Contents - Previous - Next

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

Household food security, food consumption patterns, and the quality of children's diet in a rural northern Ghana community

M. Armar-Klemesu, T. Rikimaru, D. O. Kennedy, E. Harrison, Y. Kido, and E. E. K. Takyi

Abstract

Twenty households in a rural northern Ghana community were studied to ascertain evidence of seasonality and the relationship between household food-security status and the food and nutrient intakes of preschool children. All food consumed by household members was weighed for three consecutive days, and nutrient intakes were calculated from appropriate food composition tables. Diet quality was assessed by quantitative and descriptive analysis. Mean daily intakes of selected food items reflected an over-dependence on cereals and a minimal consumption of animal products. The consumption of vegetables, especially low during the pre-harvest season, increased during the post-harvest season. Nutrient intakes, which were generally lower than recommended dietary allowances, increased during the post-harvest season, although not significantly for calories, protein, and iron (p <.05), in contrast to vitamins A and C (p < .001). There was a significant association between household food-security status and the intakes of calories (p < .001) and protein (p < .01) but not of micronutrients. Diet quality was adversely influenced by a low intake of micronutrients derived primarily from plant sources as well as by poor dietary habits. Calorie and protein intakes appear to have been more influenced by household food availability and were subject to less seasonal fluctuation, while the reverse held true for the micronutrients. In addition, household food security did not necessarily ensure the quality of children's diets in an area where food consumption patterns are monotonous.

Introduction

It has been asserted that global or national food security does not necessarily ensure household or individual food security, and that 20%-30% of the population of countries where the per capita supply of food is at or above 100% of dietary energy needs may persistently subsist on inadequate diets and be unable to meet their requirements for normal physiological functioning [1], with particularly adverse implications for the nutrition of young children [2]. Food security has been defined as access to food adequate in quantity, quality, and safety to ensure healthy and active lives for all household members [3]. However, much of the earlier work on the effects on consumption of policies to ensure household food security concentrated on calories and, to some extent, protein. This was an understandable result of the former perception initially attributing nutrition problems to a lack of protein and later of calories [4-6]. It is now recognized that an increase in household or individual calorie consumption may not necessarily ensure an adequate intake of other nutrients [7] and that the amount and bioavailability of specific nutrients in children's diets may be at least as important for growth and other functional outcomes as the total amount of available food or energy [8, 9]. In addition, although insufficient food availability may be the direct cause of low food intakes in food-insecure households, the nature of the diet, particularly its quality, may actually be the cause of low food intakes [8].

The Upper East region of Ghana, which lies in the sub-Sahelian savannah zone, is characterized by chronic food shortages as a result of adverse climato-ecological conditions. Malnutrition (< 80% NCHS median weight for age) among preschool children in the region is estimated at over 50%, with severe (i.e., clinical) cases at about 20% [10]. These figures are much higher than Ghana's current national averages of about 30% and 8% respectively [11]. In addition, vitamin A deficiency has been documented as a public health problem in the region [12], and endemic goitre is recognized as such (Assibey-Berko, personal communication). There are indications that anaemia may also be prevalent [10].

Since no comprehensive nutrition studies had been conducted in the region at the time, a full-scale survey involving anthropometric, biochemical, and dietary assessments of the nutrition status of preschool children in three rural communities was conducted by the nutrition unit of the Noguchi Memorial Institute for Medical Research between July 1990 and February 1991. The study was aimed at further clarifying the extent of nutrition problems and, more important, investigating some of the causal factors. This paper presents results of dietary studies carried out in one of the selected communities.

Methods

Study locale and population

Three villages, Tongo, Kongo, and Sumbrungu, 1115 km north, west, and east respectively of Bolgatanga, the regional capital, were selected for the study. A demographic survey of Sumbrungu, chosen for the dietary survey, indicated that preschool children made up about 20% of the total population of a typical village of 1,500 to 2,000 inhabitants. The majority of the adult population had no schooling and were subsistence farmers, working small land holdings surrounding their family homes. Other features of the study area were similar to those described extensively elsewhere [10, 12].

