Journal of Biosocial Science

http://journals.cambridge.org/JBS

Additional services for Journal of Biosocial Science:

Email alerts: Click here

Subscriptions: Click here
Commercial reprints: Click here
Terms of use : Click here

DETERMINANTS OF CHRONIC MALNUTRITION AMONG

PRESCHOOL CHILDREN IN BANGLADESH

AZIZUR RAHMAN and SOMA CHOWDHURY

Journal of Biosocial Science / Volume 39 / Issue 02 / March 2007, pp 161 - 173

DOI: 10.1017/S0021932006001295, Published online: 28 March 2006

Link to this article: http://journals.cambridge.org/abstract_S0021932006001295

How to cite this article:

AZIZUR RAHMAN and SOMA CHOWDHURY (2007). DETERMINANTS OF CHRONIC
MALNUTRITION AMONG PRESCHOOL CHILDREN IN BANGLADESH. Journal of Biosocial
Science, 39, pp 161-173 doi:10.1017/S0021932006001295

Request Permissions : Click here

Downloaded from http://journals.cambridge.org/JBS, IP address: 129.137.165.251 on 12 Mar 2015

J.biosoc.Sci, (2007) 39, 161–173,  2006 Cambridge University Press
doi:10.1017/S0021932006001295 First published online 28 Mar 2006

DETERMINANTS OF CHRONIC
MALNUTRITION AMONG PRESCHOOL
CHILDREN IN BANGLADESH

AZIZUR RAHMAN  SOMA CHOWDHURY

Department of Statistics, University of Chittagong, Chittagong, Bangladesh

Summary. This study investigated the impact of some socioeconomic,

demographic and health and community factors on chronic malnutrition or
stunting in Bangladeshi children aged less than 5 years. The analysis revealed
that the overall prevalence of stunting was 44%, of which 18% of children
were severely stunted, and the demographic characteristics appeared to be the
most significant factors for chronic malnutrition. Multinomial logistic regres-
sion analysis showed that parents’ education, household economic status,
media exposure, number of under-5 children, place of delivery, child’s age,
birth order, months of breast-feeding, birth size, mother’s BMI, mother’s
height, age of household head, measles vaccine, supplementation of diet with
liquids and regional differentials were significantly associated with severe as
well as moderate stunting.

Introduction
Ending malnutrition is an agenda for the current millennium (ACC/SCN, 2000). Each
year about 13 million infants and children under 5 years of age die in developing
countries. Most of these deaths can be attributed to malnutrition and at present over
three-quarters of the world’s malnourished children live in the South-East Asia
region, where about half of child deaths are related to malnutrition’s potential effects
and more than 80% are attributable to mild to moderate malnutrition (UNICEF,
1997). Protein–energy malnutrition signifies an imbalance between intake and the
body’s need to ensure optimal growth and function, and this imbalance over a long
period of time leads to malnutrition in the form of stunting, which has a
multi-factorial aetiology with a clear association with poverty and poor living
conditions (Vella et al., 1992). This is often related to dietary inadequacy, repeated
infections, or both (Ricci & Becker, 1996), and leads to disturbed growth and
increased morbidity and mortality rates (Pelletier et al., 1993), as well as decreased
psychological and intellectual development (Pollitt et al., 1993). Growth retardation
in early childhood is also associated with significant functional impairment in adult
life (Martorell et al., 1992) and reduces work capacity (Spurr et al., 1977), thus
161
162 A. Rahman and S. Chowdhury
affecting economic productivity in a community. There are a number of studies on the
prevalence of different levels of malnutrition and the risk factors in Bangladesh and
elsewhere (Bhuiya et al., 1986; Alam et al., 1989; Vella et al., 1992, 1994; Rahman
et al., 1993; Ricci & Becker, 1996; Jeyaseelan & Lakshman, 1997; Roy, 2000; El-Sayed
et al., 2001).
Bangladesh is one of the poorest developing countries in the world; more than
three-fifths of its population is living below the poverty line, per capita energy supply
is 2060 kcal/day and per capita total protein supply 43 g/day, which are the lowest
figures in the South-East Asia region (UNICEF, 1997). In addition, infant, child and
maternal malnutrition in Bangladesh is amongst the worst in the world; about half of
children are born with low birth weight (ACC/SCN, 2000), and more than 50% are
severe to moderately malnourished (Mitra et al., 1997, 2001) and have an increased
risk of death from infectious disease (Chen et al., 1980). Studies in Bangladesh have
shown that malnutrition leads to more severe infection and higher case fatality (Black
et al., 1984), and it is one of the major causes of morbidity and mortality in children
(Alam et al., 1989; Bairagi & Chowdhury, 1994). This paper aims to measure the
current state of chronic malnutrition in Bangladesh and investigate the associated
factors affecting the nutritional status of children aged 0–59 months.

