for the recycling pollution aggravates heart and lung disease (Kunzli et al. 2000).
of wastes. Of all available water globally, only 2.5% is fresh, and Indoor air pollution causes a major burden of respiratory diseases
less than 1% is readily available in lakes, rivers, and underground. among both adults and children (Ezzati et al. 2002; Smith and
Worldwide, almost 4% of the global burden of disease is currently Mehta 2003).
attributable to unsafe water, inadequate sanitation, and poor hy- Energy supplies are a fundamental factor in sustainable devel-
opment and are also needed to provide and maintain modern
giene. In the next century, water resources will be strongly af-
health services. The need to spend considerable time collecting
fected by trends in population, land use, and the management of
fuel can preclude proper education, especially of women, with
freshwater ecosystems. Increasing demand for food, in particular,
indirect adverse effects on health through illiteracy, lost work op-
will worsen water scarcity. It is estimated that by 2025 nearly half
portunities, family health, and large family size. More indirectly
the world population will live in river basins where water is scarce
still, energy use is linked to health effects via desertification, acidi-
and 70% of readily available water supplies will be used (WEHAB fication, ambient air pollution, and climate change.
2002d). Water scarcity can lead to use of poorer quality sources
of fresh water, which are more likely to be contaminated, tending
[Link] Nutrient Management
to cause increases in water-related diseases.
At present, 1.1 billion people lack access to safe water supplies, Application of agricultural fertilizers and organic wastes (including
while 2.6 billion people lack adequate sanitation (WHO/UNICEF sewage) can improve agricultural yields but may also lead to in-
2004; UNESCO 2003a). Lack of improved water and sanitation creased concentrations of nitrogen and phosphorus in surface wa-
is strongly associated with poverty, although this relationship var- ters and coastal sea areas (Smil 2000). This can cause certain
ies between regions (WHO 2002). Along with sanitation, water cancers (Wolfe and Patz 2002) and eutrophication in both marine
availability and quality are well recognized as important risk fac- and freshwater ecosystems, with overgrowth of bacteria, phyto-
tors for infectious diarrhea and other major diseases (Esrey 1996; plankton, macrophytes, and microalgae.
Pruss et al. 2002; Strina et al. 2003; Thompson et al. 2003). In turn, these problems can lead to increases in water-borne
The associated effects on human health are severe. Poor coun- diseases and poisoning from harmful algal blooms (UNESCO
tries, with inadequate provision of water and sanitation, will be 2003b). There are likely to be other ecological mechanisms by
most vulnerable to these effects that impact most severely on chil- which increased nutrients can lead to human diseases, but further
dren. (See Table 16.2.) In addition to direct effects, there can be research is required to clarify these (NRC 1999; Townsend et al.
indirect health effects. For example, during a water shortage, 2003).
women may have to walk further and spend additional time to
supply households with water. This additional time and energy [Link] Waste Management, Processing, and Detoxification
expenditure may affect a woman’s health and her ability to earn Well-functioning ecosystems absorb and remove contaminants.
an income and to care for household members. For example, wetlands can remove excess nutrients from runoff,
The effects of climate change on water resources are difficult preventing damage to downstream ecosystems (Jordan et al.
to forecast because of the many factors that influence rainfall, run- 2003). Inadequate management of solid waste increases human
off, and evaporation. Nevertheless the best estimates are that cli- exposure to infectious disease agents (for example, via contamina-
mate change may increase the number of people affected by water tion of water with feces, or via disease vectors). This leads to a
stress by about 0.5 billion in 2025 (Arnell 1999). Increases in tem- range of communicable diseases, especially diarrheal illness
perature would worsen water quality by increasing the growth of (WHO/UNICEF 2004; UNESCO 2003a). Of the 2.6 billion
474 Ecosystems and Human Well-being: Policy Responses

Table 16.2. Water and Sanitation-related Diseases (WHO 2003c)

Disease Burden, in
Disability-Adjusted Mortality
Disease Life-Years (thousands) (deaths per year) Relationship of Disease to Water Supply and Sanitation

Diarrheal diseases 61,966 1,797,970 strongly related to unsanitary excreta disposal, poor personal
and domestic hygiene, unsafe drinking water; 90% of deaths in
children under 5
Infection with intestinal helminths 2,882 9,360 strongly related to unsanitary excreta disposal, poor personal
(ascariasis, trichuriasis, and domestic hygiene; 133 million people suffer from high-
hookworm disease) intensity intestinal helminth infections
Schistosomiasis 1,702 15,370 strongly related to unsanitary excreta disposal and absence of
nearby sources of safe water; 160 million people infected
Trachoma 2,329 150 strongly related to lack of face washing, often due to absence
of nearby sources of safe water; 500 million people at risk; 6
million visually impaired
Malaria 46,486 1,272,390 related to poor water management, water storage, operation of
water points, and drainage; 90% of deaths in children under 5
Onchocerciasis 484 (⬍5) related to poor water management in large-scale projects
Dengue fever 616 18,560 related to drainage water organically polluted, open sewers,
eutrophied ponds
Lymphatic filariasis 5,777 417 related to poor water management, water storage, operation of
water points, and drainage

people who lack adequate sanitation, the majority live in Asia and injuries, and indirectly through economic disruption and
(Cairncross 2003). population displacement. Extreme climate events are expected to
When recycled appropriately, human waste can be a useful increase as a result of climate change (WHO/WMO/UNEP
resource that promotes soil fertility (Esrey 2002). However, 2003).
where waste contains persistent chemicals such as organochlorines Health effects of climate extremes include physical injuries,
or heavy metals, recycling onto land can lead to the accumulation increases in communicable diseases due to crowding, lack of clean
of these pollutants and increased human exposure through food water and lack of shelter, poor nutritional status, and adverse ef-
and water; this may contribute to a wide range of chronic diseases. fects on mental health (Hajat et al. 2003).
One example was the floods along the Yangtze River in 1998.
[Link] Climate Regulation For years, loggers had been cutting forests along the river’s water-
Climate regulation is an important property of Earth’s natural sys- shed, and farmers and urban developers had gradually moved to
tems. Each of the ecological services referred to above is sensitive occupy the river’s flood plains by draining lakes and wetlands.
to climate, and will be affected by climate change. Although cli- Record rains fell in the Yangtze basin in the summer of 1998, and
mate change will have some beneficial effects on human health, these degrading practices amplified the flooding, leaving 3,600
most effects are expected to be negative (IPCC 2001). people dead, 14 million homeless, and $36 billion in economic
Direct effects such as increased mortality from heat waves are losses. Restoring the ecosystem’s flood control services would
readily predicted but indirect effects are likely to predominate now take decades and billion of dollars (UNEP 2002).
(IPCC 2001; WHO/WMO/UNEP 2003). Human health is Globally, the number of people killed, injured, or made
likely to be affected indirectly by changes in productive ecosys- homeless by natural disasters is increasing (WHO/WMO/UNEP
tems and the availability of food, water, and energy supplies. 2003). An important reason for this is increasing settlement on
These changes will in turn affect the distribution of infectious coasts and floodplains that are exposed to extreme events. A num-
diseases, nutritional status, and patterns of human settlement. ber of case studies at the local scale have shown that human inter-
Changes in the geographic distribution, abundance, and behavior actions with ecosystems have also contributed to increasing
of plants and animals affect, and are affected by, biodiversity, nu- human vulnerability. Healthy ecosystems provide a buffer against
trient cycling, and waste processing. the damaging effects of climate extremes. For example, forests
Attempts have been made to estimate the global burden of absorb rainfall and reduce rapid increases in runoff, reducing
disease attributable to climate change (WHO 2002). But so far flooding and soil erosion. Coral reefs and mangroves stabilize
only a small fraction of the health outcomes associated with cli- coastlines, limiting the damaging effect of storm surges. (See MA
mate change have been included in the global burden of disease Current State and Trends, Chapters 9 and 16.)
calculations, selected on the basis of sensitivity to climate varia- In many areas the only land available to poor communities is
tion, predicted future importance, and availability/feasibility of that with few natural defenses against weather extremes. In recent
quantitative global models. (See Box 16.2.) decades, there has been a large migration to cities and more than
half the world’s population now lives in high-density urban areas.
[Link] Flood and Storm Control Such migration and increasing vulnerability means that even
Climate extremes, including floods, storms, and droughts, have without increasing numbers of extreme events, losses attributable
local and sometimes regional effects, both directly through deaths to each event will tend to increase (WHO/WMO/UNEP 2003).
 Consequences and Options for Human Health 475