Data collection

Food and dietary Intakes

The dietary survey was conducted during the pre-and post-harvest seasons, early July and late October-November of 1990. After a demographic survey during which houses were numbered, a random selection of alternate households with at least one preschool child resulted in a sample of 20 households and a total of 151 subjects, excluding breast fed infants and children. The weights of all ingredients, cooked food, and food consumed by all household members were recorded by two investigators staying in the selected house for three consecutive days. The daily intake of each item was calculated from its ratio in the cooked meal, and energy and nutrient intakes were either calculated from published food composition tables [13-15] or estimated by proximate analysis at the Institute. The adequacy of intake of selected nutrients was assessed as the ratio of nutrient intake to the recommended dietary allowance, with a ratio of 0.8 as the cut-off point.

Household food security

The degree of adequacy of dietary energy intake (in comparison with appropriate norms) for the health, growth, and activity of all individual household members is one measure of household food security [3]. To assess adequacy, the FAO/WHO/UNU recommendations [16] were used. Requirement figures that were not provided were computed employing the appropriate basal metabolic rate (BMR) multiplied by the appropriate factor for sex, age, and level of activity [16]. The household dietary energy adequacy ratio (HDEAR) was computed from the mean total household intake and the mean total household requirement. A household was considered food-secure if its HDEAR was 0.8 or above and food-insecure if it was less than 0.8.

Data analysis

Analysis of variance was used to determine seasonal variations in nutrient intakes as well as differences in intakes according to household food security. The association between a household's food-security status and the intake of selected nutrients by preschool children was investigated by linear regression analysis.

Results and discussion

Some aspects of food habits

Guinea corn, millet, and groundnuts were the main staples and energy sources of the area. Shea butter and, to some extent, groundnuts were the main and possibly sole sources of oil, although groundnuts were normally ground raw and used to prepare soups and sauces, rather than having the oil extracted. Red palm oil, which is consumed abundantly in southern Ghana, was conspicuously absent. Children had no special diets and were weaned directly to the adult diet.

The cereals were stone-ground into fine meal and cooked into a stiff porridge called tuo zaafi in the local dialect. This was eaten with sauce or soup prepared from fresh or dried vegetables if available and, depending on the season, ground raw groundnuts and millet meal. Even though almost every household raises animals such as cows, goats, sheep, and fowls, these were seldom eaten. Very few households consumed any animal products at all, and what they did eat was in the form of small quantities of dried anchovies or, on rare occasions, dry-smoked guinea fowl. Cooking was normally done once a day, in the evening. The two main meals were breakfast and supper, with breakfast more often than not consisting of leftovers from the previous evening's meal.

Children were often privileged to have some lunch of ground raw cereals soaked in water with a little shea butter, leftovers from breakfast, or fresh raw groundnuts. There was virtually no consumption of fruit.

TABLE 1. Summary of household food-security categories

TABLE 2. Mean daily intakes of selected food items (grams) by preschool children, classified by household food-security group

FS = food secure;
SFS = seasonally food secure:
CFI = chronically food-insecure.

Household food-security status

Three household food-security types were identified (table 1): Eight of the households (40%) were food-secure (FS) during both seasons, satisfying more than 90% of their energy requirements on average. A further six (30%) were seasonally food-secure (SFS) during either the pre-or post-harvest season, satisfying just about 80% of their energy requirements. The remaining six (30%) were chronically food-insecure (CFI) during both seasons, satisfying a little more than 60% of their energy requirements.

The proportion of households in the last group is in accordance with the assertion that 30% of most countries are chronically food-insecure [1]. Because of the small sample sizes, no attempt was made to establish statistical differences in the HDEARs, but the figures show a clear and definite trend that is masked when only averages are looked at.

