Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
The duration of exclusive breast-feeding and age at supplementation are highly variable among the countries reviewed. In India, 78% of urban and 88% of rural women continued exclusive breast-feeding through the sixth month. Only 6% of urban and 2% of rural mothers gave supplemental foods regularly at two to three months [18]. Milk-based foods were generally given during the first six months to the small proportion of infants receiving supplements. In the Philippines, however, extremely low percentages of urban (9%) and rural (5%) mothers breast-fed exclusively through the sixth month. Percentages of women regularly giving supplemental foods at two to three months were slightly lower among urban poor (23%) than rural women (29%). Supplemental foods were typically milk-based for the first three months, although cereals were also introduced during this time.
Data from the Malaysian Family Life Survey indicate very early infant supplementation. In a 1975-1976 cohort of breast-fed infants, only 4% were not receiving supplemental foods regularly by the third month [28]. Urbanization in Malaysia was related to earlier supplementation, and this occurred in both urban and rural areas [28].
Latin America and the Caribbean
In Mexico and Guatemala urban mothers gave supplemental foods earlier than their rural counterparts. In Guatemala 52% of urban mothers gave supplemental foods regularly in the third month compared to 12% of rural mothers. The differential was small in Chile, however, where 59% of urban and 56% of rural mothers supplemented regularly in the third month [18]. In Chile and Guatemala, milk-based foods were generally fed during the first three months, followed by cereals.
A study of infant-feeding patterns in Mexico City [29] compared 70 families who migrated to the city from a rural district and 75 families still living in the same rural villages. Rates of exclusive breast-feeding for the first three months were lower in urban (36%) than in rural (43%) areas. Urban mothers were more likely to give infant formula as a complement or substitute for breast milk and also gave solid food at an earlier age.
Africa and the Near East
Of the three African countries compared in the WHO collaborative study [18], Nigeria reported the highest percentage of urban women providing regular supplemental foods by the third month (63%) and the largest differential between urban and rural mothers (35%). Urban-rural differences were almost negligible in Ethiopia and Zaire, where about 30% of women fed supplements by the third month. Types of supplemental foods varied by country. In Ethiopia primarily milk-based liquids were used. In Nigeria milk-based foods were given for the first three months followed by cereals. In Zaire cereals were given widely during the first six months.
Women's nutrition
Women represent a particularly vulnerable population whose nutritional status is important to infant survival and household welfare in numerous ways [30, 31]. Urban areas may offer better employment opportunities for women and greater potential income from market lab our. In turn, the increased status and income of these women may be associated with improved diet [30]. Other dimensions of urban residence may also affect patterns of intrahousehold food distribution and women's dietary intake. While household consumption data comparing urban-rural areas exist for a number of developing countries [3], the only systematic study of differences in women's diets is from the Philippines and is based on a very small sample [32].
Household consumption
Urban and rural food consumption patterns in low-income countries tend to differ widely, and the nature of these differences varies with income level [33]. Comparisons have been fairly consistent in showing that the urban poor have lower caloric intake than the rural poor [34, 35]. Protein and fat intakes also vary. For example, in some low-income countries, urban dwellers consume more fat and protein than rural dwellers, particularly from animal sources. In general, the former consume increased amounts of processed foods, meat, fats, sugar, and dairy products, while the latter consume more coarse grains, roots, tubers, and pulses [33]. Increased consumption of animal products in urban areas is associated with higher intake of animal fats, vitamin A, and the more efficiently absorbed haem iron. The implications of urbanization in relation to the consumption of calcium and vitamin C are unclear [36]. While the greater diversity of foods in urban areas may provide a wide range of nutrients, increased consumption of processed foods may contribute to a less nutrient-dense diet associated with a number of chronic diseases.
Ideally, household consumption data should be related to some measure of full-time consumer equivalents and to a standard based on energy expenditures. urban households have a younger age distribution. Also, physical activity patterns differ considerably between urban and rural populations. In future comparisons of household consumption, the number of adult male consumers should be included to develop per capita profiles. Also, some attempt to develop average urban and rural energy-expenditure levels is needed. As with most other nutrition-related issues in low-income countries, it is very difficult to obtain energy-expenditure requirements for the urban resident.
An exception is the collection of weight and height data in 1980 and 1985 Tunisian surveys. Périssé and Kamoun [37] found an important contradiction between urban consumption and body-weight data. While urban residents are consistently shown to have lower energy intake, their weight curves are higher than those of rural residents from five years of age for males and eight years for females through adulthood.
This leads one to believe that the combination of the different age-sex distribution and size and energy expenditure patterns of urban and rural households may explain part of the relatively higher weights in urban areas.
Comparisons of calorie sources by income and type of food in Brazil indicated that poor urban residents obtained the highest proportion of their calories from cereals, rice, wheat, sugar, meat, fish, and fats. The rural poor obtained a large proportion of calories from maize and root crops [33]. With only minor exceptions, these urban-rural consumption patterns were similar for Turkey [38], Bangladesh [39], Pakistan [40], and Tunisia [37]. On the basis of Tunisian income and consumption data (1975,1980, and 1985) Périssé and Kamoun found that, after adjustment for income level, the portion of expenditure on food is nearly the same in urban and rural areas. While cereals, sugar, and animal products are cheaper in rural areas, rural dwellers appear to favour a more plentiful and less diversified diet. Urban dwellers select a more diversified, lower-calorie diet. Thus, with only a few exceptions, studies indicate that, while the urban diet may be more varied and include higher levels of animal protein and fats, rural diets are superior in terms of calories and total protein intake.