contact with nature need not be physical; for example, some

BOX 16.2 benefit may be obtained purely from visual or even visualized
Global Burden of Disease Attributable to Climate Change (imaginary) contact. On the other hand, it would follow that
(WHO/WMO/UNEP 2003) knowledge of the loss of valued ecosystems, even if such knowl-
• Climate change will affect the pattern of deaths from exposure to
edge is indirect, may cause a profound sense of loss, and even
harm health.
high or low temperatures. However, the effect on actual disease
burden cannot be quantified, as we do not know to what extent
16.1.3 Health in Scenarios
deaths during thermal extremes are in sick/frail persons who would
have died soon. In 2030, the estimated risk of diarrhea will be up to The effects of ecosystem change on well-being in future decades
10% higher in some regions than if no climate change occurred. are explored in this assessment using four scenarios. (See MA Sce-
Since few studies have characterized this particular exposure- narios, Chapter 11.) Each scenario describes a plausible future for
response relationship, these estimates are uncertain. the linked global socioecological system. Health is an ‘‘integrat-
• Estimated effects on malnutrition vary markedly among regions. By ing’’ outcome of the distribution and interaction of ecosystem
2030, the relative risks for unmitigated emissions, relative to no cli- and human services. Institutions—the main legal, cultural, and
mate change, vary from a significant increase in the Southeast Asia attitudinal currents that flow through society—were found to be
region to a small decrease in the Western Pacific. Overall, although a crucial determinant of the protection of ecosystem services,
the estimates of changes in risk are somewhat unstable because of human services, and human health.
regional variation in rainfall, they refer to a major existing disease In three of the four scenarios, global health was found—very
burden entailing large numbers of people. broadly—to improve, while in one scenario the health of low-
• The estimated proportional changes in the numbers of people killed income populations, which currently constitute the majority of
or injured in coastal floods are large, although they refer to low
the global population, remained unchanged or worse. Impor-
tantly, caveats were identified in the three more optimistic scenar-
absolute burdens. Impacts of inland floods are predicted to increase
ios, whereby each could have significant adverse health effects.
by a similar proportion, and would generally cause a greater acute
Society needs to balance technological and institutional devel-
rise in disease burden. While these proportional increases are simi-
opment. Belief in the ability of technology alone to solve the
lar in industrial and developing regions, the baseline rates are much
human predicament is unwarranted. On the other hand, the size
higher in developing countries. and environmental impact of the still-growing global population
• Changes in various vector-borne infectious diseases are predicted. requires the extension and deepening of many forms of technol-
This is particularly so for malaria in regions bordering current en- ogy. The incorporation of economic, social, and health costs into
demic zones. Smaller changes would occur in areas where the dis- measurements of progress is an important institutional change to
ease is currently endemic. Most temperate regions would remain facilitate this balance, by providing measurable feedback.
unsuitable for transmission, because either they remain climatically Approaches that perpetuate or worsen social inequalities may
unsuitable (as in most of Europe) or socioeconomic conditions are protect the health of privileged populations but are likely to result
likely to remain unsuitable for reinvasion (for example, in the south- in the worst global health outcome overall. Globalization proc-
ern United States). Important causes of uncertainty in these fore- esses increasingly link the health and well-being of privileged with
casts include extrapolation between regions and the factors that poor populations. A selective approach whereby the health and
translate potential into actual transmission. well-being of a small fraction of the global population is promoted
• If our understanding of broad relationships between climate and dis- at the expense of the majority entails a high risk for both popula-
ease is accurate, then climate change may already be affecting tions. Humans possess the cognitive and organizational capacity
human health. The total current estimated burden is small relative to to maintain or even improve global health in the next decades,
other major risk factors measured under the same framework. How- but this will require substantial goodwill, cooperation, and work.
ever, in contrast to many other risk factors, climate change and its
associated risks are increasing rather than decreasing over time. 16.1.4 Typologies of Response Options and How
They Apply to Health
International, national, and community responses to global eco-
[Link] Cultural, Spiritual, and Recreational Services system changes include policies aimed at stopping and reversing
the extent and rate of change (mitigation) and response options
Cultural services may be less tangible than material services but designed to effectively reduce the current and future impacts of
are nonetheless highly valued by people in all societies. People those changes (adaptation). It is recognized that an unusual degree
obtain diverse nonmaterial benefits from ecosystems. These bene- of anticipatory thinking is required to develop proactive response
fits include recreational facilities and tourism, aesthetic apprecia- options for reducing potential future ecosystem impacts. Such op-
tion, inspiration, a sense of place, and educational value. There tions should complement, not replace, mitigation policies to slow
are traditional practices linked to ecosystem services that have an or avert the process of change itself.
important role in developing social capital and enhancing social The impacts of ecosystem changes will be site-specific and
well-being. path-dependent; that is, they will depend on local circumstances
There is a hypothesis that stimulating contact with the rich (Yohe and Ebi 2005). For example, malaria epidemics occur fol-
and varied environment of ecosystems, including that of gardens, lowing rainy seasons in some regions, while epidemics occur
may benefit physical and mental health. There is limited evidence during droughts in others. Further, these impacts will not be ex-
that this may help in the prevention and treatment of depression, perienced evenly across a population; there will be particularly
drug addiction, behavioral disturbances, as well as convalescence vulnerable subgroups. Therefore, public health response options
from illness or surgery. Regular contact with pets seems to pro- (interventions) need to be designed at spatial and temporal scales
long and enhance the quality of life, especially in old age. Such appropriate to the health outcome of concern, taking into consid-
476 Ecosystems and Human Well-being: Policy Responses

eration the social, economic, and demographic driving forces, and vention. Primary prevention aims to prevent exposure to risk of
also whom the interventions should target. Interventions can disease in an otherwise unaffected population (for example, the
focus on local, national, regional, and international scales; and supply of bednets to all members of a population at risk of expo-
within these, vulnerable subgroups. sure to malaria). Secondary prevention entails preventive actions
As discussed in Chapter 3, the nature of the response options in response to early evidence of health impacts (for example,
can be legal, economic and financial, institutional, social and be- strengthening disease surveillance and responding adequately to
havioral, technological, and cognitive. As discussed in Chapter disease outbreaks such as the West Nile virus outbreak in North
19, within each of these, there may be gender issues that could America). Tertiary prevention consists of measures to reduce
affect not only the efficiency and effectiveness of interventions, long-term impairments and disabilities and to minimize suffering
but also future development. Effects on health may be complex, caused by existing disease. In general, secondary and tertiary pre-
and follow a variety of causal pathways. For example, develop- vention is less effective, and more expensive, than primary pre-
ments in agriculture that have dramatically lowered the cost of vention.
food in many countries have removed the threat of undernutri- The attributes of different risks affect the choice of response
tion, but have provided conditions for the emergence of new options, including spatial extent (the extent of land cover change
disease-causing agents (such as antibiotic-resistant Salmonella) or of an epidemic); speed of onset (how rapidly the event occurs,
(Waltner-Toews 2001). either building slowly like a drought or coming quickly like a
The vulnerability of a particular population to the potential flash flood; the slow spread of malaria or the rapid speed of an
health impacts of ecosystem change will depend on the degree to outbreak of influenza); the number of potentially affected individ-
which individuals and systems are susceptible to, or unable to uals (the response to an isolated case of plague versus an epidemic
cope with, these changes. Vulnerability depends upon the level of of dengue fever); the onset-to-peak interval (how long it takes
exposure, the sensitivity (or exposure-response relationship); and from the first detection to the maximum level of the hazard, such
the response options in place to reduce the burden of a particular as the first impacts of a flood to its peak magnitude, or the first
adverse health outcome (Ebi et al. 2005). detected cases of a disease to its maximum prevalence); and the
Populations, subgroups, and systems that cannot or will not expected frequency or return period (frequency of drought or
adapt are more vulnerable, as are those who are more susceptible floods, periodicity of disease epidemics).
to ecosystem change. Population subgroups may not have the re- Other factors affecting choice of responses include knowledge
silience to adapt because of a lack of material resources, lack of and understanding of the underlying processes or causes; capacity
relevant information, lack of effective governance and civil insti- to predict, forecast, and warn; capacity to respond (institutional
tutions, and lack of public health infrastructure (Woodward et al. and otherwise); how the risk might change over time and with
2000). The effective targeting of interventions requires under- ecosystem change; and ethical appropriateness.
standing which demographic or geographic subpopulations may Many of the possible response options to ecosystem change lie
be most at risk, the factors that contribute to their vulnerability, primarily outside the direct control of the health sector. They are
and which of these factors can be modified within the context of rooted in areas such as sanitation and water supply, education,
a particular time and location. Thus individual, community, and agriculture, trade, tourism, transport, development, and housing.
geographical factors determine vulnerability. Inter-sectoral and cross-sectoral integrated options are needed to
Response options can aim to reduce current and/or future reduce the potential health impacts of ecosystem change. These
vulnerability. Adaptive capacity describes the general ability of in- integrated interventions should address the social, economic, and
stitutions, systems, and individuals to adjust to potential damages, demographic driving forces of and responses to ecosystem change.
to take advantage of opportunities, and minimize the long-term Figure 16.2 follows an epidemiological, causal pathway ap-
consequences (Smit et al. 2001). Specific options arise from the proach (Corvalan et al. 2000). This highlights the main driving
adaptive capacity of a population. Adaptive capacity encompasses forces that are linked to health determinants (existing infrastruc-
coping capacity (what could be implemented now to minimize ture, social values, and general social, economic, and demographic
potential damage from ecosystem change) and the response op- conditions); the specific exposures at different levels (either as dis-
tions that have the potential to expand future coping capacity. tant, often indirect, or proximate, often direct, as well as ranging
Specific options arise from the coping capacity of a community, from global to local scales); the health impacts (or the positive
nation, or region. The primary goal of building adaptive capacity health consequences if seen from the point of view of ecosystem
is to reduce future premature death, avoidable disease, and disease- protection); how these links are modified by population vulnera-
related discomfort and disability in a population arising from eco- bility; and how society (or individuals) respond, in the form of
system change. Examples illustrating these various concepts are interventions at all levels (improving on the basic conditions
shown in Table 16.3. under driving forces, reducing exposures or providing health-
Response options encompass both spontaneous responses to specific interventions).
ecosystem change by affected individuals and planned interven-
tions by governments or other institutions. Examples of the latter
include watershed protection policies or effective public warning 16.2 Response Options and Actions outside the
systems for drinking water quality. In many cases, continuing and
strengthening established interventions may be the best approach
Health Sector
to reducing vulnerability and increasing adaptive capacity, while Factors that need to be considered when evaluating evidence that
in other cases, new response options will need to be developed the protection of ecosystems avoided adverse health impacts in-
(Ebi et al. 2005). Increasing the adaptive capacity of a population clude: the strength of the evidence; the plausibility of the associa-
shares similar goals with sustainable development—to increase the tion (that is, a probable or demonstrated etiologic chain); the
ability of nations, communities, and individuals to effectively and presence of supporting or contradictory evidence from non-
efficiently cope with the changes and challenges of ecosystem human systems; the extent that contextual factors and competing
change. (See Chapter 19.) Public health scientists describe re- influences could explain the adverse health impact; the policies
sponse options in terms of primary, secondary, and tertiary pre- and interventions in place that could affect the exposure-response
 Consequences and Options for Human Health 477