Methods
The data utilized for the present study were from the 1999–2000 Bangladesh
Demographic and Health Survey (BDHS). This nationally representative survey
collected information during the period early November 1999 to mid-March 2000 on
behalf of the Government of Bangladesh by the National Institute of Population
Research and Training (NIPORT) of the Ministry of Health and Family Welfare,
with funding from the US Agency for International Development (USAID) as part
of the worldwide Demographic and Health Surveys programme. A two-stage stratified
sampling design was used to collect the BDHS sample and the sampled women were
aged below 50 years at the time of interview, providing a complete birth history for
all live births. In the BDHS, all living children aged up to five years at the date of
interview and their mothers were weighed and measured to obtain anthropometric
data. The study is therefore based on a total sample of 5333 Bangladeshi children
aged 0–59 months for whom complete and plausible anthropometric measurements
were provided. Details of the survey methodology, sample design and the principal
findings can be found elsewhere (Mitra et al., 2001).
For anthropometric analysis, the standard index of physical growth ‘Height-for-
Age’ (stunting) is considered. This is expressed in terms of the number of standard
deviation (SD) units (Z-scores) from the median of the NCHS/WHO international
reference population and provides information about the nutritional status of
children, i.e. a drop in stunting signals a chronic problem – one that has persisted for
several months, but is not necessarily present when the measurement is taken. The
WHO working group’s report on measuring the nutritional status of children
recommends the use of Z-scores (WHO, 1986) as they have more advantages than
other methods. Briefly, Z-score indices are linear, sex independent and allow for
 Chronic malnutrition in Bangladeshi children 163
further computation of summary statistics such as means and standard deviations to
classify a population’s growth status.
According to the WHO recommendation, the prevalence of malnutrition should be
calculated on the basis of the proportion of individuals in the observed population
whose index is below
2 standard deviations of the index for the reference
population. In addition, the total prevalence of malnutrition can be classified as severe
and moderate if children have Z-scores below
3 standard deviations (
3SD) and
children with Z-scores between
3 standard deviations (
3SD) and below
2
standard deviations (
2SD) from the median of the NCHS reference population
respectively (Mitra et al., 2001). Thus, for analytic purposes, this study used the
cut-off points recommended by the WHO for defining malnutrition with categories,
and the general approach adopted in the analysis is to differentiate among children
who were relatively well-nourished (Z-scoresR
2SD), those who were moderately
malnourished (–3SD%Z-scores<
2SD) and those who were severely malnourished
(Z-scores<
3SD) in relation to the selected socioeconomic, demographic and health
and community factors. The multinomial logistic regression model was used to
estimate regression parameters in the multivariate analysis.
Some new variables were created by combining information for the other variables
with original codes. For the variable ‘household economic status’ the presence of each
item ‘puccua housing (house made with brick or concrete)’, ‘hygienic toilet’, ‘TV’,
‘radio’, ‘bicycle’, ‘motorcycle’ and ‘electricity’ in the household gave a score of 1, and
0 otherwise. Therefore, the total score range was 0 to 7. Within this, households were
classified as poor (0–2), middle (3–4) and rich (5–7). Respondents were asked in the
BDHS whether they usually read a newspaper, listened to the radio, or watched
television at least once a week. Using this information, an ‘exposure to mass media’
variable was indirectly computed and categorized as: ‘no mass media’, ‘one of the
three media’ and ‘at least two media’. In the BDHS, respondents were asked whether
they usually supplemented their child’s diet with ‘plain water’, ‘sugar water’, ‘fresh
milk’ and ‘other liquid’. The ‘feeding practices of liquids’ variable was computed
indirectly from this information and categorized as: ‘one liquid’, ‘at least two liquids’
and ‘none of the liquids’. Other independent variables used in the study were actual
categories from the original BDHS data. Statistical analysis was done using the
statistical software package SPSS for Windows.

