Contents - Previous - Next

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

3. The potential impact of nutrition education

Tufts University School of Nutrition, Medford, Massachusetts, USA

This presentation on the potential impact of nutrition education could also be titled, "How Much to Invest in Nutrition Communications?" Should nutrition education receive the same level of funding currently given to food aid or to major programmes in family planning, health, and agriculture ?

The answer depends not only on the cost-effectiveness of nutrition education programmes to date; it also depends on the potential impact that nutrition education could have if it received more than token levels of support. High-level policy-makers tend to have little faith in the power of nutrition education. When speaking candidly, they tend to express beliefs such as the following: "Poverty or lack of food is the real cause of malnutrition. Without increase in income, food availability, or basic services such as health, educational methods are relatively powerless to combat malnutrition. Therefore, it is not worthwhile to invest large sums of money in nutrition education. Small amounts may be needed for 'project support communications' (UNICEF's term). Large amounts would not be appropriate."

As an example of this point of view, an agency funding a multi-million dollar nutrition programme in an Asian country recently questioned plans to spend 7 per cent on nutrition education. Senior agency executives felt this was too much.

This paper attempts to answer two questions: (1) Is lack of income the main cause of malnutrition? (2) What power would communications have to affect the problem, if significant sums of money were invested in it ?

In answer to the first question, a careful look at the causes of malnutrition in young children the major type of malnutrition in developing countries - shows that this conventional point of view simply is not correct, except among the very poor. Severe endemic malnutrition occurs in almost all developing countries at the weaning age. A major proportion of this malnutrition has been proven to be caused by ignorance and incorrect food and health beliefs, linked to poor feeding and health practices. The fact that this ignorance persists reflects inequity in the distribution of the knowledge generated by modern technology.

These facts have been documented for at least 15 years [1]. The Sidney M. Cantor study of nutrition as an integrated system in Tamil Nadu, South India, quantifies the weaning age problem clearly [2]. Figure 1 shows a comparison of daily calorie need with percentage fulfilment of calorie need from birth to 70 years old in Tamil Nadu. The deep dip in the curve between birth and three years indicates an average drop in intake to below 60 per cent of caloric adequacy at about 12 months. Figure 2 presents the same percentage fulfilment figures by age for five levels of total family adequacy and shows that the dietary intake of infants between about four months and two years of age was insufficient even in families who were meeting 112 per cent of calorie requirement. Figure 3, giving the same percentages by age for three levels of family expenditure, indicates that dietary insufficiency of infants aged 12 months, amounting to an intake of less than 70 per cent of requirement, is almost equally severe in families with a total monthly expenditure of 97 rupees as in the lowest-income families with an expenditure averaging only 24 rupees. Though less completely documented, age-specific inequality in intra-family food distribution and calorie deficiency during the weaning age regardless of family wealth have been reported by nutritionists the world over.

Fig.1. Comparison of daily calorie need (upper graph) and percentage fulfilment of individual calorie needs (lower graph), Tamil Nadu, South India.

Educated elite groups, such as families of university professors, for example, have adopted modern styles of infant feeding and do not experience this weaning crisis. Children in more traditional wealthy families recover nutritionally when the weaning period is over and become susceptible to the problems of overnutrition in succeeding years.

Fig. 2. Percentage of individual calorie needs fulfilled, by age, for five levels of family calorie adequacy, Tamil Nadu, India, 1972 (redrawn by Joe D. Wray from Sidney M. Cantor Associates [2], p. 101).

A second dip, or period of dietary inadequacy, during the teens and early twenties, is shown in figures 1 to 3. This dip has consistently been reported from many areas of the developing world [3], although it may not occur in regions such as the rain-forest belt of Africa, where calorie-dense crops are plentiful at all seasons. It may also be less common in areas in North Africa where women remain mainly in the house following early marriage. Low dietary intake at this age appears to have relatively few ill effects on the teenager or young adult, but does result in high rates of low-birth-weight infants and in a reduced breast-milk supply.

