Contents - Previous - Next

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


Two main findings support the premise that the CFP programme has achieved a satisfactory level of efficiency in providing preventive health and nutrition services. First, systematic participation in the programme distributing milk-cereal to preschoolers exceeded 80% in the lower income deciles. Second, although milk-cereal is a take-home product, it was consumed mainly by preschoolers, the targeted group, with little leakage to other household members and practically no leakage to other households. The low level of leakage to adults is directly linked to the product characteristics, which are tailored to consumption by children.

Even though the rate of systematic participation by the low-income groups was as high as 80% or above, concern persists about increasing coverage of health services and food distribution, especially to families with undernourished or at-risk children. One of the main factors hindering participation among the poor is the high demand on mothers' time, particularly in acquiring regular health examinations for their children. Two findings sustaining this hypothesis were a higher average rate of participation at health centres that provided the milk-cereal for longer time intervals, therefore requiring less frequent food pick-up; and an inverse relationship observed between participation and the number of hours spent waiting for regular health examinations, a prerequisite for obtaining the supplementary food.

The hypothesis that product acceptability is correlated with the decision not to participate in the programme has been accepted. Milk-cereal, which is not commercially available in the country, was not consumed by 10% of the children under study because of its taste or perceived low quality, despite being fairly well accepted by children and considered highly nutritious by mothers.

TABLE 13. Least-square equations for daily calorie intake of children 1-5 years old

Explanatory variable Model 1 Model 2
Constant 441.9 (1.162) 481.0 (1.282)
Log monthly household expenditure per capita 74.76** (2.965) 71.49** (2.874)
Education of household head (dummy)a 2.34** (4.051) 25.50** (4.137)
Number of adults in household 4.68 (0.366) 2.52 (0.201)
Number of school-age children in household -45.85* (-2.268) -40.07* (-2.005)
Age of child 256.37** (3.243) 251.85** (3.066)
(Age of child)² -25.74* (-2.279) -23.40* (-2.096)
Sex of child (dummy)b     46.88 ( 1 .151)
Predicted participation in CFP -652.9* (- 1.896) 637.3* ( - 1.876)
Calories from on-site feeding programmes     0.441** (6.621)
Participation in on-site feeding programmes (dummy) 350.3** (5 731)    
N 386   386  
0.217   0.238  
Mean of dependent variable (kcal) 1,354.1   1,354.1  

a. This variable takes the value of 1 when the head of the household has attained secondary education.
b. This variable takes the value of 1 when the child is male.

* Significant al the 0.05 level using a two-tailed test.
** Significant at the 0.01 level using a two-tailed test

Another finding is that milk-cereal consumption by preschoolers was one of the three most important sources of calories in low-income households. The product supplied 11.2% and 9.8% of total calorie intake to children in income quintiles I and 2 respectively for households that reported consuming the product at the time of the survey. The resulting average income transfer per beneficiary in the lowest income decile varied between 3.2% and 13.1% of household income, depending on the nutrition status and age of the child.

Nevertheless, milk-cereal is a substitute for other milk, and its consumption does not imply a net increase in the calorie intake of preschoolers, as is the case with on-site feeding programmes. This conclusion is supported by regression analysis when controlling for other intervening factors, such as household income, age of child, education of household head, household composition, and participation in on-site feeding programmes such as those at day-care centres and school breakfast and lunch programmes.

The study showed that the provision of free milk-cereal to low-income families reduces household expenditure on food, allowing them to achieve a higher level of satisfaction through the consumption of other goods or through diet diversification. This does not, however, imply a higher consumption of calories.

A final conclusion is that calorie adequacy is not a good indicator of nutrition status per se in a middle-income country such as Chile. No statistical correlation was found between those variables using any of the conventional anthropometric indicators. Nevertheless, a fairly strong association was found between calorie adequacy and income, and also between nutrition status and income. These results illustrate the fact that a number of inputs are necessary for the improvement of health and nutrition, which tend to improve with income. Therefore, in the absence of purchasing power among poorer families for the acquisition of such inputs, social programmes addressing nutritional objectives should aim at providing more than just food.



