Contents - Previous - Next

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



Adequacy of child dietary intake relative to that of other family members


Rafiqul Huda Chaudhury

 

Introduction

The overall nutritional status of people in Bangladesh in general and children in particular is bleak. According to the Nutrition Survey of 1981-1982, 57% of pre-school children in rural areas were chronically undernourished (i.e., stunted) and 21% suffered from acute malnutrition.

A recent study based on a rural area of Bangladesh showed that calorie intake of a sizeable proportion - 40% to 90% - of pre-school children falls short of estimated energy requirement [1]. The infant mortality rate (deaths in the first year per 1,000 live births) was 118 in 1983 and the childhood mortality rate (deaths in the age group one to four years old per 1,000 population) was 23.8 [2]. Those who survive are likely to suffer from chronic malnutrition and frequent infections; the synergistic interaction of the two is well recognized [3]. Various reasons could be adduced to explain this poor nutritional status of children. Of these, low household income and related symptoms of poverty, such as low level of parental education and high fertility, are of significance. Unequal distribution of food within the household, favouring adults over children, is also a contributory factor.

It is often maintained that children, particularly female children, in patriarchal peasant societies like Bangladesh occupy lower status than adult members of the household, particularly the patriarch [4, 5]. In this system, age gradation is one of the key variables determining social and economic position [5]. Advancing age carries with it increased control over family resources and other privileges.

Also, given the subsistence nature of the economy, family labour is one of the most important determinants of the economic condition of a household. To ensure economic viability, the health and nutritional status of the adult members receive priority over those of other members, particularly children.

In these societies, where family resources are meagre and immediate economic viability of a household depends on, among other things, the number of able-bodied adults, marked skewing in the distribution of food in favour of adults over children is expected to be a factor contributing to high infant mortality. A partial attempt is made here to examine how children fare in meeting their nutrient needs compared to other family members, particularly adults, by employing data from a rural area of Bangladesh. This in essence examines the hypothesis of non-egalitarian distribution of food within the household.

 

Data and methods

The data employed in this study were collected from 572 members of 108 households in the village of Muyiarchar, approximately 350 kilometres north-east of Dhaka. The sample included 50% of the village households, selected from various socio-economic groups classified on the basis of land holdings and income. There were 99 household members in the age group from birth to four years old. (For details, see Chaudhury [1].)

Dependent Variables

In assessing the intake of children relative to that of other members of the family, care must be taken that the dependent variables are chosen in such a way as to control for intra-individual variations in energy requirement. The needs of children and those of adults were therefore calculated separately.

The nutrient needs of children from birth to four years old were determined by two methods, both based on recommendations of the 1973 FAO/WHO report Energy and Protein Requirements [6]. Method A uses age- and sex-specific recommendations.

Method B is based on the age, sex-, and average-weight-specific recommendations of the same report, using the mean weights reported from another study in a rural Bangladesh population [7].

The calorie needs of persons five years old and older were estimated on the basis of actual information on rest/sleeping and physical activity (for details, see Chaudhury [8]). The safe allowance for protein for those five years old and older was calculated by multiplying the average safe protein allowance per kilogram of body weight in a particular age-sex group by the average weight in kilograms of a person in the corresponding age-sex group. An adjustment was made to allow sufficient energy protein for pregnancy and lactation. All the protein estimates were corrected for 70% use [8].

Two measurements of intra-family distribution of food were used. The mean nutrient adequacy ratio is the ratio of individual nutrient (calorie-protein) intake divided by the recommended daily allowance of calories and protein respectively and multiplied by 100. The observed-expected ratio is a measurement of whether a person's consumption is more or less than or the equivalent of what would be expected if available nutrients were distributed according to relative need within the household. To find the latter, we first calculated the percentage distribution of calorie-protein need for each member of the household and applied it to the total observed intake of the household to obtain the expected consumption for each member if foods were distributed wholly according to estimated need. Finally, we obtained the ratio of observed to expected consumption for each member of the household based on estimated requirements. This provides a measurement of the appropriateness of the intra-family distribution of food.

