Summary 
The health problems of peri-urban areas have received far less attention than either rural or urban areas. There is considerable interest in promoting the development of natural resource use in peri-urban areas using the opportunities provided by the recycling of urban waste and the proximity to vast urban markets. Both research and assessment procedures are required to ensure that such development projects safeguard human health. A comprehensive review of these issues was commissioned and forms the basis for this paper. 

Cities are expanding rapidly and provide large markets for natural products. These products may be produced in peri-urban areas that are also sinks for the city's waste. The peri-urban zone can be broadly characterised as a mosaic of different land uses inhabited by communities of different economic status, in a state of rapid change with a lack of infrastructure and a deteriorating environment. It is a transition zone that is entirely rural at one end and urban at the other. The focus of the paper is the effect of natural resource development projects on the health of poor communities living in this transition zone. A more precise definition of peri-urban will be left to others. The natural resources themselves may pose health hazards during production, processing, transportation and consumption. There are also many health benefits to the poor from these activities. 

All interventions can have unexpected impacts, positive as well as negative. Such impacts can affect the environment and the community as well as human health. A development project that threatens human health may be judged unsustainable and not cost-effective. One step towards safeguarding health is to conduct rapid health impact assessments of new projects (Birley 1995). Good management is frequently concerned with identifying problems, assessing their importance and proposing solutions. In the case of health impact assessment this consists of identifying health hazards, assessing health risks and proposing risk mitigation measures. The first step, identifying health hazards, can be accomplished by studying the kind of health problems that have arisen on similar projects in other places and at other times. A detailed review has recently been commissioned of peri-urban, natural resource linked health issues and this formed the basis from which the following account is derived (Birley and Lock 1997). The second step, assessment of health risk, can be achieved by sub-dividing the problem into three components. These are the community, environmental and institutional risk factors. Each of these components divides further. For example, the assessment of community risk factors starts by identifying the stakeholders and their immunity, perception and behaviour. The assessment can then rank each health risk as likely to increase, decrease, or remain the same for each stakeholder group as a consequence of the planned development. Although crude, this ranking should be sufficient to trigger the third component of the management response. There are a wide range of possible methods for guarding health during the implementation and operation of development projects. They range from societal to individual responses. They generally have in common the requirement for inter-sectoral collaboration at the design stage. Small changes to design, construction, operation, or maintenance are often all that is required. But these cannot be planned unless appropriate research is conducted and an inter-sectoral debate takes place. 

A large number of studies have described urban health issues, but little health research has focused on peri-urban natural resource production (Bradley et al. 1992; Stephens and Harpham 1992; Atkinson and Merkle 1993; 
Satterthwaite 1993; Songsore and McGranahan 1993; Fluty and Lissfelt 1995; Goldstein et al. 1995; Harpham and Tanner 1995; Environmental Health Project 1996; Listorti 1996; Brantly et al. 1997; Hardoy and Satterthwaite 1997; World Health Organization 1997b; Yacoob and Kelly 1997). In our own review, health hazards are identified either from specific peri-urban research or, more often, by reference to urban and rural differences (Birley and Lock 1997). 

Overview 

The demand from the city for fresh fruit, vegetables, meat or biomass fuels is insatiable. Much of this produce may come from peri-urban farms, limited by the distance that produce can be transported to market. Production requires access to land, water and agro-chemicals. It often re-uses waste materials as input but also produces its own waste streams, especially during post-harvest processing. 

Competition for fresh water is especially strong. Fresh water can be used for both domestic supply and irrigation. But piped water is often unavailable. Fresh water sources include shallow and deep wells, surface water, water tankers and street vendors. Competition depresses ground water levels, while street vendors may charge high prices. Pathogens transported down boreholes pollute groundwater and inorganics are transported through soils. Surface water is polluted by domestic and industrial discharges, as well as by people who are washing, bathing, defecating and urinating. Latrines are often poorly sited and designed, over-used, overflowing, under-maintained, or non-existent. Open defecation on shaded land and by riverbanks is common place. 

The peri-urban environment receives waste from organised collections in the city, as landfill sites, refuse mountains and polluted rivers. Both indoor and outdoor air pollution is present as a consequence of road transport, energy production, industrial activity and cooking. 

People often change their work patterns as they migrate to the city. Paid work more frequently takes place away from home, for both adults and children. The cost of commuting limits the separation of living and working zones. Less time is available for caring and nurturing. Food consumption patterns change. Babies are weaned earlier. Food is cooked less frequently, stored longer, or purchased from street vendors. 

There are links between all categories of health issues and natural resource use. Some of the most evident linkages are listed in Figure 1 and described in detail in the report (Birley and Lock 1997). These include malaria and agriculture, solid waste and wastewater re-use, pathogens and heavy metal poisoning, agrochemical poisoning, biomass fuels and respiratory disease and malnutrition. Psychosocial disorders tend to cut across all other issues and are discussed below. 
 

Figure 1. Examples of health linkages with natural resource use

Rural-urban transitions 

The health issues of the rural-to-urban transition include communicable disease (e.g. malaria), non-communicable disease (e.g. poisoning), injury, malnutrition and psychosocial disorders. These can be divided into traditional and modern diseases. The traditional diseases are mostly communicable or associated with under-nutrition. They are mostly immediate and localised. The modern diseases are mostly non-communicable or associated with injury, over-nutrition or psychosocial disorder. They are mostly delayed in onset and with multifactorial causes. The change in the prevalence rate of these diseases with economic development has been referred to as the health risk transition (Smith 1997). 

The health risk transition can be compared between regions, countries and cities as well as between rural-to-urban zones. The transition is clearest at the regional level. Figure 2 illustrates the changing contribution to morbidity of communicable disease, non-communicable disease and injury in different regions, measured in disability adjusted life years (World Bank 1993). Recent comparisons of mortality between the cities of Accra, in Ghana, and Sao Paulo, in Brazil, show a similar pattern to figure 2 as an example of less and more developed cities (Stephens et al. 1994; McGranaham et al. 1996). Similar disease patterns were also apparent between different socio-environmental zones of Sao Paulo. There are also age-related patterns in poor urban areas: the urban poor must first survive the unsanitary insults of childhood, associated with diarrhoea and other communicable disease, and then the psycho-social effects of poverty in adulthood, associated with intentional injury and circulatory disease. There are similar contrasts in mortality between urban and rural communities in a single country, such as Mexico (Harpham and Tanner 1995). For example, chronic lower respiratory diseases were more common in urban areas while acute lower respiratory and gastro-intestinal diseases were more common in rural areas. Infant mortality rates also show marked differences between rural, low income urban, and other urban communities (Black 1996). The low-income urban rates tend to be much higher. 
These changes in disease patterns from traditional to modern are probably due to changes in the underlying risk factors (Smith 1997). Figure 3 is a testable model of the health risk transition for poor communities between rural, peri-urban and urban areas. The total risk may sometimes be highest in the peri-urban area: poor peri-urban communities may then live in the worst of both worlds (Ashton 1992). 
 