These figures imply that much, if not most, of the malnutrition that occurs during the weaning period is caused, not by inadequate resources, but by faulty feeding practices that can be corrected by nutrition education. Susan Scrimshaw ,[4] and others have produced evidence that such destructive feeding practices may have served a necessary if cruel function before the advent of family planning. They permitted enough infants to die to prevent population size from exceeding the agricultural carrying power of the land. Incorrect beliefs may allow these deaths to occur with no blame to the family or to society. For example, if an infant dies of dehydration, no one is to blame if there is a belief that fluids must be withheld during episodes of diarrhoea.

Fig. 3. Percentage of calorie needs fulfilled, by age, for three levels of family total expenditure, Tamil Nadu, India, 1972 (redrawn by Joe D. Wray from Sidney M. Cantor Associates [2], p. 102)

A study by Levinson [5] of 496 6- to 24-month-old children in Morinda in the Indian Punjab also illustrates dietary insufficiency at the weaning age, irrespective of income, and relates this insufficiency to nutritional status. Table 1 shows frequency distributions of percentage caloric intake allowance by per capita income and sex. Sixteen per cent of infants in the highest income group were found to be ingesting only 45 per cent of allowance or less. Table 2 compares the income elasticities of consumption of basic foods among sample children with total expenditure elasticities of demand for all age groups in rural India.

Levinson concluded [5, p. 53]:

The figures indicate clearly that the effects of income per se on the food consumption of the young children are far less pronounced than for the population as a whole. The income elasticities of consumption for calories, even among the 18- to 24-month age group with a lesser dependence on breastmilk, is only .082, meaning that if a family somehow succeeded in doubling its income (assuming everything else, including prices, remained constant) the child's caloric intake would increase by only 8 per cent. (The increase in adult's caloric intake might be 10 times that great.) The young child's protein, vitamin A, and iron intake would increase by even less. This would suggest that, for the sample population as a whole, a simple income supplementation program unaccompanied by other interventions is unlikely to have a major effect on their food intake.

Table 3 shows the nutritional status of the children according to the Gomez classification, grouped by per capita income and sex. Nearly 46 per cent of all children in the highest income category were suffering from second- or third-degree malnutrition, with a very high preponderance of malnourished females.

Table 1. Frequency distribution of percentage caloric intake allowance by per capita income and sex

Per capita monthly income (rupees) Percentage ingesting 76-100% of allowance Percentage ingesting 59-75% of allowance Percentage ingesting 46-58% of allowance Percentage ingesting 45% and below of allowance
All children
24 and below 17.5 23.3 32.0 27.2
25-39 14.9 27.2 30.7 27.2
40-50 15.2 35.9 28.3 20.7
51-75 37.9 18.9 18.9 24.2
76 and above 32.3 32.3 19.4 16.1
24 and below 25.0 25.0 29.2 20.8
25-39 13.6 20.3 32.2 33.9
40-50 19.2 44.2 21.2 15.4
51-75 41.7 22.9 14.6 20.8
76 and above 30.2 35.8 22.6 11.3

Source. Reference 5.


Table 2

Item Income elasticities of consumption among sample children aged Total expenditure elasticities of demand for all age groups in rural India






Total sample children
Supplementary milk .16 .18 .28 .24 1.66
Cereals -.16 -.10 .01 .01 0.50
Pulses -.11 .18 -.07 .03 0.71
Fruits -.05 -.08 .01 .04 1.43
Vegetables .47 .23 .12 .23 0.69
Calories .07 .04 .08 .07 -
Protein .05 .03 .03 .04 -
Vitamin A .01 .04 .07 .04 -
Iron -.06 -.06 .02 .01 -

Source: Reference 5.

Very poor communities for which nutrition education cannot be effective without simultaneous increase in real income exist in pockets in most countries and more generally in some of the low-income countries most in need of development. Nutrition education teaches better uses of resources that are already available to the family. When these resources fall below a certain level, redistributing them does not help.

Wittmann and co-workers [6], in the slums of Cape Town in South Africa, found a marked difference between the health and nutritional status of children in families receiving 21 cents per person per day versus families with 60 cents per person per day. In Maduri, South India, per capita income of the lowest 23 per cent of families with children attending the Nutrition Rehabilitation Unit at Lady Erskine Hospital was less than 5 rupees per month [7]. The low-cost diet offered in the centre, making use of groundnut cake as a source of protein, cost 60 paise per child per day. Because these families could allot no more than 10 paise a day to each pre-school child, it is very doubtful whether they had the means to feed their children.