Another consideration is how the food available to the household is distributed among its members. Relatively few studies have addressed this issue, but some authors have noted inequities in food allocation in low-income countries [32, 41-44]. The types and quantities of foods available to the household; a person's sex, age, productivity, and economic or social position within the family; and perceived nutritional needs may all influence food distribution among household members [45]. Several studies have shown that women and children, especially female children, tend to receive smaller percentages of their dietary needs than male adults [32, 41].
Even less is known about urban-rural differences in household food distribution. In a comparison of a sample of 100 urban and 97 rural households in the Philippines [32] it was found that while' overall, mothers had less adequate diets than fathers and adolescents less adequate than adults, urban diets were rated higher (in percentages of recommended daily allowances) than rural diets across all age and sex groups except male adolescents. The difference between the diets of mothers and fathers, however, was greater in urban areas (64% versus 80%, respectively) than in rural areas (56% versus 60%). A similar discrepancy between urban and rural areas was noted for female and male preschoolers, indicating that increased household consumption among urban Filipinos benefits primarily adult males and preschool boys.
In many low-income countries where household consumption is less in urban areas, age and sex biases in food distribution may be more strongly felt. There is a need for more information concerning income-related urban-rural differences in household food distribution.
Determinants of urban nutritional needs
Many researchers have felt that urban malnutrition is a manifestation of urban poverty and a vast number of associated factors [2]. We cannot readily change urban poverty, but we must understand its specific dimensions and associated conditions and causes to understand its origins and the factors to consider in designing appropriate, effective interventions. They must be considered carefully as we try either to redesign food and nutrition interventions based on rural needs to make them effective in the urban setting, or to develop new interventions.
In addition, inability to involve urban women actively in health and nutrition programmes is a central, recurring theme. This may be caused by the greater home and market production responsibilities of urban women, particularly those who are heads of households. It also may relate to the greater time costs involved in crucial activities in urban areas such as obtaining food, fuel, and water. Or it may relate to longer hours of market productions or lack of a social network to assist urban mothers.
Urban food markets
The urban household relies almost entirely on purchases to obtain food. Food may be commercially prepared and sold by small stalls, streets vendors, or larger restaurants, or as large or small purchases from stores of various sizes. Four issues stand out as we consider the food-marketing issues facing the urban poor: problems in adapting to urban areas, particularly food markets; higher prices; greater reliance on away-from-home food preparation; and greater vulnerability to changes in economic conditions.
Adjustments migrants must make
In general, migrants appear to be young, educated, and highly motivated. These characteristics often enable them to adapt quickly to the city [46], and migrants often have higher employment rates than permanent urban residents [47, 48].
At the same time, new migrants do face important initial shocks and stresses before they fit into the urban system. For example, they often encounter an entirely different price structure, frequently including a new set of commodity options. It is very difficult to change staple foods, particularly for those dealing with the stress of living in a new, alien environment [49]. The result may be that the migrants pay considerably more for their old foods rather than adapting to new ones, which may taste different and require new preparation procedures.
Another adjustment is learning where to shop and how to do it efficiently. In particular, the more poorly educated or more time-constrained migrants appear to employ inefficient shopping practices. They tend to fragment their purchases among many sources (e.g., vendors, small stores, markets), and they generally shop quite frequently, usually one or more times per day, since they lack adequate storage. They also tend to purchase smaller amounts of food during each shopping trip [50]. In the Philippines, for example, it is not uncommon to see the urban poor purchase one or two tablespoons of cooking oil or a breast-milk substitute such as condensed sweetened milk. These purchasing patterns appear to be very similar among the urban poor throughout the world.
It is generally felt that the urban poor pay more for their food than other urban income groups. Large supermarkets and central food markets are located far away from slums and squatter settlements. Between 50% and 80% of the urban poor require one or more public transport rides to reach central markets [51]. Therefore, purchases must be made in small quantities, which increases the unit price considerably.
Musgrove and Galindo's 1987 study of retail food prices [52] is the first empirical study of price differences between urban and rural areas. They found rural prices to be lower than small-town prices, which are lower than those in middle-sized and large cities. They did not find significant variations in prices based on fractional purchases (small quantities). Their study did not provide adequate evidence on the effect of credit on prices. Moreover, they focused on an analysis of prices stores said they charged rather than an analysis of costs low-income and higher-income groups actually paid. Thus they did not disprove the impact of the conditions noted above [51]. In addition, they did note that traditional foods cost more in urban areas. Overall, their work, which represents a most important addition to this topic, brings into question whether the poor pay more.
Away-from-home food preparation
Lack of fuel, poor cooking facilities, and time constraints often force the poor to purchase a large amount of their food from commercial vendors. In many cases this pattern may represent an efficient allocation of resources in which the vendor prepares large quantities of food at a lower unit cost than each poor household could. In other cases it represents the only option available to those who have minimal or no cooking facilities.