 

Malaria is pre-eminently an environmental disease as its transmission depends on the malaria mosquito. All environmental management, modification and manipulation of tropical and sub-tropical environments are likely to have either a positive or negative effect on mosquito reproduction, contact or survival. The direction of change in prevalence rates varies between ecosystems and requires careful analysis. 

In 1996, one of the influential agencies associated with tropical disease research funding published a policy analysis of future options for malaria research (Anderson et al. 1996). A striking feature of this report was an almost total absence of reference to the environment or to environmental causes of malaria transmission. The report places research emphasis elsewhere and thus illustrates the growing divide between the practical research needs of environmental institutions and the research interests of many health professionals. A number of agricultural research institutions have explicitly acknowledged health issues to be part of their sustainable development strategy, in part because of advocacy by PEEM, a joint WHO/FAO/UNEP/UNCHS committee. These institutions include members of the Consultative Group on International Agricultural Research. They also include the Natural Resources Systems Programme of the Department for International Development of the United Kingdom. 

Urban agriculture provides an example of this need for research. Urban agriculture is being vigorously promoted, at present, through the activities of the urban agriculture network and elsewhere (Smit et al. 1996). The urban agricultural movement has recognised some of the health hazards associated with urban and peri-urban agricultural production. Malaria in Africa is one example where more careful consideration is required. There is a common but mistaken belief among local communities and decision-makers that cereal crop production, such as maize, promotes malaria mosquitoes. Such mosquitoes do not breed in maize plants.  They require sources of open and relatively unpolluted water. Malaria in Africa is largely a rural phenomenon. The frequency of illness and the abundance of malaria mosquitoes changes from one district of a city to another as a result of the mobility of its population, the abundance of breeding sites for the mosquito and the quality of housing and services (Rossi-Espagnet et al. 1991; Atkinson and Merkle 1993). In particular, there are changes along the rural to urban transect (Lindsay et al. 1990; Trape et al. 1992; Adiamah et al. 1993; Lines et al. 1994). Box 1 provides an example. Promotion of urban agriculture may substantially increase the malaria risk by introducing new mosquito breeding sites, but need not do so if appropriately designed. 
 

Malaria in India is different to Africa, emphasising the need for detailed local studies. There has always been urban transmission because the local vector breeds in artificial containers, but the urban ratio is increasing (Sharma 1996). A number of specific ecotypes have been identified.  Industrial malaria is particularly relevant, as industrial complexes are frequently located in peri-urban areas. Industrial malaria can be controlled by a judicious mixture of biological and environmental methods and this is far more satisfactory than chemical control (Dua et al. 1997). 

Other vector-borne diseases associated with urban and peri-urban areas include filariasis and dengue. The breeding sites of the vectors are strongly dependent on solid and/or liquid waste disposal systems. 

Heavy metal poisoning 

Contamination of crops with heavy metals could lead to chronic poisoning of consumers (Smit et al. 1996). The extent of the problem and the validity of solutions proposed are unclear and require further research. 

Contamination of plants with heavy metals may occur through the air as well as from soils and irrigation waters. Many major roads intersect peri-urban areas. Air-borne deposition of heavy metals is frequently associated with road traffic and can substantially effect plant yield (Hassan et al. 1995). A number of hazardous pollutants are associated with road traffic but it is unclear whether respiration or ingestion is the most important exposure pathway (Mage and Zali 1992; World Health Organization 1995). Lead is a cumulative poison and has been widely monitored. Aerosol deposited lead particles, for example, do not penetrate plant surfaces and such contamination can be substantially reduced by washing (Alloway 1995). Other formulations containing lead may behave differently. A Chinese study concluded that food ingestion of lead was far more important than inhalation. In the study area, a high percentage of children in (presumed) peri-urban villages had elevated blood levels and clinical symptoms of lead poisoning, compared to urban children. The source of lead was unclear. A study of bus drivers in Bangkok also observed that most lead absorption was from food, probably purchased from street vendors and kept uncovered (World Health Organization 1995). It has been suggested that a boundary crop should be planted beside roads to protect crops from vehicle pollution (Smit et al. 1996). But some studies found little associated lead contamination of roadside crops (World Health Organization 1995). 

Concentrations of lead are much higher in the liver and kidney of animals and in crustacea and molluscs than in crops or milk (United Nations Environment Programme 1992). A global literature survey concluded that elevated levels of lead in children were more common in urban than non-urban communities, when industrial hot spots were excluded (Anonymous 1994). There was a correlation with air, soil and dust levels. Studies in Nigeria found high levels of lead in the dust from un-paved roads. Other sources of contamination were edible fish from contaminated rivers and streams and cooking salts from springs polluted by nearby mines. Ceramics using lead glazing are also an important source, especially in Mexico. 

Much of the literature on plant uptake of heavy metals is concerned with deliberate extraction in order to decontaminate soil or aqueous streams (Dushenkov et al. 1995; Watanabe 1997). This provides an example of non-food peri-urban agriculture being used to render soils and waters safe for food production. 

Industrial and domestic wastewater re-use 

Relatively little is known about current use of wastewater in peri-urban agriculture except that it is extensive and unregulated. Little is known about local availability, supply and cost of wastewater or the differences between small and large-scale wastewater re-use projects. Wastewater used for irrigation may include industrial effluents, containing heavy metals, as well as domestic waste containing pathogens. 