 

Results

Nutrient adequacy ratio of children relative to other members of the family

The first test of how children fare in meeting nutrient needs compared to other members of the household is shown in table 1. The overall calorie intake of children in the study village relative to their energy needs was low if adequacy is calculated by method A. According to this calculation, the intake of a child from birth to four years old fell 33% short of calorie requirement on average. This situation changes when energy needs are calculated by method B. which shows that such a child consumed on average 12% more calories than required. Data in table 1 indicate a deterioration in calorie adequacy after the first year of life, with improvement after 48 months. It may be observed further that when calorie needs are measured according to age-specific requirements (method A), the intake of pre-school children fell far short.

TABLE 1. Mean nutrient (calorie and protein) adequacy ratios, by age group

Age

Calorie adequacy

Protein adequacy

(years) Method Aa Method Bb Method Aa Method Bb
0-0.99 93.51 150.98 176.41 317.53
1-2.99 57.50 115.03 175.77 299.96
3-4.99 66.67 101.05 154.45 200.47
0-4.99 67.00 112.00 165.00 250.00
 

Method Cc

Method Dd

5-9.99

113.72

112.84

111.87

110.46

115.83

134.51

178.50

140.95

129.13

143.69

150.87

156.20

10-14.99
15-28.99
29-44.99
45-64.99
65
Average 105.00 113.00 165.00 196.00

Ratios are intakes divided by needs multiplied by 100.

a. Needs based on 1973 FAD/WHO age- and sex-specific recommendations |6].
b. Needs based on 1973 FAD/WHO age-, sex-, and weight-specific recommendations [6].
c. Calorie needs calculated on the basis of average weight and actual individual activity level 18]
d. Safe allowance for protein calculated by multiplying average safe protein allowance per kilogram of body weight in particular sex and age group by average weight of persons in that group [8].

The intake of other household members, however, was sufficient. When calorie requirement is estimated on the basis of actual weight (method B), this situation changes, and the intake of children was adequate, and also their calorie adequacy ratio (intake relative to need), with the exception of the oldest (i.e., those 3-4.99 years old), exceeded the ratios of many adolescents and adults. The calorie adequacy ratio of the oldest children is found to be the lowest. This may be attributed to the minor role of breast-feeding in providing the calorie needs of growing children in the absence of sufficient intake of solid food. Children meet their calorie requirement by age five, however, because by that age they may well compete with other members of the family for food, and children in Bangladesh participate in economic activities as early as age six or seven [9]. Such economic responsibility may ensure higher food intake for these children.

It may also be observed that the calorie adequacy ratio of the oldest member of the family (65 years old or older) was one of the highest. This preferential treatment in terms of food distribution may reflect one aspect of Oriental culture, which emphasizes great care of and attention to elders.

Protein intake was adequate in children, adolescents, and adults of every age group. The protein adequacy ratio for pre-school children was higher than that of other household members, including the adults, particularly when assessed by method B.

These relationships between nutrient adequacy and age remain unchanged even when the measurement of intra-family distribution of food is changed from nutrient adequacy ratio to observed:expected ratio (see FIG. 1. Observed:expected ratio of calorie adequacy for household members, by age and FIG. 2. Observed:expected ratio of protein adequacy for household members, by age).

These findings do not support the hypothesis that children are discriminated against compared with adults in the intra-family distribution of food, when allowance is made for the differential needs of the two groups. The position of children relative to that of adults varies depending on how the nutrient need is assessed.

Sex preference in the intra-family distribution of food

In traditional subsistence agrarian societies, particularly in patriarchal societies such as Bangladesh, there is a definite preference for sons over daughters [10]. Sons are considered essential for maximizing economic and non-economic utilities of the households. They contribute to the family's resources by working on the family farm and eventually assume responsibility for the household and the farm. They are also expected to take care of their parents in old age, as daughters usually move away when they marry and are less likely to provide financial support.

This preference for sons over daughters is considered one of the major reasons for explaining higher infant mortality in girls than boys. In rural Bangladesh, female mortality exceeds male mortality by as much as 50% between the ages of one and four years [11]. This could be due to the unequal distribution of food in households that exercise such sex preference.

While examining this issue, we must control for different nutrient needs of boys and girls. The average weight of a boy in Bangladesh exceeds that of a girl. This may itself lead to higher consumption of food by boys to meet their greater nutrient needs. Our measurement of intra-family distribution of food, that is, mean nutrient intake, allows for this differential need.

TABLE 2. Unadjusted and adjusted deviations from mean calorie adequacy ratios of pre-school children, by age and sex

Age
(years)
and sex
No.