In a continuing prospective study of 3,327 mother infant pairs in Cebu, Philippines [53], the foods women consumed were categorized by place of preparation. While the highest percentages were prepared at home, urban women purchased a noticeably greater amount of foods prepared in stores, restaurants, or bakeries (19.9%) than did rural women (7.2%).
Vulnerability to food price changes
The urban poor are much more vulnerable to macroeconomic and other factors affecting commercial food markets. Unlike their rural counterparts who have access to gardens, poultry, livestock, and numerous other sources of home-produced foodstuffs, the urban poor seldom have home food-production options. Because they spend such a high percentage of their budget on food, the negative impact of food price increases is much greater for them than for higher-income groups.
Urban labour markets
Labour market conditions determine the hours required for market work, its locations, and its physical and psychological stress. In turn, the household's (particularly the mother's) involvement in market economic activities affects the time and other resources available for home production. Crucial dimensions of home production include child care, food preparation and infant feeding, and use of health services. Research shows rather conclusively that the poor are employed in jobs in which productivity is much too low to provide for their basic needs [9]. This often requires them to work many more hours than their rural counterparts. Also urban jobs appear to be less compatible with child care than those in rural areas.
Urban employment and unemployment
"The principal income earners of poor urban households are found in virtually all types of employment: in large medium, and small firms; in wage employment as well as in self-employment or family employment; in the government and in the private sector; and in all urban activities" [9]. Employment figures for 578 large cities [9] show that 57%-73% of the urban poor labour force are employed by others and no more than 30% work in any one sector; manufacturing and commerce represent the largest sources of employment. Economic food markets and related distribution, preparation, and sales activities in the city involve large numbers of workers, particularly in the informal sector [49]. (The informal urban sector consists of jobs with few formal terms of reference for salary, hours, and benefits.) In fact, many feel that 15%-25% of the informal sector is involved in food marketing.
Unemployment often is higher in urban than in rural areas. It is more concentrated in the young (15 to 24 years old) than the old, among women than men, and in the more educated than the less educated [9, 54]. Migrants do not form the bulk of the unemployed or the low-wage earners [9]. The length of time a recent arrival spends unemployed while looking for work may have a serious short-term effect on food purchasing power, but evidence shows that most migrants find employment within a relatively short period of time [9, 55]. Studies suggest that, while migrant women are more likely to be employed than city-reared women, they are often engaged in the least skilled, lowest paid, and least secure work [55]. Migrant women (especially younger ones) are often willing to work longer hours at much lower wages than men. The urban poor have low, often irregular earnings, and entry into economic activities is very controlled, either by the government or by those already engaged in the occupation.
The informal sector
Earnings and mobility in the informal sector are low; work conditions are often very poor; and employer-employee relationships are based on a very dominant employer role. At best, these relationships are similar to the landlord-tenant relationships in the rural sector.
Informal-sector employment is not concentrated in marginal, predominantly service-sector activities as many have felt. It is associated with very low wage levels [54, 56] but often is used to serve the modern sector (e.g., subcontracted work from large factories). Thus it is most important to realize that employment in this sector is quite heterogeneous in its mix of occupations and has basic ties to the urban economy [9]. Urban poor men and women are concentrated in this sector, but women disproportionately so [57, 58]. They are often household heads.
In many low-income countries, some of the urban poor have no permanent niche even in the informal sector. They belong to the world of the casual, day-today labour market. This group may represent the poorest of the urban poor [59]. It is very hard to estimate the numbers of those engaging in sporadic, casual employment. What is clear, however, is the serious consequences of a few days of unemployment for the very poor.
Compatibility of market work with home production
For urban poor women, market responsibilities do not diminish their need to provide for crucial aspects of home production such as child care and food preparation. Their ability to balance these responsibilities is somewhat dependent on their occupation. Little research is available on the nature of urban or rural occupations in terms of their compatibility with home responsibilities or on the reasons women select the occupations they do. Nevertheless, most knowledgeable observers and researchers on this topic feel that urban work is quite incompatible with child care [60]. (There is also extensive literature on possible effects of market work in urban areas on breast-feeding [23, 61].)
Bunster [62] noted that urban poor women working as market sellers in Lima, Peru, selected this occupation over domestic service or factory work partly on the basis of compatibility with child care. The women brought their children to the market and used them as helpers. Unfortunately, this Peruvian study did not assess the occupation choices made by women engaged in other types of work.
Newly migrant women in Mexico City also found urban work less compatible with child care than women employed in comparable work in the Mexican villages from which they came [29].
The most useful study to date on job compatibility is the nationally representative survey of over 1,200 households in Malaysia [60]. Thirty-two per cent of the women in sales and production (small-scale retail trade, weaving, food processing) and 14% of those in service-sector jobs were accompanied to work by children under 10 years old. The same was true for 16% of women engaged in agriculture. The Malaysian urban poor are most likely employed in these occupations, and thus these figures seem to indicate that over two-thirds of Malaysian urban poor women may work in jobs incompatible with child care; unfortunately, these results were not stratified by urban-rural residence.
One crucial dimension of time use in urban market work is travel time. Urban travel is difficult and expensive. Time is great, transfers from one vehicle to another are necessary, and filth, noise pollution, and crowded conditions are the norm. These conditions make it difficult to have young children accompany the women and place more burdens on the poor.
