Child Nutrition Survey of Bangladesh 1995-96
Bangladesh, 1995 - 1996
BANGLADESH BUREAU OF STATISTICS
Identification
idno
BGD-BBS-CNS-1995-96-v01
Title
Child Nutrition Survey of Bangladesh 1995-96
Subtitle
Fourth in the series
Country
| Name | Country code |
|---|---|
| Bangladesh | BGD |
Abstract
Bangladesh Bureau of Statistics (BBS) has been periodically conducting child nutrition surveys (CNS) as part of its regular activities since 1985. The first such survey was conducted in 1985-86. The present survey which is the fourth in the series was conducted in 1995-96. Like the previous surveys, the present survey is designed to generate national level data on nutritional condition of children in the country. The survey was conducted on sub-sample (200 PSUs) of a larger national level household expenditure survey (HES) of 372 PSUs, as a special module of HES.
The survey covered 2614 (1800 rural and 814 urban) children aged 6-71 months from 132 rural and 68 urban primary sampling units (PSUs) distributed over the whole of Bangladesh. In addition to anthropometric assessment of nutritional status, information on household demographic characteristics, environmental conditions, and child feeding and caring practices were also obtained. Household socio-economic information was transcribed from HES questionnaire.
Nutritional status of the surveyed children was analyzed as standard deviation scores of NCHS (National Centre for Health Statistics, USA) reference height-for-age, weight-for-age and weight-for-height. For comparison with earlier survey results, height-for-age and weight-for-height have also been analyzed as percentages of respective reference medians. The important findings of the study are:
Nutrition situation:
* The mean height of Bangladeshi children aged 6-71 months (genders combined) is 86.67 cm and the mean weight is 11.15 kg which are respectively 91.99% and 77.43% of NCHS median height-for-age and weight-for-age.
* In terms of reduced height-for-age and weight-for-height 59.7% of all children between 6 and 71 months of age was found to be malnourished.
*Stunting which is the stable indicator of changes in nutritional status over time was found to be 51.4 % (24.2% of the children are severely and 27.2% moderately stunted).
*A decline of the order of a high magnitude, 12.8% in the prevalence of stunting from 64.2% in 1992 to 51.4% in 1995-96 was observed.
*8.2% of all the children was concurrently stunted (chronically undernourished) and wasted (acutely undernourished), 43.1% stunted but not wasted, and 8.4% wasted but not stunted.
*The prevalence of severe underweight (low weight-for-age) was 17.9% and that of moderate underweight was 39.5%.
*The prevalence of severe wasting was 2.5% and that of moderate wasting is 14.1%. The total prevalence of wasting was 16.6%.
*Compared to urban children more rural children were malnourished. The prevalences of underweight, stunting and wasting were respectively 13%, 10% and 4% higher among rural children.
*Within urban areas the slum children were worse off than their non-slum urban counterparts. More than 54% of skim children were stunted as against 42.5% of their non-slum counterparts.
*Gender differences in the prevalence of malnutrition were small but severe stunting and severe underweight were respectively 1.3% and 4% higher among females.
*Seasonal variations in the prevalence of malnutrition were -not large. Compared to first half of the year the prevalence of malnutrition was relatively higher in the second half. The overall prevalence of malnutrition was found to be the highest iii August.
Determinants of nutritional status
Food Security
*The nutritional status of children was positively related to several measures of household food security, for example, farm size, household income, per capita income and household expenditure. Household expenditure produced the best fit to the data.
*The contribution of the factor to nutritional status was however small. As household expenditure increased 3.14 (e3.14) times, the nutritional status by WAZ improved by 0.13 and by WAZ improved by 0.18 points only.
Caring Practices
*The nutritional status of children showed significant positive relationship with male child, consumption of iodized salt, consulting qualified doctors during children's illness, and level of mothers literacy, and significant negative relationship with the age of children, male headed households and household size. The relationships with the other variables were either insignificant or theoretically inconsistent.
*For the child being male the nutritional status by HAZ improved by 0.1 and by WAZ improved by 0.08 points.
*For consumption of iodized: salt the nutritional status by HAZ improved by 0.15 and by WAZ improved by 0.09 points.
