Contents - Previous - Next

This is the old United Nations University website. Visit the new site at http://unu.edu

Anthropometry

Table 13 presents the Z-score values (`x ± SEM) for length for age, weight for age, and weight for length for 19 children less than 36 months of age followed longitudinally for a period of 24 months. Figures 2 and 3 present the same information categorized for small children (less than 12 months at the initiation of the 24-month follow-up period) and older children (more than 12 months at the initiation of that period).

No main effects were detected in the ANCOVAs for years 1 and 2 (P>0.05) or for the presence or absence of rains (P>0.05). However, there was an effect of harvest (P<0.05), The analyses of interactions in ANCOVAs showed that the effect of rain was only marginally different across years (p<0.01) that the interaction of rain and harvest was also significant (p<0.05). and that the interactions were not the same across years (p<0.01)

TABLE 13. Z-score values for length for age, weight for age, and weight for length of children less than 36 months of age 1977-1979)

A pattern similar to that described in the ANCOVAs for length for age was seen in Z-score values of weight for age. There were no differences across years in Z-score values of weight for age (P>0.05). Therefore, no main effects in weight for age for harvest or rains (P>0.05) were detected by the ANCOVAs. The analyses of interactions showed that the effect of rains was different across years (p<0.01) the effect of harvest was consistent across years (P<0. 05), and the interactions of rain and harvest across years are different (P<0.01).

The ANCOVAs for weight for length, for 19 children followed for eight different periods over two years, showed no main effects of the variables being studied (P>0.05). However, the interaction of year and rain only approached significance (P<0.10). All other interactions for Z-score values of weight for length, as reflected in table 14, were consistent across seasons and/or years (P>0.05).

Table 15 summarizes the results of ANCOVAs for Z-score values of length for age, weight for age, and weight for length for the same 19 children as in table 14 divided into the following two age groups according to their age in the first trimester of the two-year follow-up period: less than 12 and 12 and more months. A pattern similar to that described for the entire cohort of children (see table 14) emerged for main effects in all anthropometric measurements when the information was divided by age groups. An exception was seen in length for age: rain was slightly significant (P<0. 10). Furthermore, the effect of harvest on the growth patterns of the different age groups was less important (P<0.10) than the effect of harvest identified for the entire cohort of children.

Values of Z-score (` x ± SEM) for length for age, weight for length and weight for age of 19 children less than 36 months

TABLE 14. Summary of analyses of co-variance (ANCOVAs), using age as a co-variable, in Z-score values of length for age, weight for age and weight for length of 19 children (1977-1979)

a. P<0.05.
b. P<0.01.
c. P<0 001.

Mean values of Z-score for length for age, weight for length and weight for age in ten children less than 12 months and nine children older than 12 months

TABLE 15. Summary of ANCOVAs, using age as a co-variable, in Z-score values of length for age, weight for age and weight for length of two groups of children, <12 and 12 and more months (1977-1979)

a. P<0.05.
b. P<0.01.

Discussion

Important and consistent seasonal fluctuations in energy and protein intakes for families and mothers were not identified in the analyses of food consumption data for families residing in Finca Mocá, Western Guatemala. Energy and protein intakes were also analysed as percentages of family requirement figures. A similar pattern of main effects and interactions emerged for the dietary information either analysed as a percentage of requirement figures or expressed in absolute quantities (grams of protein and energy). The variations detected were related to an ageing factor rather than to seasons.

The pattern of families' consumption data for the five most frequently consumed food items resembles the pattern for total family energy and protein intakes. That is, in periods when total energy and protein intakes peaked or fell, the same pattern was observed for the consumption of the most frequently consumed foods.

Children's consumption of foods did not follow a marked seasonal pattern: only a higher consumption of maize, among the five most frequently consumed foods, was observed in non-harvest as opposed to harvest months. The increased consumption of maize in children during the non harvest periods, as opposed to harvest months, did not follow the same pattern observed for the families as a whole. Families consumed more maize during the harvest months. Such a pattern may be the result of a lack of mothers' time during harvest periods: they may not be able to promote appropriate feeding practices in spite of the fact that more foods are available for consumption at the family level. Maternal care and time is also a factor of paramount importance in explaining the higher consumption of maize by children during the non-harvest months.

The lack of influence of seasons (harvest, non-harvest, rainy, and non-rainy periods) on food availability at the household level and on the intakes of female adults is consistent across years 1 and 2 and may be explained as follows. During the harvest periods, most family members, including children above six years of age, participated effectively in economic activities. Physical activity, particularly in adults, increased. Total family income also increased substantially during the same period. One would then expect, during the harvest season, important increments in food availability within households and substantially higher energy and protein intakes for mothers and other adults. However, those phenomena were not detected in the data from Finca Mocá. Further analyses of these data sets may be necessary to explore whether such inelasticity in intakes is likely to be reflected in substantial changes in absolute weights of adults, as is the case in African countries [6, 7].