Urban demographics
The urban poor are reasonably heterogeneous; nevertheless, a number of important elements must be considered. The outstanding demographic factors that have important nutritional implications are the extensiveness of female-headed households, the reduced availability of social support, and the large number of unattended or abandoned children.
The urban poor have much larger families than do others in urban areas [63]. Their average household size is quite similar to that of the rural poor. As a result, children are found disproportionately among the urban poor, and the young age distribution provides a basis for rapid population growth of these households.
Female-headed households
Very little is known about the health or nutritional status of females who head households or about their children. There is reason to believe these women and their children represent a high-risk and growing segment of the urban population in low-income countries to whom special attention should be given. Unfortunately, studies of their nutritional problems are unavailable.
Females appear to head proportionately more households in urban than in rural areas. The majority of these households are at the poverty level [64. 65]. Women household heads are most likely to work out of need. They are usually less educated and have fewer income-earning skills than male household heads. They also may have few secondary workers in the home, little time for home production, reduced access to welfare assistance, and increased social disadvantages according to cultural norms [65]. Economic and social characteristics of female-headed households may, therefore, place serious constraints on women's abilities to meet their own nutritional needs and those of other household members.
Overall, the estimated percentage of female headed households in low-income countries is substantial: 22% for sub Saharan Africa, 20% for Central America and the Caribbean, 16% for North Africa and the Middle East, and 15% for South America [64]. There is indication of a rising trend in these percentages.
Women head households because they are widowed, divorced, unmarried mothers, or permanently or temporarily separated from their spouses. In Latin America the high rate of females migrating from the country to the city has contributed to increased numbers of them heading households. In 11 Caribbean countries women headed 19%46% of the households (median 37%). Trend data for Brazil from 1950 to 1970 indicate a drastic increase in the percentage of female-headed households through separation and divorce and a large reduction in the proportion of those caused by widowhood [65].
Except for the Caribbean, the major cause of female-headed households is reported to be widowhood, with percentages ranging from a high of approximately 90% in Asia to a low of about 50% in Latin America. In many societies (e.g., the Middle East) greater institutionalized support is extended to widows than to women in consensual unions and single or deserted women. Widows with extended families may be likely to remain in rural areas where their social and economic support system is intact, while single, divorced, and separated women may prefer to migrate to urban areas in search of improved economic opportunities and reduced social ostracism. Trend data on marital and socioeconomic status of urban and rural women household heads are necessary for further understanding of the background and specific needs of the growing numbers of urban women heading households in low-income countries.
For several reasons these families represent a high-risk group. One is the high likelihood of poverty. A second is absence of proper infant and child care. Many researchers have shown a strong association between income-producing work of women, particularly in work that is not compatible with child care, and reduced nutritional status of infants and pre-schoolers [58, 62, 66, 67]. These studies have not sorted out causal relationships adequately, and it may well be that work decisions and child-feeding decisions are jointly determined by a set of underlying factors. Nonetheless, this research does establish that preschoolers who receive inadequate care and whose mothers work away from home are likely to be malnourished. Counteracting this view is the possibility that female heads of households may spend a great proportion of their resources on nutritional and health-related goods and services. [30, 68].
Social support
In general, most observers feel that the urban households lack the social support network available to rural households, particularly to the homeless and to female-headed households. Urban slums and squatter settlements do provide social networks, however- often to a much greater extent than outside observers would be led to believe.
Lack of social support has been cited as a major determinant of the decline in breast-feeding that has occurred in urban areas [68]. Inability ever to breast-feed and the so-called insufficient-milk syndrome have been associated with this lack, although in a most tenuous manner [23]. Studies of the insufficient milk syndrome have not tested the specific effect of lack of social support. Rather they have empirically tested the effects of broader social causes of this lack, such as urban residence, which obviously reflects a large number of factors. For example, inhibition of the milk let-down reflex is affected by maternal stress and anxiety levels. Social support plays a role in this relationship, as do many other factors.
Social support in rural areas plays a major role in numerous aspects of child care and household diet, such as food sharing, assistance in child care, infant feeding, and health care use. Absence of adequate social support would be expected to play a key role in reduced quality or absence of child care and some of the other difficulties the urban household faces.
Plight of children
Unattended or abandoned children represent a most important public health problem with important nutritional implications [7]. Estimates of the number of unattended and abandoned urban children vary considerably. According to UNICEF, 70 million urban children may be without families-40 million in Latin America, 10 million in Africa and the Middle East, and 20 million in Asia [69] Surveys on this issue are rare. It is clear, however, that the number of unattended or very poorly attended children is high in urban areas.
Urban environment
"Generally speaking, the environment in urban slums is much worse than in rural areas-people are far more crowded, housing is likely to be makeshift and crowded in the extreme, and waste disposal is often nonexistent" [70]. As with other aspects of life for the poor, extensive heterogeneity exists among and within each settlement. Incomes vary tremendously, as do quality of housing structure and the surrounding environment.
Housing conditions
Inadequate housing and insecurity of tenure are the rule in most squatter settlements and many slum areas. A large but unspecified percentage of the urban poor reside in highly inadequate houses. Because of the lack of land tenure and/or the unsettled legal nature of housing of the urban poor, incentives to improve its quality are lacking [71, 72].