Table 16.3. Examples of Current and Future Vulnerability and Adaptation (Kovats et al. 2003)
Definition Current example Future example

Vulnerability: degree to which individuals and populations living in areas on the fringe of the whether these populations might be vulnerable
systems are susceptible to or unable to cope current distribution of malaria are at risk for epi- in the future depends, in part, on the implemen-
with the adverse effects of climate change demics if the range of the Anopheles vector tation of effective prevention activities
changes
Adaptation Baseline: the adaptation measures the exposure-response relationship is influ- increasing access to and use of air conditioning
and actions in place in a region or community enced by the current prevention measures will decrease the percentage of the elderly pop-
to reduce the burden of a particular health out- aimed at reducing the burden of a disease; for ulation that could be adversely affected by
come example, the number of elderly adversely af- future heat waves; for example, the conse-
fected by a heat wave will depend on the num- quences of the 1995 heat wave in the midwest-
bers that have access to and use air ern United States were greater than those for a
conditioning during a heat wave similar heat wave in 1999, in part because of
programs established in the interim
Coping Capacity: the adaptation strategies, pol- a number of cities in mid-latitude countries have over time, strategies, policies, and measures
icies and measures that could be implemented the level of material resources, effective institu- can move from being possible to being imple-
now; specific adaptation plans arise from a re- tions, and quality of public health infrastructure mented (that is, being part of the adaptation
gion or community’s coping capacity to establish and maintain early warning sys- baseline); for example, providing universal ac-
tems for heat waves; until implemented, these cess to adequate quantities of clean water is
systems are within a city’s coping capacity not yet possible, although significant progress
has been made
Adaptive Capacity: the general ability of institu- adaptive capacity is the theoretical ability of a over time, it is hoped that regions and commu-
tions, systems, and individuals to adjust to region or community to respond to the threats nities will increase their adaptive ability and
potential damage, to take advantage of oppor- and opportunities presented by climate change. their resilience to what future climates will bring
tunities, or to cope with the consequences It is affected by a number of factors and encom-
passes coping capacity and the strategies, poli-
cies, and measures that have the potential to
expand future coping capacity; for example, ed-
ucation of women provides a range of benefits
to a population that results in increased ability
to deal with challenges and changes

relationship; and the timing, scale, and location of the assessment partly because of increased reporting through greater awareness
(Scheraga et al. 2003). Assessments made at one point in time or among health personnel and the general public. However, case
at one location may provide different answers when the evalua- studies from Sweden have shown that a real increase in both tick
tions are repeated over time or over larger geographic areas. population density and in disease incidence has occurred since the
early 1980s, and that ticks have expanded their distribution range
16.2.1 Case Study: Climate Change, Land Use northward (Talleklint and Jaenson 1998). These changes have
Changes, and Tick-borne Diseases—Illustrative been associated with milder and shorter winters (Lindgren et al.
Example from Sweden 2000; Lindgren and Gustafson 2001).
Research findings have enabled preliminary predictions to be
Diseases transmitted by blood-sucking ticks are especially sensitive made every year in early spring; that is, prediction of whether the
to changes in the local environment, particularly alterations coming year is a potentially high-risk year for tick bites. Swedish
caused by land use or by land cover changes and changed climatic newspapers, radio, and television news now address the risk of
conditions. (See MA Current State and Trends, Chapter 14.) The tick-borne encephalitis and Lyme disease repeatedly each year
climate sets the limit for both the altitude and latitude distribution when the tick-activity season starts. High-risk areas are shown,
of ticks and is important for tick population density. Biodiversity new risk areas pointed out, and effective preventive measures are
and species composition may affect the transmission of pathogens mentioned, such as removal of thick undergrowth vegetation in
in nature and, hence, the risk of disease in an area (LoGiudice et parks and gardens, and daily body inspection for rapid detection
al. 2003). and removal of ticks. The latter decrease the risk for Lyme disease
Ixodid ticks, which live for up to three years, may transmit but do not protect against the transmission of tick-borne encepha-
several diseases, of which the most important are Lyme disease litis. Before the high-risk season begins, tick-borne encephalitis
and the severe form of tick-borne encephalitis. The latter is en- vaccination is made easily accessible for people living or working
demic in Europe and in most western parts of Eurasia, whereas in or visiting endemic areas.
Lyme disease is prevalent throughout the temperate zones of the
Northern Hemisphere. About 85,000 cases of Lyme disease are 16.2.2 Case Study: Responding to the Risk of
reported in Europe annually compared to 15–20,000 cases in the Water-borne Campylobacteriosis
United States (Steere 2001).
Over the last two decades, the incidence of Lyme disease and From hunter-gatherer societies through agricultural societies to
tick-borne encephalitis has increased in endemic regions. This is industrial societies, human settlements have always centered on a
478 Ecosystems and Human Well-being: Policy Responses

Figure 16.2. Causal Pathway from Driving Forces, through Exposures to Health Outcomes, in the Context of Ecosystem Change.
The impacts are modified by the population’s vulnerability and the interventions implemented.

reliable supply of good quality fresh water. When supplies have and domesticated animals (Chin 2000). Although compliance
been disrupted, the effects of thirst upon health are immediate with these measures is difficult to formally assess, there is little
and can be rapidly fatal. When water quality has been compro- question that they contribute significantly to a reduction of the
mised, we have seen some of the largest disease outbreaks the disease burden, and should be maintained and encouraged on that
world has known. Human settlements have, therefore, always basis. However, they have failed to arrest the rapid rise of campy-
been dependent on healthy freshwater ecosystems to supply pota- lobacteriosis. It is appropriate, therefore, to also consider public
ble water, and water catchment protection is so ingrained in pub- health interventions based on restoring the health of freshwater
lic health culture that it is often taken for granted. In modern ecosystems.
times, water treatment plants have fulfilled a ‘‘magic bullet’’ role Slowing runoff is important because of the limited survival of
and have arguably taken the edge off the perceived importance of fecal pathogens, whose half-lives are more likely to be exceeded
catchment protection—that is, until outbreaks of waterborne ill- before human exposure occurs. Waters from catchments with
ness in rich countries started to seriously shake public confidence native vegetation are least likely to contain viable pathogens; re-
in public water supplies. vegetation could therefore be advocated as a public health inter-
Campylobacterosis is a gastrointestinal disease that may be vention. Importantly, it is not only the direct transmission of
spread by food or by water and was first recognized as an ‘‘emerg- Campylobacter in drinking or recreational water exposure that will
ing’’ human disease in the late 1970s. Campyobacteriosis is now be affected. If livestock infections are also decreased as a result of
the most commonly reported infectious disease in rich countries. regrowth of native plants in water catchments, the number of
The disease is prevalent among domesticated animals such as human infections acquired occupationally (farm, abattoir) and by
poultry, sheep, and cattle, and transmission to humans depends on the food-borne route (animal products) will also be reduced.
‘‘survival trajectories’’ followed by the pathogen between excre- The lesson from this case study is that, in many cases, scien-
tion from the reservoir and ingestion by the case (Skelly and tifically based public health interventions can be devised only
Weinstein 2003). The life-cycle of this organism can be complex with an understanding of the ecology of the disease.
and its survival in the environment is subject to the influence of a 16.2.3 Case Study: Linking Ecosystems and Social
variety of abiotic factors. Pastoral farming has a major impact on
both water flow and quality. As vegetation is lost from hillsides Systems for Health and Sustainability—River
and riverbanks, the volume and speed of runoff increases. The Catchments
natural purification of water percolating through soil and vegeta- The management of river catchments poses an emerging ‘‘up-
tion is also reduced. This exposes both livestock and humans stream’’ public health issue—spanning concerns regarding the
downstream to a variety of zoonotic pathogens, including Campy- safety and sustainability of freshwater ecosystems, socioeconomic
lobacter, Cryptosporidium, and Giardia. development, and multistakeholder governance processes. As
Current preventive measures for controlling transmission and such, river catchment management has implications for both the
infection with Campylobacter include food and farm hygiene, thor- environmental and socioeconomic determinants of health and ex-
ough cooking (or irradiation) of food, use of pasteurized milk and emplifies the importance of response options and actions outside
chlorinated water supplies, and control of the disease in domestic the health sector.
 Consequences and Options for Human Health 479

