c. An adaptation agenda with children in mind In cities in high-income nations, adaptation can focus on preventing disasters. Protective infrastructure and good quality housing are in place, serving almost everyone, and there is the potential for larger investments in adapting protective infrastructure. But most urban centres in low- and middle-income countries lack protective infrastructure and the adaptive capacity for disaster-prevention, and risk levels are much higher. In these urban areas, adapting to climate change-related risks means attention to measures that can reduce the impacts of disasters. Better protective infrastructure is critical, but also disaster-preparedness plans and the capacity for disaster-response and long-term rebuilding.
An adaptation agenda developed with children in mind broadens the terms of discussion in several ways. Where infrastructure, housing and basic services are concerned, it means reconsidering a range of standards and objectives to take account of the concerns of children and those who care for them. Ensuring adequate water supplies, for instance, would mean paying closer attention to the quantities of water that are necessary to protect the health of small children, and the distances that caregivers can reasonably be expected to carry these quantities. Upgrading a road to ensure it is not washed away by flooding means considering the increased traffic that would then pass through a community and the speed of that traffic; when existing roadways are the primary play space for children, this can have implications that are not generally considered.
But the issues go well beyond infrastructure. To some degree, the lack of attention to children reflects a generally lower level of attention to the social and human implications of climate change – as compared, for instance, to the environmental and economic implications. A sharper focus on children within the context of climate change could bring along with it a better awareness of many of the human realities that could use closer consideration in the development of an adaptation agenda – the social dynamics as well as the economic implications, the mental health effects as well as the implications for survival and physical morbidity. Theory and practice regarding children has long stressed the necessity for an integrated approach to development and well being,16 and this same standard could be more broadly applied. Adaptation, in these terms, would mean considering, among other things, how to strengthen and support children’s capacity to cope with the full range of risks associated with climate change, as well as that of the families and communities on which they depend.
This paper first discusses why and how children of different ages are at particular risk from certain aspects of climate change, whether directly or indirectly, explaining some of the impacts for their health and safety, learning, psychological well being and social support. It then explores the implications for adaptation, focusing on risk reduction and preparation for extreme events, as well as responses to the immediate losses and threats of extreme weather events and rebuilding to reduce future risks.
II. SOME BACKGROUND ON CHILDREN a) Risk factors and protective factors Discussion of children’s well being and development is often couched in terms of vulnerability and resilience. Features of their daily lives can be seen either as risk factorswhich pose threats to their well being and optimal development, or as protective factors,which buffer them from these threats and contribute to their resilience and capacity to cope and thrive.17 These concepts can be particularly useful in considering children who routinely face challenges and hardships, and they are very apt for a discussion of the implications of climate change.
This approach is conceptually compatible with an assets and vulnerability approach to considering households in poverty.18 The accumulation of risks or of protective factors is an important concern with children, as it is with vulnerable households. When children are in good health and well nourished, secure and well supported in the world, with opportunities to develop and exercise their competence, they are more likely to meet challenges with resilience and to recover quickly from hardships or shocks. But the likelihood of poor outcomes for children has been found to increase cumulatively with the number of risks that they face.19 This is true whether we are talking about physiological stresses like malnutrition or about psychological challenges. Children on the edge, like families on the edge, have fewer assets to draw on in every sense of the word, and are more likely to be adversely affected by the various challenges imposed by climate change. Children’s level of health, confidence and competence as well as their standing within the family, their family’s capacity to cope, the social support they receive, the quality of the physical surroundings and the range of services available within a community will all help to determine how they respond to both extreme weather events and to the more slow onset deterioration of living conditions. It is important to bear in mind that both risks and protective factors are mediated by the meaning that they hold for children and those around them; it is a mistake to take too mechanical an approach to the relationship between children and their conditions they face.
b) Children’s disproportionate vulnerability to the impacts of climate change Why are children as a group more vulnerable to many of the challenges associated with climate change? Children, and especially very young children, are in a stage of rapid development and are less well equipped on a number of fronts to deal with deprivation and stress. While it is risky to try to draw hard conclusions about their relative vulnerability in areas that are more socially and culturally constructed, their more rapid metabolisms, immature organs and nervous systems, developing cognition, limited experience and particular behavioural characteristics are all at issue here. Table 2 provides a biomedical perspective on these differences. Children are more vulnerable to certain risks, and exposure to these risks may be more likely than it is with adults to have long term repercussions. It is overly simplistic to lump boys and girls of different ages together. The concerns for a two-year-old boy are not the same as those of a 14-year-old girl (who may, in fact, be functioning as an adult woman in some settings.) Their concerns also vary in acuteness, which is not to say that the two-year-old is always the one at greatest risk.
Table 2: Modalities and mechanisms by which children may be more susceptible to climate change than adults.