The instability of land and housing ownership and the poor quality of housing do not mean that the urban poor, even those residing in illegal squatter settlements, do not pay rent. On the contrary, they often pay rent representing a considerable portion of their income and unrelated to the quality of the housing or its environment [73]. Rent would appear to be more a function of supply and demand, linked often to proximity to employment opportunities, water sources, or transportation and other important urban services, than of quality of life or health [9, 73].
Housing density
The density of living conditions for the urban poor appears to be much greater than for the rural poor [9, 15, 72]. It is assumed that this increased density is responsible for a higher frequency of infectious diseases such as measles in Cité Simone, Haiti [74]. In earlier research, however. an eminent epidemiologist [75] reviewed literature and was not able to show that increased density had the profound effects on the spread of infectious disease that many associate with residence in urban slum areas. Rather, it may be the combination of poor water, poor sanitation, overcrowding, and malnutrition that causes these problems. Recently, two well-conducted studies in two low-income cities in the Middle East found improved dwelling quality to be associated with reduced infant mortality [11]. Future research must separate the effect of available infrastructure and services from effects of the shelter structure per se.
Water and sanitation
The relationship between diarrhoeal disease and water and sanitation has been studied extensively. On the basis of a review of the literature, one can say with certitude that disease is reduced as water supply and sanitation services improve. It is not clear, however, whether, with respect to health, these services should be provided at high levels and therefore at high cost, or whether reduced standards in such areas would free money to be spent on drainage and housing as well [11].
Health services
Availability of facilities and personnel
Health facilities are concentrated in urban areas; thus urban dwellers usually are better provided for medically than those in rural areas.
But this does not apply to the urban poor. The distribution and cultural and financial accessibility of facilities are such that most of the marginal groups are patently underserved. Hospitals have traditionally been the main vehicle for delivery of health care to urban population. However, hospitals are generally removed (physically and socially) from the urban poor, emphasize sophisticated technology and are often overcrowded, although under-utilization is not uncommon. There is a relative scarcity of more accessible ambulatory and social services, referral is difficult and the emphasis in general is on curative medicine. Yet most of the health problems of the urban poor would be more appropriately dealt with by preventive and well integrated social action, both of which are largely deficient. [1]
Statistics to document these relationships are distinctly lacking; however, few who have written on this topic would disagree with the quotation [41. On the basis of our research, we can say that the health care provided the urban poor may be better than that provided the rural poor [76], but it is not adequate.
In several related studies, we examined the patterns in facilities and personnel available in the Philippines [53, 7678]. On a national level, medical facilities and personnel, in particular physicians, are concentrated not just in urban areas but particularly in Manila. In 1983, 61% of the public hospital beds and 67% of private hospital beds (but only 31% of the public hospitals and 46% of the private hospitals) were located in Manila. Similarly, in the late 1970s 31% of the physicians were in Manila and 42% were in adjacent areas. At the same time, the trend in the Philippines from 1973 to 1983 pointed toward dispersal of public and private facilities away from the Manila region.
In the second largest Filipino urban region, the Cebu City metropolitan area, detailed data on accessibility and quality of health facilities available to urban and rural populations show the following [11]:
- Urban households have much more accessible public and private, modern and traditional prenatal and delivery services (rural travel times are three to eight times greater than urban times).
- Urban household members are much more able to walk to private modern prenatal services (93% versus 32%) and have almost the same ability to walk to traditional midwives (98% versus 80%).
- Rural household members are significantly more able to walk to public facilities (the rural health unit or city health office and primary health care facilities).
- Rural public facilities are much more likely to have trained midwife public practitioners who provide deliveries at home (59% versus 41%) and away from home (95% versus 32%). Similarly, rural prenatal care in public facilities is provided much more by midwives (54% versus 7%).
- Urban facilities, in particular public ones, are open many more hours per week.
- Drug supplies are quite similar in urban and rural public and private, traditional and modern facilities.
We have found that even in the case of a country with accessible urban health care, increased availability of primary health care services would significantly improve prenatal care patterns. If travel time were used as a criterion, urban households would be better off. However, travel time has not been found to affect the behaviour of urban Filipino households; rather. quality and economic or other factors are more important. In fact, there appears to be a dual urban health market in terms of demand for prenatal care services. One group desires more accessible primary health care workers and a second group wants physicians [53].
Accessibility to appropriate health care appears to be a more important issue in other low-income countries. Few African or Asian countries have the extensive public and private medical care systems with the widespread coverage of high-quality personnel and facilities one finds in the Philippines [76, 77].
Preventive health care use
Primary health care is essentially a rural phenomenon, although there are scattered examples of urban primary health care projects [79]. The important issue with respect to primary health care is how it does or could prevent important causes of malnutrition. Health care can improve nutrition in several ways [80-83]: by preventing gastrointestional and infectious diseases affecting nutrient absorption, use, and intake; by promoting adequate diet and weight gain of women during pregnancy and lactation; and by promoting appropriate infant and child feeding through use of information, education, and communication, monitoring growth, and other programmes.
Few studies have assessed urban-rural differences in vaccination coverage, oral rehydration therapy use, breast-feeding or supplemental feeding education campaigns, and growth-monitoring programmes. The urban poor do not appear to be adequately served in any of these areas. Immunization coverage in urban areas is greater than in rural areas in the few studies that have examined these patterns for comparable samples [84]. Nevertheless, it is very low, particularly in Africa and Asia among the urban poor. One would suspect that the same holds true for other nutrition-related health-care initiatives.