Method Aa

Method Bb

Mean

Deviation from mean

F
value
d
Mean

Deviation from mean

F
value
d
Unadjusted Adjustedc Unadjusted Adjustedc
0-4.99   67.05     5.08 112.41     3.65
male 54   -3.67 -2.84 (.02)   -4.80 -2.95 ( .05)
female 45   4.41 3.41     5.46 3.54  
0-0.99   93.50     0.051 150.98     0.051
male 6   2.24 -4.87 (NS)   3.62 7.86 (NS)
female 7   - 1.92 4.17     -3.11 6.74  
1-2.99   57.50     2.69 115.03     0.847
male 18   -2.97 -3.65 (NS)   -3.24 -4.71 (NS)
female 16   3.31 4.10     3.65 5.29  
3-4.99   66.67     17.09 101.05     10.51
male 30   -4.49 -2.33 (.001)   -5.36 -1.98 (.002)
female 22   6.12 3.18     7.30 2.70  

a. Based on 1973 FAD/WHO age- and sex-specific recommendations [6].
b. Based on 1973 FAO/WHO age-. sex-, and weight-specific recommendations [6].
c. Adjusted from the Brand mean for the following variables: age, per capita expenditure on food, number of children below adult age, education of father, and education of mother.
d. Figures in parentheses indicate significance level of F value; NS = not significant.

Table 2 presents unadjusted and adjusted [12] relationships between sex and calorie adequacy ratios of pre-school children by methods A and B. The values shown in the "unadjusted" and "adjusted" columns are deviations from the grand mean. For example. -3.67 at the top of the "unadjusted" column under method A means that pre-school-age male children as a whole had a 3.67% lower calorie adequacy ratio than the average. According to the unadjusted data for method A, for the sample of pre-school children as a whole, the calorie adequacy ratio of female children was higher than that of male children by 8 percentage points. The females' calorie adequacy ratio exceeded that of males in all age groups of preschool children except the youngest. The calorie adequacy ratio of male infants (0-11 months old) slightly exceeded that of female infants. This difference is not statistically significant.

The pattern of relationship between sex and calorie adequacy ratio observed at the zero-order level remains unchanged, although somewhat attenuated, even when adjustment is made for the effect of other variables. Although the nutrient adequacy ratio of girls exceeded that of boys in three of four age groups, this difference is statistically significant only for the oldest pre-school children (three to four years old) and for pre-school children considered as a whole.

TABLE 3. Unadjusted and adjusted deviations from mean protein adequacy ratios of pre-school children, by age and sex

Age
(years)
and sex
No.

Method Aa

Method Bb

Mean

Deviation from mean

F valued Mean

Deviation from mean

F valued
Unadjusted Adjustedc Unadjusted Adjustedc
0-4.99

male

female

  164.66     3.02 250.01     5.53
54   -6.57 -4.35 (.08)   -17.83 -12.07 (.021)
45   7.89 5.22     21.40 14.48  
0-0.99

male

female

  176.41     0.042 317.53     0.042
6   5.55 0.87 (NS)   9.99 1.57 (NS)
7   -4.76 -0.75     -8.57 -1.35  
1-2.99

male

female

  175.77     1.83 299.96     7.45
18   -9.53 -11.13 (NS)   -36.03 -37.82 (.011)
16   10.72 12.52     40.54 42.55  
3-4.99

male

female

  154.45     12.59 200.47     7.86
30   -6.04 -2.52 (.001)   -6.41 - 1.02 (.007)
22   8.23 3.43     8.74 1.39  

Notes as for table 2.

TABLE 4. Ratio of male to female nutrient adequacy for pre-school children, by age and socio-economic characteristics of the household