> respiratory rate
> metabolic rate
> water demand per unit body mass
Air pollution, allergens
Malnutrition, thermal extremes
Gastrointestinal disease, dehydration
> outdoor time
> vigorous activity
< ability to avoid unhealthy situations
< swimming capacity
Infectious diseases, air pollution, UV radiation, thermal extremes, allergens
Weather extremes, UV radiation, thermal extremes
> less surface area:volume
< detoxifying capacity
< skin development
infectious diseases, UV radiation
air pollution, infectious diseases, thermal extremes
Infectious diseases, allergens/mycotoxins
> latency for genetic/long term effect
> lifetime exposure time
UV radiation, allergens, malnutrition
Malnutrition, stunting, psychosocial trauma
Morbidity, quality of life
Source: Bunyavanich, S, C Landrigan, A J McMichael, P R Epstein (2003) The impact of climate change on child health, Ambulatory Pediatrics 3(1) page 47
c) Children as active agents
An important concept here is that of children as active agents. This is a vital component of a rights-based approach to children20 and also an accepted reality within theory on children’s development. Children do not just passively experience the process of development, but are actively engaged in it in purposeful ways, even from their earliest days.21 On some essential level, they know what they need – be it nourishment, attention from caregivers, or opportunities to expand their knowledge and competence – and they help to create the circumstances of their own development. Even the youngest children can make their needs known to caregivers.
Older children have the capacity not only to play a valuable role in identifying and securing the conditions necessary for their own well being; they also routinely make practical contributions to their households and communities. In some cases, they are already surviving on their own without the support of adults. When children’s priorities and perspectives are ignored, it may be more difficult to come up with solutions that reliably meet their needs and those around them. This means that the practical every day realities faced by children and those who care for them, and their perception of these realities, must be clearly understood in order to be effectively responded to. Ideally, for older children this will go beyond consultation to include active engagement in developing and managing solutions. There is ample documentation of the benefits of having children and young people active, informed and involved in responding to the challenges in their lives, not only for their own learning and development as responsible citizens, but for the energy, resourcefulness and depth of knowledge that they can bring to local issues. 22
III. UNDERSTANDING THE IMPACTS FOR CHILDREN OF FACTORS RELATED TO CLIMATE CHANGE This section of the paper considers the range of risks that climate change-related factors pose for children, both directly and indirectly, and the potential impact of these risks. It considers several areas – children’s health and survival, their learning and competence, their emotional security and the social support they receive within families and communities. There are considerable overlaps here – none of these areas can truly be considered in isolation. Poor health affects cognitive development, for instance, and a lack of love and attention may well be reflected in diminished health.
There is not enough hard knowledge about the implications of climate change for children to present a comprehensive picture. Even where there are projections for the more general impacts of climate change, these are not able to specify the likely changes for particular localities and the figures are seldom disaggregated to reflect the specific implications for people of different ages. But it is possible to extrapolate from existing knowledge in a number of related areas. This includes work on environmental health in urban areas, on household strategies to cope with weather variability, the extreme weather that can cause disasters, and their aftermath. It also includes work on the range of effects for children of urban poverty, on the resilience of children and the beneficial effects of their participation in various efforts, all contribute to a broad sense of the potential implications – both of climate-related disasters and responses to disasters, as well as of more gradual change and the adaptations likely to be made at various levels. Nor is this simply an academic exercise, given that, at least for the next few decades, the risks posed by climate change are by and large extensions of the everyday risks that are already experienced in many cities..
a. Health and survival Droughts, floods, cyclones or hurricanes, heat stress and an expansion in the range of various disease vectors take a physical toll on people of all ages. But the disproportionate health burden for children of challenging conditions is well documented. A recent study, calculating the extent to which environmental factors are responsible for the burden of death and disease worldwide, provides an overview of this disparity.23 According to this report’s very conservative estimates, which include only disease burdens that can be reliably measured, 25 percent of deaths in the population at large can be attributed to environmental factors. Among children under 14, however, this rises to 36 percent. The same kind of gap exists in terms of morbidity according to this report, (although the WHO has estimated that two thirds of all preventable ill health due to the environment occur in children.24) The biggest killers for young children are diarrhoeal disease, malaria and respiratory infections, all threats that may be exacerbated by climate change, whether in the context of post-disaster situations or the increasing routine challenges presented by more gradual change.
When these disease burdens are considered in terms of the loss of healthy life years, the figures become even more telling: “Globally, the per capita number of healthy life years lost to environmental risk factors was about 5-fold greater in children under five years of age than in the total population. The difference was even greater (7to10 -fold greater) for major diseases, such as upper and lower respiratory infections, diarrhoea, malaria and malnutrition ... Although these statistics are alarming, they do not capture the longer term effects of exposures that occur at a young age, but do not manifest themselves as disease until years after the exposure.”25
These conservative estimates demonstrate that the disproportionate health burdens for children are not simply a minor matter of degree. As a group, young children in particular suffer dramatically greater health risks than do adults from challenging environmental conditions. While these health risks exist for many poor urban populations even in the absence of climate change, they are likely to be exacerbated in many cases by climate change, and realistic responses to these conditions must take this into account.
Mortality related to extreme weather events
Small children, along with women and the elderly, are generally considered the most likely to be victims of such extreme weather events as flooding, high winds and landslides. This makes sense given their lesser size and strength and capacity to move rapidly. However, there also are studies indicating higher mortality for adult men than for all other groups during these events.26 A study analyzing the causes and circumstances of flood disaster deaths in Europe and the USA, for instance, found that the risk-taking behaviour of males contributed significantly to flood disaster deaths.27 Another study in the USA found that men between 50 and 60 had the highest incidence of hurricane-related injury, largely related to activities after the hurricane (such as removing fallen trees).28
However, these kinds of figures appear primarily in high income countries where adequate housing and infrastructure prevent most potential disaster-related mortality and injury. What these studies indicate more than anything is the huge potential that exists for preventing death (and injury) in the face of extreme events. In low income countries, and especially among the poor, the loss of life is repeatedly demonstrated to be disproportionately high among women and children.