During the 1990s, water governance priorities shifted from ability research. ECO-PAR is founded on interaction between
their developmental focus on infrastructure provision (domestic knowledge, participation, and action and facilitates a unified ap-
water supply, sanitation, and irrigation) to recognize the critical proach to ecosystems, social systems, health, and sustainability
need for an ecosystems approach that manages freshwater re- (Parkes et al. 2003).
sources as an integral part of natural cycles (UNCSD 1998; World
Water Forum 2000; Helming and Kuylenstierna 2001). Priorities 16.2.4 Case Study: Ciguatera (Fish Poisoning) and
for water resource management at the turn of the twenty-first Ecological Change
century include recognition and maintenance of: (1) catchments
as critical to the management of freshwater ecosystems—enabling Certain marine algae produce potent toxins that cause illness
fresh waters to be viewed within a landscape or systems context; when consumed via contaminated fish or shellfish. The number
(2) the socioeconomic, ecological, and human health values of and geographic distribution of harmful algal blooms appears to
freshwater ecosystems, their services, and functions; (3) processes have increased in recent decades, in parallel with other changes
that support freshwater ecosystem integrity, structure, function, in marine ecosystems, nutrient contamination of waterways, and
and adaptive capacity, including quality, quantity, and timing of climatic change (van Dolah 2000). There are several clinical syn-
flow (Baron et al. 2002); and (4) protecting the determinants of dromes associated with these events. The most common is ciguat-
health through catchment management. era (fish poisoning) caused by consuming reef fish contaminated
The place-based links between environmental and socioeco- with algal toxins.
nomic determinants of health were examined in a case study of Traditional environmental health practice has focused on di-
catchment (ecosystems) and community (social systems) in New rect effects of pollutants on human health. Ciguatera is an exam-
Zealand’s Taieri River catchment. In the Taieri Catchment & ple of a different kind of environmental problem. Morris (1999)
Community Health Project, public health issues of concern writes, ‘‘Harmful algal blooms are an example of an alternative
ranged from the direct health impacts associated with the ecologi- paradigm, in which human-induced stress on complex living sys-
cal determinants of water-related disease to the indirect health tems leads to the emergence of new, potentially harmful microor-
impacts of catchment management, freshwater ecosystem change, ganisms (or the reemergence of ‘old’ pathogens from previously
and rural sustainability—mediated through socioeconomic deter- restricted environmental niches), which, in turn, cause human
minants of health (Duncanson et al. 2000; Hales et al. 2003; Skelly disease.’’
and Weinstein 2003). Figure 16.3 illustrates some of the potential social and ecologi-
The Taieri catchment case study combined knowledge gener- cal drivers of ciguatera and the pathways to health impacts. Indi-
ation with actions to address the social and ecological dimensions rect drivers of change (population increase and resource
of catchment and community health issues. The multi-method consumption) affect direct drivers (global climate change and land
study examined the links between ecosystem change and the de- use change). Rise in sea surface temperature, contaminated run-
terminants of health through socioecological analysis of knowl- off, and other anthropogenic factors lead to disturbance of the
edge strengths and deficits in the catchment; community-oriented marine ecosystem (including coral bleaching), increased growth
participatory action research with diverse catchment stakeholders; of toxic algae, and contamination of reef fish. This, in turn, causes
and selected collaborative research initiatives—including a whole
catchment questionnaire survey and specific biophysical studies.
All phases of the research were based on building collaborative
relationships with community reference groups (including resi-
dents living throughout the 5,650 square kilometer rural catch-
ment) and co-researchers (included agencies, researchers, and
indigenous organizations involved with science and decision-
making regarding environment, health, development, and con-
servation issues in the catchment).
The catchment case study drew attention to the linked role of
ecosystems and social systems as a mutually reinforcing basis for
health, experienced as healthy living systems, livelihoods, and life-
styles. There was a transition from a research-initiated project
through a ‘‘Community-University Partnership’’ to the ‘‘Taieri
Alliance for Information Exchange and River Improvement’’ (the
TAIERI Trust). This trust represents a shift from separate univer-
sity and community interests to an integrated organization com-
bining the interests of community, academic, and agency
stakeholders to foster the health and sustainability of the river and
local communities. This collaborative approach to knowledge,
participation, and action demonstrates the application of success-
ful decision-making processes into the research setting.
This case study strengthens the argument that place-based
actions outside the health sector can respond to environmental
and socioeconomic concerns—building resilient ecosystems and
social systems that provide a double dividend for health and sus-
tainability. Research and experience in the Taieri catchment case
study led to the recommendation for ECO-PAR (Ecosystem- Figure 16.3. Potential Ecological Pathways in Fish Poisoning
based Community-oriented Participatory Action Research) as a (Ciguatera)
generic approach to integrated, collaborative health, and sustain-
480 Ecosystems and Human Well-being: Policy Responses

ciguatera in people consuming the reef fish, or alternatively, it likely public health consequences from any environmental
causes people in island communities to avoid this important pro- change. Important policy decisions such as legislation on environ-
tein source, potentially leading to malnutrition. mental lead, asbestos, and secondary tobacco smoke are largely
dependent on health scientists measuring the links between these
exposures and health outcomes, reaching a reasonably broad con-
16.3 Response Options and Actions by the sensus, and presenting these findings to policy-makers. In these
Health Sector cases, the demonstration of a clear and significant health risk has
taken precedence over other competing influences. Although
In order to respond effectively to threats from ecosystem change,
most of the success stories are for environmental factors acting at
the health sector must be able to carry out effective monitoring
a local level, examples such as the Montreal Protocol on CFC
and surveillance of disease and risk factors for disease; interpret
emissions show that health considerations can also be important
data provided by surveillance systems; use surveillance data in
in influencing decisions on global environmental issues.
conjunction with environmental and other data to develop mod-
els to predict disease occurrence; link changes in disease rates to
specific environmental factors; and intervene to remove the 16.3.2 Methods for Measuring and Prioritizing
causes of disease or to lessen the damage they cause (Wilson and Environmental Influences on Health
Anker 2005). In recent years, there have been important methodological devel-
Tracking death registrations through periods of extreme opments in the linkages between environment and disease data-
weather is an example of the first condition for effective response bases and in quantitative analytical techniques demonstrating
(Hajat and Kovats 2002). An example of the second is the capacity relationships between them. (See Box 16.3.)
to relate changing patterns of communicable disease to climate These linkage methods could potentially be applied to wide-
variability (Hales et al. 1999a; Hales et al. 1999b). The 2003/ area ecological measures other than climate. One such study cor-
2004 epidemic of severe acute respiratory syndrome showed how related World Resources Institute measures of ‘‘ecological disin-
quickly new pathogens can spread around the world. The source tegrity’’ against data on life expectancy, infant mortality, and
of SARS is not known but organisms of the kind that caused percent low-birth-weight babies for 203 countries (Sieswerda et
SARS frequently emerge from human disruption of biota-rich al. 2001). There was a ‘‘modest relationship’’ between the ecolog-
ecosystems. In this instance, public health systems in a large num- ical and health measures, but Sieswerda et al. pointed out that
ber of countries responded effectively to the threat of a global these relationships are inconsistent, the data are of uneven quality,
epidemic and provide an example of the third category of re- and that other factors (such as GDP) appear to have a stronger
sponse options (WHO 2003b). influence. Another linkage study found no evidence of a negative
Pressures on the health sector as a result of ecosystem distur- relationship between loss of biodiversity and human health at the
bance are likely to be most acute in developing countries. Ways global scale (Huynen et al. 2004).
in which these pressures could be reduced include: In the last decade, the World Health Organization promoted
• strengthening environmental health services; the use of ‘‘burden of disease’’ assessments. These measures ex-
• providing technical and financial assistance to implement the press the total health effect (including both mortality and morbid-
Health for All strategy, including health information systems ity) of any disease or risk factor. The most widely used units of
and integrated databases on development hazards; disease burden are DALYs, the sum of years of life lost from pre-
• strengthening advocacy and health communications at all lev- mature death (taking into account the age of death compared to
els; reviewing delivery of basic health services at the local level natural life expectancy) and the number of years of life lived with
to ensure that priority problems of poor people are adequately a disability (taking into account the duration of the disease and
addressed; weighted by a measure of the severity of the disease) (Murray
• making essential drugs affordable and available to the world’s et al. 1994); One advantage of these measures in the context of
poorer nations, including (where necessary) alterations in the environmental change is that they allow impacts of different
multilateral trade system, national policies, and institutional causal pathways to be combined, such as the combined effects of
drug supply management; climate change on infectious diseases, malnutrition, and the im-
• implementing long-range health and human resource plan- pacts of natural disasters (WHO/WMO/UNEP 2003). This po-
ning to train, recruit, and retain staff and developing codes of tentially allows direct comparisons of the effects of different
conduct for international recruitment of health professionals; ecological changes and can therefore help set priorities.
• strengthening health services for displaced communities and Burden-of-disease assessments depend on access to sufficient
those affected by war or famine or environmental degradation; quantitative data to relate changes in the risk factor to the inci-
• implementing health impact assessment of major development dence of specific diseases. In the environmental health field, they
projects, policies, and programs and monitoring indicators for have therefore been most successfully applied to discrete and rela-
health and sustainable development (WEHAB 2002c). tively localized environmental factors with well-characterized
The following sections examine in more detail some of the health effects, such as air pollution and environmental lead. It is
actions that can be taken by the health sector to lessen harmful more difficult to apply these assessments to ecosystem changes
effects of ecosystem damage on human populations. acting through more diffuse causal pathways. For example, it is
16.3.1 Improved Decision-making in the Health plausible, or even probable, that the reduced availability of fresh
water would adversely affect health by increasing a range of
Sector water-borne diseases and through effects on agriculture, therefore
Decisions affecting ecological systems, whether by politicians or negatively impacting food availability. It is, however, impossible
private organizations and individuals, are determined by a wide to make accurate quantitative measurements of their contribution,
range of inputs. These include empirical evidence, value systems, in the context of the multitude of other causal factors, such as
and financial constraints. Despite this complexity, the health com- human behavior and economic influences on agricultural produc-
munity has an important role to play in presenting evidence of tion.
 Consequences and Options for Human Health 481