Promoting breast-feeding is one area in which there appears to be a clear basis for improving knowledge, attitudes, and practices within the health sector. Elsewhere we review surveys on the poor knowledge, attitudes, and practices of public and private, traditional and modern health professionals [85]. A number of studies in urban areas seem indirectly, although not conclusively, to show that training and other changes in the practices of health facilities can have an important impact on the extent and duration of breast feeding [86-88].
Mass media: use of commercial products
Clearly, coverage by all forms of mass media is greatest in urban areas, and most commercial food advertising is concentrated there. Many of these media appear oriented to middle- and higher-income populations; however, some consider that commercial advertising adversely affects nutrition of the urban poor. Most emphasis has been placed on the effects of the mass media on infant feeding, particularly on the use of breast-milk substitutes.
Surprisingly few studies have documented mass media penetration among the urban or rural poor. It has been suggested that factors associated with formula brand-name recognition are associated with earlier supplementation and weaning [89]. Other studies that have tried to control for confounders have not found a clear effect [90]. Moreover, the 1981 WHO-UNICEF Code of Marketing of Breast Milk Substitutes bans all use of mass media for promoting these products, and most food companies appear to have complied with this.
At the same time, few persons have studied the effects of commercial advertising of evaporated or condensed sweetened milk, both of which are used as breast-milk substitutes and whose promotion is not affected by the WHO-UNICEF code. Similarly, there is little understanding of the role of advertising commercial supplemental foods such as Cerelac on breast-feeding and infant health.
Another way mass media could adversely affect nutrition is by increasing the consumption of food of low nutrient density such as soft drinks. The patterns and consequences of consumption of such products by the urban poor remain to be considered.
It is difficult to separate the mass-media effect from many social and behavioural factors associated with urban residence. For instance, the low-income urban resident, particularly the new migrant, usually faces disparities in income, wealth, life-style, and diet often not seen in rural areas. The desire to emulate practices of the rich, or at least to adopt various consumption patterns perceived to be followed by the rich such as using infant formula rather than breast milk, may cause some nutritional problems. Once again, documentation of factors influencing consumption patterns and the magnitude of their impact is unavailable.
Economic adjustments
Balance of payments and other economic problems in the 1970s and early 1980s forced an increasing number of countries to develop economic adjustment policies that included reduced fiscal deficits and monetary expansion, reforms in government expenditures and revenues, mote realistic exchange rates, and realignments in prices (particularly increases in food prices). While the goal of such policies is to increase economic growth, important short-term social costs can occur, particularly for the poor who do not benefit from these programmes. The rural homeless and urban poor are the two major groups such programmes could hurt [91].
The poor in urban areas rely on cash for survival much more than those in rural areas [69]. Changes in the availability of employment, shifts in food prices or the prices of other basic commodities such as petroleum products, and cuts in basic services the government provides affect the urban poor much more than the rural poor [9]. This finding does not necessarily mean that the nutritional status of all age and sex groups will be hurt by inflation and increased poverty. The severe inflation in Brazil after 1978 did not seem to be associated with increased malnutrition among preschoolers despite increased impoverishment of the children studied in one favela (shanty town) [92]. Increased poverty was associated with an increased extent and duration of breast-feeding between 1980 and 1983.
There are numerous examples of the association between economic recession and increased poverty' malnutrition, and poor health in the past decade [91]. No research has linked this association to economic adjustment policies or the effects of selected components of such policies. Given the increasing promotion of economic adjustment policies by the World Bank, the International Monetary Fund! and other agencies, it is essential to understand how price inflation and increased unemployment associated on a short-term basis with devaluation and other adjustments affect diet and the nutritional status of the urban poor.
Programmes and policies for nutritional needs of the urban poor
On the basis of the information reviewed above, it is apparent that the nutritional needs of the urban poor are associated with a complex weave of demographic, environmental, market, employment, social, and service factors. An understanding of these is essential for the design of programmes and policies to meet the needs of the urban poor. Numerous factors for consideration include the development of community based programmes in urban settings where there may be little community cohesion and where time constraints on the poor especially women heading households, limit participation. While urban physical infrastructure may facilitate market interventions and the integration of nutrition- and health-related services, the lack or disorganization of administrative infrastructure may create serious barriers to effective programme design and implementation.
Community participation
Community involvement in programme design and implementation has been considered most important to the continued success of urban basic-services programmes [4, 12, 93]. Urban poor communities lack the social cohesion and administrative infrastructure of rural communities and thus have a greater need to be involved in programme design. Rossi-Espagnet [1] noted that community involvement and participation are important not only to develop effective health programmes but to provide people with "social connective tissue lost in the migration process." While urban poor communities lack structure compared with most rural ones, the potential for action once they are organized appears to be good [69].
UNICEF has placed great emphasis on implementing community-based approaches. A review of experiences from 70 countries, including case studies prepared for 9, found such approaches were workable [69]. They allow services and facilities to reach the poorest households, are designed in a manner sensitive to the needs of each community, are more likely to be maintained over time, and are very cost-effective.