Results
The characteristics of the study’s children are summarized in Table 1. A socio-
economic profile of the study population shows that the standard of living in
Bangladesh is relatively low. Nearly three-quarters (3628) of the children lived in
households of poor economic class. The lack of education is striking: 45% of mothers
and 42% of fathers had no formal education. More than half (53%) of the children’s
mothers had not been exposed to any mass media and about 90% (4767) of children
were delivered at home without medical facilities. In terms of demographic charac-
teristics, only 16% of children had a birth size larger than average, which may be due
to poor maternal nutrition. Forty-two per cent (2249) of mothers had a low BMI and
most of the household heads (56%) were older than 35 years. More than a quarter
164 A. Rahman and S. Chowdhury
Table 1. Socioeconomic, demographic and health and community characteristics of
studied children aged 0–59 months and their percentage distribution by severe and
moderate stunting in Bangladesh, 2000

Stunting
No. of Percentage Total
Characteristic and categories childrena of children Severe Moderate (severe+moderate)

Socioeconomic
Mother’s educational level**
None 2384 44·7 23·8 28·8 52·6
Primary 1547 29·0 18·4 27·5 45·9
Secondary to higherb 1402 26·3 7·2 20·0 27·2
Father’s educational level**
No education 2238 42·0 22·9 29·2 52·1
Incomplete primary 813 15·2 24·0 26·8 50·8
Primary to incomplete secondary 1365 25·6 13·6 26·7 40·3
Secondary to higherb 908 17·0 6·4 16·6 23·0
Household economic status**
Poor 3628 68·0 21·6 28·6 50·2
Middle 1039 19·5 12·5 23·9 36·4
Richb 572 10·7 4·5 15·0 19·5
Exposure to mass media**
None 2822 52·9 21·9 29·3 51·2
One of three types 1456 27·3 16·9 23·8 40·7
At least two typesb 1025 19·7 8·4 20·7 29·1
No. under-5 children**
One 2667 50·0 14·4 25·5 39·9
>Oneb 2666 50·0 21·3 26·8 48·1
Place of delivery**
Respondent’s home 3503 65·7 20·5 26·8 47·3
Other home 1264 23·7 15·8 28·9 44·7
Hospitalb 561 10·5 6·1 15·7 21·8
Demographic
Age of child in months**
35–59 2147 40·3 22·6 27·9 50·5
22–34 1138 21·3 20·5 30·8 51·3
12–21 966 18·1 18·4 31·0 49·4
0–11b 1082 20·3 5·3 13·5 18·8
Birth order**
1 1512 28·4 14·4 27·8 42·2
2–9b 3821 71·6 19·3 25·5 44·8
Months of breast-feeding**
25+ 1724 32·7 20·8 29·5 50·3
7–24 2721 51·0 19·5 28·3 47·8
0–6b 870 16·3 6·9 12·8 19·7
 Chronic malnutrition in Bangladeshi children 165
Table 1. Continued

Stunting
No. of Percentage Total
Characteristic and categories childrena of children Severe Moderate (severe+moderate)

Demographic continued
Size of child at birth**
Larger than average 851 16·0 12·1 22·0 34·1
Average 3469 65·0 16·9 26·5 43·4
Smaller than averageb 1004 18·8 26·3 28·4 54·7
Mother’s BMI (kg/m2)**
%18·50 2249 42·2 22·1 28·1 50·2
18·51–20·50 1637 30·7 18·0 28·2 46·2
>20·50b 1416 26·6 10·9 20·3 31·2
Mother’s height**
% 45 cm 865 16·2 31·0 32·0 63·0
>145 cmb 4455 83·5 15·3 25·0 40·3
Age of household head (years)*
>35 2991 56·1 18·5 24·8 43·3
26–35 2029 38·0 16·5 27·7 44·2
%25b 313 5·9 20·4 28·1 48·5
Health and community
Measles vaccine**
No 2006 37·6 17·4 22·3 39·7
Yesb 3325 62·3 18·2 28·5 46·7
Feeding practice of liquids**
One of four 3073 57·6 21·5 28·9 50·4
At least two 1789 33·5 15·0 24·1 39·1
Noneb 424 8·0 4·7 14·6 19·3
Division of residence**
Barisal 470 8·8 21·9 23·4 45·3
Chittagong 1163 21·8 18·9 25·4 44·3
Dhaka 1283 24·1 17·5 26·7 44·2
Khulna 792 14·9 11·1 25·8 36·9
Rajashahi 941 17·6 16·7 24·5 41·2
Sylhetb 684 12·8 23·5 31·0 54·5
Overall 5333 100 18 26 44

a
Total number of children may differ due to missing data.
b
Reference category in the multivariate analysis.
*p<0·05; **p<0·001 (based on chi-squared test).