The risk posed by industrial contaminants will depend on their dilution and uptake pathways. The rivers flowing out of cities often receive mixtures of both domestic and industrial waste. Tanneries are natural resource processing industries that are often found in peri-urban areas discharging  substantial quantities of chromium salts into surface waters (JICA 1996). Chlorination of such effluent may change the salts from a less to a more toxic form (Tarcher 1992). However, some heavy metals precipitate in sludge and so concentrations in treated wastewater may be very small. Other mutagenic materials may be enhanced by activated sludge treatment (Blevins and Brennan 1990). Wastewater effluent is not usually destined to be drunk. However, the chemicals may percolate to groundwater, accumulate and be extracted for drinking from wells. Some chemicals accumulate in soils. 

Guidelines for limiting human exposure to hazardous chemicals in wastewater used in agriculture are at an early stage of development. There are two approaches: prevent any pollutant accumulation in waste-receiving soil; or, more realistically, take advantage of soil’s capacity to assimilate, attenuate and detoxify pollutants. In order to derive acceptable loading, it is necessary to determine intake through consumption of plants grown in contaminated soils. A tentative list of acceptable concentrations of various organic and inorganic compounds in soil is available but is regarded as a first approximation requiring further research (Chang et al. 1995). 

Liquid waste from domestic sewage is a valuable commodity in the peri-urban environment that can be used for irrigation, biogas production, and fertiliser for field crops and fishponds. However, the health hazards associated with such waste include many communicable diseases. 

There has been considerable progress in the development of systems for safe wastewater production and use, culminating in publication of the WHO (Engelberg) Guidelines (Mara and Cairncross 1989; World Health Organization 1989). The Guidelines relaxed standards for faecal bacteria but introduced new standards for nematode eggs. Recent studies have concluded that they are about right for protecting consumers providing that wastewater treatment systems are stable, secondary contamination with untreated wastewater does not occur and wild vegetables are not harvested and consumed by field workers (Blumenthal et al. 1996). The need for a series of barriers to infection was identified. These include crop type, method of irrigation and working practices. Crop types include food eaten raw, food cooked, tree crops and non-food crops. Untreated wastewater can be used to cultivate valuable ornamentals, seedlings and wood crops provided that the workforce is protected. Methods of irrigation include field flooding, localised and spraying. Together, these determine the degree of contamination of both the workforce and the produce. A number of issues remain unresolved. 

Many protozoan parasites have a sufficiently long half-life so that they can contaminate the harvested crop and be transferred through the market chain to the consumer’s kitchen. Produce can be decontaminated by peeling, cooking and, to a small extent, by soaking in antiseptic fluids. But further contamination may occur from soil and debris in the kitchen. 

The WHO guidelines concentrated on standards for bacteria and helminths and not viruses. A recent review of the effect of viruses on human health from wastewater used in agriculture and aquaculture suggests that current bacteriological standards do not provide good indicators of viral content or the efficiency of disinfection procedures and that more research is needed (Schwartzbrod 1995). Particular concern was expressed about decontamination procedures that use chlorination. The risks associated with sprinkler or spray irrigation were not considered to be a significant source of infection except during epidemics when the viral concentration was very high. Viruses also accumulate in the sludge from waste treatment plants and proper drying and long term storage was required to de-activate them, suggesting greater persistence than previously believed. 

Solid waste re-use 

Increased use of urban solid waste as a fertiliser requires policy changes and improved management  (Lewcock 1995; Schillhorn and O'Connell 1997). Collection and disposal of refuse can consume up to 50% of a municipal operating budget. In many cities, only 50-70% of refuse is regularly collected. Where markets for solid waste exist there is usually a thriving trade which supports many destitute people who may live or work on refuse dumps that are often sited in peri-urban areas (Furedy 1996). Many consist of abandoned children and destitute families. They may be exposed to extensive health risks, which are largely undocumented, and suffer exploitation and deprivation. Health hazards include raised levels of infant mortality, hand and leg injuries, intestinal and respiratory infections, eye infections, lower back pain, malnutrition, skin disorders and poisoning. Rates of intestinal worm infection, scabies and upper respiratory tract infection were higher amongst waste picking versus non-waste picking children in India (Hunt 1996). These hazards are relatively under-researched compared to wastewater (Furedy 1996). Water supply, for drinking and washing, and sanitation facilities are usually very poor at dumpsites. 
Composting organic waste for use as soil improver has the positive health benefit of sanitising by heat destruction of pathogens. However, this is only successful if high temperature composting is used or waste is stored for periods of about one year. There is a risk of introducing unacceptable concentrations of heavy metals into the food chain (Jeevan Rao and Shantaram 1995). Composted solid waste can cause injury to farmers as sharp objects are not always properly removed (Nicolaisen et al. 1988; Allison and Harris 1996). 

A South African study of household risk factors for diarrhoea in peri-urban areas identified not owning a refuse receptacle as a significant factor (Atkinson and Merkle 1993). Houseflies are important in the transmission of enteric infections (Chavasse et al. 1996). Disease transmission by houseflies is greatest where inadequate refuse storage, collection and disposal is accompanied by inadequate sanitation. 

Once collected in poorly designed or poorly operated disposal sites, rubbish may contaminate groundwater with nitrates, heavy metals and other chemicals. Incineration of wastes may pollute the air with particulates and oxides of sulphur and nitrogen. The slag and ashes from incinerators may result in leachates that are rich in heavy metals and other potentially toxic substances (World Health Organization 1985).  Combustible gases will be generated from waste tips for more than 20 years and these travel under roads and through ducts to create an explosion hazard in buildings, as happened in Turkey (Associated Press 1993). 

Agrochemical poisoning 

More than 1000 chemical compounds, biological and physical agents are used as insecticides, fungicides, herbicides, rodenticides, fertilisers and antimicrobials. They have been responsible for substantially increasing food production and also for the control of some important human diseases such as malaria and typhus. However, these agrochemicals also cause a wide range of health problems varying from straightforward topical irritant reactions to complex systemic illness which can have both acute and chronic clinical effects (Forget et al. 1993; O'Malley 1997). 

The variation in pesticide use along the rural-urban transition is a matter of debate. It has been suggested that small-scale urban farmers may use little (Smit et al. 1996; Anonymous referee ). A study from Lusaka observed a household usage rate of 23% in rural areas, 64% in peri-urban area and 36% in urban areas (Drescher 1997). There were also gender specific differences. In horticulture there are believed to be many opportunities to reduce the amount and frequency of spraying, although it may be over-optimistic to suggest that use of pesticides can be entirely eliminated. 