*For consulting qualified doctors during children's illness the nutritional status by HAZ improved by 0.10 and by WAZ improved by 0.08 points.
*For each step rise in mothers' level of literacy from illiteracy to primary, primary to high school, and above high school level the nutritional status both by HAZ and WAZ increased by 0.14 points.
*For One month increase in children's age nutritional status by WAZ decreased by 0.012 and by WAZ decreased by 0.002 points.
*For one member increase in household size the nutritional status by WAZ decreased by 0.04 and by WAZ decreased by 0.02 points.
Health environment
*The nutritional status of children showed significantly positive relationship with the ownership of hygienic latrines, and significantly negative relationship with the attack by diarrhea, fever and cough. The relationships with the other variables are either insignificant or theoretically inconsistent.
*For househokjs owning hygienic latrines the nutritional status by WAZ increased by 0.25 and by WAZ increased by 0.13 points.
*For each episode of diarrhoea the nutritional status by HAZ decreased by 0.21 and by WAZ decreased by 0.15 points.
*For each episode of prolonged fever and cough the nutritional status by WAZ decreased by 0.14 and by WAZ decreased by 0.23 points.
Program me implications
*it appears that an effective strategy for addressing child malnutrition in the country would be behavioural change in respect of health environment and child care practices, particularly to contain child morbidity down, feeding iodized salt, consulting qualified doctors during illness, and using hygienic latrine, together with greater empowerment of women by more female literacy and more female authority in household affairs.
*The efforts should accompany alleviation of poverty by raising incomes of the poor. The overall strategy should receive greater bias in favour of girls, receive special attention during the second half of the year, and target children at an early age.
The survey covered 2614 (1800 rural and 814 urban) children aged 6-71 months from 132 rural and 68 urban primary sampling units (PSUs) distributed over the whole of Bangladesh. In addition to anthropometric assessment of nutritional status, information on household demographic characteristics, environmental conditions, and child feeding and caring practices were also obtained. Household socio-economic information was transcribed from HES questionnaire.
Nutritional status of the surveyed children was analyzed as standard deviation scores of NCHS (National Centre for Health Statistics, USA) reference height-for-age, weight-for-age and weight-for-height. For comparison with earlier survey results, height-for-age and weight-for-height have also been analyzed as percentages of respective reference medians. The important findings of the study are:
Nutrition situation:
* The mean height of Bangladeshi children aged 6-71 months (genders combined) is 86.67 cm and the mean weight is 11.15 kg which are respectively 91.99% and 77.43% of NCHS median height-for-age and weight-for-age.
* In terms of reduced height-for-age and weight-for-height 59.7% of all children between 6 and 71 months of age was found to be malnourished.
*Stunting which is the stable indicator of changes in nutritional status over time was found to be 51.4 % (24.2% of the children are severely and 27.2% moderately stunted).
*A decline of the order of a high magnitude, 12.8% in the prevalence of stunting from 64.2% in 1992 to 51.4% in 1995-96 was observed.
*8.2% of all the children was concurrently stunted (chronically undernourished) and wasted (acutely undernourished), 43.1% stunted but not wasted, and 8.4% wasted but not stunted.
*The prevalence of severe underweight (low weight-for-age) was 17.9% and that of moderate underweight was 39.5%.
*The prevalence of severe wasting was 2.5% and that of moderate wasting is 14.1%. The total prevalence of wasting was 16.6%.
*Compared to urban children more rural children were malnourished. The prevalences of underweight, stunting and wasting were respectively 13%, 10% and 4% higher among rural children.
*Within urban areas the slum children were worse off than their non-slum urban counterparts. More than 54% of skim children were stunted as against 42.5% of their non-slum counterparts.
*Gender differences in the prevalence of malnutrition were small but severe stunting and severe underweight were respectively 1.3% and 4% higher among females.
*Seasonal variations in the prevalence of malnutrition were -not large. Compared to first half of the year the prevalence of malnutrition was relatively higher in the second half. The overall prevalence of malnutrition was found to be the highest iii August.