(28%) of children had first birth order and nearly three-quarters of the total children
were breast-fed up to two years of age; among them 16% of children were breast-fed
for less than 7 months. About one-fifth of the children (1082) were aged less than one
year and the mean age of the study sample was approximately 29 months. Two
166 A. Rahman and S. Chowdhury
thousand and six (38%) children had not been given measles vaccinations and only 8%
of children were not given the selected supplementary liquids. Nearly a quarter of the
children (24%) lived in the Dhaka division, while 9%, 18% and 13% resided in the
Barisal, Rajashahi and Sylhet divisions respectively, which are relatively poor regions
of Bangladesh.
Table 1 also shows the percentage distribution of children according to the
bivariate relationships between the set of studied covariates and the outcome variables
‘severe’ and ‘moderate’ stunting. Overall, 18% of the study children were observed to
be severely stunted and almost 26% of children were moderately stunted, i.e. the
prevalence of total stunting of children in Bangladesh was about 44%. This figure of
total stunting is smaller than the figure of 55% reported by Mitra et al. (1997).
The percentage of children with severe as well as moderate stunting decreases with
increasing levels of mother’s education, household economic status, mass media
exposure, size of child at birth, mother’s BMI, mother’s height and age of household
head (Table 1). The results show that incomplete primary educated fathers have the
highest percentage of severely stunted children (24%), but the percentage of children
with moderate stunting increases with decreasing levels of father’s education.
Households with more than one child have a higher percentage of severely and
moderately stunted (21% and 27%) children compared with households with only one
child. Respondents whose children were delivered in hospital have a lower percentage
of severely and moderately stunted (6% and 16%) children relative to those whose
children were delivered at the respondents’ home or at someone else’s home. The
percentage of severely stunted children increases with age, and the bivariate results
indicate that the percentage of moderate stunting is highest among children aged
12–21 months, and total stunting is highest among children aged 22–34 months. After
that, increases in age represent a decreasing percentage of chronic malnutrition.
Moreover, the prevalence of moderate stunting is higher with first birth than higher
order births, and the percentage of stunting increases with months of breast-feeding.
Prevalence of stunting is higher among the children who received measles
vaccinations at the bivariate level. Both types of stunting are lower in children
supplemented with liquids and highest in children supplemented with only one liquid.
The percentage of children with severe and moderate stunting is highest in the Sylhet
division (24% and 31%) and the percentage of children with severe stunting, and total
stunting, is lowest in the Khulna division (11% and 37%). However, the percentage
of moderately stunted children in the Khulna division (26%) is somewhat higher than
that in the Chittagong division (25%), the Rajashahi division (24·5%) and the Barisal
division (which contains the lowest percentage of moderately stunted children, i.e.
23%). That is, differences by division show that children in the Sylhet division are
somewhat more likely, and children in the Khulna division are somewhat less likely
to be chronically malnourished than children in other divisions.
The results of the multivariate analysis are presented in Table 2. Cumulative
deficient growth is measured by stunting associated with long-term factors. In the
present study, most of the selected socioeconomic characteristics were significantly
associated with severe stunting. Among them, mother’s education and exposure to
mass media did not show any significant effect on moderate stunting. In particular,
low household economic status and parental education were found to be generative
 Chronic malnutrition in Bangladeshi children 167
for chronic malnutrition. Households having only one child under 5 had a reduction
in the likelihood that the child was severely as well as moderately stunted by 39% and
19% respectively. Children delivered at respondents’ homes with traditional delivery
systems were 1·5 times and 1·4 times more likely to be severely and moderately
stunted than children born in hospital.
Demographic characteristics have a large impact on stunting. The results revealed
that retarded growth increased with a child’s age and the likelihood of the child being
stunted decreased with increased birth size and especially with maternal health. More
specifically, children who had a bigger birth size than average were 70% and 48% less
likely to be severely and moderately stunted compared with children who had a
smaller than average birth size. Furthermore, the children of nutritionally at-risk
(height %145 cm) mothers ran a 3·4 times and 2·1 times higher risk of becoming
severely and moderately stunted than their better-fed counterparts. Also, increased
months of breast-feeding and initial birth order enhanced the moderate stunting rate.
Severe stunting was more affected than moderate stunting by the health and
community characteristics. Children who did not receive measles vaccinations ran a
1·4 times higher risk of becoming severely stunted; notwithstanding the bivariate
analysis, the effect of this vaccine on severe and moderate stunting was encouraging.
Children whose diets were supplemented with one liquid had the highest risk of severe
stunting and they were 2·2 times more likely to be severely stunted than children
whose diets were not supplemented by liquids. Moreover, the analysis shows that
measures of cumulative deficient growth or stunting were high and significantly
associated with residential differential factors (see Table 2). Thus, the study seems to
show that stunting in children is one of the best indicators of community inequality,
and this is in agreement with the study by Michel et al. (1989).