A recently published paper drew on data more than ten years old to demonstrate the significant disparities in the distribution of flood related deaths (see Box 1). These findings, which used an existing database to verify residency prior to the flood, indicated among other things that preschool girls were five times more likely to die than adult men, and that the relative risk of those in poor households was over six times higher than that of high income households.29
Box 2: Flood fatalities in Nepal: highest for children In 1993, a severe flash flood devastated the district of Sarlahi in the southern plains of Nepal. After an unprecedented 24 hour rainfall, a protective barrage on the Bagmati River was washed out during the night, sending a wall of water over 20 feet high crashing through communities, killing over 1600 people. Two months later, a follow up survey assessed the impact of the flood. This survey was unusual in that an existing prospective research database was available to verify residency prior to the flood. As part of a large community-based nutrition programme, longitudinal data existed on children between the ages of two and nine and their parents from 20,000 households, about 60 percent of households in the study area. The survey was able to establish age and gender specific flood-related deaths among over 40,000 registered participants,(including deaths due to injury or illness in the weeks after the flood. Flood related fatalities were 13.3 per thousand for girls between two and nine, 9.4 per 1000 for boys, 6.1 per 1000 for women and 4.1 per 1000 for men. The difference between boys’ and girls’ fatalities existed primarily among children under five. This possibly reflects the gender discriminatory practices that are known to exist in this poor area – the fact that when hard choices must be made in the allocation of resources,, boys are more often the beneficiaries. This could be reflected in rescue attempts as much as it is in nutritional levels. There was also a significant difference in the relative risk of high and low income households, a reflection of the sturdiness of their homes. The homes of the poor were more than 5 times as likely to wash away as the homes of the rich, and those whose homes were swept away were 57 times more likely to die.
Source: Pradhan, Elizabeth Kimbrough, Keith P West, Joanne Katz, Steven C LeClerq, Subarna K Khatry, and Sharada Ram Shrestha (2007) “Risk of flood-related mortality in Nepal”, Disasters 31(1) 57-70.
The distribution of deaths related to the 2004 Indian Ocean tsunami followed a similar pattern. (The tsunami was not related to climate change, but this calamity provides important lessons about both the impacts of a large scale disaster, regardless of cause, and the effectiveness of responses.) A survey of 859 households in evacuation camps in Ampara, Sri Lanka, for instance, found that mortality among children was three to four times that of young adults; that mortality for children under five was double that for adults over 50; and that mortality for females of all ages was double that for males. (See figure 1) The higher numbers of deaths for small children and women were attributed in part to the fact that they were more likely to be indoors when the tsunami came.30
Figure 1: Age and gender distribution of tsunami-related deaths
In slower onset disasters such as droughts and famines, mortality rates also tend to be more extreme for young children – as is reflected in the indicators used to define the severity of an emergency. For a situation to be considered an emergency, it is common to expect that the death rate for children under five be twice as high as that for the population at large.31 This should be put into context. The relationship between crude mortality rates and under-five mortality rates varies a great deal from country to country, and is generally related to the wealth or poverty of that country. Roughly speaking, in a wealthy country, under-five mortality rates are lower than crude mortality rates (see Table 3). The poorer or more trouble-torn a country is, the higher the under-five mortality rate is compared to the crude rate. (Cuba is well known for being an anomaly in this regard.) In some cases, these figures may be exaggerated by governments in order to stimulate higher levels of aid. But certainly, under-five child mortality rates in poor countries are routinely much higher than those of adults.
Table 3: Under-five and crude mortality rates in a range of countries
U-5 mortality rate
Crude mortality rate
Japan (high income)
Finland (high income)
Canada (high income)
China (lower middle income)
Guatamala (lower middle income)
Bangladesh (low income)
Mozambique (low income)
Somalia (low income)
Source. Figures taken from United Nations Children’s Fund (2007) The State of the World’s Children, UNICEF: New York
Given these figures, an emergency threshold mortality rate for under-five children twice that of the crude rate is far from being unrealistic. But it does highlight a dismal reality – that a higher mortality rate for young children, unthinkable in high income countries, should be so routinely accepted as a baseline indicator of normalcy. A Mozambique study illustrates this bias. When research found that food aid programmes in drought stricken Tete province resulted in slightly lower death rates for children under five than for the population at large, the conclusion was that the food aid had been poorly targeted.32 The rate of child deaths commonly used to define an emergency (2/10,000/day) translates into an annual rate (73 per 1,000) that is reported to be exceeded in over a third of the world’s countries.33 An alarming proportion of children in the world, then, are routinely living in a state of emergency by this definition. Overall death rates for young children continue to drop in most parts of the world due to improved nutrition, health care and immunization rates, as well as better environmental health . But for many of the children who are most at risk from diarrhoeal disease, respiratory illness, malaria and malnutrition (the most significant causes of mortality for children) the situation is likely to worsen for most places in low- and middle-income nations with some of the effects of climate change. Many of the most disaster prone countries are also those that already have extremely high infant and child mortality rates.