There are many potential exposure pathways, both occupational and non-occupational (World Health Organization Commission on Health and Environment 1992b). Some 3 million people annually suffer ill-health from single short-term exposure to pesticides. Approximately 1 million are serious unintentional poisonings and 2 million are suicide attempts (World Health Organization 1986). Fatality rates vary from 1% to 9% and over 700,000 people a year suffer from the chronic effects of long-term exposure. The symptoms of pesticide poisoning may be incorrectly ascribed to other causes. 

The acute clinical effects of pesticide poisoning are very varied. For example, organophosphates can cause diarrhoea, nausea, central nervous system excitation, irritation of the skin and upper respiratory tract, decreased pulse rate leading to dizziness and collapse, headache, memory impairment and loss of sensation (Rosenstock et al. 1990; Amr et al. 1993; Stephens et al. 1995). Other pesticides can cause convulsions, cognitive impairment, liver and kidney impairment, lung fibrosis and coma (O'Malley 1997). Many older and more toxic pesticides are still available in stores and market places. 

Occupational agrochemical poisoning 

Unintentional acute and chronic pesticide poisoning is an occupational hazard of agricultural workers (Loevinsohn 1987; McCracken and Conway 1987; Pingali and Marquez 1990; Stephens et al. 1995). The level of risk of exposure to chemicals is usually higher in intensive farming and horticulture than in traditional farming (World Health Organization Commission on Health and Environment 1992b).  For example, there is evidence of increased adult male mortality among rural intensive rice cultivators in Philippines attributed to chronic exposure (Loevinsohn 1987; Pingali and Marquez 1990). A preliminary study on pesticide exposure among female floriculturalists in Bogota determined that large numbers of pesticides were being used and there were significant differences in foetal loss, prematurity and congenital malformation among women according to their exposure (Restrepo and et al. 1994), 

Lack of knowledge and improper practices are the main causes of poisonings (Forget et al. 1993). Low levels of literacy and education with poor access to training increase the risk. Application is often by itinerant, unskilled, unsupervised operators. It is common to observe storing, mixing, application and disposal without adequate safety precautions. Protective clothing is expensive and hot. There is poor access to water and soap for decontamination. Operatives frequently eat, drink and smoke during spraying operations. Aerial spraying of insecticides often contaminates operatives, casual bystanders and local fauna, resulting in serious levels of exposure. Lack of knowledge of pesticide toxicity does not completely explain dangerous application practices (Yvon 1997). 

Occupational exposure in pesticide factories and stores is often very intense and these facilities may frequently be located in peri-urban or urban areas (Amr et al. 1993). 

Non-occupational agrochemical poisoning 

Rare, unintentional mass poisoning occurs when people consume treated grains or contaminated stored produce or use pesticide containers as cooking utensils, for water storage and collection. Empty containers are often sold in the markets. Chronic illness has also been associated with agrochemicals in the food chain, including red meat, poultry, vegetables and eggs (Forget et al. 1993). Residues are found in human milk and the levels ingested, particularly by nursing infants, are often many times greater than the internationally acceptable daily intake (Food and Agriculture Organization and World Health Organization 1988). The variation along the rural-urban transect is unknown. 

Pesticide residues in locally grown vegetables are frequently far in excess of the acceptable limits. Green leafy vegetables are especially at risk (Conway and Pretty 1991). Many species of wild food, including fish, molluscs, crustacea, insects and vegetables, are harvested among cultivated crops that have been sprayed with chemicals. Such foods are especially important for the poor. Unacceptable concentrations are often found in drinking water (World Health Organization Commission on Health and Environment 1992b). 

Urban and peri-urban communities often depend on groundwater sources that are readily contaminated with nitrogen that originates in fertilisers (Environmental Health Project 1995). Excess nitrogen poses a health risk especially to bottle-fed babies. Increased use of urban solid waste as a fertilising material in peri-urban areas may reduce this risk (Lewcock 1995). 

Biomass fuels: respiratory diseases, injury and malnutrition 

Provision of energy has many direct and indirect benefits to health. For example, electricity is used in refrigeration of food and medical supplies and fossil fuel is used in fertiliser manufacture. There are several major reviews of the negative health impacts of the energy sector (Cooper Weil et al. 1990; World Health Organization 1992; Smith 1993). Hazards identified include: indoor air pollution from biomass and fossil fuel combustion; exposure of urban and peri-urban populations to high levels of ambient air pollution; and serious injuries associated with extraction, storage and power generation. There is an energy transition, similar to the health transition, as people move from more rural to more urban environments and from less developed to more developed economies (ETSU 1996). This is a transition from more traditional to more modern energy sources, also called an energy ladder. Cheaper fuels tend to be more polluting. As income increases people tend to move up the energy ladder. Fuel use can show seasonal variation and purchased fuel may be supplemented with gathered fuel. Households under energy stress employ a range of management strategies that affect their nutrition and food safety. Food vendors may extend their cooking times and hence their exposure to air pollutants. Working women may shorten their cooking time, producing undercooked foods that are less safe. 

Urban and peri-urban poverty may reinforce exposure to pollution at the household level. The poor may have less access to less polluting technologies, pay more for fuel than the more wealthy and live in areas more affected by industrial or traffic pollution (Leitmann 1991). On the other hand, the total pollutant emissions from one large remote fossil-fuelled power station may be less than the pollution from a large number of small wood-burning plants. Energy expenditure occupies a prominent place in poor households. For example, in Kenya very low income communities spent 30% of their income on food and 10-30% on fuel (Harrison 1987; ETSU 1996). Energy collection, production and use is gender related. Household cooking on an open fire has been described as the largest single occupational health problem of women in the world (World Health Organization Commission on Health and Environment 1992a). Infants and young children are exposed to smoke for long periods. The smoke from biomass fuels tends to cause acute respiratory disease in children and chronic obstructive lung disease in adults while the smoke and heat may cause eye diseases (World Health Organization 1992). Stoves are often at floor level causing injuries, especially burns to children, and jeopardising food hygiene (World Health Organization Commission on Health and Environment 1992a; Listorti 1996). 

Respiratory ailments are strongly associated with fuel use and energy efficiency. Fifty percent of the burden of disease in poor countries has been attributed to indoor air pollution (World Bank 1993). In the period 1984-94, some 1.3% of spending in sub-Saharan Africa was devoted to improving efficiency of cooking stoves but it was done for ecological not health reasons (Listorti 1996). Improved cooking stoves may reduce particulate concentrations and carbon monoxide by 50% or more and improved lung function has been associated with a switch to less polluting fuels (Leitmann 1991; Dutt et al. 1996). There is a widespread agreement that a more integrated approach is required that considers improvements in kitchen conditions more generally. Fuelwood is one of the safest crops to grow with wastewater because it minimises the danger of contamination with pathogens or hazardous chemicals. 