Household food security and preschool child nutrition

Food consumption patterns

There was a general tendency toward a higher mean daily intake of staples by preschool children from the FS households (table 2). For households of all food-security types, the intake of rice and groundnuts, probably freshly harvested, increased in the post-harvest season, while the relative proportion of millet decreased. The consumption of guinea corn appeared to be stable across the seasons. The pre-harvest intake of meat and fish was highest for the FS group, followed in order by the CFI and SFS groups; no consistent trends were observed in the post-harvest season, although intakes were generally lower for the FS group. There was a definite trend toward higher intakes of fresh green vegetables in the post-harvest season and a corresponding decrease in the intake of dried vegetables. No consistent group trends were observed.

The data in table 3, showing percentages of children consuming the selected food items at least once during the three-day study period, confirm the observation of seasonal substitution among staples, with a higher proportion of children consuming higher quantities of rice and groundnuts in the post-harvest season. Similar consumption patterns have been reported in the Philippines [17], where consumers reacted to the non-availability of the main staples of rice and corn (because of price increases) by substituting other energy-dense staples, and in Chad [18], where millet, the main cereal available at the beginning of the harvest period, was substituted for the maize and wheat that were eaten at other seasons. In the community we studied, rice and groundnuts were readily available in the immediate post-harvest period, while millet was more available during the pre-harvest season as a result of an early millet crop, which was harvested before the main cropping season.

Fish was eaten by more children than meat although at a lower per capita consumption. Conversely, however, only a small proportion of the children ate any meat during the three-day study periods. Pre-harvest consumption of dried vegetables was universal, in contrast to the post-harvest season, when the proportion of children eating fresh vegetables was higher. Generally, the consumption of dried vegetables (pepper, okra, leaves) was more consistent all year round.

TABLE 3. Percentages of children consuming selected food items at least once a day

Nutrient intake

Table 4 summarizes the adequacy in comparison with recommended dietary allowances of the intakes of selected nutrients by preschool children from the three food-security types of households in the two seasons, as well as the proportion of children failing to satisfy their requirements. The nutrient intakes generally increased in the post-harvest season, and this was clearly reflected in a corresponding decrease in the proportion of children failing to satisfy their requirements. There was a general tendency toward inadequate intakes except for protein, with the majority of children unable to meet their requirements for most nutrients.

Analysis of variance shows significantly differences between the groups for the intakes of calories (p > .001) and protein (p < .01) but none for iron or vitamin A and C ( p < .05). On the other hand, the intakes of vitamins A and C increased significantly in the post-harvest season (p < .001), while those for calories and protein did not. The lack of a significant seasonal effect on the intake of calories and protein is not surprising, as consumption patterns generally reflect consumers' apparent awareness of fluctuations in the calorie content of their diet and their attempt to maintain adequate levels of calories (as far as possible) by substituting among the staples [17].

To analyse further the relationship between household food-security status and preschool-child nutrient intake, linear regression of the nutrient-adequacy ratios for the preschool children on the HDEARs of the corresponding households was carried out (table 5). Household food-security status positively influenced preschool children's intake of calories and protein but not of micronutrients. Indeed, a study in Kenya [7] has shown that, as children's food and energy intake increased, their consumption of vitamin A, for instance, actually decreased.

TABLE 4. Adequacy levels of calories, protein, iron, and vitamins A and C, and percentages of children with intakes below 80% of requirements

Significant group differences: *p < .01; **p < .001.
Significant seasonal trends: *p < .001.
NS = no significant group or seasonal trends (p > 05).

TABLE 5. Correlation between adequacy ratios of nutrients consumed by preschool children and household dietary energy adequacy ratios

*p < .01.
**p < .001. NS = not significant.

Aspects of diet quality

Having established that adequate diet quality is a general problem in the community, we merged the groups in investigating aspects of the diet that tend to influence its quality, since micronutrient intake is more of a measure of diet quality than are calories.

