This is the old United Nations University website. Visit the new site at

Contents - Previous - Next

Water-health relationships in el-obeid: an example from an urban semi-arid area in western Sudan

Sharaf el-Din Ibrahim Bannaga
University of Khartoum

John Pickford
Loughborough University of Technology, UK

A careful investigation of the factors associated with the water supply of a town in the Sudan could lead to a reconsideration of water requirements in other developing countries. In the past, governments, international agencies, and consulting engineers have been equally uncertain about the choice of suitable consumption figures for urban and rural water supply. The el-Obeid study indicates that, from the health point of view, a figure of 60 litres per person per day may be ample.

The provision of adequate water supplies in the Third World has been a recurring theme of recent United Nations gatherings. United Nations conferences on the environment at Stockholm in 1972, on habitat at Vancouver in 1976, and on water at Mar del Plata in 1977 raised hopes that water would soon be available for all, but there has been little discussion about how much water should be provided.

For more than a century those wishing to promote public works for the improvement of water supplies have given the expected gains in public health as the chief justification for expenditure.

Sir John Snow's investigation in the 1850s led to the incrimination of the Broad Street pump as the cause for high cholera incidence (Longmate, 1970),and it has been followed by a number of other cases where polluted water has been proved to be the cause of infection. In 1955-56 several tens of thousands of people were afflicted by infectious hepatitis in the Delhi area after drinking inadequately treated water from the River Jamuna (WHO, 1973). The pollution of surface water used for drinking caused many deaths in the villages of the Mulanje in Malawi in 1973-74 (Pineo and Subrahmanyam, 1975).

One frequently quoted case of health benefits of safe water is the gradual eradication of typhoid fever in the United States after municipal water supplies were extended (Wolman and Boch, 1963; Fair et al., 1966; Steel, 1960). In India, the death-rate at Burdwan dropped from 42 per thousand to 20 per thousand after piped water was provided in the 1880s (Wallace, 1893). More recently, five years after the construction of water supply schemes in Uttar Pradesh, the average death-rate from cholera had fallen by 74 per cent, from typhoid by 63 per cent, and from diarrhoea by 43 per cent ( Lee, 1969). In Kenya the general level of health improved dramatically only four years after a piped water supply system became operational (Carruthers, 1973).

Nineteenth-century workers stressed the importance of water quality, and it became accepted practice for civil engineers to design for absolutely safe potable water. British water engineers have been especially assiduous in maintaining the highest standards of quality. For example analysis of thousands of samples of water supplied to London have shown zero E. cold counts (Southgate,1969).

In the past decade or so, a growing number of people have questioned the wisdom of this 100 per cent safe policy when applied to rural water supplies in developing countries. It is argued that, when resources are limited, it is better to give access to a reasonable quantity of reasonably safe water for a large number of people rather than first-class water for only a few people. Bradley's disease classification (Bradley and Emurnon, 1968; White et al., 1972) set the scene for this by pointing out that many water-related diseases are associated with scarcity of water rather than with its quality.

In fact, most of the diseases which may be classified as truly water-carried can be transmitted by other means. Personal hygiene may be crucial in preventing the spread of cholera and diarrhoea, and personal hygiene requires water. Therefore, sufficient water for washing is the major health requirement, and quantity may be more important than quality.

Ignorance, poverty, and diseases are the basic factors hampering development in the developing world. The elimination of poverty and diseases can only proceed if an adequate supply of water sustains agriculture and industry and maintains a healthy environment. The majority of urban centres in the developing world have water supplies which are overloaded, working at inadequate pressure, intermittent, or delivered through contaminated distribution systems (Pineo and Subrahmanyam, 1975; Dietrich and Henderson, 1963; Damme, 1973). Furthermore, the inadequacy of municipal water services in many cities has become more acute under the weight of population growth and urban migration.

With these ideas in mind, the authors during a water study at el-Obeid endeavoured to find out just what connection exists between the amount of water consumed and the health of the people who use it. The study also considered a variety of other matters concerned with the present and future water supplies in the area.

El-Obeid is a town of some 140,000 people in western Sudan, 320 km southwest of Khartoum across the semi-arid zone, which extends from el-Obeid in all directions (Fig. 1). Annual rainfall is low, and in some years only 100 mm or so falls during a short season of high-intensity precipitation. With maximum daytime temperatures of up to 40 C, losses by evaporation are high.