Discussion and Conclusions

Between the 1996–97 and 1999–2000 BDHS surveys, the total prevalence of stunting
in Bangladesh decreased by 11% – from 55% to 44% – in children under the age of
five. The level of severe stunting within this time period decreased rapidly by about
10%, but the moderate stunting rate declined by only 1%. In the period 1996–97 the
prevalence of severe stunting was 1% higher than that of moderate stunting (Fig. 1).
Thus the prevalence of stunting in Bangladesh shows a noteworthy declining trend,
yet the fall in prevalence has not been as rapid as the rise in the population.
Education is the pre-condition of progress and provides a source of knowledge.
More educated mothers can make better use of health services, provide better
childcare, have increased knowledge of appropriate child rearing, have more hygienic
household practices and personal habits and also have higher status in the family. In
the present research, children of uneducated or primary educated mothers had a
higher risk of severe stunting. Lower maternal education has also been highlighted as
a risk factor for malnutrition in rural areas (Roy, 2000). Both moderate and severe
stunting are strongly associated with father’s education. In Bangladesh, father’s
education should be more closely related to household income than mother’s
education. For example, the father’s education is a pre-requisite for sufficient access
to food in a household and contributes a higher proportion of wealth to children’s
168 A. Rahman and S. Chowdhury
Table 2. Effects of socioeconomic, demographic and health and community charac-
teristics on severe and moderate stunting of preschool children in Bangladesh, 2000
(number of study children=5333)a

Severe stunting Moderate stunting

Coefficient Odds Coefficient Odds
Characteristicb (B) SD ratio (B) SD ratio

Socioeconomic
Mother’s educational level
None 0·564**** 0·160 1·76 0·177 0·121 1·19
Primary 0·421*** 0·154 1·53 0·127 0·114 1·14
Father’s educational level
None 0·607*** 0·193 1·84 0·412*** 0·144 1·51
Incomplete primary 0·755**** 0·201 2·13 0·379** 0·155 1·46
Primary to incomplete secondary 0·270 0·185 1·31 0·335*** 0·130 1·40
Household economic status:
Poor 0·832**** 0·243 2·30 0·565**** 0·162 1·76
Middle 0·667*** 0·243 1·95 0·445*** 0·158 1·56
Exposure to mass media
None 0·259* 0·156 1·30 0·120 0·117 1·13
One of three types 0·226 0·156 1·25
0·092 0·116 0·91
No. under-5 children
One
0·492**** 0·090 0·61
0·217*** 0·075 0·81
Place of delivery
Respondent’s home 0·370* 0·213 1·45 0·312** 0·147 1·37
Other home 0·334 0·223 1·40 0·390*** 0·154 1·48
Demographic
Age of child in months
35–59 2·225**** 0·202 9·26 1·156**** 0·151 3·18
22–34 2·053**** 0·211 7·79 1·235**** 0·158 3·44
12–21 1·850**** 0·210 6·36 1·171**** 0·154 3·22
Birth order
First birth
0·013 0·110 0·99 0·223*** 0·087 1·25
Months of breast-feeding
25+ 0·225 0·205 1·25 0·491*** 0·164 1·63
7–24 0·308 0·193 1·36 0·495**** 0·150 1·64
Size of child at birth
Larger than average
1·218**** 0·149 0·30
0·660**** 0·124 0·52
Average
0·760**** 0·105 0·47
0·332**** 0·094 0·72
Mother’s BMI (kg/m2)
%18·50 0·508**** 0·118 1·66 0·274**** 0·095 1·32
18·51–20·50 0·434**** 0·124 1·54 0·314**** 0·098 1·37
Mother’s height
%145 cm 1·209**** 0·106 3·35 0·754**** 0·097 2·13
Age of household head
>35 years
0·307* 0·184 0·74
0·168 0·156 0·85
26–35 years
0·544*** 0·186 0·58
0·156 0·157 0·86
 Chronic malnutrition in Bangladeshi children 169
Table 2. Continued