1. ODEPLAN. Faculty of Economic and Administrative Sciences, University of Chile. Impacto redistributivo de gasto social. Santiago: ODEPLAN, 1986.

2. Atalah E, Puentes R, Castillo C, Radrigan M. Programa de Alimentación complementaria, 1965-1985. Rev Chil Pediatr 1985;56(5):362-68.

3. Ministry of Health, Chile. Tabla de referencia antropométrica NCHS: una alternative pare la evaluación nutricional de menores de 6 años en Chile. Santiago: Ministry of Health, 1986.

4. ODEPLAN. Encuesta de Caracterización socioeconómica nacional (CASEN). Santiago: Ministry of Finance, 1985.

5. Muchnik E, Vial I. Impacto del PNAC en preescolares de Santiago. Santiago: Ediciones Mar del Plata, 1990.

6. Schiff M, Valdés A. Nutrition: alternative definitions and policy implications. Econ Dev Cult Change 1990; 38(2):281-92.


Food and nutrition interventions in Brazil

Yony Sampoio and Antonio Carlos Campino

Malnutrition and mortality: An overview

What global effects can be attributed to Brazil's nutrition and health interventions? General indicators are not easily explained, since causalities can run in different directions. Some economic data are presented in table 1. Income distribution, for example, has become more skewed since 1960. Brazil's income concentration is one of the highest in the world. Average income has generally increased since 1960 as a consequence of high rates of growth of real product; however, a severe recession during 19811983 caused it to drop sharply, while unemployment rates reached 10% in some state capitals. Income transfers and supplementary programmes may have alleviated some of the effects of worsening economic conditions during this period

Production of basic food crops has lagged in relation to population growth, particularly since 1974 [2]. This has forced prices up, and the cost of food index shows higher increases than the general cost of living index.

Infant mortality shows a declining trend (table 2), but this is attributed mainly to sanitation and health programmes. An evaluation of health activities in a rural development programme showed statistically significant decreases in morbidity and mortality [4]. Nutrition interventions may have played a role in counterbalancing the worsening income conditions during 1981-1983, but it is not possible to state this definitively. What can be said is that the country's social conditions have not improved greatly and are far below the level in countries with similar economic conditions [5].


Food and nutrition interventions

Interventions prior to INAN

Food shortages and famine have been a constant problem in Brazil since at least early colonial times. In this export-oriented agricultural economy, few resources were diverted to the production of basic foods for internal consumption. Colonial authorities tried unsuccessfully to solve the problem by decreeing (but without enforcement) that a fixed proportion of land should be devoted to export and food crops, such as sugar-cane and manioc.

In the first 50 years of this century the combined effects of urbanization and periodic droughts in the north-eastern area worsened the food shortages, resulting in several episodes of unrest. The government took action only in the worst years, implementing emergency imports, food distribution, and other programmes. Only in the early 1940s were effective measures taken to improve the workers' lot in a policy clearly intended to smooth relations between emerging industrial firms and the unions. A minimum wage was established in 1940, and in the same year the food service of the Social Security System was instituted with the major objective of supplying low-cost meals to workers protected by the programme [6-9].

Between that time and 1972, a school feeding programme, usually called the school-lunch programme, was created and expanded, and food was distributed free or sold at subsidized prices [6; 10]. It should be remembered that in the 1950s and 1960s the literature emphasized protein deficiencies and limitations in the production of basic foods as major causes of malnutrition, although Brazil had a positive record in these areas [1; 11; 12]. This approach to the problems of malnutrition motivated the creation of food technology departments in universities and also led to experiments with food fortification.