Housing in the municipality is divided into four classes based on plot size. Buildings in the first and second classes are of good quality with water provided by house connections from the public supply. The medium-density third class is subdivided into three groups: class 3A has piped water to the compounds, class 3B has mixed supplies, and class 3C relies on public standpipes. The fourth class areas are congested areas with the majority of the houses of sub-standard or temporary structure, mostly with mud walls. Their inhabitants rely on public standpipes, unprotected local sources, or water vendors. In the municipality as a whole only 41 per cent of the houses have piped water connections.

In order to relate health to water consumption a total of 235 households were selected as being representative of the different types of households in the town. Pre-coded schedules of measurable questions were prepared for obtaining data by household interviews, and return visits were made over a period of six months to ascertain the incidence of illness and changes in the household. Data were obtained about the present water supply, including the actual quantity used and its cost. The age, educational standard, occupation, and other personal details of members of the household were noted during the first visit, together with information indicating the standard of living. An important group of questions was drawn up to check the health and disease pattern. Of particular interest was the number of children of the householder who had died before reaching adulthood. During return visits, information was obtained about the type and duration of illness for all members of the household.

To supplement the household survey, information about diseases was obtained from hospital and clinic records. Table 1 shows the proportion of patients at el-Obeid General Hospital in 1974 whose illness might be related to water or to a shortage of water.

For the whole of the Sudan during 1974/75 about 560,000 patients were treated for inflammatory diseases of the eye, over one million for skin diseases. and some 600.000 for diarrhoea. Moreover, it is likely that these patients included in the health statistics were only a small proportion of those suffering from these diseases. Diarrhoea, eye infections, and skin diseases are particularly associated with poor personal hygiene due to lack of water.

Examination of children by the School Health Division of the elObeid General Hospital demonstrates the connection between eye infections and water supply. Table 2 summarizes information from annual reports for 1971/72, 1973/74, and 1974/75.

FIG. 1. Location of el-Obeid

TABLE 1. Percentages of Out-Patients and Deaths at elObeid General Hospital Associated with Water-Related Diseases

Disease Out-patient
Deaths of in
Diarrhoea and dysentery 13.6 31
Malaria and "fever" 26.1 7
Kidney diseases, etc. 6.4 2
Eye infections 12.7 -
Skin infections 4.3 -
Bilharzia 0.2 -
Totals 63.3 40

TABLE 2. Incidence of Eye Diseases among el-Obeid Schoolchildren


Catchment area type

Average water use in catchment area (litres per day)

Percentages of children examined

Personal cleanliness at inspection

Eye disease at inspection

Good Average Poor Trachoma Inflammatory diseases Others
Southern 1st and 2nd class 100-250 100 0 0 4 10 0
el-Guba, Amir, Quarters, and Petrol 3rd class A 55-70 66 28 6 19 12 0
el-Mayrum, el-Radif, and Falustin 3rd class B   45 39 16 22 25 5
el-Nazir and el-Radif Extension 3rd class C 20 33 42 25 31 27 8
Fellata andTayba 4th crass 16 4 71 25 70 30 9

Eye diseases are also very prevalent in the rural areas around elObeid which are included in the el-Bideira Rural Council. The incidence of various water-related diseases, as estimated by nurses in charge of some village clinics, is shown in Table 3. Water is a very scarce commodity in these villages except during the short rainy season.

The survey of 235 el-Obeid households was carried out between April and September 1975. Table 4 summarizes the incidence of some water-related diseases, as reported by the householders during the survey, together with the average daily water-use. The number of child deaths was obtained by asking the householder how many of his own and other dependent children had died before reaching the age of 20.

The data obtained from the household survey were subjected to Pearson correlation, partial correlation, chi-square, and contingency tests. The study data met all the requirements of the partial and the simple Pearson correlations. Although the partial correlation is more accurate, the simple correlation could be more appropriate since it gives the same indication as the partial correlation. Table 5 shows that there was a significant inverse relationship between water-use and certain water-related diseases.

However, the main concern is to show which of the waterrelated diseases is associated with water-use more than the others. It is by no means a direct relationship, since the effects of other factors which might contribute to the prevalence of these diseases are not eliminated. It is interesting to note that diseases which are thought to be due to lack of personal hygiene (eye and skin) are more associated with water-use than others.