BOX 16.3
Developments in Linking Disease to Environmental Factors
Advances in computing power and software have facilitated linkages be- ing the effect of specific ecological characteristics (such as the proportion
tween environmental and disease databases, and have therefore made of land area covered by forest) and therefore allow estimation of the dis-
epidemiological analyses of large-scale ecological change considerably ease effects of alterations in these ecological conditions.
more feasible. Exposure and disease data can be linked either in time or On a global scale, most attention has been focused on investigating
space. Time series methods are particularly well developed for studying the link between climate (and therefore climate change) and vector-borne
the effects of air pollution. In essence, the process involves linking obser- disease. For example, maps of climate variables have been linked to
vations of temporal (typically daily) variations in exposure with a disease maps of the distribution of both malaria (Rogers and Randolph 2000) and
outcome measure at the same point, or with an appropriate time-lag after- dengue (Hales et al. 2002) in order to define the climatic conditions under
wards. The quantitative relationship between the pollutant and disease which each disease is most likely to occur. These statistical models can
outcome of concern can be defined by regression techniques (after speci- then be applied to scenarios of future climate change, to project plausible
fying an appropriate error structure, and controlling for the effects of con- climate-driven changes in disease distribution into the future.
founders such as seasonal variations) (Corvalan et al. 1997). Linkage methods have been tried recently with broad area ecological
Geographic information system software can be used to link predictor measures, other than climate (Sieswerda et al. 2001). There was a ‘‘mod-
and disease outcome data in space as well as over time. In addition, a est’’ but inconsistent relationship between the ecological and health mea-
wide range of satellite sensors provide detailed information on ecological sures, the data were of uneven quality, and other factors (such as GDP)
characteristics such as vegetation, altitude, and climate, with complete appear to have a stronger influence. Another linkage study found no evi-
global coverage at low or no cost. As for time series studies, regression dence of a negative relationship between loss of biodiversity and human
techniques (again taking account of potential confounders and spatial health at the global scale (Huynen et al. 2004). Soskolne and Broemling
auto-correlation between data points) can be used to quantify the relation- (2002), in recognizing the importance that the health sector contributes,
ship between ecological characteristics and disease outcomes. emphasize that methods are needed for developing sensitive measures
The majority of studies of this type have been applied to specific dis- capable of linking ecological degradation with health outcomes. Herein
eases on a sub-national scale. Many of them are designed to generate lies a challenge for eco-epidemiologists.
predictive maps for disease control, but are equally applicable to measur-

Considerations of time scale are important: the burden of dis- This chapter covers economic costing, environmental health indi-
ease attributable to climate change is modest compared to other cators developed by WHO and subsequently applied in a variety
risk factors over the short time scales for which most political of settings, and health impact assessment (Corvalan et al. 1999;
decisions are taken (a five-year horizon, at most), but is more Confalonieri 2001). When policy-makers contemplate decisions
significant when impacts are considered over several decades that impinge on human health they must make choices, and HIA
(WHO/WMO/UNEP 2003). The discount rate chosen for is a means of laying out these choices so that significant conse-
DALY calculations has a very large effect on the rankings of long- quences are not overlooked. These might include, for example,
term problems like climate change. The rate at which future gains the effects on the health of communities and individuals of large-
and losses are discounted can be modified for the DALY formula, scale transport planning (Freeman and Scott-Samuel 2000).
but the burden-of-disease framework fails to take into account HIA is a cousin of environmental impact assessment; both are
that some environmental changes, such as biodiversity loss, are related to integrated impact assessment (Hubel and Hedin 2000,
irreversible. There is no means of weighting effects from which Milner 2004). None require major changes to be applied to assess-
there is no recovery. Finally, such frameworks do not account for ments of ecosystem change. For instance, Mutero (2002) adapted
the different valuation that people give to health risks over which this approach to examine the effect of irrigation projects along the
they have direct individual control, compared to those controlled Tana River in Kenya on rates of schistosomiasis. HIA is not a
by the community as a whole or by other agencies. For example, ‘‘black box’’ for generating policy—it does not avoid the need for
there is greater concern over deaths among passive smokers rather assumptions, approximations, improvisations, and value judg-
than active smokers. Ecological changes usually fall into the cate- ments; but it offers a systematic approach to collecting and ap-
gory of externally imposed change. praising information, and for this reason, has the potential to
Burden-of-disease assessments are therefore appropriate for improve the quality of decisions that affect the state of ecosystems
aggregating health impacts through a range of mechanisms and and human health.
can potentially aid in priority setting and decision-making in the Cost-effectiveness analysis is increasingly used to select among
context of ecosystem change. However, they must be considered different interventions to improve public health. Costs of inter-
as only one component of evidence, as they do not take full ac- ventions (usually measured in monetary terms) are considered
count of features such as complex causal pathways, long time- alongside their resulting health gains (usually measured as deaths,
scales, potential irreversibility, and individual versus community or DALYs, averted). Outcomes from these analyses are quoted as
responsibility (WHO/WMO/UNEP 2003). These important cost-effectiveness ratios (for example, DALYs per dollar) as a mea-
properties need to be included in the final considerations about sure of the value for money of the intervention, often along with
any response to ecological change. aggregate costs and benefits, to represent the overall impact of the
16.3.3 Methods for Selecting Interventions to intervention. When applied in a rigorous and standardized man-
ner, cost-effectiveness analysis can provide an objective ranking
Protect Health of the efficiency of different interventions. This allows policy-
Chapter 3 of this volume reviewed ways in which the effects of makers to select those that provide the greatest health gains for
the environment on health and well-being may be measured. any specified level of resources.
482 Ecosystems and Human Well-being: Policy Responses

Cost-effectiveness analysis requires quantitative data on all sig- 16.4 Cross-sectoral Response Options and
nificant costs and benefits, which in turn requires an understand-
ing of all the important links between the intervention and Actions
eventual health outcomes. Cost-effectiveness analysis has been
employed where the intervention is clearly and directly linked to 16.4.1 Health, Social Development, and
a health outcome, with relatively complete quantitative data on Environmental Protection
the relationships, such as selecting different options to improve Trends in inequality, resource consumption and depletion, envi-
water supplies to reduce diarrhea. Conceptually, it could equally ronmental degradation, population growth, and ill health are
be applied to decisions that act higher up the causal chain, such as closely interrelated (McMichael 1995). This means that better
the effect of land use policies on child health. This is seldom done, health, in the long term, will depend on cross-sectoral policies
however, because the links are more diverse and complex, intro- that promote ecologically sustainable development and address
ducing greater uncertainty into the analysis. There are ways to underlying driving forces. Agenda 21 and the Rio Declaration on
determine the monetary value of nonmarket systems but these are Environment and Development describe a comprehensive ap-
not widely agreed upon. proach to ecologically sustainable development incorporating
cross-sectoral policies (McMichael 2000). The broader topic of
16.3.4 Addressing Risk Perception and sustainable development is discussed further in the next chapter.
Communication Examples of specific relevance to health are the following strate-
gies, developed for the World Summit on Sustainable Develop-
In order for any research on the health effects of ecological change ment (WEHAB 2002c):
to affect either official policy or individual behavior, it is necessary • mitigation strategies that reduce drivers of ecosystem change
to take into account how risk is perceived. A deliberate and well- while simultaneously improving human health;
informed approach to community risk will maximize the chance • adaptation strategies to reduce the effect of ecosystem disrup-
of effective changes through policy interventions that enjoy pop- tion on health (addressing direct, mediated, and long-term
ular support (Slovic 1999). health impacts);
Any assessment of ecological change and health should be in- • integrated action for health, such as health impact assessment
fluenced by the risk perceptions of those communities that are of major development projects, policies, and programs, and
most likely to be affected. That is, ecological assessments should indicators for health and sustainable development;
involve open and frequent stakeholder participation from the be- • inclusion of health in sustainable development planning efforts
ginning of the process rather than as an afterthought (Parkes et al. such as Agenda 21, in multilateral trade and environmental
2003). This approach of community engagement in the process agreements and in poverty reduction strategies;
serves the purpose of accessing local knowledge about the effects • improvement of inter-sectoral collaboration between different
of ecological factors, ensuring that the assessment addresses issues tiers of government, government departments and NGOs;
of greatest concern to those affected and maximizing the proba- • international capacity-building initiatives, that assess health
bility that any recommended change in policy or behavior will be and environment linkages and use the knowledge gained to
adopted. If a source of information is not widely trusted, it is create more effective national and regional policy responses to
unlikely that recommended changes will be accepted. Commu- environmental threats; and
nity surveys have shown that some groups tend to be regarded as • dissemination of knowledge and good practice on health gains
highly trustworthy, while others (such as government agencies) from inter-sectoral policy.
are treated with caution (Maeda and Miyahara 2003). Healthcare The conventional indicators of population health, such as life
providers tend to be one of the ‘‘high trust’’ groups, underlining expectancy, suggest that we have made considerable progress over
again the important role they can play in explaining the signifi- the last hundred years in many parts of the world. Economic de-
velopment and environmental protection are responsible for
cance of healthy ecosystems.
much of this improvement. An important lesson from history is
Any such consultation should make the best use of the exper-
that economic growth is a double-edged sword. On the one
tise of both stakeholders and researchers. Stakeholders may have
hand, it is the engine that generates wealth and opportunity; on
expert local knowledge but may have inaccurate ideas of the true
the other hand, economic growth has tended to be socially dis-
nature of risks associated with different factors; researchers should ruptive and environmentally damaging. The experience of coun-
have more exact knowledge of disease processes and relative risks tries that industrialized early is that, initially, harmful effects
but may inappropriately estimate the applicability of general con- predominated (Wohl 1983). What was needed to turn economic
cepts to local situations. growth into social benefit was the development of robust, inclu-
Accurate and accessible reporting of assessment results can sive political processes and strong public institutions such as public
remedy inaccurate risk perceptions and can enhance the public’s health and local government (Szreter 1997).
ability to evaluate science/policy issues; the individual’s ability to What present-day summary indicators of health status fail to
make rational personal choices is enhanced. In the past, poor re- reveal is the gross inequalities within and among nations, between
porting misled and disempowered a public that is increasingly af- rural and urban areas, and among population subgroups. In some
fected by applications of science and technology (Myers and regions (such as southern Africa), life expectancy remains low and
Raffensperger 1998). Stakeholder engagement will make it more in, some instances, is falling further. Where gains have been made,
likely that the research is credible and is translated into practice. they may be relatively fragile, as shown by the rapid deterioration
Technically intensive, externally driven interventions may of health statistics in Eastern Europe after the break-up of the
produce rapid results but at the risk of marginalizing local com- Soviet Union. Underlying social and political factors include the
munities. Interventions that engage local communities and trans- change from politico-military colonialism to economic depen-
fer expertise are more likely to result in ecologically sustainable dence, and migration from rural areas to urban centers resulting in
improvements. unemployment, poverty, and social disruption (Avila-Pires 2003).
 Consequences and Options for Human Health 483