Determinants of nutritional status
Food Security
*The nutritional status of children was positively related to several measures of household food security, for example, farm size, household income, per capita income and household expenditure. Household expenditure produced the best fit to the data.
*The contribution of the factor to nutritional status was however small. As household expenditure increased 3.14 (e3.14) times, the nutritional status by WAZ improved by 0.13 and by WAZ improved by 0.18 points only.
Caring Practices
*The nutritional status of children showed significant positive relationship with male child, consumption of iodized salt, consulting qualified doctors during children's illness, and level of mothers literacy, and significant negative relationship with the age of children, male headed households and household size. The relationships with the other variables were either insignificant or theoretically inconsistent.
*For the child being male the nutritional status by HAZ improved by 0.1 and by WAZ improved by 0.08 points.
*For consumption of iodized: salt the nutritional status by HAZ improved by 0.15 and by WAZ improved by 0.09 points.
*For consulting qualified doctors during children's illness the nutritional status by HAZ improved by 0.10 and by WAZ improved by 0.08 points.
*For each step rise in mothers' level of literacy from illiteracy to primary, primary to high school, and above high school level the nutritional status both by HAZ and WAZ increased by 0.14 points.
*For One month increase in children's age nutritional status by WAZ decreased by 0.012 and by WAZ decreased by 0.002 points.
*For one member increase in household size the nutritional status by WAZ decreased by 0.04 and by WAZ decreased by 0.02 points.
Health environment
*The nutritional status of children showed significantly positive relationship with the ownership of hygienic latrines, and significantly negative relationship with the attack by diarrhea, fever and cough. The relationships with the other variables are either insignificant or theoretically inconsistent.
*For househokjs owning hygienic latrines the nutritional status by WAZ increased by 0.25 and by WAZ increased by 0.13 points.
*For each episode of diarrhoea the nutritional status by HAZ decreased by 0.21 and by WAZ decreased by 0.15 points.
*For each episode of prolonged fever and cough the nutritional status by WAZ decreased by 0.14 and by WAZ decreased by 0.23 points.
Program me implications
*it appears that an effective strategy for addressing child malnutrition in the country would be behavioural change in respect of health environment and child care practices, particularly to contain child morbidity down, feeding iodized salt, consulting qualified doctors during illness, and using hygienic latrine, together with greater empowerment of women by more female literacy and more female authority in household affairs.
*The efforts should accompany alleviation of poverty by raising incomes of the poor. The overall strategy should receive greater bias in favour of girls, receive special attention during the second half of the year, and target children at an early age.
Kind of Data
Sample survey data [ssd]
Version
Version Date
1997-12
Coverage
Universe
Table 2.4 'preseiits the number of PSU's, number of households, and number of children enumerated by time period. The nutrition survey teams identified if a 6-71 months old child was available for weighing and measuring in the selected households, and if a parent or guardian was available for interview.
The mean coverage of the survey was 92.5% varying from 90.5% to 95.0%. The mean non-response was 7.5% varying from 5.0% to 9.5%. In the 1992 .CNS the mean non-response was 13.2%. The non-response .was mainly due to children not being available at home,during data collection orthe families having moved Out of the place rather than unwillingness of the households to participate. In the 1995-96 CNS in .129 households children were not at 'home and 17 households were found to have moved to other places.
The mean coverage of the survey was 92.5% varying from 90.5% to 95.0%. The mean non-response was 7.5% varying from 5.0% to 9.5%. In the 1992 .CNS the mean non-response was 13.2%. The non-response .was mainly due to children not being available at home,during data collection orthe families having moved Out of the place rather than unwillingness of the households to participate. In the 1995-96 CNS in .129 households children were not at 'home and 17 households were found to have moved to other places.
Producers and sponsors
Authoring entity/Primary investigators
| Agency Name | Affiliation |
|---|---|
| BANGLADESH BUREAU OF STATISTICS | Statistics Division, Ministry of Planning |
Funding Agency/Sponsor
| Name | Abbreviation |
|---|---|
| Statistics Division | SD |
Sampling
Sampling Procedure
The survey used the BBS's "Integrated Multipurpose Sample (IMPS)". The IMPS design included the entire Bangladesh. The design is based on the 1991 population census area frame and consists of 372 primary sampling units (PSUs), of which 252 rural and 120 urban. This design was used by BBS for three surveys. household expenditure survey (HES), labour force survey (LFS), and health and demographic survey (HDS). The present child nutritional status survey (CNS), 1996 was carried out on a sub-sample of HES 1996 i.e. a sub sample of IMPS.