An increasing percentage of women’s available time may be spent on fuel gathering because of increasing scarcity. Fuelwood often has to be transported long distances on heads and backs promoting occupational injury. A study in India suggested that the energy cost of collecting fuelwood, water and other domestic chores represented one third of a woman’s daily energy expenditure (Cooper Weil et al. 1990). Removal of large quantities of biomass from a given locality will produce changes in soil, forestation, groundwater recharge, surface runoff and aquatic biota that could adversely affect the productivity of fisheries and farms. This degradation casts an "urban shadow" over the peri-urban areas, especially along transportation routes (Leitmann 1991). 

In addition to domestic air pollution, relatively little research has been done on the occupational health and safety of workers in cottage industries and institutions that use biomass fuels. These include brickmakers, ceramic factories, schools and hospitals. 

Malnutrition, food security and food safety 

Much natural resource development is concerned with food production, processing, delivery and consumption.  The main health problems are contamination of food and lack of food. 

A process of nutritional transition is associated with the rural-urban transition (Popkin 1996). At its simplest this consists of a change from under-nutrition to over-nutrition. Urban diets are considered to differ broadly from rural diets by showing tends towards: "superior grains" such as rice and wheat rather than corn or millet; more milled and polished grains; food higher in fat; more animal products and sugar; food prepared away from home; and more processed food. Average infant and child mortality rates and childhood malnutrition rates appear to be lower in urban compared with rural communities (International Food Policy Research Institute 1996). Some evidence suggests that urban diets are more diverse and the availability of both energy and micro-nutrients is greater in urban diets than rural diets. On the other hand, the diets of the urban and peri-urban poor may sometimes be worse than their rural counterparts as intra-urban differentials are very great. Malnutrition rates are much higher in poorer areas and the differences between poor and wealthy areas are larger than urban/rural differences (Rossi-Espagnet et al. 1991; International Food Policy Research Institute 1996). For example, urban slum communities in Thailand were more under-nourished than rural communities. 

Food security is one determinant of malnutrition and it is a complex problem dependent on  factors such as food quality, quantity, regularity, affordability and with rural, peri-urban and urban components (Hutabarat 1994; Food and Agriculture Organization 1995; International Food Policy Research Institute 1996; Ngleshie-Amanfro Study Team 1996). The price and availability of food often depends on urban purchasing patterns, distribution systems and transport. Wholesale markets are often run-down, too small and badly managed. Small and scattered retail markets cater to the needs of the urban and peri-urban poor who must make frequent purchases of tiny amounts. 

Diarrhoea is still the major cause of morbidity and mortality in children and food contamination is an important cause (World Health Organization Commission on Health and Environment 1992b). Some foodborne pathogens, such as Listeria and Toxoplasma, are dangerous during pregnancy as infection of the foetus can cause death or serious malformations. Under-nourished children are more susceptible to communicable disease and more likely to die than the better nourished (Rossi-Espagnet et al. 1991). The effects of under-nutrition include less than average weight or height, blindness, cretinism, anaemia, bone diseases and poor skin condition. Women and children are especially vulnerable because of differential entitlements that occur within the household. 

One cause of childhood malnutrition is believed to be inadequate maternal care such as breast-feeding, food preparation and hygiene. Infants that were breast-fed for less than 6 months in Latin America had a greatly increased risk of dying. Working mothers often have to rely on artificial feeding that is provided by other child carers so that reduced breast-feeding seems to be a feature of urbanisation. Opportunities for breast-feeding infants in the urban workplace are limited. 

In some countries, 25% of urban household budgets are spent on street foods (International Food Policy Research Institute 1996). This provides a cheap source of food for the poor urban population as well as a source of income for the vendors, who are often disadvantaged women. Studies of street foods in Africa suggested that street food was no less nutritious than modern commercial food and was no less safe than other foods available in the household, providing it was eaten soon after purchase. The purchasers made savings in preparation time, fuel costs and costs of the foodstuffs (Atkinson and Merkle 1993). 

Food preparation and storage in poor domestic environments provides many opportunities for contamination (Motarjemi et al. 1995; World Health Organization 1996). For example, peri-urban migrants in Lima, Peru, prepared their food in the morning, ate some at lunch and held the rest until supper when it was reheated (Bryan et al. 1988). Infants in the households had diarrhoea and contamination of food by faecal matter was observed. Thermos and containers for babies’ milk were not properly sterilised and baby foods were kept too long after opening or preparation. Domestic animals had access to food preparation areas. Reheating procedures were relatively ineffective. Use of refrigerators was poor or they were unavailable. A study, in Liberia, observed higher rates of Campylobacter in urban versus rural children. The water quality was better in urban areas but food was prepared in bulk and stored for several days. The two main reasons for long food storage times were the employment of women outside the home, which reduced time available for cooking, and the high price of charcoal in urban areas (Atkinson and Merkle 1993; World Resources Institute 1996). 

Stored food products are susceptible to contamination by mycotoxins, such as aflatoxin. The many adverse effects can include reduced effectiveness of immunisation programmes, increased susceptibility to communicable diseases such as measles, malaria and HIV, acute fatal poisoning and long term risks of liver cancer (Hendrickse 1991). Aflatoxins cross the placenta and are excreted in the milk of both women and domestic animals. Infants are often exposed to it. Street sellers may obtain their raw materials from a range of uncontrolled sources and these may often be low grade and contaminated. 

Use of contaminated animal faeces, for example from intensively reared poultry, may spread drug-resistant pathogens through the food chain (World Health Organization 1997c). Widespread use of antibiotics in aquaculture may also lead to antibiotic resistance in human pathogens when people eat products that contain high residues, although there are no studies confirming this. Little is known about post-harvest decontamination of vegetable crops grown with solid or liquid waste and a review is under preparation (World Health Organization 1997a). 