Household
characteristics

Calorie adequacy

Protein adequacy

Method A

Method B

Method A

Method B

0-0.99

years

1-2.99

years

3-4.99

years

0-0.99

years

1-2.99

years

3-4.99

years

0-0.99

years

1-2.99

years

3-4.99

years

0-0.99

years

1-2.99

years

3-4.99

years

Landholding statusa
landless 0.79 0.83 0.94 0.80 0.87 0.96 0.82 0.83 0.98 0.83 0.74 0.98
< 1 - 0.88 0.72 - 0.92 0.76 - 1.02 0.81 - 1.00 0.85
1-2.5 - 0.73 0.88 - 0.79 0.91 - 0.67 0.94 - 0.50 0.95
> 2.5 0.66 1.31 0.84 0.66 1.34 0.85 0.65 0.85 0.93 0.65 0.66 0.90
Incomeb
< 149 - 1.33 0.74 - 1.44 0.77 - 0.77 0.82 - 0.57 0.87
149-225.9 0.62 0.80 0.91 0.62 0.83 0.94 0.76 0.82 0.95 0.75 0.71 0.93
226 612.9 1.12 1.01 0.79 1.12 1.05 0.82 1.06 1.02 0.89 1.06 0.94 0.93
613 - 0.92 0.79 - 0.97 0.83 - 1.07 0.82 - 0.79 0.83

A dash implies the absence of members of one or both sexes in the given category.
a. Landholding in acres.
b. Per capita income in takes.

The finding of a higher calorie adequacy ratio for girls than for boys, according to both the unadjusted and adjusted series, in all but the youngest age group (method A) persists when calorie adequacy is calculated by method B. The same pattern holds for protein adequacy ratios and within different socio-economic groups. A similar picture is obtained when the intra-family distribution of food is measured by observed:expected ratios (FIG. 3. Observed:expected ratio of calorie adequacy for pre-school children, by age and sex).

The findings do not support the hypothesis of son preference in the intra-family distribution of food in a rural area of Bangladesh. This differs from another study in a rural area of Bangladesh [11], which found a higher calorie adequacy ratio for male than for female children. This difference may be attributed, among other factors, to the scope of the two studies. The study by Chen et al. was based on observations of dietary intake of children for a few months only and therefore did not take into account the tremendous seasonal variation in food intake in Bangladesh. Our data, however, were collected over one full year. Moreover, the other study was based on fewer cases. The criteria used in selecting the sample and measuring nutrient needs in the two studies may also have led to the difference in findings. Unfortunately, this cannot be evaluated here since Chen et al. did not adequately discuss the methodologies employed in selecting their sample and estimating children's nutrient needs.

Evidence is strong, however, that the calorie adequacy ratio of male children, particularly the oldest in the pre-school group, is significantly lower than that of their female counterparts. These findings suggest that the explanation for the higher infant mortality for girls must be found in factors other than the alleged non-egalitarian distribution of food in households in favour of male children.

Sex discrimination in other aspects of life

The preference for males in areas other than food distribution in the household may also have a negative bearing on the survival of female children. Some of these aspects are quality of food, mother's child-care time and quality, and expenditure on health care. By employing data from the present study, we found some evidence of preferential treatment for boys over girls, and this may partially explain the prevailing higher infant mortality for girls in Bangladesh.

Table 5 provides data on the intake of calories by sources and sex. It may be observed that the boys tended to consume a more protein-rich diet than the girls. The differences with respect to the consumption of quality food are not very large, however, with the exception of calories obtained from milk and milk products. These items account for a small fraction of the total calories consumed.

TABLE 5. Mean calorie intake per person per day from various food sources by pre-school children (0-4 years old), by sex

Source

Mean intake

Male

Female

Ratio M: F

Rice

533

527

1.01

Pulses

25

21

1.19

Wheat

39

40

0.98

Fish/meat

68

65

1.05

Milk/milk products

9

7

1.29

The per capita expenditure on food per day was little higher for a pre-school male child than for his female counterpart. The ratio is 1.04, but it changes to 0.95 (i.e., the food expenditure for girls exceeded that for boys) when adjustment is made for the relative energy needs of these children. The sex ratio of food expenditure per day is deflated by the sex ratio of the estimated energy needs of pre-school males and females. When the sex ratio of food expenditure for children is analysed by age groups, it is favourable for the youngest males, but it reverses in the higher age groups, both before and after adjustments for relative energy needs (table 6).

TABLE 6. Expenditure on food (takes per person per day) for pre-school children, by age and sex

Age

(years)

Mean expenditure

Ratio M:F

Male

Female

Unadjusted

Adjusteda

0-2.99

1.25

1.09

1.15

1.07

3-4.99

2.68

2.83

0.95

0.90

0-4.99

2.02

1.94

1.04

0.95

a. Obtained by dividing the sex ratio for food expenditure by the sex ratio for estimated energy needs of pre-school children.