The method of selection involves a two stage stratification. In the first stage 5 administrative divisions are i:reated as super-strata and within these super-strata there is a second stage of stratification comprising (i) rural areas, (ii) statistical metropolitan areas and (iii) municipalities. Statistical metropolitan areas, and municipalities constitute urban sample areas. Thus there are 14 strata altogether (there is no SMA in Barisal division). The number of Primary Sampling Units (PSU) are allocated among the 14 strata as follows:
Details of sampling procedure are given in Appendix B
The method of selection involves a two stage stratification. In the first stage 5 administrative divisions are i:reated as super-strata and within these super-strata there is a second stage of stratification comprising (i) rural areas, (ii) statistical metropolitan areas and (iii) municipalities. Statistical metropolitan areas, and municipalities constitute urban sample areas. Thus there are 14 strata altogether (there is no SMA in Barisal division). The number of Primary Sampling Units (PSU) are allocated among the 14 strata as follows:
Details of sampling procedure are given in Appendix B
Response Rate
The child nutrition survey was carried out on ,a sub-sample cOnstituting 50 percent of the HES/ IMPS, 1996 sample under the restriction that at least one PSU would be selected from each of the 64 districts. The number of PSUs selected is 200. Among these 200 PSUs, 68 PSUs were urban and 132 were rural. HES selected 20 households from each PSU. The CNS covered only those households which had children aged 6-71 months. The following table shows the distribution of selected PSUs by division and type.
Weighting
Since 1985-86 the Bangladesh Bureau of Statistics (BBS) conducted three national level Child Nutrition Surveys (CNS). This report presents the third CNS conducted in 1992. The survey covered 2110 (1325 rural and 785 urban) children aged 6-71 months from 196 Primary Sampling Units (PSU's) distributed over the whole of Bangladesh, except the hill districts of Chittagong. In addition to anthropometric assessment of nutritional status, information on household demographic characteristics, environmental conditions, and child feeding and caring practices were also obtained. Household socio-economic information was transcribed from the HES questionnaire.
Nutritional Status of individual children was analyzed by height-for-age, weight-for-age and weight-for-height SD scores of standard values (NCHS reference population) and also as percentages of respective standards (reference medians). The important findings of the study are:
Nutrition Situation
* The mean height of Bangladeshi children aged 6-71. months is 85.58 cm and the mean weight is 10.86 kg, which are respectively 90.57% and 75.24% of standard values (NCHS medians).
* Taking both height-for-age and weight-for-height into consideration, only 31% of all children between 6 and 71 months of age could be classified as "not malnourished0. The remaining 69% of the children are victims of malnutrition of one kind or the other.
* Twelve percent of all the children are both stunted (chronically undernourished) and wasted (acutely undernourished), 52% are stunted but not wasted, and 5% are wasted but not stunted.
* Thirty three percent of all children are severely stunted (low height-for-age), and 31.2% are moderately stunted. The total prevalence of stunting is 64.2%.
* The prevalence of severe underweight (low weight-for-age) is 24.9% and that of moderate underweight is 43.4%.
* The prevalence of severe wasting (low weight for existing height) is 1 .7% and that of moderate wasting is 15%. The total prevalence of wasting is 16.7%
* Compared to urban children the prevalence of malnutrition is higher in rural children. The prevalence of stunting, wasting and underweight are respectively 13%, 1.8% and 12.6% higher among rural children.
* Within the urban areas slum children are worse off than the non-slum children. About 80% of slum children are chronically undernourished (stunted) as against 50.6% of non-slum children.
* Rural-urban differences and urban slum and non-slum differences in the prevalence of malnutrition are statistically significant.