It is important to emphasize that the study population and its environment had certain features that made it difficult to identify marked detrimental seasonal effects on health and nutrition. For example, the study area did not exhibit dramatic fluctuations in rainfall (prolonged dry periods). Furthermore, the heads of families in coffee plantations had a permanent job all year round. Access to the basic staple, corn, and to other foods did not fluctuate greatly throughout the year. This finding differed from food-access patterns for areas and communities in Africa, where report! have shown that seasonality negatively affected the health and nutritional status of the population. Subsistence farming areas of Guatemala or other Central American countries showed a different pattern to that of the study population. The seasonality questions addressed in the study environment of coffee plantations in Guatemala are: (a) Does a seasonal increase in income lead to a substantial increase in food consumption? (b) Does a seasonal increase in morbidity lead to further deterioration of children's growth (c) How is the combination of these events reflected in children's growth?

The following factors may be important determinants of decisions made at the household level with respect to the preparation and consumption of more food during the periods when family income is increased. The harvest period is characterized by excessive additional demands on women's time. In some plantations women's days may start as early as 2 to 3 a.m., when they take the corn cooked the night before to the mill. After queueing at the mill for an hour, they return home to prepare tortillas for breakfast, make lunch for all family members, and get the children ready-the small as well as the older ones-to go with them to the field site assigned for harvesting coffee beans.

While they work, mothers carry small children (less than 24 months) on their backs for most of the day. Older siblings, besides being engaged in the harvest, may help their mothers by taking care of younger children and carrying out the lunch which is consumed by all family members during a work-break in the field. Some sites assigned for picking coffee may be located an hour's walking distance from the women's households. The family usually returns to the administration building of the plantation at around 4 p.m., carrying with them from 50 to 100 pounds of coffee. From August to October this return trip is usually made in heavy rain. Coffee is weighed at the administration building and queueing there may take an hour. After the coffee is weighed, women return home to prepare tortillas and other foods for dinner, as well as the corn to be consumed the next day. This work pattern during harvest periods may not provide women with enough time, energy, and motivation to prepare more of the same staples and/or other foods.

Other reasons for the lack of an important and consistent seasonal effect on family and mothers' intakes have been given in the introductory section of this article, and have also been discussed, in more detail, by Valverde [20]. The first reason is that the price of corn on Finca Mocá is fixed throughout the year. Most corn needs for the entire family are obtained through this subsidized system, and corn consumption accounts for 60 per cent of the total daily family energy intake. The second reason is that the plantation administration provides loans to their labourers during any period of the year, and these loans are discounted in instalments on pay-days (every two weeks). Most of the debts with the plantation are paid off during the harvest season when total family income is increased. Similarly, there is a system of indebtedness with the local shops.

Even though family income is increased during the harvest period, there is another factor in addition to loan repayments that may preclude the possibility of higher family energy intakes. Investments in household goods are made during the harvest period, when merchants come to the plantation to offer a wide variety of products. However, these factors may also act as buffers, smoothing the potential detrimental effects in intakes of seasonal periods when total family income is substantially lower.

The results presented on energy and protein intakes, morbidity, and growth patterns of children studied longitudinally also deserve further discussion. As reported by Delgado et al. [2], there is a clear seasonal morbidity pattern which is, for most signs and symptoms analysed, dictated by the initiation of the rainy period (nasal secretion, cough, diarrhoea) and the presence of rains (stools with blood and mucus). While statistically significant peaks for morbidity events are identified in the rainy periods, it is important to point out that the prevalence of these signs and symptoms is high throughout the year.

During the non-harvest and rainy period, when most morbidity events reach a peak, mothers may have more time for taking their children to the clinic or for other activities that will improve their children's health. During the harvest season the health clinic in Finca Mocá opened late in the afternoons so that mothers could take their children to the clinic upon returning from picking up and weighing the coffee harvested. Delgado et al. [2] have reported the positive nutritional effects that took place in children with diarrhoea when they sought medical care from a clinic.

Although no consistent pattern of seasonal effects was identified, there were a few variations in energy and protein intakes and in the nutritional status of children that were related more to the interaction of years and ageing factors than to season.

For addressing issues such as the health and nutritional impacts of social, economic, and biological factors that interact in communities in developing nations, longitudinal analyses are more powerful tools than cross-sectional data-analysis approaches. However, in interpreting longitudinal data careful attention should be given to the ageing factor, as this may either obscure effects or, conversely, show effects that are, in fact, analytical artefacts. Thus, in analysing longitudinal children's growth and food intake data in this report, ANCOVAs were estimated by age groups.

continue

Contents - Previous - Next