Water and sanitation-related illnesses
A lack of easy access to sufficient supplies of clean water and to adequate sanitation is a major factor in the health risks associated with climate change – both the extreme events and the slower onset changes. After disasters (whether related to extreme weather or not), a breakdown in provision can be the most serious threat to contend with (Box 2). According to WHO, “Indeed, diarrhoeal diseases, as a result of contamination of water supplies, breakdown of sanitation facilities and the need to scavenge for food, often take a larger toll of life than the original disaster.”34 These diarrhoeal diseases will, in almost all cases, primarily affect small children.
Box 3 : The coping strategies of the urban poor in Bangladesh during the 1998 floods
In 1998, Bangladesh experienced the worst floods in recorded history. The floods lasted over 2 months and 33 million people were marooned. Some of those worst affected were urban slum dwellers in Dhaka. Over half the city’s population lives in settlements with inadequate water supplies, poor sanitation and drainage and few paved streets or lanes. Continuous rain and blocked drains meant floodwaters had risen to four feet in some areas. Most families moved into relief camps and shelters, driven out by the stench of stagnating water, the nuisance of mosquitoes, snakes, leeches and rats, and fears that babies would fall into water or that someone would be electrocuted because of loose electrical lines littering the area But despite the appalling conditions, some families stayed in their homes, fearful of losing valuable possessions. They stayed on rooftops, or raised their beds on bricks or makeshift platforms. “I stay awake all night to guard our household things now that our house is under water,” said one woman, “and the children sleep next to me.” Women stood in line for hours at a time to get safe drinking water, and had to walk long distances to reach a usable latrine. Often they and their children were forced to urinate and defecate in their homes, using newspapers or plastic bags which they then threw into the water; others simply stood in the water. People complained of fevers and diarrhoeal illnesses, skin infections and funguses on their legs and feet. Young children, who played in the dirty water out of boredom, even drinking and bathing in it, suffered from repeated diarrhoea. Most families had to cut down their food intake because costs had sky rocketed and they were without earnings. Unlike their rural counterparts, they had no food stocked up. The lack of work and anxieties about every aspect of life led to increasing tension and sometimes violence between husbands and wives.
Source: Rashid, SF (2000) “The urban poor in Dhaka city: their struggles and coping strategies during the floods of 1998”, Disasters 24(3) 240-253
But slower onset climate-related changes will also take a toll. The IPCC projects that climate change will increase the burden of diarrhoeal disease in low income countries by approximately 2 to 5 percent by 2020.35 This is likely to be felt especially in poor urban settlements, where greater extremes of weather will exacerbate the already considerable difficulties experienced by those without proper provision for water, sanitation and drainage. Sewage and latrine overflows, for example, are far more likely to occur in underprovided urban communities than anywhere else, rural or urban.
Small children are without question the principal victims of sanitation-related illnesses (diarrhoeal disease primarily). In part this is because of their less developed immunity, but also because of their behaviour. They want to play and explore, they touch everything, and they have little consciousness of hygiene. This means they are more likely to come into contact with excreta and pathogens. Children under five are estimated to bear over 80 percent of the global burden of diarrhoeal disease, while school-aged children have by far the highest rates and intensity of helminth infections.36 Child mortality and morbidity rates are more highly correlated with inadequate access to potable water and proper sewerage connections than they are to any other commonly cited variables, such as the number of households in poverty within a community, or the lack of access to health services.37
For small children’s health, water quantity is considered, in most circumstances, to be a more important factor than its quality.38 Children, food, utensils, floors, cooking surfaces are all less likely to be kept clean when water supplies are inadequate or difficult to reach, and this contributes to higher levels of endemic illness. Studies have long indicated, for instance, that both mortality and morbidity for young children go up dramatically when water must be accessed from community wells or standpipes rather than in the house or compound, because households simply make do with less water than they need.39 The amount considered necessary for basic health (about 40 litres per day per person) multiplied by four or five or more family members, adds up to an extremely heavy amount to carry, and a lot of time can be spent waiting at water points and walking back and forth. During times of drought, when water supplies may be especially low or erratic, households are more likely than ever to make do with less water than they need to meet their children’s routine health needs. This problem will be exacerbated in some areas by climate change.
The IPCC’s Fourth Assessment projects that in Africa, by 2020 between 75 and 250 million people will be exposed to increased water stress due to climate change.”40 Freshwater availability is also projected to decrease in Central, South, East and South-east Asia, particularly in large river basins; “along with population growth and increasing demands arising from higher standards of living (this) could adversely affect more than a billion people by the 2050s.”41 During the 20th century, mean precipitation in all seasons has tended to decrease in all the arid and semi-arid regions: Northern Chile, the Brazilian Northeast and Northern Mexico, West Africa and Ethiopia, the drier parts of Southern Africa, and Western China.42 If these trends continue, water-resource limitations will become more severe precisely where they are already critical.43
But the quality of water is also a significant issue, and problems with contamination are likely to be exacerbated by various features of climate change. For urban areas on the coast, rising sea levels, for instance, may cause saline intrusion into groundwater, and may cause latrine pits to flood, contaminating water sources. When water supplies become scarce or difficult to access, families are also more likely to have to store their water, a practice which greatly increases the potential for contamination (for instance, through children’s grubby hands reaching into pails for drinking water.) In a low-income neighbourhood in Abidjan, Côte d’Ivoire, for instance, E. coli was found in only 1 per cent of water samples taken at the community source, but in 41 per cent of samples that had been stored at home.44.