* The prevalence of malnutrition is higher among the female children. Chronic undernutrition
(stunting) and current undernutrition (low weight for age) are respectively 2.8% and 5.4% higher among females. The gender difference in the prevalence of stunting and underweight is statistically significant. Due to higher prevalence of chronic under-nutrition among the female children the gender difference in acute undernutrition (wasting due to low, weight for existing height) is less apparent.
* Seasonal variations in the prevalence of malnutrition are apparent. Higher prevalence of weight dificiericy in April-May is followed by height dificiency in August. Seasonal differences in the prevalence of wasting (low weight for existing height) and underweight (low weight for age) are statistically significant, but the apparent seasonal difference in stunting (chronic undernutrition) is not. This is obvious because stunting is the result of long-term undernutrition rather than the effect of short term seasonal fluctuations.
Nutritional Status of individual children was analyzed by height-for-age, weight-for-age and weight-for-height SD scores of standard values (NCHS reference population) and also as percentages of respective standards (reference medians). The important findings of the study are:
Nutrition Situation
* The mean height of Bangladeshi children aged 6-71. months is 85.58 cm and the mean weight is 10.86 kg, which are respectively 90.57% and 75.24% of standard values (NCHS medians).
* Taking both height-for-age and weight-for-height into consideration, only 31% of all children between 6 and 71 months of age could be classified as "not malnourished0. The remaining 69% of the children are victims of malnutrition of one kind or the other.
* Twelve percent of all the children are both stunted (chronically undernourished) and wasted (acutely undernourished), 52% are stunted but not wasted, and 5% are wasted but not stunted.
* Thirty three percent of all children are severely stunted (low height-for-age), and 31.2% are moderately stunted. The total prevalence of stunting is 64.2%.
* The prevalence of severe underweight (low weight-for-age) is 24.9% and that of moderate underweight is 43.4%.
* The prevalence of severe wasting (low weight for existing height) is 1 .7% and that of moderate wasting is 15%. The total prevalence of wasting is 16.7%
* Compared to urban children the prevalence of malnutrition is higher in rural children. The prevalence of stunting, wasting and underweight are respectively 13%, 1.8% and 12.6% higher among rural children.
* Within the urban areas slum children are worse off than the non-slum children. About 80% of slum children are chronically undernourished (stunted) as against 50.6% of non-slum children.
* Rural-urban differences and urban slum and non-slum differences in the prevalence of malnutrition are statistically significant.
* The prevalence of malnutrition is higher among the female children. Chronic undernutrition
(stunting) and current undernutrition (low weight for age) are respectively 2.8% and 5.4% higher among females. The gender difference in the prevalence of stunting and underweight is statistically significant. Due to higher prevalence of chronic under-nutrition among the female children the gender difference in acute undernutrition (wasting due to low, weight for existing height) is less apparent.
* Seasonal variations in the prevalence of malnutrition are apparent. Higher prevalence of weight dificiericy in April-May is followed by height dificiency in August. Seasonal differences in the prevalence of wasting (low weight for existing height) and underweight (low weight for age) are statistically significant, but the apparent seasonal difference in stunting (chronic undernutrition) is not. This is obvious because stunting is the result of long-term undernutrition rather than the effect of short term seasonal fluctuations.
Data Collection
Dates of Data Collection (YYYY/MM/DD)
| Start date | End date |
|---|---|
| 1995 | 1996 |
Mode of data collection
Face-to-face [f2f]
Supervision
Training of survey personnel was conducted at three stages. At the first stage, all Regional Statistical Officers (RSO) were given two days training at Dhaka head Office who were acted as regional coordinators during field data collection. At the second stage four days training before each of four round data collection was imparted to 16 master Trainers/HQ supervisors at Dhaka Head Office. All the master trainers/HQ supervisors were the core staff of CNS project. The master trainers, in term, imparted two days training to the local enumerators and team leaders at Regional Statistical Offices at the third stage. Both the first and second stage training were imparted by the Project Director and Survey coordinators. The training course included class room training devoted to questionnaire and data collection procedures, age assessment technique of survey child and practice of anthropometric measurement of children. All trainees had to undertake filling up of questionnaires including getting anthropometry of children through field visit.fohlowed by discussion and evaluation on field, work. The field/training manual was made available to each enumerator and team leader so that they could consult it for any problem faced during field enumeration.
