Cover Image
close this bookEarly Supplementary Feeding and Cognition (Society for Research in Child Development, 1993, 123 pages)
View the document(introductory text...)
View the documentAbstract
View the documentI. Introduction
Open this folder and view contentsII. Methodology and findings of the longitudinal study
Open this folder and view contentsIII. Conceptual rationale for the follow-up
Open this folder and view contentsIV. Methodological and substantive considerations
Open this folder and view contentsV. Methods of the cross-sectional follow-up
Open this folder and view contentsVI. Results from the cross-sectional follow-up
Open this folder and view contentsVII. Discussion
View the documentAppendix A: Average nutrient intakes of Atole and Fresco subjects
View the documentAppendix B: Descriptions of tests used in the analysis of the preschool battery
View the documentReferences
View the documentAcknowledgments
View the documentCommentary - Going beyond nutrition: Nutrition, context, and development
View the documentCommentary - Early supplementary feeding and cognition: A retrospective comment
View the documentReply - Nutrition and development: Considerations for intervention
View the documentContributors
View the documentStatement of editorial policy

Commentary - Early supplementary feeding and cognition: A retrospective comment

Nevin S. Scrimshaw

Three common nutritional deficiencies are now recognized to have the potential for permanent adverse effects on learning and behavior, those of protein-energy deficiency, iron, and iodine. The lasting effects of protein-energy malnutrition began to be recognized from studies in the 1950s and those of iron in the late 1960s. Although feeble-minded dwarfs have been associated with maternal iodine deficiency for more than 50 years, convincing evidence of lesser degrees of neurological damage due to iodine deficiency during the first and second trimester of pregnancy was first presented only in the 1970s. For both iron and chronic protein-energy deficiencies, permanent damage can occur in infancy and early childhood.

The severe acute deficiency of protein relative to calories that results in kwashiorkor is characterized by a profound apathy that responds relatively rapidly to a therapeutic diet. However, cognitive tests of children who had recovered from kwashiorkor gave evidence of continuing impairment. These long-term effects seemed to be due, not to the acute episode of kwashiorkor, but to the duration and severity of the marasmus on which the acute episode of kwashiorkor is usually superimposed (Yatkin & McClaren, 1970). Most cases of kwashiorkor occur after the first year of life as a consequence of poor weaning practices and concurrent infections.

When marasmus develops in infants and young children as the result of a chronic lack of both dietary energy and protein, it produces lasting effects on cognition, although these can be moderated by refeeding and psychosocial stimulation (Grantham-McGregor, Stewart, & Schofield, 1980). Decreased cognitive function with partial starvation in adults was a consistent finding in the World War II studies of Keys, Brozek, Henschel, Mickelsen, and Taylor (1959). Among the many countries from which lasting cognitive effects have been described as the result of kwashiorkor and marasmus are Chile (Monckeberg, 1968), India (Champakam, Srikantia, & Gopalan, 1968), Jamaica (Grantham-McGregor, Powell, Walker, & Himes, 1991), Lebanon (Hoorweg & Piaget, 1977), Mexico (Gomez, Velasco, Ramos, Cravioto, & Frenk, 1954), Peru (Pollitt & Granoff, 1967), Uganda (Gerber & Dean, 1967), and Yugoslavia (Yatkin & McClaren, 1970).

Cravioto and his colleagues (Cravioto & DeLicardie, 1968; Cravioto, DeLicardie, & Birch, 1966) deserve credit for demonstrating that subclinical malnutrition manifested only by impaired growth can affect behavior. Among lower socioeconomic groups in both Mexico and Guatemala., preschool children in the lowest quartile of weight for age had significantly poorer performances on various tests of intersensory perception than did those in the highest quartile. For children of university faculty and other professionals in middle- and upper-income groups in the two countries, there was no relation with weight quartile. The extent to which the differences were due to the lesser sensory stimulation of children of the poorest families rather than the biological consequences of a poor diet and -infections could not be determined at the time.

In March 1967, the international conference "Malnutrition, Learning and Behavior," held in Cambridge, Massachusetts, brought together for the first time four overlapping disciplinary groups - nutritionists and behavioral scientists, some working with experimental animals and others with human subjects (Scrimshaw & Gordon, 1968). The evidence presented at the meeting that malnutrition in the infant animal could impair performance on behavioral tests was incontrovertible. So was the experimental evidence for a permanent effect of early malnutrition on the brain structure and functioning of these animals. It was also clear that lack of the stimulation could have similar effects and that stimulation could, at least partially, prevent the effects of malnutrition.