Severe stunting Moderate stunting

Coefficient Odds Coefficient Odds
Characteristicb (B) SD ratio (B) SD ratio

Health and community

Measles vaccine
No 0·307*** 0·098 1·36 0·055 0·089 1·06
Feeding practice of liquids
One of four 0·768*** 0·270 2·16 0·182 0·179 1·20
At least two 0·639** 0·277 1·89 0·086 0·185 1·09
Division of residence
Barisal
0·164 0·177 0·85
0·448*** 0·160 0·64
Chittagong
0·261* 0·142 0·77
0·319*** 0·124 0·73
Dhaka
0·366*** 0·142 0·69
0·311** 0·123 0·73
Khulna
0·811**** 0·174 0·45
0·392*** 0·137 0·68
Rajashahi
0·536**** 0·151 0·59
0·517**** 0·130 0·60
Model fitting information

2 Log likelihood 8680·76
Chi-squared (df) 1256·36 (66)†

a
Reference category of dependent variable is well-nourished children.
b
Omitted categories not shown.
*p<0·10; **p<0·05; ***p<0·01; ****p<0·001, †p<0·0000.

Fig. 1. Trends in prevalence of chronic malnourished children in Bangladesh.

welfare, and better economic conditions increase the living standards of families,
allowing them to take essential care of the children, and provide support to the
mother for baby care-giving. All of this depends on the father’s education because the
main household earner in the country is predominantly male. In Bangladesh, the risk
of chronic malnutrition increases with age and the findings generally show a peak in
the prevalence of moderate stunting during the second year of life and downward
170 A. Rahman and S. Chowdhury
thereafter. The prevalence of total stunting on the other hand increases during the
second year of life and remains stable thereafter. This finding is consistent with those
of Mitra et al. (2001).
The findings suggest that exclusive breast-feeding up to about 6 months, avoidance
of supplementation with selected liquids, and perhaps the introduction of solid foods
appear to contribute significantly to the improvement in children’s health. After about
6 months of age the nutrients provided by breast-milk become inadequate to sustain
growth and the supplementation is necessary. In general, Bangladeshi children are
breast-fed for a long period of time; supplemental liquid foods are introduced fairly
early and solid foods are introduced quite late (Nazneen et al., 1997). Moreover, plain
water is traditionally given to children after feeding, and this may be of poor quality.
Supplementary feeding by bottle can result in bacterial contamination due to lack of
sterilization, and the use of plain water or sugar water can be a cause of diarrhoea.
Thus careful introduction of adequate, healthy and hygienic supplementations after
the first 4–6 months is vital.
Bangladeshi children delivered at home under traditional delivery systems without
medical facilities have a higher risk of stunting than children delivered in hospital
under appropriate medical care. Certainly it is true that mothers who have sought
medical services up to delivery gather different knowledge of complications, child and
maternal nutrition, breast-feeding practices and childcare, which may have positive
effects on the child’s health and negative effects on maternal and child mortality.
This study’s findings show that children whose mothers were not exposed to any
mass media had a higher risk of becoming severely stunted. Children in the lower
economic classes, and who did not receive measles vaccinations had a significantly
higher risk of developing chronic malnutrition. Vaccination against diseases such as
measles is very important to children’s health and this may depend on household
economic conditions as well as awareness. Also children living in a household with
only one child have a lower risk of both types of stunting than children in households
with more than one child, and this confirms the findings of a study by Roy (2000).
Children of first order births have a higher risk of moderate stunting than those of
higher order births. The primary providers of care for young children in all cultures
are predominantly mothers, and therefore care of the child is inextricably linked to
the situation of the household and the mother, and all first order births in Bangladesh
face inexperienced adolescent mothers due to early marriage. Most of these mothers
have a limited awareness of proper childcare and adequate nutrition practices.
Two indicators of a mother’s nutritional status, i.e. mother’s BMI and height,
have a strong link with child nutrition. The results show that the risk of severe and
moderate stunting in children decreases with increase in maternal nutrition. The
problem of maternal malnutrition in a country like Bangladesh is a complex one with
several underlying causes. Poverty is a direct cause of maternal undernutrition
because mothers from poor families are nutritionally deprived during childhood and
adolescence, and this does not improve after marriage. Social causes, like early
marriage, frequent childbirth, lack of proper birth spacing, and discrimination of
intra-household food distribution in a male-dominated family, all exacerbate the poor
state of maternal malnutrition. Besides this, the preferential treatment of male
children is common in many South-Asian cultures, and this would increase the
 Chronic malnutrition in Bangladeshi children 171