Table 3. Frequency distribution of Gomez classification groupings by per capita income and sex

Per capita monthly income Percentage normal (over 90% of Harvard reference weight for age) Percentage with 1st-degree malnutrition (76-90%) Percentage with 2nd-degree malnutrition (60-75%) Percentage with 3rd-degree malnutrition (below 60%)
All children
24 end belong 9.7 29 1 51.5 9.7
25-39 11.6 33.0 41.1 14.3
40-50 18.1 38.3 35.1 8.5
51-75 16.1 44.1 35.5 4.3
76 end above 19.1 35.1 43.6 2.1
24 and below 14.0 40.0 42.0 4.0
25-39 14.0 47.4 36.8 1.8
40-50 30.2 41.5 24.5 3.8
51-75 25.0 Solo 25.0 0.0
76 and above 30.2 43.4 26.4 0.0
24 and below 5.7 18.9 60.4 15.1
25-39 9.1 18.2 45.5 27.3
40-So 2.4 34.1 48.8 14.6
51-75 6.7 37.8 46.7 8.9
76 and above 4.9 24.4 65.9 4.9

Source: Reference 5.

Adeline Andre, nutrition educator with the Department of Agriculture in Haiti, tried to live from the market for the US $0.09 that low-income families average daily (per capita) for food and concluded that it is simply not possible to eat adequately for this amount unless one can collect firewood and grow green leafy vegetables at home [8]. Similarly, Pellerin [9] concluded that the 47.6 per cent of Haitian farmers with farms less than 0.5 hectares (1.25 acres) in size probably were unable to redistribute food resources in a manner than would make it possible to prevent malnutrition among pre-schoolers.

Dr. P.M. Shah [10] estimated from a domiciliary rehabilitation programme in Kasa, India, that "only one-third of the severely malnourished, who are extremely poor, need nutrition from outside sources." In the concept-testing exercise of the World Bank-supported nutrition project in Indonesia, Griffiths and co-workers [11] discovered that about 10 per cent of families were too poor to try suggestions for improving nutrition even at low cost.

Table 4. Profile of nutrition education activities by funding source in the United States

Term used for nutrition Nutrition education education Nutrition education Risk reduction Nutrition counselling/ weight control Food advertising
Stated goals Behaviour change/ nutritional status change Informed choice Behaviour change/ improved risk Behaviour change/ improved risk indicators Behaviour change = purchase of healthful foods
USS/yeara 0.75 100 5 (for nutrition component only) Not estimated 2,000 (only for foods reducing risk of cardiovascular disease)
Target groups Developing country MCH Low income MCH High-risk adults, entire population Sick or high risk, MCH High-risk adults,
entire population
Role of nutrition professionals Significant Important Minimal Moderately important Minimal
Estimated impact on national dietary practices High return on investment Low High return on investment High Very high

a. Approximate estimates by informed individuals.
b. S Agency for International Development.
c. Department of Agriculture.
d. Health and Human Services (National Institutes of Health).
Source: Reference 12.

The poverty threshold below which the family cannot afford to redistribute food to the weaning-age child is low. Only 3 to 5 per cent of the total family food budget needs to be shiRed to the youngest. This is calculated by noting that the average 12-monthold child needs 400 calories per day to increase his intake from 60 to 100 per cent of requirement. This 400 calories is less than 5 per cent of the total daily intake of a family of six - two adults and four children - who consume 70 to 80 per cent of requirement. This calculation demonstrates clearly that the malnutrition problem among young children is not due to income constraints in most cases.

As for the second question: What would be the potential of nutrition education if substantial funds were invested in it? Since this has rarely, if ever, occurred in developing countries, it may be informative to look at the nutrition education situation in the United States. Table 4 [12] gives a profile of nutrition education expenditures by funding source in the US. It shows that an estimated $2 billion are spent per year by the food industry to advertise foods that are good for health. This advertising, which can be termed "nutrition education," is about one-fourth of a total of $8 billion spent yearly on food advertising.