Despite increasing recognition of the importance of the quality of children's diet, there is currently no simple way of assessing it, because dietary quality remains vaguely defined and not easily measurable. Among proposed guidelines employing specific characteristics to define or evaluate dietary quality are macronutrient and micronutrient density, digestibility and bioavailability, the proportion of calories and protein from animal sources, dietary diversity, absence of anti-nutritional factors, and appropriate organoleptic characteristics for age. To assess the quality of preschool children's diets, we chose diversity and the proportion of macronutrients and micronutrients from plant and animal sources as a proxy for bioavailability.

The diets of the children in the community studied have very little diversity, the food eaten being predominantly cereals and legumes with little or no intake at all of animal products. Fish appears to be more widely eaten than meat, but the quantities are really negligible. It is not surprising that, as is usually observed in predominantly cereal-consuming populations, although the intakes of calories and protein are more or less acceptable, those of the micronutrients are woefully inadequate, especially during the pre-harvest season. Further, even though protein requirements are apparently being satisfied, this assessment is made using food composition tables that contain values from the chemical analysis of foods but make no allowance for their biological use [19] and is questionable since cereals are the main and probably sole source of protein for most of these children and the qualitative limitations of cereal proteins are well known.

Table 6 shows the predominance of plant sources for the nutrients selected for the attempt to assess bioavailability on the basis of the proportion of intake from animals and plants. Given that cereals form the basis of the children's diets, several anti-nutritional factors could affect the availability and utilization of the already inadequate micronutrient intakes. As fibre and phytate are particularly high in this diet consisting predominantly of millet and guinea corn, iron absorption is likely to be inhibited. The extremely low intake of vitamin C, which enhances iron absorption and utilization, does not help either. Moreover, groundnuts, which might improve protein quality if appropriately processed, are more often than not consumed raw, and their anti-nutritive factors further compound an already adverse situation.

The case with vitamin A is very much the same. About 99% of its intake is from plant sources in the form of pro-vitamin A or beta-carotene, derived mainly from dark green leafy and other vegetables. Our data show that vegetables (including leafy vegetables) are more consistently consumed dried than fresh, and drying, regardless of the method used, significantly reduces the carotene content in vegetables [20]. Sun-drying, the method employed in this community, specifically destroys up to 60% of the carotenoid content of green leaves [21]. Therefore, vitamin A consumption is likely to have been overestimated. In addition, fat, which improves the utilization of retinol (pre-formed vitamin A, found solely in animal foods) and beta-carotene [22], was lacking in the diet.

The question then is what are the implications of diet quality for nutritional outcomes? Further analysis of our data to relate child anthropometric indices to household food security status, and hence household and individual dietary energy adequacy, not presented here, did not indicate a significant association. However, other studies [9, 18] suggest that vitamin A appears to be a better predictor for child growth than either dietary energy or protein quantity, possibly because diets high in vitamin A reflect greater diversity and are also excellent sources of other nutrients and therefore are of better quality [9].

TABLE 6. Percentage intake of selected nutrients from animal and plant sources

Conclusion

Generally, for poor rural households, food availability translates into increased intakes of high-calorie staple foods. Thus, such households often cannot obtain the necessary dietary variety to meet their nutrient needs, a primary cause of micronutrient deficiency. Our findings from this study of a rural community in one of the most deprived regions of Ghana, although based on a rather small sample of households, clearly suggest that the myriad nutritional problems associated with the area are a result of poor diet quality coupled with inappropriate dietary habits. Efforts to alleviate malnutrition through improved programmes for food security should also be aimed particularly at ensuring diet quality in conjunction with promoting appropriate dietary behaviour, and national policies directed only at promoting the use of local cereal-legume mixtures as weaning diets for young children-as in the Weaning Food Programme in Ghana should be reviewed.