TABLE 1. Economic indices, Brazil 1960-1985

  % of income to Real income as proportion of minimum real wage
Lower 40% of population Top 10% of population Average Lower 40% Top 10%
1960 12.2 37.4 1.69 0.52 6.30
1970 11.1 43.8 2.10 0.58 9.20
1980 9.2 46.7 3.77 0.87 17.64
1981 9.3 45.5 3.80 0.88 17.28
1983 8.1 46.2 3.16 0.64 14.61
1985 8.3 47.2 3.20 0.66 15.12
  Gini coefficient Real productiong rowth rate Population below poverty line (%) Open unemployment
Millions Rate
1960 0.496 9.7 - - -
1970 0.547 8.8 - - -
1980 0.580 7.2 - - -
1981 0.579 -1.6 25.1 2.0 5.8
1983 0.597 -3.2 34.5 2.4 6.3
1985 0.608 8.3 6 29. 2.2  

Source: Ref. 1.
a. Those earning less than one minimum wage. In 1985 the average value of the minimum wage was Cz$322,80 per month, corresponding to US$52.

TABLE 2. Infant mortality rates. Brazil. 1977-1984

Year Brazil North-east North South-east South Centre-west
1977 104.3 160.1 127.3 77.3 68.8 75.2
1978 86.8 145.6 107.6 72.3 58.4 69.1
1979 87.6 131.2 100.1 66.8 52.1 58.5
1980 81.1 122.5 93.0 59.9 48.9 51.6
1981 74.1 111.6 91.0 56.0 42.4 49.7
1982 65.8 93.1 81.3 53.4 38.0 42.4
1983 68.6 103.2 98.2 50.1 37.2 45.0
1984 73.7 116.1 98.6 52.0 39.7 41.4

Source: Ref. 3.


Following the general trend in 1972, the government created the National Institute for Food and Nutrition (INAN), a multidisciplinary agency affiliated with the Ministry of Health but charged with formulating the National Food and Nutrition Programme (PRONAN) [10; 13-15]. Not until 1974, however, was an inter-ministerial board added that provided the means for integrate and formulating policies intended to increase production and improve the marketing of basic food crops, and to expand supplementary food programmes for priority target groups: pregnant and nursing women, and preschoolers [6; 10]. The experience of INAN reflected a world trend and the influence of international agencies, and is quite similar to that of the Colombian National Food Programme (PAN) and the Mexican Food System (SAM) [16-19].

The first PRONAN, which covered 1973-1974, was never effective. Under a new director, the second PRONAN was planned for 1976-1979 but was extended because the third programme, submitted in 1980, never received final approval. Thus, during two presidential periods (1979-1984 and 1985 to the present) no new national programmes were introduced.

In 1975 INAN signed a US$71 million agreement with the World Bank for the purpose of testing new intervention strategies, developing technical resources, and acquiring planning, managerial, and evaluation expertise [6; 20; 21]. Similar World Bank agreements were signed in Indonesia, Colombia, and India [22]. Table 3 shows a summary of food and nutrition interventions in Brazil for the period 1975 - 1985.

TABLE 3 Food and nutrition interventions, Brazil, 1975-1985

Programme Target group Dates
PSA (PNS) supplementary feeding programme pregnant and nursing women: children under 7 yr 1975-
PNAE school-lunch programme first-graders in public and philanthropic schools 1954-
Improvement of small producer agriculture    
PROCAB rationalization of production of basic food crops small producers, cooperatives 1977-
PAT workers" food programme employees 1977-
PINS food subsidy and health-service programme pregnant and nursing women; children under 6 yr 1977-1980
PRAMENSE small-agriculture and primary health care low-income rural producers 1977-1980
PROAPE integrated education and food supplementation preschoolers 1977-1984
PCA complementary feeding programme pregnant and nursing women; children 3-36 mo 1977-
Food subsidy urban poor 1979-
— PROAB (through private retail outlets) urban poor 1979-
— wheat subsidy consumers and producers 1951-1987
PNL national milk programme families with children under 7 yr 1985-
PAP people's food programme urban poor 1985-

The second PRONAN attempted to define a more global strategy directed at the medium term and was in clear contrast to short-run interventions such as food supplementation. Its approach, strikingly similar to those of the PAN and the SAM, was to focus its activities on low-income families, rather than to target individuals. In addition, it was tied to income-generating policies, such as credit and technical assistance to poor farmers, and to policies aimed at correcting structural inefficiencies, particularly in land ownership and the marketing system. PRONAN also emphasized the use of basic rather than processed foods [23; 24].