Type of Research Instrument
The draft questionnaire for the child nutrition survey was designed on the basis of objectives of the survey. The main objective was to finalize the questionnaire and develop instruction and training manual, control forms, survey procedures etc. Two field pretests on draft Questionnaire were carried out, one in Dhaka city area and other in rural area of Dhamrai Thana. The pretest on questionnaire was done in order to test the suitability of the survey questionnaire and to estimate the time required for filling of the questionnaire.
The pretested questionnaires were thoroughly reviewed and analyzed and placed in the meeting of Technical Committee. The Technical Committee suggested certain change in question and approved the questionnaire. Then the survey questionnaire, training manual and control forms were finalized incOrporating the modification as suggested by the Technical Committee.
The pretested questionnaires were thoroughly reviewed and analyzed and placed in the meeting of Technical Committee. The Technical Committee suggested certain change in question and approved the questionnaire. Then the survey questionnaire, training manual and control forms were finalized incOrporating the modification as suggested by the Technical Committee.
Data Collectors
| Name | Abbreviation | Affiliation |
|---|---|---|
| Statistics Division | SD | Ministry of Planning |
Data Processing
Cleaning Operations
Field editing
The field supervisors collected the completed questionnaires from the local enumerators and checked them for any inconsistencies before departing from the PSU. The field supervisors, in turn, deposited the questionnaires to the respective Regional Statistical Officers (RSOs). The RSOs checked the questionnaires for the second time and asked the field supervisors to ,re-check doubtful figures. Finally the questionnaires were returned to the central office.
Editing at BBS
At the central office the completed questionnaires were again thoroughly checked and edited by the HQ staff. After.data entry computer print outs were compared with the questionnaires by the HQ staff.
Computer analysis
Data were finally analyzed using dBase. Anthrb and SPSS software packages. Cross tabulations were created from the data where the unit of analysis was the survey child. Regression analyses were also done using SPSSPC+ software package.
The field supervisors collected the completed questionnaires from the local enumerators and checked them for any inconsistencies before departing from the PSU. The field supervisors, in turn, deposited the questionnaires to the respective Regional Statistical Officers (RSOs). The RSOs checked the questionnaires for the second time and asked the field supervisors to ,re-check doubtful figures. Finally the questionnaires were returned to the central office.
Editing at BBS
At the central office the completed questionnaires were again thoroughly checked and edited by the HQ staff. After.data entry computer print outs were compared with the questionnaires by the HQ staff.
Computer analysis
Data were finally analyzed using dBase. Anthrb and SPSS software packages. Cross tabulations were created from the data where the unit of analysis was the survey child. Regression analyses were also done using SPSSPC+ software package.
Data Appraisal
Data Appraisal
The objectives of the 1995-96 child nutrition survey are same as those of the three earlier surveys, which are reproduced below for completeness of the report:
1. To document a consistent national level time series data on nutritional status of children in Bangladesh.
2. To identify the nutritional status of children aged 6-71 months by age, gender, residence, season and other relevant 'factors\ and characteristics.
3. To identify the relationship of common factors with the nutritional status of children and to identify the determinants of malnutrition in the country.
4. To highlight issues which the policy ii akers and programme managers need to address in order to improve the nutrition situation of children in the country.
1. To document a consistent national level time series data on nutritional status of children in Bangladesh.
2. To identify the nutritional status of children aged 6-71 months by age, gender, residence, season and other relevant 'factors\ and characteristics.
3. To identify the relationship of common factors with the nutritional status of children and to identify the determinants of malnutrition in the country.
4. To highlight issues which the policy ii akers and programme managers need to address in order to improve the nutrition situation of children in the country.
Metadata production
Document ID
DDI-BBS-CNS-1995-96-v01
Producers
| Name | Abbreviation | Affiliation | Role |
|---|---|---|---|
| BANGLADESH BUREAU OF STATISTICS | BBS | Statistics Division, Ministry of Planning | Documentation of the study |
Date of Production
2019-09-01