Acknowledgements

The authors acknowledge with gratitude the support for the conduct of this study from the Japan International Cooperation Agency and the assistance of Mr. James Asedem and his staff of the Upper East Regional Nutrition Office during data collection. We also appreciate the comments and suggestions from Prof. F. K. Nkrumah, Director of the Noguchi Memorial Institute for Medical Research, and Mr. George Mensah's assistance in data analysis.

References

1. FAO. Food and nutrition: creating a well-fed world. Rome: FAO, 1992.

2. Teklu T. von Braun J. Zaki E. Drought and famine relationships in Sudan: policy implications. Food Nutr Bull 1992;14(2):153-5.

3. Gillespie SR, Mason JB. Nutrition-relevant actions: some experiences from the eighties and lessons for the nineties. ACC/SCN State-of-the-Art Series, Nutrition Policy Discussion Paper no. 10. Geneva: ACC/SCN, 1 991.

4. Williams CD. A nutritional disease of childhood associated with a maize diet. Arch Dis Child 1933;8:42333.

5. McLaren DS. The great protein fiasco. Lancet 1974: 2:93-6.

6. Payne PR. Safe protein-calorie ratios in diets: the relative importance of protein and energy intake as causal factors in malnutrition. Am J Clin Nutr 1975;28:281-6.

7. Kennedy E. Oninag'o R. Household and preschooler vitamin A consumption in southwestern Kenya. J Nutr 1993;123(5):841 -6.

8. Brown KH. The importance of dietary quality versus quantity for weanlings in less developed countries: a framework for discussion. Food Nutr Bull 1991; 13(2): 86-94.

9. Allen LH, Black AK. Backstrand JR et al. An analytical approach for exploring the importance of dietary quality versus quantity in the growth of Mexican children. Food Nutr Bull 1991;13(2):95-104.

10. Noguchi Memorial Institute for Medical Research. Nutritional status of pre-school children in the Upper East Region. NMIR Nutrition Unit Research Report. Legon, Ghana: NMIR, ]992.

11. UNICEF. Situation of women and children in Ghana. Accra: UNICEF, 1990.

12. Ghana Vitamin A Supplementation Trial Study Team. Vitamin A supplementation in northern Ghana: effects on clinic attendances, hospital admissions and child mortality. Lancet 1993;342:7-12.

13. Eyeson KK, Ankrah EK. Composition of foods commonly used in Ghana. Accra: Food Research Institute, 1975.

14. Science and Technology Agency. Japanese food composition table. 4th ed. Tokyo: Finance Ministry Publishing Office, 1982.

15. Leung WTW, Busson F. Jardin C. Food composition tables for use in Africa. Bethesda, Md, USA: Department of Health, Education and Welfare, 1968.

16. WHO/FAO/UNU. Energy and protein requirements. WHO Technical Report Series. Geneva: WHO, 1985.

17. Bouis HE. Dietary patterns, income and food prices: an analysis of micronutrient intakes of Philippine farm households. SCN News 1991;no.7:23-5.

18. Bégin F. Alladoumgué M, Nandjingar K, Delisle H. Household dietary adequacy and individual nutrition status: relationships and seasonal effects in a Sahelian community in Chad. Food Nutr Bull 1992;14(4):304-13.

19. Ferrando R. From analysis to reality: bioavailability in nutrition and toxicology -a misunderstood concept. World Rev Nutr Diet 1987;53:26-68.

20. Renquist VH, DeVreeze AC, Evenhuis B. The effect of traditional cooking methods on carotene content in tropical leafy vegetables. Indian J Nutr Diet 1978;15: 154-8.

21. Butao RP. A study to determine the carotene content of local used vegetables, both wild and domesticated, and the effect of traditional sun-drying. Master's thesis. Lilongwe, Malawi: Bunda College of Agriculture, University of Malawi. 1981..

22. Dimitrov NV, Meyer C, Ullrey DE, Chenoweth W. Michelakis A. Bioavailability of beta carotene in humans. Am J Clin Nutr 1988;48:298-304.

Contents - Previous - Next