Psychosocial disorder 

Mental and psychosocial disorder has only recently started to receive concerted research attention in relation to urbanisation. Mental disorders increase as a component of modern diseases in the health transition (Stephens et al. 1994; Guerrero 1996; McGranaham et al. 1996). A substantial burden of non-communicable diseases has been attributed to this source (World Bank 1993). Rates are about twice as high in women because they are often victims of violent or alcoholic family members. The economic cost is unknown but it contributes significantly to lost productivity, hospitalisation and early retirement. Intentional injury was identified as an important component of mortality in Sao Paulo. 

A model of mental disorder recognises a set of stressors, or risk factors, including the following related to the peri-urban/urban environment (Reichenheim and Harpham 1991; Harpham and Blue 1995; Blue and Harpham 1996). 

o Poor physical environment, including lack of open space, overcrowding, noise; 
o Switch from subsistence to cash cropping; 
o Insecure tenure; 
o Women's labour force participation; 
o Under-employment; 
o High levels of violence and accidents; 
o Rural-urban migration; 
o Lack of control over events and lack of community support; 
o Negative life events such as unemployment. 

The evidence is mixed regarding differentials between rural and urban environments, there are many confounding factors. An important set of risk factors includes life stresses in combination with the corresponding lack of resources to resolve them. But this is not to imply that only external events are relevant. The vulnerability of individuals is very variable. Many people show remarkable resilience and coping strategies when faced with extreme situations. Others may suffer considerably from more subtle manifestations of disorder. There are also cultural aspects to the expression and resolution of emotion. 

Conclusion 

There are many more health issues associated with peri-urban natural resource development than those illustrated above. Examples include zoonoses and dust induced lung diseases. These are described in the main report (Birley and Lock 1997). Consequently, there are many areas where joint research is required between natural resource and health specialists. But such research is of little consequence unless it is translated into action. One area where this is possible is the prospective analysis of the health impacts of proposed development projects. Once the risk is identified there are often many opportunities for mitigation. 

Acknowledgements 

This study was funded by the Department for International Development of the United Kingdom who can accept no responsibility for any information provided or views expressed. We would like to thank the numerous staff of both national and international organisations who spared time to answer our questionnaire and to meet with us. The full list is included in the main report (Birley and Lock 1997) and they will each receive a copy. We thank Dr Chris Lewcock of NRI who managed this project and remained encouraging throughout. 