When community-based nutrition programmes de signed for rural areas have been attempted in urban areas, the results have been mixed. For example, the Indonesian nutrition effort is unique in the scope coverage, and success of growth-monitoring and related activities [4, 94], but it has had little success in mobilizing community support and reaching its target levels in urban areas [95]. We feel that more is needed than simply adapting existing organizations in urban poor neighbourhoods to provide the functions of traditional community organizations found in rural areas or' in some cases, helping to develop new organizations.
Potential for citywide macro-interventions
To date, much of the emphasis by UNICEF, WHO and others seeking to improve urban nutritional and health status has been on more focused community based interventions. Given the difficulty of expanding coverage to reach all the urban poor with this approach, targeted macro-initiatives should also be considered.
The potential for developing many wide-ranging broader-scale interventions in urban areas appears to be excellent, particularly in very large cities. The large population base and physical infrastructure allow citywide solutions to be developed efficiently Examples include food price subsidies, fair price shops, food fortification, and social marketing activities. The last may use the informal food sector (small stores, stalls, food vendors, etc.) as well as the more formal food sector.
Weak administrative infrastructure
Many persons actively involved with the urban sector in providing basic services have found urban planners and administrators unable to cope with the problems they face [96]. Urban government has been unable to implement effectively many of the large development efforts the World Bank and others have planned with them [97]. These deficiencies in the human-resource skills and the administrative structures of cities must be considered in selecting and designing nutrition programmes and policies.
Few metropolitan areas have coherent administrative structures. Most often individual cities have separate bureaus, including health departments. In a few cases, city health offices are not even under a national ministry of health. Planning and implementing health and nutrition activities are undertaken separately The results are difficulty in initiating new health and nutrition programmes, in particular, those requiring cross-departmental linkages, and low priority for reaching poor neighbourhoods when resources are limited.
The difficulty is compounded by the low priority city administrators give to nutrition issues [95, 98]. We also feel that a major problem is the resistance to change of city health-care professionals and hospitals. Very strong advocacy and incentives will be needed to redirect urban resources in the necessary manner [98].
Food market options
The income and time constraints the urban poor face, in conjunction with the availability of commercial channels for programme implementation, provide an attractive rationale for several food-market interventions.
Food fortification
Research shows it is easy to find food carriers, particularly for urban populations. The food must be centrally processed, be consumed regularly, and be one that can be fortified economically without major alteration of its organoleptic properties. Fortification is most feasible in urban areas for numerous foodstuffs, and can be done for vitamin A, iodine, and numerous other vitamins and minerals [99].
Fair price (ration) shops
There are stores that distribute staples at subsidized prices in urban areas in numerous countries. They may be operated by the private or the public sector or as community co-operatives. All can alter market mechanisms to provide lower-cost staples in urban low-income areas.
Price subsidies
Numerous scholars and international agencies have developed a very strong case for having most governments remove the majority of food subsidies [100, 101] (see the article by Reutlinger on page 24). Most low-income countries use food imports and various price and other budget subsidies to provide protection for all urban consumers [100] These subsidies could be targeted to protect both the urban and the rural poor [102]. This seems best achieved by selecting the type of commodity or food ration shop or by providing food stamps (see Reutlinger). Food subsidies could benefit both the urban poor (e.g., sorghum in Bangladesh, cassava in Indonesia) and higher-income groups (e.g., rice in Indonesia or Bangladesh) [103, 104]. Extensive research exists in this area, and clearly it is possible to design appropriate interventions. In north-eastern Brazil, subsidies significantly reduced prices charged by urban stores [52].
Feeding programmes
Traditionally, on-site or take-home feeding programmes have focused on ways to benefit pre-school children and pregnant or lactating mothers. They have been designed to meet income constraints and intra-household food-allocation problems [102, 105]. It is also possible to design these programmes to address constraints of time, fuel, and cooking facilities of urban households as well as to provide households with the benefit of large-quantity food purchases. The comedora populares programmes of Peru and Bolivia, in which urban households combine resources to provide higher-quality and lower-cost meals to entire households in the morning and the evening, are an example [106]. Community kitchens to feed preschool children have been organized by slum or squatter neighbourhoods in a number of other countries [51, 107].
Street vendors
Small stalls and vendors are an important source of food for the urban poor. Research on hygienic practices of this sector is sparse, and we are unaware of programmes that have worked to improve sanitation, reduce costs, or improve nutritional quality of the food they prepare. Given the size of this sector and its importance as a source of nutrients, improved food safety may be important.
Health care services
The primary health care (PHC) model as first prepared at Alma Ata and as implemented in numerous countries appears most suitable for the urban setting. It helps to ensure that multisectoral issues are addressed. It also can improve the flow of information from target families to the health system, and it should be most cost-effective. Density of population and transportation infrastructure mean that a PHC worker should be able to reach the target population much more efficiently in an urban than a rural setting. The poor urban setting seems to demand a PHC programme that provides some focus on environmental sanitation and water, expanded programmes of immunization and oral-rehydration therapy, and growth-monitoring (with emphasis in the last on improvements in infant-feeding practices) [7, 79, 108].