Problems related to inadequate water supplies are complicated further by poor provision for sanitation, which can contribute to the contamination of water supplies, and which greatly increases the need for hygiene. Ironically, sanitation problems and the associated illnesses can result from too much as well as too little water. In Dhaka, for instance, morbidity from diarrhoeal disease was found to increase by 4 percent for every 10mm that rainfall decreased below a certain threshold, but it also increased by 5 percent for every 10 mm that rainfall increased above this threshold.45 This, again is especially an issue in low-income urban settlements, where there are so often high concentrations of both people and wastes without proper provision for sanitation, drainage and waste removal. A considerable body of research relates the quality of local provision to the incidence of diarrhoeal illness in children. Research in Salvador, Brazil, for instance, comparing low-income urban settlements with and without improvements to drainage and sewerage, found that the incidence of children’s diarrhoea in neighbourhoods with drainage and sewerage was one third that in otherwise similar neighbourhoods with neither.46 A related study in the same city found that significant risk factors for infection with Giardia duodenalis were the number of children under five in the household, the absence of a toilet and the presence of visible sewage nearby.47 Yet another Brazilian study points to the connection between diarrhoea in children under five, and the presence of wastewater in the streets and flooding in the house lot.48
As the incidence and intensity of rainstorms increases in many locations as a result of climate change, these conditions are likely to become increasingly prevalent.49 During heavy or prolonged rains, flooding is far more likely to occur in these under-served communities – which, are, in addition, the communities most often located in low-lying and flood prone areas. Blocked drains and flooded latrines can cause feces and waste of all kinds to be spread throughout a settlement, and fecal contamination can be difficult to avoid in densely settled areas. In Lagos, for instance, where many homes are built on stilts above the swamps that are natural flood basins, residents claim that flooding is happening more frequently. They say they are concerned not only about property damage but about the impact on child health, since flood waters carry all sorts of organic waste into their homes.50 In Peru, after the floods related to the 1997-98 El Nino, hospital admissions of young children for diarrhoea were reported to triple.51
In summary then, diarrhoeal illness for young children is a serious problem in many post-disaster scenarios, but the risks in poor urban settlements are also likely to increase as a result of both drought conditions and heavier rains and flooding. The implications of these increased risks may be seen not only in increased mortality and morbidity directly related to diarrhoeal illness, but also in higher levels of malnutrition, increased vulnerability to other illnesses, and effects for overall development, which will be discussed further.
Malaria and other vector-borne and infectious diseases
Altering weather patterns, especially increased temperatures and changes in precipitation, are increasing the incidence and range of various vector-borne diseases. This, too, has particular implications for children. Mosquito and tick borne encephalitis, for instance, are becoming more prevalent, and their incidence is estimated to be twice as high among children between 5 and 10 as among adults. Dengue, another mosquito-borne disease, affects older children and adults more often, but young children are more likely to experience severe symptoms.52 Changes in temperature and rainfall affect the range of disease vectors, but extreme weather events can also contribute to situations where disease vectors (as well as bacteria, viruses, mildew and mold) can thrive and spread.53 Diseases spread by rodents (such as leptospirosis), for instance, appear to increase during flooding or heavy rainfall, especially where there are open sewers and drains.54 After the floods in Mumbai in 2000, 2001, and 2005, the incidence of leptospirosis increased as much as 8 fold. Outbreaks were associated primarily with children playing in flood water and wading through it to get to school. 55
The most serious vector-borne threat, however, is malaria, exposure to which is increasingly prevalent. Up to 50 percent of the world’s population is now considered to be at risk, an increase of almost 10 percent in the last decade.56 This increase is reflected in higher mortality and morbidity rates. Each year now there are up to three million deaths from malaria throughout the world, and about five billion episodes of non-fatal illness. More than 90 percent of the burden is in Africa, and approximately 65 percent of the mortality is among children under five.57 In Kinshasa, the capital of the Democratic Republic of Congo, the P. falciparum parasite was found in the in blood smears of 17 percent of sampled schoolchildren in 1983; by 2000 it had doubled.58 High rates of malaria in places like Kinshasa are related more to poverty and poor management than to changing climate. By 2002, climate change was considered to be directly responsible for only about 6 percent of the burden of malaria in some countries.59 But given the expanding populations at risk of malaria, it is important to consider existing knowledge about the implications for children, and the success of efforts to prevent and manage the disease.
Figure 2: The age distribution of deaths from malaria in Africa (%)
Source: Figures taken from World Health Report (2002) Reducing Risks, Promoting Healthy Life. World Health Organization, Geneva
The disproportionate numbers of children affected by malaria, taken alone, fail to capture the larger implications of malaria for children. An analysis of 48 African demographic surveillance studies found that a higher prevalence of malaria parasites also contributes to deaths from other causes, more than doubling overall mortality for children under five. In areas with a low prevalence of malaria parasite infection (under 25 percent), mortality was 13.7 per 1000 per year for children under five years of age; where parasite prevalence was 25-49 percent, mortality increased to 35 per 1000.60 Malaria results in chronic anemia and increases the severity of other diseases, making mortality from all other causes more likely.