References 

Adiamah, J., K. Koram, M. Thomson, S. Lindsay, J. Todd and B. Greenwood (1993). "Entomological risk factors for severe malaria in a peri-urban area of The Gambia." Annals of Tropical Medicine and Parasitology 87: 491-500. 
Allison, M. and P. Harris (1996). A review of the use of urban waste in peri-urban interface production systems, The Henry Doubleday Research Association, Coventry. 
Alloway, B. J., Ed. (1995). Heavy metals in soils. London, Blackie Academic and Professional. 
Amr, M. M., M. El Batanouni, A. Emara, N. Mansour, H. H. Zayat, G. A. Atta and A. Sanad (1993). Health profile of workers exposed to pesticides in two large-scale formulating plants in Egypt. Impact of pesticide use on health in developing countries. G. Forget, T. Goodman and A. de Villiers. Ottawa, International Development Research Centre: 118-130. 
Anderson, J., M. Maclean and C. Davies (1996). Malaria research: an audit of international activity. London, Wellcome Trust Unit for Policy Research in Science and Medicine. 
Anonymous (1994). The global dimensions of lead poisoning: an initial analysis. Washington DC, Environmental Defense Fund, Alliance to End Childhood Lead Poisoning. 
Anonymous referee (1998). Personal Communication. 
Ashton, J., Ed. (1992). Healthy cities. Milton Keynes, Open University Press. 
Associated Press (1993). Refuse kills 13. The Times. London: 15. 
Atkinson, S. and A. Merkle, Eds. (1993). Urban health in Africa. London, IIED. 
Birley, M. H. (1995). The health impact assessment of development projects. Liverpool, HMSO, London. 
Birley, M. H. and K. Lock (1997). A review of the health impacts of peri-urban natural resource development. Chatham, Natural Resources International. 
Black, M. (1996). Thirsty cities, water sanitation and the urban poor. London, WaterAid. 
Blevins, R. D. and L. A. Brennan (1990). "Fate of mutagenic activity during conventional treatment of municipal wastewater sludge." Archives of Environmental Contamination and Toxicology 19(5): 657-664. 
Blue, I. and T. Harpham (1996). "Mental health matters." World Health 49(1): 34-35. 
Blumenthal, U., D. Mara, R. Ayres, E. Cifuentes, A. Peasey, R. Stott, D. Lee and G. Ruiz-Palacios (1996). "Evaluation of the WHO nematode egg guidelines for restricted and unrestricted irrigation." Water Science Technology 33(10-11): 277-283. 
Bradley, D., C. Stephens, T. Harpham and S. Cairncross (1992). A review of environmental health impacts in developing country cities. Washington DC, The International Bank for Reconstruction and Development/The World Bank. 
Brantly, E., R. Hetes, B. Levy, P. Clydette, L. Whiteford and M. Yacoob (1997). Comparative health risk assessment: a method for setting priorities in environmental health. International perspectives on environment, development and health, towards a sustainable world. G. Shahi, B. Levy, A. Binger, T. Kjellstrom and R. Lawrence. New York, Springer: 388-409. 
Bryan, F. (1992). Hazard analysis critical control point evaluations, a guide to identifying hazards and assessing risks associated with food preparation and storage. Geneva, World Health Organisation. 
Bryan, F., S. Michanie, N. Fernandez, M. Vizcarra, D. Taboada, O. Navarro S., A. Alonso and E. Requejo (1988). "Hazard analyses of foods prepared by migrants living in a new settlement at the outskirts of Lima, Peru." Journal of Food Protection 51: 323. 
Chang, A., A. Page and T. Asano (1995). Developing human health-related chemical guidelines for reclaimed wastewater and sewage sludge applications in agriculture. Geneva, World Health Organization. 
Chavasse, D., N. Ahmad and T. Akhtar (1996). "Scope for fly control as a diarrhoea intervention in Pakistan: a community perspective." Social Science and Medicine 43(8): 1289-1294. 
Conway, G. R. and J. N. Pretty (1988). "Fertilizer risks in the developing countries." Nature 334(6179): 207-208. 
Conway, G. R. and J. N. Pretty (1991). Unwelcome harvest: agriculture and pollution. London, Earthscan Publications. 
Cooper Weil, D. E. C., A. P. Alicbusan, J. F. Wilson, M. R. Reich and D. J. Bradley (1990). The impact of development policies on health: A review of the literature. Geneva, World Health Organization. 
Drescher, A. W. (1997). Urban agriculture in the seasonal tropics of Central Southern Africa. Management strategies in African homegardens & the need for new extension approaches, www.cityfarmer.org/AxelA.htm, City Farmer, Canada's Office of Urban Agriculture. 
Dua, V., S. Sharma, A. Srivastava and V. Sharma (1997). "Bioenvironmental control of industrial malaria at Bharat Heavy Electrical Ltd. Hardwar, India - results of a nine-year study (1987-95)." Mosquito Control 13(3): 000. 
Dushenkov, V., P. Nanda Kumar, H. Motto and I. Raskin (1995). "Rhizofiltration: the use of plants to remove heavy metals from aqueous streams." Environmental Science and Technology 29(5): 1239-45. 
Dutt, D., D. K. Srinivasa, S. B. Rotti, A. Sahai and D. Konar (1996). "Effect of indoor air pollution on the respiratory system of women using different fuels." National Medical Journal of India 9(3): 113-117. 
Environmental Health Project (1995). Strategic framework 1995-1999. Washington DC, Environmental Health Project (USAID). 
Environmental Health Project (1996). Health and the environment in urban poor areas: avoiding a crisis through prevention. Environmental Health Project Capsule Report. Washington DC, Environmental Health Project (USAID). 1: 1-8. 
ETSU (1996). A review of energy utilisation in peri-urban production systems. Harwell, ETSU. 
Fluty, H. and J. Lissfelt (1995). USAID's experience in urban health and directions for the future. Urban health in developing countries, progress and priorities. T. Harpham and M. Tanner. London, Earthsan: 153-171. 
Food and Agriculture Organization (1995). Street foods. Calcutta, Food and Agriculture Organisation. 
Food and Agriculture Organization and World Health Organization (1988). "Guidelines for predicting the dietary intake of pesticide residues." Bulletin of the World Health Organization 66: 429-434. 
Forget, G., T. Goodman and A. de Villiers, Eds. (1993). Impact of pesticide use on health in developing countries. Proceedings of a symposium held in Ottawa, Canada, 17-20 September 1990. Ottawa, IDRC. 
Furedy, C. (1996). Solid waste reuse and urban agriculture - dilemmas in developing countries: the bad news and the good news. Association of Collegiate Schools of Planning and Association of European Schools of Planning, joint international congress, Myerson Polytechnic University, Toronto. 
Goldstein, G., A. Rossi-Espagnet and I. Tabibzadeh (1995). How the World Health Organization supports urban health development. Urban health in developing countries, progress and priorities. T. Harpham and M. Tanner. London, Earthsan: 110-122. 
Guerrero, R. (1996). "Cali: the right priorities." World Health 49(1): 10. 
Hardoy, J. and D. Satterthwaite (1997). Health and environment and the urban poor. International perspectives on environment, development and health, towards a sustainable world. G. Shahi, B. Levy, A. Binger, T. Kjellstrom and R. Lawrence. New York, Springer: 123-162. 
Harpham, T. and I. Blue, Eds. (1995). Urbanisation and mental health in developing countries. Aldershot, Avebury. 
Harpham, T. and M. Tanner, Eds. (1995). Urban health in developing countries, progress and priorities. London, Earthsan. 
Harrison, P. (1987). The greening of Africa: breaking through in the battle for land and food. London, Paladin. 
Hassan, I., M. Ashmore and J. Bell (1995). "Effect of ozone on radish and turnip under Egyptian field conditions." Environmental pollution 89: 107-114. 
Hendrickse, R. G. (1991). "Clinical implications of food contaminated by aflatoxins." Ann Acad of Med (Singapore) 20(1): 84-90. 
Hunt, C. (1996). "Child waste pickers in India: the occupation and its health risks." Environment and Urbanisation 8(2): 111-118. 
Hutabarat, L. (1994). Street foods in Bangkok, the nutritional contribution and the contaminants content of street foods. Rome, Food and Agriculture Organization. 
International Food Policy Research Institute (1996). Urban challenges to nutrition security: a review of food security, health and care in the cities. Washington DC, International Food Policy Research Institute. 
Jeevan Rao, K. and M. V. Shantaram (1995). "Contents of heavy metals in crops treated with urban solid wastes." Journal of Environmental Biology 16(3): 225-232. 