Child care
Child-care programmes, like PHC programmes, can provide a wide range of benefits to participants and their households. Improvements in child growth, social and intellectual development, and the training and employment of low-income women are possible benefits. UNICEF reports active involvement in providing child-care services in 21 field offices. These community services can focus on selected high-risk groups-as do, for instance, India's mobile creches organized at construction sites, or child-care centres serving children of female-headed households in Venezuela [51, 109].
Social marketing of nutritional education
The general feeling among researchers seems to he that mass media can be used quite effectively to deliver focused messages to low-income households. This approach seems most effective when it reinforces or works in close collaboration with direct delivery systems [110-112]. The promotion of breast-feeding, appropriate supplemental feeding practices, and growth-monitoring have all been enhanced effectively with use of social-marketing techniques. Most documented nutrition efforts of this kind have addressed rural populations; however, they also appear very useful for the urban population.
Summary
Accelerated urban population growth in low-income countries poses a tremendous challenge for those concerned with nutrition. The large increases in the numbers of urban poor residing in densely populated slum and squatter settlements occur at a time in which few resources have been allocated to their nutritional needs. In general, nutrition research, programme development, and programme evaluation have focused on rural areas. Because of the sparsity of relevant nutrition literature, it is necessary to draw on related economic, health, and demographic literature to consider the magnitude of urbanization and its implications for nutrition programme and policy design.
Urban areas in all developing countries are growing at a tremendous pace; by the end of this century they will contain over 40% of the population. The growth rate of urban areas exceeds that of rural areas in all regions, and this staggering increase, in absolute and relative terms, places an enormous burden on urban social services.
Increases in the number of poor people living in cities are particularly important to note. It is estimated both that the absolute number of the poor will grow rapidly and that, by the year 2000, 57% of the poor people in low-income countries will be living in cities. This will represent a major shift in the locus of poverty from the first few decades of this century. Urban poor people often live under very unhealthy environmental conditions; insufficient housing and lack of access to adequate water supplies and sanitation appear to be the rule.
In addition to the economic and health indications of nutritional problems in low-income urban areas, breast-feeding, supplemental infant feeding, and household consumption data provide direct information about the extent of nutritional risk in urban areas. While we cannot determine the number or percentage of households at risk because of deficient nutrient intake, or of infants at risk because of poor feeding practices, these data provide a clear picture of the significant nutrition problem existing in urban areas.
Larger proportions of infants are never breast-fed and the duration of breast-feeding is shorter in urban than in rural areas of most low-income countries. This decline of breast-feeding may be one of the most important nutrition-related consequences of urbanization. Urban-rural differences are greatest in Latin America and the Caribbean, but they also occur in selected countries in Asia and Africa, particularly those with large cities. Infants are also given supplementary foods early in many urban areas, particularly in Latin America.
There are too few data available to draw any conclusions on the intake of urban and rural women.
Comparisons of household consumption data tend to reinforce the impression of important urban nutritional needs. Urban poor households in a number of countries have lower caloric intake than do their rural counterparts. At the same time urban households consume greater proportions of energy from animal protein, fats, and processed foods and lower proportions from cereal and tubers. As a result, while animal protein and fat intakes are more likely to be deficient among the rural poor, the more Western diet of the urban poor is likely to be deficient in energy and total protein, and to be associated with a number of chronic diseases, including diabetes, hypertension, atherosclerosis, and various types of cancer.
While a broad range of factors associated with urban poverty appear to be important causes of nutritional deprivation of the urban poor, several major ones stand out. One overriding issue is the time constraints under which urban women live and the resultant difficulty of involving them in health and nutrition programmes. Urban women, particularly in Latin America and Asia, appear likely to head households; they may lack the social support network of rural women; they require more time for market employment; and crucial home activities appear to be more time-intensive.
The urban poor face different food-market conditions than do the rural poor. They are more vulnerable to economic adjustments and inflation, are much more reliant on away-from-home food purchases, and Two potentially important social dimensions facing the urban poor-the proportion of female-headed households, and the location and compatibility of urban work-have not been adequately evaluated. We have no large studies on differences in urban and rural areas with respect to these factors.
Urban social services, particularly health, housing, water, and sanitation, do not appear to assist the urban poor greatly. Squatter settlements frequently must be located in the worst areas, which results in very unhygienic conditions associated with a vast range of communicable and infectious diseases.
These social, economic, and environmental factors provide the rationale for considering programme and policy options for addressing urban nutrition problems. Issues that stand out as important considerations include environmental conditions, time constraints, inadequate food-market structure, high food prices, and lack of information (and related adjustment problems for new migrants). At the same time, agencies and groups involved in designing, implementing, and evaluating urban programmes have posed a number of considerations for developing social services. One is the need to assist many urban communities in developing organizations for facilitating programme design and implementation and for using community participation to develop lower-cost, more effective interventions. Another is the potential for developing areawide food interventions.
A number of programmes and policies appear feasible within these two sets of concerns. They include interventions in the food sector by fortification, food subsidy, and related efforts, some unique urban feeding options, refocusing the urban health system toward more preventive primary health care activities, and providing child care and various educational activities.
While the past has seen us ignore the needs of the urban sector, the increasing magnitude of the urban problem will not allow this oversight in the future. We hope that this review and the other articles in this issue of the Food and Nutrition Bulletin will provide some guidelines for making rapid progress on the nutritional needs of the urban poor in the future.
References