Although the links remain poorly defined, there is also evidence that malaria contributes to impaired development, performance and behaviour in children (including severe motor deficit, visual and hearing impairments, behavioural difficulties, learning problems, language impairment and epilepsy). These potential effects result directly from the insult to the brain during acute episodes of malaria, but can also be mediated by the effects of anemia, repeated illness and undernutrition associated with the disease.61
There are also other indirect effects for children, through the impact that malaria has on household coping strategies. African families may spend a significant part of household income on malaria prevention and control, in addition to the indirect costs of missing work through illness or care for sick family members. The consequences over time can have a serious impact on a family’s capacity to cope. 62 When resources are tight, children’s nutrition and schooling are often the first things to be affected.
Although malaria transmission is considered to be less intense in urban areas because of the greater likelihood of a history of control, the urban population can also be at higher risk because of its lower level of immunity. The danger of epidemics may therefore be higher, and the resulting strain on health services could arguably lead to much higher fatality rates.63 However, epidemics aside, the urban rates are dependent on living conditions. Rates can be especially high in communities where a lack of good drainage creates breeding places for mosquitoes.64 Response measures in poor areas are seldom adequate. A survey in Kampala of families in all income groups, for instance, found that children in 36 percent of surveyed households had experienced an episode of fever in the last two weeks, but less than 1 percent of these received adequate treatmentat the correct dose within 24 hours of onset of fever. Only 11 percent of households used treated nets, and these were all in the wealthiest households (see Box 3).65 Even this level of net use is high for Africa. According to a sample survey of several African countries, insecticide treated nets are used only 2 percent of the time with children under five.66
Box 4: Preventing and treating malaria in children in Kampala, Uganda
There is some evidence that in urban areas in Africa, children, the primary victims of malaria, are likely to receive better preventive care and better treatment when they are ill. A community-based study in Mulago III parish, Kampala, however, suggests that the level of care may be far from meeting recommended standards.
A large swamp extends along the northern border of Mulago III parish, which has poor drainage and frequent flooding during rainy seasons. Malaria is considered to be meso-endemic here. The area is well served with health facilities however. Mulago Hospital, Uganda’s main public hospital, is no more than 2 kilometers from any part of the parish, and in addition there are 9 private clinics and 16 drug shops. A survey was undertaken in 339 randomly selected households, all of which included at least one child between 1 and 10 years of age. Based on household occupation, housing materials and access to electricity, the sample was considered to be slightly better off than the average household in the area. Households were typically one room, occupied by 5 family members. Most caregivers were literate, and nearly half had finished secondary school. Yet despite these protective factors, over a quarter of caregivers reported that a child in their care had died, and in over half the households, a child had been hospitalized. While ownership of bed nets was relatively common in this sample, only 11 percent owned an insecticide treated net, and many of these had not treated the net in the last six months. Most of those who owned treated nets were in the highest income group. This is not surprising, given that the cost of a treated net may be the equivalent of a week’s wages for many households.
In parts of Africa where malaria is endemic, it is widely advocated that children with fever be treated presumptively with anti-malarials, and that treatment with either quinine or combination therapy be initiated within 24 hours. Of the 339 caregivers surveyed in Mulago III parish, 39 percent reported that a child in their care had had an episode of fever in the previous two weeks. The illnesses lasted between 4 and 7 days on average, and most caregivers spent an average of two days caring for sick children rather than dealing with their other responsibilities. .
Almost all the affected children had been treated with commercially produced pharmaceuticals and none were taken to conventional healers. About a third were taken to a private clinic and just a very small number (4 percent) went to the hospital. Care from clinics and hospital tend to involve long waiting times, and is also about four times as costly as simply obtaining medication from a pharmacy – the first option that was chosen in more than half the cases. However, when children did not improve (this was the case for more than a third of them), they were then most often taken to a clinic or the hospital. Overall, only 15 percent of the children were treated within 24 hours, as is recommended. Only 5 percent received the recommended kind of medication within 24 hours, and only a single child received the correct dose of that medication in that important window of time. Even at the hospital and in the clinics, then, the recommended treatment was seldom reliably dispensed.