The accelerating rates of change brought about by high tech- Policies addressing human health needs in relation to food
nology demand urgent solutions. On the positive side, the associ- and nutrition, water and sanitation, and energy services have been
ation of basic research with technological development proved to developed as part of the ‘‘water-energy-health-agriculture-
be a key factor in progress. But we need to find creative ways of biodiversity’’ process and are summarized in Box 16.4. Imple-
extending its benefits to all. Technological progress implies social mentation of these policies will depend on national and local cir-
change and we must stimulate a corresponding effort from sociol- cumstances. For example, in industrialized countries, integrating
ogists and philosophers to help us understand and cope with the national agriculture and food security policies with the economic,
swift pace of change. social, and environmental goals of sustainable development could
be achieved, in part, through taxes on food products to ensure
16.4.2 Linking Health and Ecosystem Responses that the environmental and social costs of production and con-
For each category of ecosystem services, we have extracted from sumption are fully reflected in the price. Taxes should be one
earlier sections of the report a sample of recommended responses. element in a package of policies designed to protect the environ-
(See Table 16.4.) In each instance, we have listed some of the ment without jeopardizing food security for the most vulnerable
possible effects that these responses could have on human health. groups in society. With that proviso, a full-cost approach to food
For simplicity these are illustrative lists, not intended to be ex- pricing may bring major benefits to health and ecosystems, for
haustive. instance through reduced consumption of animal products
Table 16.4 makes the case that in almost every category of (WHO 2003a). Improvements to traditional fuels and cooking
ecosystem response the consequences for health may be either devices could lead to the prevention or at least reduced emissions
positive or negative. The balance will depend on how the policy of local air pollutants, while implementing better transportation
or regulation is framed and what account is taken of contingencies practices and systems could lead to increased physical activity in
and local circumstances. Using trade and economic levers to sedentary populations as well as reductions in greenhouse gas
widen food markets, for instance, has been successful in some emissions (Von Schirnding and Yach 2002; WEHAB 2002b).
instances and, of course, increased food supply can lead to better
health (FAO 2003). However, in other settings, ‘‘globalizing’’
policies have led to deepening poverty, diminished food security,
16.5 Conclusion
and deteriorating standards of public health. This illustrates the Ecosystem disruption damages health in a variety of ways and
fact that national strategies to protect ecosystem services and through complex pathways. The links between ecosystem change
human health can be successful only if the global policy context and human health are seen most clearly among impoverished
is supportive. communities (who lack the ‘‘buffers’’ that the rich can afford).

Table 16.4. Examples of Potential Health Implications of Sectoral Responses

Ecosystem services
under threat Possible Responses Possible Consequences for Health

Floods and storm control waste-water management ▲ improved water quality (fewer enteric infections)
vegetation of water catchments ▼ disease vector proliferation (e.g., urban wetlands)
Food production economic and trade policies to increase reach of global ▲ more food choices—improved nutrition
markets ▲ decreased poverty, consequent improvements in health
▼ reduced food security—especially for the most vulnerable
groups (deepening poverty and reduction in health status)
Climate regulation reduce greenhouse gas emissions (e.g., vehicle ▲ improved air quality
emission standards) ▲ improved water quality
carbon sequestration (e.g., reforestation) ▼ decreased access to health services for the poor
▼ increased fire risk
▼ displaced populations
▼ reduced food production
Wood, woodfuel and economic incentives for re-forestation ▲ reduced flood risk
fiber ▼ increased fire risk
Freshwaterwater charges to reduce wasteful consumption ▲ improved access to sectors in the population
infrastructure (e.g., dams and dikes) ▼ decreased access for low income groups—water-related
diseases
▼ new habitat for disease vectors
Wastes increase recycling ▲ decreased toxic emissions (e.g. from incinerated waste)
reduce amounts of waste ▼ vector-breeding sites—more mosquito-borne disease

Key: ▲ Improved health ▼ Impaired health

484 Ecosystems and Human Well-being: Policy Responses

BOX 16.4
Examples of Responses to Improve Human Health (WEHAB 2002a, 2002b, 2002d)
Food and nutrition responses that can improve human health include: cacy and training programs that contribute to improved household
• integrate national agriculture and food security policies with the eco- hygiene practices for the poor;
nomic, social, and environmental goals of sustainable development; • identify best practices and lessons learned based on existing proj-
• ensure equitable access to agriculture-related services and prod- ects and programs related to provision of safe water and sanitation
ucts, with particular focus on food security and sustainable livelihood services focused on children;
needs of the poor; • create multistakeholder partnership opportunities and alliances at all
• orient market forces toward environmentally optimal solutions levels that directly focus on the reduction of child mortality through
through appropriate policies and regulations; diseases associated with unsafe water, inadequate sanitation, and
• exploit and expand locally available resources for improved food secur- poor hygiene;
ity and promoting diversification for more effective risk management; • develop national, regional, and global programs related to the provi-
• focus on needs of rural areas through decentralized cooperative sion of safe water and improved sanitation services for urban slums
initiatives and improvements in rural infrastructure; and in general, and to meet the needs of children in particular; and
• strengthen regional and international cooperation for food security • identify water pollution prevention strategies adapted to local needs
and market stability. to reduce health hazards related to maternal and child mortality.

Water and sanitation responses to improve human health: Energy and fuel responses to improve human health:
• assign the role of water-related public awareness to the agency • reduce poverty by providing access to modern energy services in
responsible for integrated water resource management at the coun- rural and peri-urban areas;
try level; • minimize the environmental impacts of traditional fuels and cooking
• institute gender-sensitive systems and policies; devices;
• raise awareness and understanding of the linkages among water, • improve air quality and public health through the introduction of
sanitation, and hygiene and poverty alleviation and sustainable de- cleaner vehicular fuels; and
velopment; • implement better transportation practices and systems in mega-
• develop in partnership with all relevant actors community-level advo- cities.

This extends to subpopulations within wealthier communities rich countries’ food overconsumption) or postponed (as in the
who have relatively less access to ecosystem resources. case of long-term consequences of climate change or desertifica-
Poor communities are the most directly dependent upon pro- tion). Decisions about health and ecosystems must consider how
ductive ecosystems for their health. Measures to promote ecologi- one is related to the other. Choices that are made about the man-
cal sustainability will (by definition) safeguard ecosystem services agement of ecosystems may have important consequences for
and therefore benefit health in the long term. This means that the health, and vice versa. Healthy ecosystems protect human health;
poorest and most disadvantaged individuals and communities can healthy people protect their ecosystems.
be among the first to benefit from ecosystem protection, leading Decision-makers need to consider the connections between
to improvements in health equity. health and other sectors. Where there are ‘‘win-win’’ options,
A healthy community is more capable of sustainable develop- these will be attractive to policy-makers; where there are trade-
ment than an unhealthy one. Therefore, where a population is offs, it is important for politicians, regulators, and the public to
weighed down by diseases related to poverty and lack of entitle- understand the consequences of taking one path in preference to
ment to essential resources such as shelter, nutritious food, or another. The health sector bears responsibility for revealing the
clean water, the provision of these resources should be the first links between ecological services and health and indicating which
priority for healthy public policy. interventions are needed: this is despite the fact that responses and
Where disease is caused by unhealthy levels of consumption interventions to protect human health are often carried out in
(especially of food or energy), substantial reductions in this over- other sectors.
consumption would have major health benefits as well as reducing Consideration of ecosystem change enlarges the scope of
pressure on ecosystems. Both human health and the environment health responses by highlighting ‘‘upstream’’ causes of disease and
are likely to benefit from a redistribution of resources if this leads injury. This implies that health considerations should weigh heav-
to basic entitlements being distributed more equitably and a re- ily in decisions on ecosystem responses. History shows that health
duction in overconsumption. Such changes could improve health is one of the most highly valued social outcomes.
in the short term as well as contribute to long-term ecological
sustainability. Win-win outcomes of this kind depend on how References
Allen, B., 2002: Birth weight and environment at Tari, Papua New Guinea Medi-
these changes in resource use and management are achieved.
cal Journal, 45, pp. 88–98.
Local conditions are critical in shaping the health manifesta- Arnell, N.W., 1999: Climate change and global water resources, Global Environ-
tions of ecosystem disruption. Empirical evidence supporting the mental Change, 9(S1-S2), pp. S31–51.
link between ecosystems and health is difficult to find. Our Avila-Pires, F., 2003: Health, biodiversity, and sustainable development: Mak-
knowledge is increasing but there are still many gaps. One reason ing globalization sustainable, M. Pallemaerts (ed.), Brussels University Press,
for this is the many confounding factors (associated with environ- Brussels, Belgium, pp. 111–27.
Baron, J., N. Poff, and P. Angermeier, 2002: Meeting ecological and societal
mental change and also determinants of health) that are hard to needs for freshwater, Ecological Applications, 12, pp. 1247–60.
measure and to separate from the effect of interest. Berkes, F. and C. Folke, 1998: Linking Social and Ecological Systems: Management
The effects of ecosystem disruption on health are frequently Practices and Social Mechanisms for Building Resilience, Cambridge University
displaced, either transferred geographically (such as the costs of Press, Cambridge, UK.
 Consequences and Options for Human Health 485