Child-care time

Strong evidence in support of a preference for sons is found in the amount of time mothers spend in child care (minutes per child per day). Table 7 clearly shows that this varies with the sex of the child. Mothers pay more attention to male than to female children, and this holds true for every age group of pre-school children. This preferential treatment may expose female children to infectious diseases more frequently than males. This could lead to higher mortality for female children.

TABLE 7. Time devoted by mothers to caring for pre-school children (minutes per child per day), by age and sex of the youngest child

Age

(years)

Male

Female

No.a

Mean

SD

No.a

Mean

SD

< 1

12

62

10.19

13

56

12.84

2

8

54

11.75

8

47

17.07

3

9

41

14.36

-

-

-

4

6

35

15.77

1

31

-

Time includes time spent feeding, caring for, playing with. and reading to children.
a. Number of mothers.

Expenditure on health care

TABLE 8. Average expenditure on health care for pre-school children (takes per child per month), by sex of child and socio-economic characteristics of the household

Household
Characteristics

Male

Female

Ratio

M:F

No. Mean No. Mean
Landholding statusa
Landless 32 76 19 50 1.52
< 1 8 84 10 50 1.68
1-2.5 5 100 6 33 3.03
> 2.5 9 64 10 47 1.36
Incomeb
< 149 24 32 17 36 0.88
149-225.9 10 106 11 30 3.53
226-612.9 6 106 6 78 1.36
613 14 121 11 64 1.89
Education of head of household
illiterate 31 54 26 44 1.22
grades 1-4 10 65 8 31 2.09
grade 5 7 190 5 133 1.42
> grade 5 6 83 6 8 10.37
Education of mother
illiterate 35 48 28 46 1.04
grades 1-5 14 114 14 41 2.78
> grade 5 5 180 3 86 2.09
Family size
1-4 9 167 3 51 3.27
5-7 23 70 26 56 1.25
8 22 48 16 32 1.50

a. Landholding in acres.
b. Per capita income in takas.

A glaring example of sex discrimination is found in expenditure on health care. The average monthly expenditure on health care was 63 takes for a pre-school child. When broken down by sex, the figures are Tk 77 for boys and Tk 47 for girls, a difference of 64%. This unequal expenditure persists even when we control for socio-economic background, measured by household characteristics. At each level of household characteristics, the health expenditure for a male child was higher than that for a female child.

There may be two possible interpretations of these findings. First, the higher health expenditure for boys may mean that their exposure to disease is higher. Second, the difference may indicate different use of health services by sex. The latter seems the more plausible explanation, given the available information on the subject. Although we have not collected data on disease frequency, information obtained from rural areas of Bangladesh by other investigators provides no evidence of different disease exposure according to sex [11, 13]. If this is indeed the case, the difference in health expenditure implies a greater use of health services by males than by females. This was verified by a study that found that the users of a treatment centre in rural Bangladesh included a disproportionate number of males, which was not explained by differences in disease prevalence [13]. In addition, despite nearly comparable occurrence of field diarrhoea, the treatment rate for male children exceeded that for female children by 66% [11]. This form of sex bias may explain, at least partially, the finding of higher female than male infant mortality among pre-school children in rural Bangladesh.

 

Overview, discussion, and policy implications

The findings in this study clearly show that the calorie intake of a sizeable proportion of pre-school children falls short of estimated energy requirements, whether the need is assessed according to age, sex, and average weight (method B) or simply age and sex (method A). When the need is assessed by the most logical means, method B. at least 40% of pre-school children fail to meet their calorie requirement, and this figure rises to 92% when the need is assessed by method A. This low calorie adequacy ratio is an important factor contributing to the poor nutritional status and high mortality among pre-school children in rural Bangladesh. Children appear to satisfy their protein requirements, whether assessed by method A or B.

The hypothesis that the adult members of the household, particularly males, received preferential treatment over young children in terms of distribution of food in patriarchal societies like Bangladesh is not unanimously supported by data when allowance is made for the difference in nutrient needs of the two groups. The position of children relative to other members of the household varies with the method of assessing needs of children. The calorie adequacy ratio of pre-school children was higher than that of other household members when the energy needs of children were assessed by average-weight-specific recommendations. This changed when the energy needs were assessed by age-specific requirements. In this case, children seem to be systematically deprived in meeting their caloric needs compared to adolescents and adults. It has also been shown consistently that the nutrient adequacy ratio of the oldest member of the household is one of the highest. Reverence and care for the old persons in the society may account for this finding.