Source: Kemble, Sarah K, Jennifer C. Davis, Talemwa Nalugwa, Denise Njama-Meya, Heidi Hopkins Grant Dorsey, and Sarah G Staedke (2006) “Prevention and treatment strategies used for the community management of childhood fever in Kampala, Uganda”, American Journal of Tropical. Medicine and Hygiene 74(6). 999–1007
Respiratory illnesses remain a major killer and cause of morbidity for children, causing almost 20 percent of all under-five deaths. A number of factors are involved here, not all of which are directly affected by climate change. Children’s vulnerability to respiratory disease may be related, for instance, to their proximity to traffic, the level of crowding in their homes, and the cooking or heating fuel burned within their homes. But respiratory ailments can also be related to ambient outdoor air quality, According to WHO, 5 million children die annually from diseases linked to air pollution.67 and various factors related to climate change that end up affecting air quality may also have an effect on the prevalence these illnesses. Changes in temperature and precipitation, for instance, are expected to increase the number of forest and bush fires, which can affect air quality for thousands of miles, and which are generally accompanied by increased numbers of people experiencing respiratory difficulty of various kinds.68 Changing pollen counts, fungal growth and moulds related to flooding, increases in ozone and other pollutants, can also play a part in increasing the rates of such respiratory problems as pneumonia, upper respiratory diseases and asthma. Asthma is noted to have doubled worldwide over the last 15 years, for instance, with the greatest increases for children,69 and deaths from asthma are expected by WHO to increase by 20 percent by 2016.70
However, the associations between different factors affecting air quality and the prevalence or severity of different respiratory ailments are far from clear. Factors that appear to be closely related in one situation may not show the same relationship in another – the variables are simply too numerous.71 Nor are children always those most affected. For instance, while children under 14 appeared to be more seriously affected than their elders by elevated rates of air pollution in Hong Kong (based on hospital admissions)72, they were the age group least likely to be admitted as a result of smoke exposure during forest fires in Kuching, Malaysia.73 This could be related to the proximity of fewer children to these fires, however, since a number of physiological and behavioural factors are known to make children more vulnerable to respiratory illness related to pollutants: they breathe more rapidly, and because of their play behaviour, tend to take in more air than adults, and more often through the mouth, all leading to relatively more unfiltered air and potentially damaging air going to the lungs.74
The evidence is that acute respiratory infection (the number one killer of young children) is more closely tied to the quality of indoor air than to outdoor air pollution, especially in the homes of the poor, where the use of smoky cooking or heating fuels in confined spaces can increase the risk of severe respiratory infection by a factor of two to six . In houses with open fires or poorly vented stoves, concentrations of particulates have been estimated to be up to 100 times higher than health standards allow for, and to be many times higher than the outdoor urban concentrations.75 The children most at risk are those that spend the greatest amount of time indoors – often tied to the backs of caregivers as they cook.76 Although indoor air pollution is unlikely to be affected by factors related to climate change, the number of hours spent indoors arguably could be. Where rain storms are increasing in severity and duration, for instance, many children in poor urban communities without adequate drainage are likely to be trapped indoors by flooding for more of the time.
Global climate change is likely to mean an increase in the frequency and intensity of heat waves, and those most at risk of heat stress are the elderly and the very young, who sweat less and have a greater surface area to body mass ratio.77 According to a recent review, “Maximum daily temperature is strongly associated with emergency presentations of fever and gastroenteritis among young children, with UV index negatively associated with gastroenteritis.”78 In São Paulo, an investigation into the impact of temperature on mortality found that for every degree increase above 20°C, there was a 2.6 percent increase in overall mortality in children under 15 – very similar to the increase found in those over 65 years of age (2.5 percent). 79 For younger children, this increase is likely to be higher.
The impact is likely to be most serious in cities, where the “urban heat island” effect can mean differences of as much as 10°C by the end of the night between the city and surrounding areas.80 There can also be statistically significant differences in temperatures between neighbourhoods in the same city during heat waves, with lower socioeconomic groups more likely to live where temperatures are higher. Research in the United States found that higher settlement density, less vegetation, and a lack of open space in a neighbourhood were significantly correlated with higher temperatures. In addition, people living in the warmer neighborhoods had fewer resources to cope with the extreme heat.81 This disparity is also likely in urban areas in low income countries. Although the report from São Paulo did not demonstrate this effect , it acknowledged that results were probably obscured by the disaggregation only by large and socially heterogeneous city districts.82 In low income countries, the urban poor have even less chance of being able to cope adequately with extreme heat. The public health advice that heat stress be responded to by cool baths83 has little chance of being observed in the poorest urban settlements where water is at a premium.
Another effect of heat waves is the natural tendency of those who do strenuous work (most often the poor) to slow down or take more breaks. The lower resulting earnings could have critical implications for some poor households, and for the nutritional levels of their young children.84 Malnutrition
The available literature on the impacts of climate change for children gives greatest attention to the nutritional implications – a reasonable focus given that nutritional status can easily be checked and is also related to many factors associated with climate change. An increase in malnutrition rates is primarily related to food shortages, either because of reduced rainfall and other changes that negatively affect agriculture, or, in the case of sudden acute events, because of interruptions in food supplies.
But malnutrition is also tied to unsanitary conditions and to children’s general state of health. Data collected from over 80 countries indicated that the best predictor of malnutrition, next to the household’s capacity to pay for food, is the level of access to water.85 Frequent bouts of diarrhoea and infestations of worms mean impaired absorption and a loss of nutrients. There is a contention, also, that when children are raised in dirty surroundings, calories that should go towards growth are spent instead supporting their challenged immune systems.86 In other words, children may become malnourished even when there is sufficient food to go around. When they are malnourished, their vulnerability to opportunistic infections is greatly increased, and so a vicious cycle is set in place, one that can have long term effects for their health and development.
Malnutrition may take the form of wasting (acute malnutrition), which is a reflection of children’s current intake or of the failure to absorb nutrients because of diarrhoea or malabsorption. It can also take the form of stunting, the slowing down of linear growth that is the consequence of chronic malnutrition. Generally speaking, in a disaster or emergency, the nutritional risk for children tends to be low if they were previously well nourished.87 If the acute malnutrition associated with the event does not go on for too long, children can recover. They have been found, for instance, to catch up well after seasonal fluctuations.88 But if they are already undernourished, or if the situation continues for too long, it may contribute to stunting and a long term failure to catch up.