The effectiveness of advertising in changing food behaviour is beyond dispute. Food companies can expect to earn in profits an average of $1,000 for every dollar spent on advertising. The amount spent by the food industry on research to develop new food products and advertising materials to sell them is many times more than amounts given for nutrition education development. A US food company thinks nothing of spending $200,000 on research for product development alone before designing a new candy bar that will cost $0.50. Sophisticated computer modelling is used to figure out what new products the public will buy and how to sell them.

The point here is that these amounts would not be spent if human behaviour were not highly influenced by such research and marketing. The Indonesian case study presented by Mr. Ruslan Adji elsewhere in this volume is an example of a nutrition education project in which the concept-testing methods of the US food industry were adapted and applied in rural Indonesia to develop and promote nutrition education messages. This project made a major impact on young child malnutrition in five subdistricts in Indonesia.

In the United States the food industry started roughly 30 years ago to market products, such as margarines low in saturated fats, advertised to reduce the risk of heart disease. Over the years this nutritional marketing increased in volume in proportion to public demand. About 15 years ago the death rates from nutrition-related degenerative diseases began decreasing nationally. Recent reports indicate that the rate of death from stroke, for example, fell by 40 per cent in the past decade.

Such lag time between marketing campaigns and beneficial effects may not be necessary in developing countries. Degenerative diseases take decades of bad eating habits to develop. The young child's nutritional status can deteriorate or improve in weeks or months.


This paper presents evidence suggesting that the potential impact of nutrition education in developing countries would be very high if significant funds were invested in changing food habits. Existing evidence should justify at least one trial of large-scale funding of nutrition communications on a national scale in a developing country with an investment of from 10 to 20 million dollars. These funds should be used for scientific message development by means of field trials and market research and for expert design of campaign materials.


1. D.B. Jelliffe, Infant Nutrition in the Tropics and Subtropics (World Health Organization, Geneva, 1968).

2. Sidney M. Cantor Associates, Inc., "An Operations Oriented Study of Nutrition as an Integrated System in the State of Tamil Nadu,' The Tamil Nadu Nutrition Study, vol. I, report to USAID/nesa399 (Haverford, Pa., 1973).

3. A. Lechtig, J.-P. Habicht, H. Delgado, R.E. Klein, C. Yarbronsh, and R. Martorell, "Effect of Food Supplementation during Pregnancy on Birth-weight," Pediatrics, 56: 508-520 (1975)

4. S.C.M. Scrimshaw, "Infant Mortality and Behavior in the Regulation of Family Size," Population and Development Review, 4(3): 383-403 (1978).

5. F.J. Levinson, Morinda: An Economic Analysis of Malnutrition among Young Children in Rural India, Cornell/MIT International Nutrition Policy Series (Cambridge, Mass., 1914).

6. W. Wittmann, A. Moodie, J. Hansen, and J. Brock, "Studies on Protein Calorie Malnutrition and Infection," Ciba Foundation Study Group, no. 31, in G.E.W. Wolstenholme and M. O'Connor, eds., Nutrinon and Infection (Little, Brown & Co., Boston, 1967).

7. G. Venkataswami and S.A. Kabir, "Report on Nutrition Rehabilitation Centre and Village Child Care Centres with Evaluation" (Government Erskine Hospital, Maduri, India, 1975).

8. A. Andre, personal communication, 1977.

9. A. Pellerin, "Some Challenges for Training Auxiliary Nutritionists for the Ministry of Health of the Republic of Haiti," in K.W. Shack, ea., Teaching Nutrition in Developing Countries, or the Joys of Eating Dark Green Leaves (Meals for Millions Foundation, Santa Monica, Calif., 1977), pp. 54-66.

10. V. Reddy, "Nutrition Rehabilitation Units," Indian Pediat., 12(1): 103 (1975).

11. M. Griffiths, R.K. Manoff, T.M. Cooke, and M.F. Zeitlin, Volume 1: Concept Testing, Nutrition Communication Project Paper, Nutrition Communication and Behaviour Change Component, Indonesian Nutrition Development Programme (Manoff International, Washington, D.C., 1980).

12. M.F. Zeitlin, The Two Worlds of Nutrition Education: Differences and Similarities in Industrialized versus Developing Countries, New Developments in Nutrition Education, Nutrition Education Series, issue 11 (Unesco, Paris, 1985).

Contents - Previous - Next