Fig. 2. Percentage of stunted children by division of residence and level of economic

status.

prevalence of nutritionally poor mothers. Rahman et al. (1993) report that under-
nourished mothers tend to have little or no education and mostly come from poor
families. Undernourished mothers probably cannot breast-feed their children
adequately, which can contribute to the poor nutrition of their children and they
usually give birth to low birth weight babies. Admittedly, this study suggests that the
factors influencing maternal nutrition as discussed above would be more pronounced
since the nutritional status of mothers emerges as the most important determinant of
their children’s nutritional status.
Half of the total newborns in Bangladesh have a low birth weight (ACC/SCN,
2000). This study also suggests that a greater size at birth reduces the risk of severe
as well as moderate chronic malnutrition of children. But size of the child at birth
depends on various maternal and socioeconomic factors such as food intake during
pregnancy, mother’s health and nutritional condition, mother’s education, mother’s
age, household economic status and maternal check-ups during pregnancy. Only a
small percentage of Bangladeshi children are born with a larger than average birth
size and they have a lower risk of severe and moderate stunting. Therefore, special
attention should be given to increasing the birth size of children.
Children living in the Khulna and Rajashahi divisions have a low risk of severe
stunting and moderate stunting and children living in the Sylhet division have a high
risk of stunting. Stunting in children is one of the best indicators of social inequality.
Figure 2 shows the stunting rate of children by division of residence for the three
classes of household economic status. For children in the Khulna and Rajashahi
divisions the economic differentials in stunting rates are relatively small. In fact, in the
Barisal and Sylhet divisions, children from rich classes have lower stunting rates
compared with those of other divisions, while the proportions of stunted children in
this economic class and also in the middle class are lowest in the Barisal region, but
within the poor class stunting rates are higher in the Sylhet division. In different
divisions, however, the differentials fluctuate and a pattern of stunting rates in
accordance with different classes of household economic status comes into view. In
172 A. Rahman and S. Chowdhury
the Khulna and Rajashahi divisions, children in the poor and middle economic classes
are in relative good health compared with those in the rich economic class and those
residing in other divisions. The findings of the study suggest that the relatively high
magnitude and unsteadiness of the economic differentials in stunting rates in the
Sylhet, Barisal, Chittagong and Dhaka divisions can be ascribed in large part to
socioeconomic and demographic differentials, and in vaccination and feeding
practices. Most of the households in the Sylhet and Barisal divisions fall in the
lower economic class. In addition, socioeconomic, demographic, immunization and
supplementation conditions are poor in the Sylhet division compared with other
divisions. This echoes the findings of the 1999–2000 BDHS (Mitra et al., 2001).
In conclusion, the study suggests that severe and moderate chronic malnutrition
is a public health problem among children in Bangladesh. Poor maternal and
demographic situations, poor socioeconomic conditions, poor feeding and
immunization practices and also divisional differentials are the most important factors
associated with the higher prevalence of severe and moderate stunting. Consequently
it is recommended that policymakers should give attention to improving those factors.
Furthermore a prospective study is needed to find out the specific set of determinants
of child and maternal nutrition since the two are strongly linked.

Acknowledgments
The authors wish to acknowledge the help of the National Institute of Population
Research and Training (NIPORT) of Bangladesh for supplying the data used in this
study. They are also grateful to Dr S Ahmed, Professor in the Department of
Mathematics and Physics, East West University, Dhaka, Bangladesh, for his friendly
collaboration in the work.