But PRONAN could not avoid political compromises with other ministries. The school feeding programme, under the Ministry of Education, was added to PRONAN, a supplementary feeding programme (PNS, later PSA) was established, and a new workers' lunch programme (PAT) was negotiated with the Ministry of Labour.

In addition several experimental programmes were financed by the World Bank and implemented during PRONAN II [20]. These were an integrated food-subsidy and health-service programme for low-income families in the Recife area (PINS), a combined agricultural project with primary health care in rural Sergipe (PRAMENSE), and an integrated education and food-supplementation programme for preschoolers (PROAPE) [20]. As a result of these, two additional programmes were established, although the basic ideas were already present in PRONAN II. One was a scheme to purchase agricultural products directly from small producers (bypassing intermediaries), operating in some areas of the northeast (PROCAB). The other was a geographically targeted food subsidy implemented through private retail outlets, a kind of fair-price shop programme that started in Recife and was later expanded to other state capitals in the north-east (PROAB).

Other operating programmes provided supplementary food to pregnant and nursing women and children under 7 years of age (PNS), and pregnant and nursing women and children from 3 to 36 months of age (PCA). The PNS, called PSA after 1985, distributes food through the state and municipal government health centres under the coordination of the Ministry of Health. The PCA distributes food-mixture formulas and is coordinated by the Brazilian Foundation League of Assistance, an agency of the Ministry of Social Welfare.

From its inception in 1976, PRONAN II was never effectively integrated with national development plans because agreement could not be reached with the Ministry of Agriculture or other ministries. For this reason, programmes for the agricultural sector (PROCAB and PRAMENSE)-a key area for PRONAN II-achieved meagre results and did not reach the intended number of beneficiaries.

In 1981 a controversy over PRONAN developed within the government, and the Ministry of Planning postponed approving the third PRONAN under pressure from the National Council for Social Development. What followed was similar to some of the difficulties experienced by the PAN and the SAM.

No new programme was created, nor was there was a precise assessment of continuing projects. Because of severe credit restrictions, due to the external debt problems and the IMF demand to reduce the budget deficit, total real resources allocated to food and nutrition policies decreased, although traditional subsidies (e.g., wheat) were not greatly affected. Also, because of decreases in real wages and increases in the unemployment rate, labour unions and the opposition party focused on economic policies, rather than on INAN or PRONAN. A point worth emphasizing is that food and nutrition policies are directed to those most in need, but these groups have no advocates who can put pressure directly on the government [16].


Post-INAN interventions

It was expected that with the election of the first civilian government in 21 years the power of INAN and similar institutions would be enhanced, but this was not the case. Although the government spoke of reducing the social debt, still greater emphasis was placed on economic policies, without any attention to their distributive consequences. It was no surprise that with mayoral elections in all capitals in 1985 and general elections in 1986, populist policies that included general subsidies for commodities such as milk were adopted and implemented without prior studies. INAN's resistance to changing its initial philosophy unfortunately led to a decline in its influence, and ultimately resulted in its isolation.

Without the patronage of INAN, new programmes were planned and coordinated by the Secretariat for Community Affairs under the office of the President of Brazil. A fair-price shop programme with no subsidy, the People's Food Programme (PAP), and a non-targeted programme of milk distribution to families with preschool children (PNL) were created in 1985. In 1987 a food-basket subsidy and a bread subsidy to low-income families (to counteract the end of the wheat subsidy) were announced but neither was implemented.

A reflection on the INAN experience shows many similarities to multidisciplinary nutrition institutions established in other countries during a 'decade of illusion" [25; 15; 14]. INAN followed the same course as the PAN and the SAM. All three survive but have much less power than previously; the PAN is under the Ministry of Agriculture, the SAM is within Mexico's new National Food Programme (PRONAL), and INAN is still alive, although PRONAN has ceased to operate. However, some of PRONAN's activities have been picked up by other agencies. All three programmes (Colombia, Mexico, and Brazil) exhibited a similar trend of initial enthusiasm followed by decline.

Contents - Previous - Next