TABLE 3. Water-Related Diseases in Some Villages of the el-Bideira Rural Council: Number of Cases per Person

Gaibat Kazgeil Taggat Umm
Eye infections 0.53 0.33 0.42 0.49 0.43
Skin infections 0.03 0.03 0.03 0.18 0.12
Diarrhoea and
0.07 0.20 0.17 0.25
Kidney diseases 0.07 0.04 0.08 0.32 0.13
Malaria 0.30 0.13 0.13 0.23 0.17

A connnection between availability of water and the incidence of diarrhoea or dysentery in the United States was reported by Hollister et al. (1955) and Stewarts et al. (1955). A detailed study of diarrhoeal diseases in seven developing countries, including the Sudan, was reported by van Zijl (1966) who pointed out that accessible water and sanitation result in less diarrhoea. White et al. (1972) found in East Africa that 19 per cent of the households without piped water had had a case of diarrhoea in the week preceding the survey, whereas only 3.1 per cent of households with water connections had suffered from diarrhoea in the same period.

Although the incidence of disease increases with scarcity of water, there are, of course, many other factors which make illness more likely in poor-quality housing areas. In the worst districts, sanitation is bad, housing is over-crowded, children are less educated, the people are poorer, and, there fore, malnutrition is more likely. In the fourth class areas, few households have their own latrines, and the children excrete in the open spaces between houses.

TABLE 4. Illness Reported during el-Obeid Household Survey

Class and
(litres per day)
Cases of disease per person Days
Child deaths
per house
and dysentery
and fever
1st class 242 none none 0.0 0.2 2.0 0.01
2nd class 95 0.3 0.0 0.2 0.5 3.8 0.03
3rd class A  
Petrol 58 0.1 0.0 0.4 0.8 5.6 0.07
el-Guba 68 0.1 0.0 0.3 0.6 6.0 0.12
Quarters 66 0.1 0.0 0.5 0.6 8.2 0.16
3rd class B  
Falustin 42 0.5 0.3 0.8 1.2 13.2 0.24
el-Radif 41 0.3 0.1 0.7 0.9 10.8 0.17
el-Mayrum 48 0.2 0.1 0.7 0.6 10.1 0.16
Fellata 17 0.8 0.5 1.2 0.8 19.5 0.43
3rd class C  
el-Nazir 18 0.5 0.3 0.9 0.7 14.4 0.24
Radif Ext. 22 0.4 0.2 0.7 0.4 12.1 0.21
el-Ashara 26 0.4 0.1 0.7 0.5 11.2 0.17
Mayrum Ext. 18 0.3 0.3 1.0 0.8 13.1 0.26
4th class  
Fellata 16 0.6 0.5 1.1 1.0 19.4 0.50
India 14 0.5 0.3 1.1 0.7 19.7 0.38
Kareima 19 0.4 0.2 0.8 0.7 15.5 0.39

TABLE 5. Correlation Coefficients of Disease Factors with Water Use per Person-el-Obeid
Household Survey

Pearson *
Coeff. Sig. Chi
Days illness per person -0. 77 -0.58 0.001 598 392 0.85
Child deaths per -0. 73 -0.51 0.001 413 280 0.80
Eye disease -0.42 -0.32 0.001 261 196 0.73
Skin diseases -0.42 -0.29 0.001 196 168 0.67
Diarrhoea -0.36 -0.24 0.003 296 252 0.75
Dysentery -0.25 -0.21 0.005 146 112 0.62
Fever other than malaria -0.24 -0.18 0.011 110 112 0.57
Malaria -0.21 -0.12 0.016 310 252 0.75

*Level of significance is 0.001 for all factors except malaria, for which it is 0.003. n = 235

TABLE 6. Correlation Coefficients of Various Factors with Number of Days Illness per Person-el-Obeid Household Survey

Partial correlation Chi-square test Contingency
Coefficients Significance Chi-
Degre of
Water-use per person -0. 77 -0.47 0.001 598 392 0.85
Standard of -0.72 -0 39 0.001 428 240 0.80
person -0.69 -0.27 0.001 393 264 0.79
Waste removal -0. 63 -0.22 0.001 266 84 0.73

*Level of significance is 0.001 for all factors.

Some of these other factors influencing the health/disease pattern were studied in the household survey, and their relationship with disease was correlated. Table 5 shows that the most significant measure of disease was the number of days of illness per person, so the statistic was related to other factors, as shown in Table 6.

In this analysis, "standard of living" was given a numerical value to indicate the conditions within the home. The number of days of illness per person during the survey period (April-September 1975) was plotted against average water-use per person, as shown in Fig. 2. A curve was obtained by the least-squares-fit method. This is shown on the graph as a hyperbola of equation:

where d represents the number of days of illness per person during the study, and w is the daily water-use per person.

FIG. 2. Relationship between Days of Illness and Water Use

The chi-square test was used to check the accuracy of this curve, giving the following results: chi-square 468 degrees of freedom 165 level of significance 0.001 contingency coefficient 0.82

However, the graph shows the gross relationship between water-use and the incidence of disease and, therefore, makes no allowance for the effects of the other factors, because it is difficult to isolate these effects when constructing a simple graph. The inclusion of the graph in this paper is to give the reader a visual impression of the effects of water-use on the incidence of disease.