The coexistence of social deprivation and malnutrition, characteristic of children in low income groups, was not confounding in the work; with experimental animals. However, in studies of children, the influence of the two factors could not be separated, and the conference was unable to resolve this dilemma. Moreover, the anatomical evidence of neurological change in malnourished children was weak except for a correlation with head circumference.

An early study frequently cited at the time, that of Stoch and Smythe (1968) in South Africa, was invalid because the malnourished and control groups were from quite different social and economic strata. Even in those studies in which the malnourished and control groups were carefully matched for socioeconomic status, the stimulating effects of the supplement were not taken into account.

The 1967 conference stimulated a number of now classic studies in developing-country populations. In Bogota (More, Herrera, Sellers, & Ortiz, 1981), children from poor families received either food supplementation for the entire family, from mid-pregnancy until the target child was 3 years old, or a twice-weekly home visiting program, from birth to 3 years, to promote cognitive development. At 3 years of age, children who received the food supplementation averaged 2.6 cm and 642 grams larger than controls. Supplementation resulted in small but significant improvement in Bayley test scores. Similar findings were reported from a study by Bacon Chow in Taiwan (Adair & Pollitt, 1985).

In Cali, preschool slum children received 1-4 years of combined supplementation and stimulation daily (Sinisterra, McKay, McKay, Gomez, & Korgi, 1977). Although each additional year brought a significant improvement in cognitive test performance, final scores were 20% lower than in children of middle- and upper-income families not enrolled in the program, even after 4 years (McKay, Sinisterra, McKay, Gomez, & Lloreda, 1978). The experimental design did not permit identifying the role of malnutrition alone.

Two important studies were begun at that time, one in Guatemala at the Institute of Nutrition of Central America and Panama (INCAP) (Klein, Irwin, Engel, & Yarbrough, 1977), described in this Monograph, and the other in Mexico by the National Institute of Nutrition (Chavez & Martinez, 1982). Both were well designed, and both strongly confirmed that poor rural breast-fed children who received a protein-calorie supplement up to 2 years of age were taller, had fewer infections, and performed better on appropriate cognitive tests than children receiving only social stimulation.

In Guatemala, the experimental and control groups were in different villages, introducing a possible confounding factor. However, in Mexico, after the base-line studies, half the children in the village of Tezonteopan, selected randomly, received a supplement of flavored milk and added vitamins and minerals. All the children in this study received stimulation visits and medical care. In Guatemala, the comparison was between groups receiving a beverage rich in protein and energy (Atole) and groups receiving a protein-free, low-calorie drink (Fresco). In both the Guatemalan and the Mexican studies, growth and performance on behavioral tests were significantly greater in the supplemented group. It is unlikely, therefore, that the results of the initial study in Guatemala described in this Monograph are an artifact due to chance differences among the study villages.

Both studies were planned to minimize the confounding effects of social stimulus during delivery of the nutritional supplement. However, the design of these two seminal studies made it impossible to know what the situation would have been with no intervention at all. It is highly probable that the benefits of nutritional supplementation (compared with doing nothing at all) were underestimated because any supplementation also provides increased stimulation. It should also be noted that the differences between the Atole and the Fresco control groups in this study were due to the higher protein content of the Atole. The Fresco was very low in calories as well as protein free, but, since about three times as much of it was consumed per person, caloric intakes were similar.

Demonstrating the improvement in cognitive performance of infants and young children whose mothers had been supplemented and who have been given additional nutritious food during and after weaning is in itself an important achievement. It certainly provided support for the supplementary preschool feeding programs of international, bilateral, and private voluntary agencies and for such programs as Headstart in the United States. But there were always those who questioned the lasting effects of such programs on their beneficiaries.

In this Monograph, Pollitt et al. now present solid evidence that the benefits of correcting early malnutrition that were considered to be small at 2 years of age could be characterized as medium to large by the time the children had become adolescents and young adults. Moreover, the effects are evident in the kinds of tests and measures known to correlate with social and economic achievement. This Monograph and the similar results of the follow-up (Chavez, Martinez, & Soberanes, in press) of the Mexican study should dispel these doubts. This important Monograph should give new motivation and impetus to efforts to avoid the tragic permanent damage of early malnutrition on many children of underprivileged families everywhere.


Adair, L. S., & Pollitt, E. (1985). Outcome of maternal nutritional supplementation: A comprehensive review of the Bacon Chow study. American Journal of Clinical Nutrition, 41, 948-978.