The data do not support preference for sons in the intra-family distribution of food when adjustment is made for the different nutrient needs of boys and girls. However, a strong preference for sons exists with regard to expenditure on health and mothers' time spent in child care. Some differences by the sex of the child were noted with respect to the consumption of quality foods, with males consuming more of such items than females. The differences were not very large, however, with the exception of calories obtained from milk and milk products. Some of these discriminatory practices, particularly those in the use of health care, may explain the higher rate of infant mortality for female than for male children in rural Bangladesh.

From the findings of the present study, one may draw several policy conclusions. The calorie adequacy ratios obtained by methods A and B clearly demonstrate drastic calorie deprivation beyond the second year of life. This may be attributed to the diminished role of breast-feeding to meet the calorie needs of growing children without sufficient intake of complementary solid foods. The importance of solid foods is also underscored by the findings that breast-fed infants, particularly in the second six months of life (method A) or first year of life (method B), fail to receive adequate calories from breast milk as judged by estimated requirements, unless they receive sufficient solid food along with it [14]. Results obtained from regression analysis using data of the present study also showed a strong inverse relationship between the timing of introduction of solid foods on one hand and the nutrient intake and nutrient adequacy ratio of infants on the other [1]. In other words, the longer the delay in introducing solid foods, the worse the effect on nutrient intake and nutrient adequacy ratios.

These findings reinforce the need for designing nutritional education programmes that emphasize the timely introduction of solid foods together with breast milk to ensure the calorie adequacy of infant diets. Interventions capable of inducing families to start feeding their babies solid foods could yield significant improvement in meeting the calorie needs of infants. This depends also on the economic conditions of the household, apart from the need for nutritional education.

 

References

  1. Chaudbury RH. Determinants of dietary intake and dietary adequacy for pre-school children in Bangladesh. Food Nutr Bull 1984;6(4):24-33.
  2. Bangladesh Bureau of Statistics. Statistical year hook of Bangladesh 1983-84. Dhaka: Government Printing Press, 1984.
  3. Scrimshaw NS, Taylor CE, Gordon JE Interactions of nutrition and infection. World Health Organization monograph series no. 57. Geneva: WHO, 1968.
  4. Caldwell JC. A general theory of fertility. Canberra, Australia: Dept. of Demography, Australian National University, 1979.
  5. Cain M et al. Class, patriarchy, and women's work in Bangladesh. Popul Dev Rev 1979;5(3):405-438.
  6. FAO-WHO. Energy and protein requirements. Report of a joint FAO/WHO ad hoc expert committee. World Health Organization technical report series no. 522. Geneva: WHO, 1973.
  7. Chen LC. An analysis of per capita food grain availability, consumption, and requirements in Bangladesh: a systematic approach to food planning. Bangladesh Dev Stud 1975;3(2):93-126.
  8. Chaudhury RH. Determinants of nutrient adequacy for lactating and pregnant mothers in a rural area of Bangladesh. Food Nutr Bull 1985;7(1):26-32.
  9. Chowdhury FK. The economic value of children. Doctoral dissertation, Dhaka: Dhaka University, 1983.
  10. Chaudhury RH. The effect of mother's work on child care, dietary intake, and dietary adequacy of pre-school children. Bangladesh Dev Stud 1982;10(4):64-88.
  11. Chen LC, Hug E, D'Souza S. Sex bias in the family allocation of food and health care in rural Bangladesh, Popul Dev Rev 1981;7(1):55-70.
  12. Andrews FM, Morgan JN, Sanquist JA. Multiple classification analysis. Ann Arbor, Mich, USA: Institute for Social Research, University of Michigan, 1967.
  13. Horton S. Claquin P. A cost-effectiveness study of hospital and ambulance services at Matlab Treatment Centre. Dhaka: International Center for Diarrhoeal Disease Research, 1980.
  14. Chaudhury RH. The duration of breast-feeding adequacy in a rural area of Bangladesh. Food Nutr Bull 1984;6(1):44 49

Contents - Previous - Next