JICA (1996). Study on the development of a water supply system for Damascus city. Phase 1. Damascus, Nippon Koei Co and DAWASSA. 
Leitmann, J. (1991). Energy-environment linkages in the urban sector, UNDP. 
Lewcock, C. (1995). "Farmer use of urban waste in Kano." Habitat International 19(2): 225-234. 
Lindsay, S., H. Campbell, J. Adiamah, A. Greenwood, J. Bangali and B. Greenwood (1990). "Malaria in a peri-urban area of  The Gambia." Annals of Tropical Medicine and Parasitology 84: 553-562. 
Lines, J., T. Harpham, C. Leake and C. Schofield (1994). "Trends, priorities and policy directions in the control of vector-borne diseases in urban environments." Health Policy and Planning 9(2): 113-129. 
Listorti, J. (1996). Bridging environmental health gaps. Washington DC, The World Bank. 
Loevinsohn, M. E. (1987). "Insecticide use and increased mortality in rural Central Luzon, Philippines." The Lancet 332(8546): 1359-1362. 
Mage, D. and O. Zali, Eds. (1992). Motor vehicle air pollution, public health impact and control measures. Geneva, World Health Organization. 
Mara, D. and S. Cairncross (1989). Guidelines for the safe use of wastewater and excreta in agriculture and aquaculture. Geneva, World Health Organization. 
McCracken, J. A. and G. R. Conway (1987). Pesticide hazards in the third world: new evidence from the Philippines. London, International Institute for Environment and Development. 
McGranaham, G., J. Songsore and M. Kjellen (1996). Sustainability, poverty and urban environmental transitions. Sustainability, the environment and urbanization. C. Pugh. London, Earthscan: 103-133. 
Motarjemi, Y., F. Kaferstein, G. Moy, K. Miyagishima, S. Miyagawa and A. Reilly (1995). Food technologies and public health. Geneva, World Health Organization. 
Nanda Kumar, P., V. Dushenkov, H. Motto and I. Raskin (1995). "Phytoextraction: the use of plants to remove heavy metals from soils." Environmental Science and Technology 29(5): 1232-38. 
Ngleshie-Amanfro Study Team (1996). Ngleshie-Amanfro: A participatory rapid appraisal of food security in a peri-urban community, Noguchi Memorial Institute for medical Research, Centre for Community Studies, Action and Development, May Day Rural Project, International Food Policy Research Institute. 
Nicolaisen, D., U. Plog, E. Spreen and S. B. Thapa (1988). Solid waste management with people's participation. An example in Nepal. Eschborn, GTZ. 
O'Malley, M. (1997). "Clinical evaluation of pesticide exposure and poisonings." The Lancet 349: 1161-1166. 
Pingali, P. L. and C. Marquez (1990). Health costs of long term pesticide exposure in the Philippines - a medical and economic analysis. Annual General Meeting of the American Agricultural Economics Association, Vancouver BC, Canada, Social Science Division, International Rice Research Institute, Manila, Philippines. 
Popkin, B. (1996). "Understanding the nutrition transition." Urbanisation and health newsletter 30(September): 3-19. 
Reichenheim, M. and T. Harpham (1991). "Maternal mental health in a squatter settlement of Rio de Janeiro." British Journal of Psychiatry 155: 44-47. 
Restrepo, M. and et al. (1994). Pesticides exposure and reproductive outcomes. Women, health and development, an anthology. J. Sims. Geneva, World Health Organization. WHO/EHG/94.11: 162. 
Rosenstock, l., M. Keifer and W. Daniell (1990). "Chronic central nervous system effects of acute organophosphate pesticide poisoning." The Lancet 338: 223-227. 
Rossi-Espagnet, A., G. Goldstein and I. Tabibzadeh (1991). "Urbanisation and health in developing countries: a challenge of health for all." World Health Statistics Quarterly 44(4): 187-244. 
Satterthwaite, D. (1993). "The impact on health of urban environments." Environment and Urbanization 5(2): 87-111. 
Schillhorn, T. and P. O'Connell (1997). Urban waste and rural soil management. Agriculture Technology Notes, Agricultural and Forestry Systems Division, World Bank. 17. 
Schwartzbrod, L. (1995). Effect of human viruses on public health associated with the use of wastewater and sewage sludge in agriculture and aquaculture. Geneva, World Health Organization. 
Sharma, V. (1996). "Re-emergence of malaria in India." Indian Journal of Malaria Research 103: 26-45. 
Smit, J., A. Ratta and J. Nasr (1996). Urban agriculture, food, jobs and sustainable cities. New York, United Nations Development Programme. 
Smith, K. (1993). "Fuel combustion, air pollution exposure, and health: the situation in developing countries." Annual Review of Energy and the Environment 18: 529-566. 
Smith, K. (1997). Development, health and the environmental risk transition. International perspectives on environment, development and health, towards a sustainable world. G. Shahi, B. Levy, A. Binger, T. Kjellstrom and R. Lawrence. New York, Springer: 51-62. 
Songsore, J. and G. McGranahan (1993). "Environment, wealth and health: towards an analysis of intra-urban differentials within the Greater Accra Metropolitan Area, Ghana." Environment and Urbanization 5(2): 10-34. 
Stephens, C. and T. Harpham (1992). "Health and environment in urban areas of developing countries." Third World Planning Review 14(3). 
Stephens, C., I. Timaeus, M. Akerman, et al. (1994). Environment and health in developing countries: an analysis of intra-urban differentials using existing data. Accra, Ghana and Sao Paulo, Brazil, London School of Hygiene and Tropical Medicine, Fundacao SEADE, Brazil. 
Stephens, R., A. Spurgeon and I. Calvert (1995). "Neuropsychological effects of long-term exposure to organophosphates in sheep dip." The Lancet 345: 1135-1139. 
Tarcher, A. B., Ed. (1992). Principles and practice of environmental medicine. New York, Plenum. 
Trape, J. and A. Zoulani (1987). "Malaria and urbanization in Central Africa: the example of Brazzaville." Transactions of the Royal Society of Tropical Medicine and Hygiene 81(Supplement 2): 1-33. 
Trape, J.-F., E. Lefebvre-Zante, F. Legros, G. Ndiaye, H. Bouganali, P. Druilhe and G. Salem (1992). "Vector density gradients and the epidemiology of urban malaria in Dakar, Senegal." American Journal of Tropical Medicine and Hygiene 47(2): 181-189. 
United Nations Environment Programme (1992). The contamination of food. Nairobi, United Nations Environment Programme. 
Watanabe, M. (1997). "Phytoremediation on the brink of commercialization." Environmental Science and Technology 31(4): 182A-186A. 
World Bank (1993). World development report 1993. New York, Oxford University Press. 
World Health Organization (1985). Environmental pollution control in relation to development, World Health Organization. 
World Health Organization (1986). Informal consultation on planning for the prevention of pesticide poisoning, World Health Organization. 
World Health Organization (1989). Health guidelines for the use of wastewater in agriculture and aquaculture. Geneva, World Health Organization. 
World Health Organization (1992). Epidemiological, social and technical aspects of indoor air pollution from biomass fuel. Geneva, World Health Organization. 
World Health Organization (1995). Human exposure to lead. Bangkok, World Health Organization. 
World Health Organization (1996). Guidelines for strengthening a national food safety programme. Geneva, World Health Organization. 
World Health Organization (1997a). Personal Communication. 
World Health Organization (1997b). Health and environment in sustainable development: five years after the Earth Summit. Geneva, World Health Organization. 
World Health Organization (1997c). "Multi-drug resistant Salmonella typhimurium." WHO Factsheet 139: 4. 
World Health Organization Commission on Health and Environment (1992a). Report of the Panel on Energy. Geneva, World Health Organization. 
World Health Organization Commission on Health and Environment (1992b). Report of the panel on food and agriculture. Geneva, World Health Organization. 
World Resources Institute (1996). World resources 1996-97. New York, Oxford University Press. 
Yacoob, M. and M. Kelly (1997). "Creating sustainable environmental health conditions by redefining municipal roles and responsibilities: experiences from Tunisia and Ecuador." Natural Resources Forum 21(1): 39-50. 
Yvon, J. (1997). Farmers'perceptions: pests and their management for tomato and garden egg production, IFCSP.