Cairncross, S., 2003: Sanitation in the developing world: Current status and Hales, S., P. Weinstein, Y. Souares, and A. Woodward, 1999a: El Nino and the
future solutions, International Journal of Environmental Health Research, 13, pp. dynamics of vector-borne disease transmission, Environmental Health Perspec-
S123–31. tives, 107, 99–102.
Chin, J., 2000: Control of Communicable Diseases Manual, American Public Health Hales, S., P. Weinstein, and A. Woodward, 1999b: Ciguatera (fish poisoning),
Association, Washington, DC. El Niño, and Pacific sea surface temperatures, Ecosystem Health, 5, pp. 20–5.
Confalonieri, U., 2001: Environmental change and health in Brazil: Review of Haug, G.H., D. Gunther, L.C. Peterson, D.M. Sigman, K.A. Hughen, and
the present situation and proposal for indicators for monitoring these effects. B. Aeschlimann, 2003: Climate and the collapse of Maya civilization, Science,
In: Human Dimensions of Global Environmental Change, Brazilian Perspectives, 299(5613), pp. 1731–5.
D.J. Hogan and M.T. Tolmasquin (eds.), Brasileira De Ciencias, Rio de Ja- Helming, S. and J. Kuylenstierna, 2001: Water: A Key to Sustainable Develop-
neiro, Brazil, pp.43–77. ment, International Conference on Freshwater, Bonn, Germany.
Corvalan, C., F. Barten, and G. Zielhuis, 2000: Requirements for successful Hubel, M. and A. Hedin, 2000: Developing health impact assessment in the
environmental health decision-making, Decision-making in Environmental European Union, Bulletin of the World Health Organization, 81, pp. 463–64.
Health: From Evidence to Action, C. Corvalan, D. Briggs, and G. Zielhuis Huynen, M., P. Martens, and R. De Groot, 2004: Linkages between biodiver-
(eds.), WHO, London, UK. sity loss and human health: A global indicator analysis, International Journal of
Corvalan, C., T. Kjellstrom, and K. Smith, 1999: Health, environment and Environmental Health Research, 14, pp. 13–30.
sustainable development: Identifying links and indicators to promote action, IFRCRCS (International Federation of Red Cross and Red Crescent Societies),
Epidemiology, 10(5), pp. 656–60. 2002: World Disasters Report, Eurospan, London, and IFRCRCS, Geneva,
Corvalan, C., M. Nurminen, and H. Pastides (eds.), 1997: Linkage Methods for Switzerland.
Environment and Health Analysis, UNEP, Nairobi, Kenya/ US EPA, Washing-
IPCC (Intergovernmental Panel on Climate Change), 2001: Climate Change
ton, DC/ WHO, Geneva, Switzerland.
2001: Impacts, Adaptation and Vulnerability, Contribution of Working Group
DFID/EC/UNDP/World Bank (Department for International Development/
II to the third assessment report of the IPCC, Cambridge University Press,
European Community/United Nations Development Programme), 2002:
Cambridge, UK.
Linking Poverty Reduction and Environmental Management: Policy Challenges and
Jordan, T., D. Whigham, K. Hofmockel, and M. Pittek, 2003: Nutrient and
Opportunities, DFID, London, UK/ Directorate General for Development,
EC, Brussels, Belgium/ United Nations Development Programme, Nairobi, sediment removal by a restored wetland receiving agricultural runoff, Journal
Kenya/ The World Bank, Washington, DC. of Environmental Quality, 32, pp. 1534–47.
Duncanson, M., N. Russell, and P. Weinstein, 2000: Rates of notifiable crypt- Kilbourne, E., 1997: Heat Waves and Hot Environments: The Public Health Conse-
osporidiosis and quality of drinking water in Aoteroa New Zealand, Water quences of Disasters, E. Noji (ed.), Oxford University Press, New York, NY.
Research, 34, pp. 26–34. Kovats, R., K. Ebi, and B. Menne, 2003: Methods for Assessing Human Health
Durie, M., 2001: Mauri Ora. The Dynamics of Maori Health, Oxford University Vulnerability and Public Health Adaptation to Climate Change, WHO, Geneva,
Press, Auckland, New Zealand. Switzerland/WMO, Geneva, Switzerland/UNEP, Copenhagen, Denmark.
Ebi, K., J. Smith, I. Burton, and S. Hitz, 2004: Adaptation to Climate Variability Kunitz, S.J., 1994: Disease and Social Diversity: The European Impact on the Health
and Change from a Public Health Perspective: Integration of Public Health with Ad- of Non-Europeans, Oxford University Press, New York, NY.
aptation to Climate Change: Lessons Learned and New Directions, K.L. Ebi, J.B. Kunzli, N., S. Medina, M. Studnicka, O. Chanel, P. Filliger, et al., 2000: Pub-
Smith, and I. Burton (eds.), Taylor & Francis Group PLC, London, UK, pp. lic-health impact of outdoor and traffic-related air pollution: A European
1–7. assessment, Lancet, 356, pp. 795–801.
Epstein, P., E. Chivian, and K. Frith, 2003: Emerging diseases threaten conser- Leitzmann, C., 2003: Nutrition ecology: The contribution of vegetarian diets,
vation, Environmental Health Perspectives, 111, pp. A506–7. American Journal of Clinical Nutrition, 78(3 Suppl), pp. 657S–9S.
Esrey, S., 1996: Water, waste, and well-being: A multicountry study, American Lindgren, E. and R. Gustafson, 2001: Tick-borne encephalitis in Sweden and
Journal of Epidemiology, 143, pp. 608–23. climate change, Lancet, 358, pp. 16–18.
Esrey, S., 2002: Philosophical, ecological and technical challenges for expanding Lindgren, E., L. Tälleklint, and T. Polfeldt, 2000: Impact of climatic change on
ecological sanitation into urban areas, Water Science Technology, 45, pp. the northern latitude limit and population density of the disease-transmitting
225–58. European tick, Ixodes ricinus, Environmental Health Perspectives, 108, pp.
Ezzati, M., A.D. Lopez, A. Rodgers, S. Vander Hoorn, and C.J. Murray, 2002: 119–23.
Selected major risk factors and global and regional burden of disease, Lancet, LoGiudice, K., R. Ostfeld, K. Schmidt, and F. Keesing, 2003: The ecology of
360(9343), pp. 1347–60. infectious disease: Effects of host diversity and community composition on
FAO (Food and Agriculture Organization of the United Nations), 2003: The Lyme disease risk, Proceedings of the National Academy of Science USA, 100, pp.
State of Food Insecurity in the World, FAO, Rome, Italy. 567–71.
Freeman, N. and A. Scott-Samuel, 2000: A prospective health impact assess- Maeda, Y. and M. Miyahara, 2003: Determinants of trust in industry, govern-
ment of the Merseyside Integrated Transport Strategy, Journal of Public Health ment, and citizen’s groups in Japan, Risk Analysis, 23, pp. 303–11.
Medicine, 22, pp. 268–74. McMichael, A., 1995: Nexus between population, demographic change, pov-
Frumkin, H., 2001: Beyond toxicity: Human health and the natural environ-
erty and environmentally sustainable development, Paper for the Third An-
ment, American Journal of Preventive Medicine, 20, pp. 234–40.
nual World Bank Conference on environmentally sustainable development,
Graber, D., W. Jones, and J. Johnson, 1995: Human and ecosystem health:
London, UK.
The environment-agriculture connection in developing countries, Journal of
McMichael, A., 2000: Human Health: Methodological and Technological Issues in
Agromedicine, 2, pp. 47–64.
Technology Transfer, B. Metz (ed.), IPCC, Cambridge, UK.
Greenwood, B. and T. Mutabingwa, 2002: Malaria in 2002, Nature, 415, pp.
670–2. McMichael, P., 2001: The impact of globalization, free trade and technology
UNEP/GRID-Arendal, 1997: Mapping Indicators of Poverty in West Africa, on food and nutrition in the new millennium, Proceedings of the Nutrition
Technical Advisory Committee working document, Consultative Group on Society, 60(2), pp. 215–20.
International Agricultural Research, Washington, DC/ Food and Agriculture Mellor, J., 2002: Poverty Reduction and Biodiversity Conservation: The Complex
Organization, Rome, Italy. Role for Intensifying Agriculture, World Wide Fund for Nature, Washington,
Hajat, S., K. Ebi, S. Edwards, A. Haines, S. Kovats, and B.M., 2003: Review DC.
of the human health consequences of flooding in Europe and other industrial- Milner, S., 2004: Using health impact assessment in local government, Health
ized civilizations, Applied Environmental Science and Public Health, 1, pp. 13–21. Impact Assessment, S. Palmer (ed.), Oxford University Press, Oxford, UK.
Hajat, S. and R. Kovats, 2002: Impact of hot temperatures on death in London: Morris, J., 1999: Harmful algal blooms, Annual Review of Energy and the Environ-
A time series approach, Journal of Epidemiology and Community Health, 56, ment, 24, pp. 367–90.
367–72. Murray, C.J.L., A.D. Lopez, and D.T. Jamison, 1994: The global burden of
Hales, S., W. Black, and C. Skelly, 2003: Social deprivation and the quality disease in 1990: Summary results, sensitivity analysis and future directions,
of community water supplies in New Zealand, Journal of Epidemiology and Bulletin of the World Health Organization, 72(3), pp. 495–509.
Community Health. In press. Mutero, C., 2002: Health impact assessment of increased irrigation in the Tana
Hales, S., N. de Wet, J. Maindonald, and A. Woodward, 2002: Potential effect River Basin, Kenya. In: The Changing Face of Irrigation in Kenya: Opportunities
of population and climate changes on global distribution of dengue fever: An for Anticipating Change in Eastern and Southern Africa, International Water Man-
empirical model, Lancet, 360(9336), pp. 830–4. agement Institute, Colombo, Sri Lanka.
486 Ecosystems and Human Well-being: Policy Responses