It is therefore concluded that there was little improvement in health standards when water-use was increased beyond 60 litres per person per day. it is fair to say that this figure is an approximate one and by no means represents an absolute value.

Main Recommendations

It is imperative that research be conducted into the provision of water in developing countries, particularly on the relationship between the adequacy and inadequacy of water services and the health and well-being of the people, and the extent to which sufficient provision of these services is inhibited by social, economic, and environmental factors. This is necessary because designs of water schemes are generally based either on existing provisions with an annual percentage growth factor, or on current design criteria in industrial countries. Neither of these bases is satisfactory.

There is a need to conduct studies which can identify the benefits of a safe and adequate water supply in order to quantify the economic burden of water-related diseases as represented by loss of life-capital through premature deaths, loss of production from absenteeism, etc. The value of water includes both personal benefits-such as freedom from disease and feeling miserable, and from a lowered potential energy output in adult working life caused by childfood infection-and its monetary value in commerce and industry. There is also a need to re-appraise the position of water supply compared with alternatives in social investment and with investment in more directly productive activities.

When considering the future of semi-arid regions, prompt consideration should be given to the scarcity of water. The first priority in training or research programmes should be given to studying the existing water sources, investigating accessible resources, and conducting research into suppression of evaporation at reasonable cost.

The planning of water schemes should be based on an under standing of the role of water systems in a wider socio-economic context and should, therefore, occupy an important position in urban planning. Current policy, however, tends to consider water utilities in isolation. The idea that the installation of a permanent water supply system in a town would stimulate other activities leading to further community development, has not yet taken root.


Bradley, D.J., and P. Emurnon. 1968. "Predicting the Epidemiologal Effects of Changing Water Sources: I. A Quantitative Approach." E. African Med. J., 45, No. 5, pp. 284-91.

Carruthers, I.D. 1973. Impactand Economics of Community Water Supply: A Study of Rural Water Investment in Kenya. Agr. Dept. Studies Report No. 6, Ashford, Kent, UK: Wye College.

Damme, J.M.G.V. 1973. "Needs and Problems in Water Supply in Developing Countries." Proc. Conf. Environmental Health Eng. in Hot Climates and Developing Countries. Loughborough.

Dietrich, B.H.,and J.M. Henderson, 1963. Urban Water Supply Conditions and Needs in Seventy-five Developing Countries. WHO Public Health Paper No. 23.

Fair, G.M., J.C. Geyer, and D.A. Okun. 1966. Water and Waste water Engineering. Vol. 1: Water Supply and Wasterwater Removal. New York: John Wiley.

Hollister, A.C., et al. 1955. "influence of Water Availability on Shigella Prevalence in Children of Farm Labour Families." Amer. J. Publ. Hlth., 45, pp. 35462.

Lee, T.R. 1969. Residential Water Demand and Economic Development. University of Toronto Press.

Longmate, N. 1970. Alive and Well. Harmondsworth: Penguin.

Pineo, C.S.. and D.V. Subrahmanyam, 1975. Community Water Supply and Excrete Disposal Situation in the Developing Countries. Geneva: WHO.

Southgate, B.A. 1969. Water: Pollution and Conservation, London: Thunderbird Enterprises.

Steel, E.W. 1960. Water Supply and Sewerage, 4th ed. New York: McGraw-Hill.

Stewarts, W.H., et al. 1955. "Diarrhoeal Disease Control Studies IV: The Relationship of Certain Environmental Factors to Prevalence of Shigella Infection." Amer. J. Trop. Med Hyg., 4, pp. 718-24.

Wallace, J. 1893. Sanitary Engineering in India. Bombay: Educ. Soc. Steam Press.

White, G.F., D.J. Bradley, and A.U. White, 1972. Drawers of Water. Chicago: University of Chicago Press.

Wolman, A., and H.M. Boch. 1963. "US Water SUPPLY Lessons Applicable to Developing Countries." J. Amer. Water Works Assoc, 55, No. 8. pp. 941-56.

World Health Organization. 1973. Reuse of Effluents Methods of Wastewater Treatment and Health Hazards. WHO Technical Report Ser. No. 517.

Zijl, W.J. van. 1966. "Studies on Diarrhoeal Diseases in Seven Countries by the WHO Diarrhoeal Advisory Team." Bull. WHO 35, pp. 249-61.


Contents - Previous - Next