Champakam, S., Srikantia, S. G., & Gopalan, C. (1968). Kwashiorkor and mental development. American Journal of Clinical Nutrition, 21, 844 - 855.

Chavez, A., & Martinez, C. (1982). Growing up in a developing community. Mexico City: National Nutrition Institute.

Chavez, A., Martinez, C., & Soberanes, B. (in press). Effect of malnutrition on infant development. In N. S. Scrimshaw (Ed.), Longitudinal community based studies of the impact of early malnutrition on child health and development. Boston: INFDC.

Cravioto, J., & DeLicardie, E. R. (1968). Intersensory development of school-age children. In N. S. Scrimshaw & J. E. Gordon (Eds.), Malnutrition, learning, and behavior. Cambridge, MA: MIT Press.

Cravioto, J., DeLicardie, E. R., & Birch, H. G. (1966). Nutrition, growth and neurointegrative development: An experimental and ecologic study. Pediatrics, 38 (Suppl. 2, Pt. 2), 319-372.

Gerber, M., & Dean, R. F. A. (1967). The psychological changes accompanying kwashiorkor. In F. R. Wickert (Ed.), Readings in African psychology from French language sources. East Lansing: Michigan State University Press.

Gomez, S. F., Velasco, A. J., Ramos, G. R., Cravioto, J., & Frenk, S. (1954). Studies on malnourished children: 17. Psychological manifestations. Boletin M├ędecina Hospital Infantil (Mexico), 2, 631-641.

Grantham-McGregor, S. M., Powell, C. A., Walker, S. P., & Himes, J. H. (1991). Nutritional supplementation, psychological stimulation and development of stunted children: The Jamaican study. Lancet, 338, 1-5.

Grantham-McGregor, S. M., Stewart, M. E., & Schofield, W. N. (1980). Effect of long-term psychological stimulation on mental development of severely malnourished children. Lancet, 2, 785 - 789.

Hoorweg, J., & Piaget, S. J. (1977). Intellectual abilities and protein-energy malnutrition: Acute malnutrition vs. chronic malnutrition. In J. Brozek (Ed.), Behavioral effects of energy and protein deficits. Washington, DC: U.S. Department of Health, Education, and Welfare.

Keys, L. A., Brozek, J., Henschel, A., Mickelsen, O., & Taylor, H. L. (1959). The biology of human starvation. Minneapolis: University of Minnesota Press.

Klein, R. E., Irwin, J., Engel, P. L., & Yarbrough, C. (1977). Malnutrition and mental development in Guatemala: An applied cross-cultural research study. In N. Warren (Ed.), Advances in cross-cultural psychology. New York: Academic.

McKay, H., Sinisterra, L., McKay, A., Gomez, H., & Lloreda, P. (1978). Improving cognitive ability of poorly nourished children. Science, 200, 270-278.

Monckeberg, F. (1968). Effect of early marasmic malnutrition on subsequent physical and psychological development. In N. S. Scrimshaw (Ed.), Malnutrition, learning and behavior. Cambridge, MA: MIT Press.

Mora, J. O., Herrera, M. G., Sellers, S. G., & Ortiz, N. (1981). Nutrition, social environment and cognitive performance of disadvantaged Colombian children at three years. In Nutrition in health and disease and international development: Symposia from the XII International Congress of Nutrition. New York: Alan R. Liss.

Pollitt, E., & Granoff, D. (1967). Mental and motor development of Peruvian children treated for severe malnutrition. Review of Interamerican Psychology, 1, 93-102.

Scrimshaw, N. S., & Gordon, J. E. (Eds.). (1968). Malnutrition, learning, and behavior. Cambridge, MA: MIT Press.

Sinisterra, L., McKay, A., McKay, H., Gomez, H., & Korgi, J. (1977). Response of malnourished preschool children to multidisciplinary intervention. In J. Brozek (Ed.), Behavioral effects of energy and protein deficits. Washington, DC: U.S. Department of Health, Education, and Welfare.

Stoch, M. B., & Smythe, P. M. (1968). Undernutrition during infancy, and subsequent brain growth and intellectual development. In N. S. Scrimshaw (Ed.), Malnutrition, learning and behavior. Cambridge, MA: MIT Press.

Yatkin, U. S., & McClaren, D. S. (1970). The behavioral development of infants recovering from severe malnutrition. Journal of Mental Deficiency Research, 14, 25-32.