|Community-Based Longitudinal Nutrition and Health Studies : Classical Examples from Guatemala, Haiti and Mexico (International Nutrition Foundation for Developing Countries - INFDC, 1995, 184 pages)|
There is no substitute for the longitudinal community based study as an approach to understanding both the multiple causative factors and the functional consequences of disease in underprivileged populations. When the classical studies described in this volume were planned and conducted, there was only limited precedent for them. John Gordon, Head of the Department of Epidemiology at the Harvard School of Public Health (HSPH), had carried out one of the first, 1934-1938 in Rumania and recognized the power of this approach. In 1953 he had initiated the 7-year Khanna Population and Health Study in Northern India (Wyon and Gordon, 1971 ) that provided unexpected and valuable insights into the health-related behaviors and their consequences in representative communities.
When I came to the Harvard School of Public Health in 1968, John Gordon was teaching a small seminar course designed to teach the planning and implementation of longitudinal epidemiological field studies. We both became enthusiastic about the possibility of a field study of the interactions between nutrition and infection in Guatemalan communities. Planning for such a study began at once and Gordon came to Guatemala the following summer to assist in the preliminary reconnaissance and completion of the experimental design. The actual study, carried out from 1959-1964 and summarized in Chapter 1, involved three similar villages, one with supplementary feeding of mothers and young children, one with intensive medical care, and the third with no health intervention. It not only demonstrated a reduction in the burden of diarrhea! and respiratory disease with nutritional supplementation but also revealed the limited value of curative medicine in the absence of public health measures to improve nutrition and control infection.
Joaquin Cravioto, who had carried out the original pioneering cross sectional studies in Mexico demonstrating a relationship between stunting and intrasensory perception in Mexico (Cravioto and DeLicardie, 1968), saw an opportunity to use the feeding village, Santa Catarina Barahona of the Guatemalan three-village study, to look for an effect of the supplementary food on intersensory perception and selected another nearby village as control. Unfortunately, the use of the village for this purpose in the final two years of the five-year study illustrates the risk to such studies of confounding factors.
The obvious deficiency in the design of the three-village study was the lack of a community in which young children received both supplementary food and enhanced medical care. Funds were not available for this because the funding agency considered that a favorable result from both interventions was obvious. Subsequently Carl Taylor, who was also influenced by John Gordon, organized the Narangwal study in Northern India (Kielmann et al., 1978; Taylor et al., 1 978a, 1978b) which used the original four-village design planned for Guatemala and added additional elements and villages. Its findings confirmed those of the three village study described in Chapter 1 and will be of interest to readers of this volume.
After the three-village study was finished, Leonardo Mata, then Chief of the Division of Microbiology of INCAP, saw an opportunity to use the rapport with the community and the clinic facilities already constructed in Santa María Cauqué to initiate a prospective investigation of the causes and consequences of diarrhea and other infectious diseases in infants and preschool children. This study is reported in Chapter 2. It provides the most detailed information of the ecology of infectious disease in a community ever obtained and stands as a model of intensive longitudinal health-related research.
Adolfo Chávez, who spent time at INCAP from 195 5-195 8, returned to Mexico stimulated by the experience and initiated an extraordinary 14 year study in the village of Tezonteopan, the longitudinal investigation. Because children within the same village were randomly assigned to supplementary feeding or control groups, this design assured better control of nonnutritional factors than possible with separate experimental and control villages. It became evident early that the supplemented children were more active, had fewer illnesses, and performed better on cognitive tests. In order to document better mother child interactions and breast-feeding practices his long term assistant, Celia Martínez spent long hours, day and night, observing individual mother-child dyads. These same investigators have been able to continue to observe these children for 18 years and the results, described in Chapter 3, confirm the long term benefits of the nutritional supplementation over and above those of stimulation and enhanced medical care received by all children in the village.
In 1969 INCAP embarked on an ambitious 8-year "four village" study in mestizo ("Ladino") communities in the Pacific lowlands of Guatemala. Known as the "Oriente" study, it compared the health and cognitive performance of children in two villages who received complementary feeding of atole, a protein-rich beverage for the first three years after birth, compared with those in two similar villages who received fresco, a low-calorie, protein-free beverage to balance the stimulation effect that is inevitable with any feeding program. The beverages were also offered to their mothers during pregnancy. Those in the atole village grew more and performed better on appropriate cognitive tests. Two of the original investigators organized a follow-up study 14 years later when the participating children were adolescents and young adults. This follow-up study is of great public health significance because it demonstrates that years later, with no intervening intervention, the differences in cognitive and physical performance favor the atole group, are greater than at the close of the original study, and statistically more robust. It is a striking demonstration of the value of early nutritional supplementation and stimulation of underprivileged infants and preschool children. Chapter 4 summarizes both the original and follow-up studies and their implications.
The final chapter deals with observations in the Deschapelle valley of Haiti by the same principal investigators extending over a period of 25 years. When Warren and Gretchen Berggren first came from the Harvard School of Public Health to the Albert Schweitzer Hospital in 1967, they found that the pediatric services of the hospital were swamped with cases of severe malnutrition and infections. Their approach was to develop preventive programs. The impact of the village "mothercraft" centers and the administration of tetanus toxoid to pregnant women had a dramatic impact on hospital case load and case fatality. The community programs that they initiated have been monitored ever since and Chapter 5 describes their evolution and effectiveness.
The original motivation for this book was my experience in attempting to familiarize students in the Harvard School of Public Health with the techniques and potentials of long-term prospective community-based nutrition and health studies. Since each was described either in an entire book or monograph or in an extensive series of papers, students could not be expected to become familiar in depth with more than one study. Moreover, few institutions have all of these publications available. It was recognized that there was a need to make condensed versions of these studies widely and inexpensively accessible to students, teachers, and policy makers that would highlight the lessons learned and contribute to the design and interpretation of future community-based longitudinal studies. This volume is designed to meet this need. Analysis of the strengths and weaknesses of these five studies will provide a critical introduction to the prospective community-based approach to the evaluation of potential nutrition and health interventions.
If time permitted analysis of the studies summarized in this volume could usefully be supplemented by critical reviews of the Khanna (Wyon and Gordon, 1971), Narangwal (Kielmann et al., 1978; Taylor et al., 1978a, 1978b) and Poshak, (Gopaldas et al., 19751, 1975b) studies in India; the Bogotá (More et al., 1981; Super et al., 1990) and Cali (Sinisterra et al., 1977; McKay et al., 1978) studies in Colombia, and the Bacon Chow (Adair and Pollitt, 1985) study in Taiwan. Unfortunately, the original reports of most of these studies are widely scattered and inaccessible. If this volume proves useful, support will be sought to prepare a companion volume of studies on other continents.
Nevin S. Scrimshaw