References
ACC/SCN (2000) Final Report to the ACC/SCN by the Commission on the Nutrition Challenges
of the 21th Century. United Nations, February 2000, pp. i–v.
Alam, N., Wojtyniak, B. & Rahman, M. M. (1989) Anthropometric indicators and risk of death.
American Journal of Clinical Nutrition 49, 884–888.
Bairagi, R. & Chowdhury, M. K. (1994) Socioeconomic and anthropometric status, and
mortality of young children in rural Bangladesh. International Journal of Epidemiology 23,
1179–1184.
Bhuiya, A., Zamicki, S. & D’Souza, S. (1986) Socioeconomic differentials in child nutrition and
morbidity in rural area of Bangladesh. Journal of Tropical Pediatrics 36, 17–23.
Black, R. E., Brown, K. H. & Becker, S. (1984) Malnutrition is a determining factor in
diarrhoeal duration, but not in incidence, among rural children in a longitudinal study in
rural Bangladesh. American Journal of Clinical Nutrition 37, 87–94.
Chen, L. C., Chowdhury, A. K. M. A. & Huffman, S. L. (1980) Anthropometric assessment of
energy-protein malnutrition and subsequent risk of mortality among preschool aged children.
American Journal of Clinical Nutrition 33, 1836–1845.
El-Sayed, N., Mohammed, A. G., Nofal, L., Mahfuz, A. & Zeid, H. A. (2001) Malnutrition
among pre-school children in Alexandria, Egypt. Journal of Health, Population and Nutrition
19(4), 275–280.
 Chronic malnutrition in Bangladeshi children 173
Jeyaseelan, L. & Lakshman, M. (1997) Risk factors for malnutrition in South Indian children.
Journal of Biosocial Science 29, 93–100.
Martorell, R., Rivera, J., Kaplowitz, H. & Pollitt, E. (1992) Long-term consequences of growth
retardation during early childhood. In Hernandez, M. & Argente, J. (eds) Human Growth:
Basic and Clinical Aspects. Elsevier Science Publishers, Amsterdam, pp. 143–149.
Michel, C., Anne, M. & Masse, R. (1989) Nutritional Status: The Interpretation of Indicators.
Children in the Tropics, No. 181/82. International Children’s Centre, Paris, pp. 16–35
Mitra, S. N., Al-Sabir, A., Anne, R. C. & Knata, J. (1997) Bangladesh Demographic and Health
Survey 1996–1997. NIPORT, Mitra and Associates, Dhaka, Bangladesh, and Macro
International Inc., Calverton, Maryland, USA, pp. 133–137.
Mitra, S. N., Al-Sabir, A., Saha, T. & Kumar, S. (2001) Bangladesh Demographic and Health
Survey 1999–2000. NIPORT, Mitra and Associates, Dhaka, Bangladesh, and ORC Macro,
Calverton, Maryland, USA. pp. iii–xii & 140–145.
Nazneen, C., Ataharul, M. I. & Nitai, C. (1997) Infant and child feeding practices in
Bangladesh: Evidence from BDHS 1993–94. Demography India 26, 275–286.
Pelletier, D. L., Fronqillo, E. A. Jr & Habicht, J. P. (1993) Epidemiologic evidences for a
potentiating effect of malnutrition on child mortality. American Journal of Public Health 83,
1130–1133.
Pollitt, E., Gorman, K. S., Engle, L. Martorell, P. & Rivera J. (1993) Early supplementary
feeding and cognition: effects over two decades. Monograph of the Society for Research in
Child Development 58, 1–99.
Rahman, S. K. M., Roy, M., Ali, A. K., Mitra, A., Alam, N. & Akbar, M. S. (1993) Maternal
nutritional status as a determinant of child health. Journal of Tropical Pediatrics 39, 86–89.
Ricci, J. A. & Becker, S. (1996) Risk factors for wasting and stunting among children in Metro
Cebu, Philippines. American Journal of Clinical Nutrition 63, 966–975.
Roy, N. C. (2000) Use of mid-upper arm circumference for evaluation of nutritional status of
children and for identification of high-risk groups for malnutrition in rural Bangladesh.
Journal of Health, Population and Nutrition 18, 171–180.
Spurr, G. B., Barac-Nieto, M. & Maksud, M. G. (1977) Productivity and maximal oxygen
consumption in sugar cane cutters. American Journal of Clinical Nutrition 30, 316–321.
UNICEF (1997) Malnutrition in South ASIA: A Regional Profile. UNICEF, November 1997,
pp. 1–50.
Vella, V., Tomkins, A., Borghesi, A., Migliori, G. B. & Oryem, V. Y. (1994) Determinants of
stunting and recovery from stunting in northwest Uganda. International Journal of
Epidemiology 23, 782–785.
Vella, V., Tomkins, A., Borghesi, A., Migliori, G. B., Adriko, B. C. & Crevatin, E. (1992)
Determinants of child nutrition and mortality in northwest Uganda. Bulletin of the World
Health Organization 70 (5), 637–643.
WHO (1986) Use and interpretation of anthropometric indicators of nutritional status. Bulletin
of the World Health Organization 64, 929–941.