Myers, N.J. and C. Raffensperger, 1998: When science counts: A guide to Townsend, A., R. Howarth, F. Bazzaz, M. Booth, and C. Cleveland, 2003:
reporting on environmental issues, The Networker: Media and Environment, Human health effects of a changing nitrogen cycle, Frontiers in Ecology and
3(2). Environment, 1, pp. 240–6.
National Research Council, 1999: From Monsoons to Microbes: Understanding UNCSD (United Nations Commission on Sustainable Development), 1998:
the Ocean’s Role in Human Health, National Academies Press, Washington, Report of the Export Group of Strategic Approaches to Freshwater Management,
DC. UNCSD–6th Session of the CSD, New York, 20 April–1 May 1998.
Nurse, L. and G. Sem, 2001: Small island states, Climate Change 2001: Impacts, UNEP (United Nations Environment Programme), 2002: Global Environmental
Adaptation and Vulnerability, Contribution of Working Group II to the third Outlook, UNEP, Nairobi, Kenya.
assessment report, J.J. McCarthy, O.F. Canziani, N.A. Leary, D.J. Dokken, UNESCO (United Nations Educational, Scientific and Cultural Organization),
and K.S. White (eds), Cambridge University Press, Cambridge, UK, p. 1032. 2003a: Water for People, Water for Life: UN World Water Development Report,
UNESCO,Division of Water Sciences, Paris, France.
Ostfeld, R. and F. Keesing, 2000: The function of biodiversity in the ecology of
UNESCO, 2003b: Manual on Harmful Marine Macroalgae, UNESCO, Paris,
vector-borne zoonotic diseases, Canadian Journal of Zoology, 78, pp. 2061–78.
France.
Parkes, M., R. Panelli, and P. Weinstein, 2003: Converging paradigms for en-
van Dolah, F., 2000: Marine algal toxins: Origins, health effects, and their in-
vironmental health theory and practice, Environmental Health Perspectives, 111, creased occurrence, Environmental Health Perspective, 108(Suppl. 1), pp.
pp. 669–75. 133–41.
Pruss, A., D. Kay, L. Fewtrell, and J. Bartram, 2002: Estimating the burden of von Schirnding, Y. and D. Yach, 2002: Unhealthy consumption threatens sus-
disease from water, sanitation, and hygiene at a global level, Environmental tainable development, Revista de Saúde Pública, 36, pp. 379–82.
Health Perspectives, 110(5), pp. 537–42. Waltner-Toews, D., 2001: An ecosystem approach to health and its applica-
Reijnders, L. and S. Soret, 2003: Quantification of the environmental impact tions to tropical and emerging diseases, Cadernos de Saude Publica, 17(Suppl.),
of different dietary protein choices, American Journal of Clinical Nutrition, 78(3 pp. 7–36.
Suppl.), pp. 664S–8S WCD (World Commission on Dams), 2000: Dams and Development: A New
Reuer, M. and D. Weiss, 2002: Anthropogenic lead dynamics in the terrestrial Framework for Decision, Earthscan, WCD, London, UK.
and marine environment, Philosophical Transactions of the Royal Society London, WEHAB (Water, Energy, Health, Agriculture, and Biodiversity), 2002a: A
Series A (Mathematical, Physical & Engineering Sciences), 360, pp. 2889–904. Framework for Action on Agriculture, World Summit on Sustainable Develop-
Rogers, D.J. and S.E. Randolph, 2000: The global spread of malaria in a future, ment, Johannesburg, South Africa.
warmer world, Science, 289, pp. 1763–5. WEHAB, 2002b: A Framework for Action on Energy, World Summit on Sustain-
Scheraga, J., K. Ebi, J. Furlow, and A. Moreno, 2003: From science to policy: able Development, Johannesburg, South Africa.
Developing responses to climate change, Climate Change and Human Health: WEHAB, 2002c: A Framework for Action on Health and Environment, World Sum-
mit on Sustainable Development, Johannesburg, South Africa.
Risks and Responses, A. McMichael (ed.), WHO, Geneva, Switzerland.
WEHAB, 2002d: A Framework for Action on Water, World Summit on Sustain-
Sieswerda, L., C. Soskolne, S. Newman, D. Schopflocher, and K. Smoyer,
able Development, Johannesburg, South Africa.
2001: Towards measuring the impact of ecological disintegrity on human WHO (World Health Organization), 2000: The African Summit on Roll Back
health, Epidemiology, 12, pp. 28–32. Malaria, WHO, Geneva, Switzerland.
Skelly, C. and P. Weinstein, 2003: Pathogen survival trajectories: An eco-envi- WHO, 2001: Macroeconomics and Health: Investing in Health for Economic Develop-
ronmental approach to the modeling of human campylobacteriosis ecology, ment, Report of the Commission on Macroeconomics and Health, WHO,
Environmental Health Perspectives, 111, pp. 19–28. Geneva, Switzerland.
Skrabski, A, M. Kopp, and I. Kawachi, 2003: Social capital in a changing soci- WHO, 2002: The World Health Report 2002, WHO, Geneva, Switzerland.
ety: Cross sectional associations middle-aged female and male mortality rates, WHO, 2003a: Diet, Nutrition and the Prevention of Chronic Diseases, WHO, Ge-
Journal of Epidemiology and Community Health, 57, pp. 114–19. neva, Switzerland.
Slovic, P., 1999: Trust, emotion, sex, politics, and science: Surveying the risk- WHO, 2003b: A multicentre collaboration to investigate the cause of severe
assessment battlefield, Risk Analysis, 19, pp. 689–701. acute respiratory syndrome, Lancet, 361, pp. 1730–3.
Smil, V., 2000: Feeding the World: A Challenge for the 21st Century, The MIT WHO, 2003c: The World Health Report 2003, WHO, Geneva, Switzerland.
Press, Cambridge, MA. WHO/UNICEF (UN Children’s Fund), 2004: Joint Monitoring Programme for
Smit, B, O. Pilifosova, I. Burton, B. Challenger, S. Huq, R. Klein, et al., 2001: Water Supply and Sanitation: Meeting the MDG Drinking Water and Sanitation
Adaptation to climate change in the context of sustainable development and Target: A Mid-term Assessment of Progress, WHO and UNICEF, Geneva, Swit-
zerland.
equity, Climate Change 2001: Impacts, Adaptation and Vulnerability, Contribu-
WHO/WMO/UNEP (World Meteorological Organization), 2003: Climate
tion of Working Group II to the third assessment report, J.J. McCarthy, O.F.
Change and Human Health: Risks and Responses, WHO, Geneva, Switzerland/
Canziani, N.A. Leary, D.J. Dokken, and K.S. White (eds), Cambridge Uni- WMO, Geneva, Switzerland/ UNEP, Nairobi, Kenya.
versity Press, New York, NY. Wilson, M. and M. Anker, 2004: Disease surveillance in the context of climate
Smith, K. and S. Mehta, 2003: The burden of disease from indoor air pollution stressors: Needs and opportunities, Integration of Public Health with Adaptation
in developing countries: Comparison of estimates, International Journal of Hy- to Climate Change: Lessons Learned and New Directions, K.L. Ebi, J.G. Smith,
giene and Environmental Health, 206, pp. 279–89. and I. Burton (eds.), Taylor & Francis Group PLC, London, UK, pp. 191–
Soskolne, C. and N. Broemling, 2002: Eco-epidemiology: On the need to 214.
measure health effects from global change, Global Change and Human Health, Wohl, A., 1983: Endangered Lives: Public Health in Victorian Britain, Harvard Uni-
3, pp. 58–66. versity Press, Cambridge, MA.
Steere, A., 2001: Lyme disease, New England Journal of Medicine, 345, pp. Wolfe, A. and J. Patz, 2002: Reactive nitrogen and human health: Acute and
115–25. long term implications, Ambio, 31, pp. 120–5.
Strina, A., S. Cairncross, M.L. Barreto, C. Larrea, and M.S. Prado, 2003: Woodward, A., S. Hales, N. Litidamu, D. Phillips, and J. Martin, 2000: Pro-
Childhood diarrhea and observed hygiene behavior in Salvador, Brazil, Amer- tecting human health in a changing world: The role of social and economic
ican Journal of Epidemiology, 157(11), pp. 1032–8. development, Bulletin of the World Health Organization, 78(9), pp. 1148–55.
Woodward, A., S. Hales, and P. Weinstein, 1998: Climate change and health
Szreter, S., 1997: Economic growth, disruption, deprivation, disease and death:
in the Asia Pacific region: Who will be most vulnerable? Climate Research,
On the importance of the politics of public health for development, Popula-
11, pp. 31–8.
tion and Development Review, 23, pp. 693–728. World Water Forum, 2000: Ministerial Declaration on Water Security in the 21st
Talleklint, L. and T.G. Jaenson, 1998: Increasing geographical distribution and Century, Second World Water Forum, The Hague, The Netherlands.
density of Ixodes ricinus (Acari: Ixodidae) in central and northern Sweden, Yohe, G. and K. Ebi, 2004: Approaching adaptation: Parallels and contrasts
Journal of Medical Entomology, 35, pp. 521–6. between the climate and health communities, Integration of Public Health with
Thompson, T., M. Sobsey, and J. Bartram, 2003: Providing clean water, keep- Adaptation to Climate Change: Lessons Learned and New Directions, K.L. Ebi,
ing water clean: An integrated approach, International Journal of Environmental J.B. Smith, and I. Burton (eds.), Taylor & Francis Group PLC, London, UK,
Health Research, 13, pp. S89–94. pp. 18–43.