In a study in Bangladesh after the severe 1998 floods, children under five who had been exposed to the flood were compared to those in their neighbourhoods who had not. Data were collected at 2, 8 and 15 months after the end of the flood. The data indicated that the linear growth of the flood-exposed children was interrupted and did not fully recover, at least not in the study period. Although children returned to normal growth rates by the end of the study period, they did not experience the “catch-up” growth that is common after a shock, and they remained shorter than the unexposed children. Households had been unable, over time, to compensate for the shortage of food and the general deterioration in the health environment during the flood.89 Research in Zimbabwe found a similarly significant impacts for children’s growth following a drought. In this case, the difference in growth was still evident 4 years after the drought. 90
There is some evidence that malnutrition for children is a greater risk among displaced families. In Sri Lanka after the tsunami, the prevalence of both acute and chronic malnutrition among children in relief camps was found to be significantly higher than the national average.91 In Honduras, after Hurricane Mitch, similarly, children in shelters and resettlement camps were found to be significantly more malnourished than exposed children who had not been resettled.92 It is possible that this is related at least in part to the very poor levels of sanitation that exist in many temporary shelters. Also at higher risk are children in households where adults are ill. A community-based study in an urban neighborhood in Bangladesh found that illness or incapacitation on the part of wage earners was strongly associated with the prevalence of severe undernutrition in under-five children.93.
An important nutritional consideration for infants is the effect that emergencies may have on breast feeding. The benefits of breastfeeding for both physical and emotional nurturance are well established. Breastfeeding has the additional advantage of protecting infants from infection, and hence it is all the more critical to continue the practice in environments without safe water and good sanitation.94 However, the stresses related to a crisis, together with the possibly diminished health and nutrition of the mother, may affect the production of breast-milk in the aftermath of a disaster. Since breast milk substitutes are a common component of the nutritional supports provided in an emergency, it is quite likely that many mothers, in the absence of adequate support and advice, will feel it is practical in this situation to switch to the use of these substitutes. But the preparation and storage of breast milk substitutes, especially in unsanitary environments, can present real health risks to children, since there is ample opportunity for contamination. Without education, mothers may also be unclear about the quantities to be mixed, which may place children at risk of malnourishment. While some infants may require substitutes – those who are separated from mothers for instance, or those in need of temporary supplementation – it is critical that adequate information, clean water and fuel be available for preparation, that every effort be made to restore the mother’s health and nutrition, and that mothers be encouraged to continue breastfeeding even while their supplies are depleted in order to maintain and stimulate their supply.95
After extreme events, injury rates are likely to go up, especially for children. Debris, damaged housing, muddy ground, broken power lines and overcrowded emergency camps with inadequate storage for such materials as fuel, pesticides or medicines, can all pose significant risks. Some research from high income countries, as noted above, has pointed to higher injury rates for men in the immediate aftermath of a disaster, both because of their involvement in clean up activities and because of general risk taking behaviour.96 But children, because of their size and developmental immaturity, are also known to be particularly susceptible to injuries of various kinds. They are curious and driven to explore, yet lack the capacity to understand and respond well to danger. Falls and burns, along with drowning, disproportionately affect children under five.97 In the post-disaster context, with the general confusion and disruption of routines, the level of oversight needed to keep children from harm is much heightened.
There is very little research focused on injuries after extreme events, however, except in the context of high income countries, and even there it is quite limited. One compelling USA study of admissions at pediatric hospitals found that the rate of accidental (“non-inflicted”) traumatic brain injury for children under two went up more than 10-fold in the six months after a hurricane in severely affected areas. The number of admissions for these injuries dropped somewhat after six months, but remained seven times higher than the pre-hurricane levels. Reasons for this increase were speculated to include the greater presence of environmental hazards due to displacement to temporary housing and reduced parental supervision.98 Another USA study in the aftermath of Hurricane Katrina has pointed to the greater likelihood of injury for children in evacuee shelters, and the need for pro-active measures to ensure their safety. 99 The absence of similar post-disaster studies from low income countries does not mean that injury is a less common problem there. It is likely that both the research agenda and the public health measures are focused on the disease-related impacts of disaster, since these remain such a critical threat in low income countries. More general research points to far higher rates of injury for children in poor countries and communities.100 The literature also associates injury with a range of factors that contribute to complexity in the environment and to higher levels of preoccupation or stress on the part of adults.101 This suggests the likelihood of far higher injury rates in the aftermath of disaster.
It is not just the sudden, extreme events or their aftermath that are a concern with regard to children’s injury however. Even adaptations to more gradual change are likely to have implications for children’s safety. As sea levels rise, for instance, or floods become more frequent or intense, more and more poor households are likely to live in areas that are hazardous for children. Housing on stilts and with raised walkways, to give just one example, present a much increased risk of falling and drowning for children.102 Injuries for small children are closely related to overcrowded, challenging living conditions, and their prevalence appears to be much higher in poor urban settlements. In Johannesburg, South Africa, for instance, a study undertaken in six neighbourhoods in a low income area found sharply elevated rates of injury in the informal settlements as compared to those with formal council housing.103 The risks for many households will undoubtedly go up in the absence of measures to provide safe land for secure housing.
In addition to being more susceptible to injury, children are likely to experience more serious and long term effects from injuries because of their size and physiological immaturity (see table 4). For instance, because their bones are still growing, fractures in children are more likely to result in permanent disfigurement; because of their thin epidermis and larger body surface area relative to volume, burns are more likely to be serious; because of the immaturity of their nervous systems and organs, they are more likely to be susceptible to toxins.104 Table 4: Percentage of the total burden of injury experienced