How does mode of delivery associate with double burden of malnutrition among mother–child dyads?: a trend analysis using Bangladesh demographic health surveys | BMC Public Health


For the purpose of analysis, data were extracted from the last four Bangladesh Demographic Health Survey (BDHS) conducted in 2007, 2011, 2014 and 2017 and then combined. BDHS survey is a nationally representative survey that collects current information on the major indicators of maternal and child health-related issues. The survey was implemented by the National Institute of Population Research and Training (NIPORT), Health Education and Family Welfare Division of the Ministry of Health and Family Welfare. United States Agency for International Development (USAID) provided financial assistance in conducting the survey [19,20,21,22].

BDHS follows two stage stratified sampling plan where the enumeration areas from the Population and Housing Census of the People’s Republic of Bangladesh, provided by the Bangladesh Bureau of Statistics are considered as the primary sampling unit and a systemic sample of households within the survey is counted as the secondary sampling unit. The ever-married women in reproductive age are interviewed from the selected households in the sample for necessary information regarding maternal and child health indicators. The anthropometry measures of the respondents and their children under the age of five years are collected in these surveys [19,20,21,22]. The interviewers used lightweight SECA scale with a digital screen manufactured under the authority of UNICEF for measuring the weight. The height was measured by height boards specially produced by Shorr Production according to study settings. Recumbent length for children less than 2 years and standing height for the elder children are recorded in the survey. The detail of the survey methodology can be found elsewhere [19,20,21,22].

The study considered the mother–child pairs from four waves of BDHS survey where mothers are non-pregnant women aged 15–49 years and they had children who were born in the last 3 years preceding the survey. The details of the number of cases considered in this study along with the criteria that result in the exclusion of cases are explained in Fig. 1. After considering all desired criteria, we included 14,975 mother–child pairs combining the aforementioned BDHS surveys.

Fig. 1
figure 1

Flow chart for sample size selection

Outcome variable

The binary outcome variable of interest in this study is the double burden of malnutrition (DBM) status at the household level defined considering the nutrition status of the mother and her child. The presence of DBM at the household level (taking value 1) arises if a mother is overweight or obese and her child is malnourished [5]. A mother is identified as overweight or obese if her BMI is 25 kg/m2 or more [23]. A child is considered to suffer from under-nutrition if s/he is stunted or wasted or underweight [24]. Stunting, wasting and underweight are assessed following the measurement of the WHO Child Growth (WHO) Standards reference population [25]. The definition of DBM is illustrated in Fig. 2.

Fig. 2
figure 2

Diagram for definition of DBM

Independent variables

Several covariates are included in the study based on the suggestions of the previous studies. The prime focus of this study is on the mode of delivery which is categorized as “C-section” and “normal” based on the procedure followed during the child’s birth. The other independent variables that are included in the study are current breastfeeding status (yes, no), division (Barisal, Chattogram, Dhaka, Khulna, Rajshahi, Sylhet), place of residence (rural, urban), wealth index (poor, middle, rich), media exposure (yes, no), mother’s education level (no education, primary, secondary, higher), father’s education level (no education, primary, secondary, higher), mother’s working status (yes, no), mother’s decision-making capacity (yes, no), attitude towards violence (yes, no), mother’s age (15 to 24, greater than 24), received antenatal care (ANC) (yes, no), wanted child (yes, no), initiation of breastfeeding (within 1 h, after 1 h), child’s sex (female, male), birth order (first birth, otherwise), child’s age (in months) (< 18, 18 +), vitamin A consumption (yes, no).

It is to be noted that some variables are reconstructed to ensure consistency among variables’ categories in four consecutive surveys. We considered the categories of division as mentioned in BDHS 2007 whereas BDHS 2011 and BDHS 2014 consist of 7 categories for division and BDHS 2017–2018 subclasses division into 8 categories. Therefore, we merged Rangpur with Rajshahi for BDHS 2007, BDHS 2011, BDHS 2014 and BDHS-2017–18 and Mymensingh with Dhaka for BDHS 2017–18. To create the variable named “attitude towards violence” we considered respondent’s attitude towards wife beating being justified if she (1) burns food (2) argues with her husband (3) goes out without telling him, (4) neglects the children, and (5) refuses to have sex with him for BDHS 2011, BDHS 2014 and BDHS 2017–18. BDHS 2007 recorded the responses only for the first four reasons. If the respondent justifies at least one reason, she is considered in “yes” category, otherwise “no”. A woman is categorized as “yes” in the context of women’s decision-making capacity if she participates any of the following decisions on (1) her own health care, (2) major household purchases, and (3) visits to her family or relatives. If a mother received antenatal care at least once during her pregnancy is coded as “yes” for the variable “received antenatal care (ANC)”, otherwise the individual is coded as “no”. For constructing the variable “media exposure”, the respondent is categorized as “yes” if she is in touch with any of the following media at least once in every week (1) newspaper/ magazine, (2) radio, (3) television.

Statistical methods

The study used the pooled dataset from the aforesaid BDHS surveys for examining the trend of DBM and the effect of mode of delivery on DBM by survey years. Firstly, the profile of background characteristics of the mother–child pairs is reported by survey years. Percentage distribution of DBM by several characteristics is depicted by survey years and latterly in the pooled dataset. Chi-square test is used in the pooled dataset to measure the association between DBM and the included independent variables. The independent variables that are significantly associated with DBM are considered further for regression analysis. Bivariate logistic regression analysis is used to pooled dataset to determine the adjusted association of covariates on DBM. The interaction effect between mode of delivery and survey year is allowed in regression model to assess the change in DBM at the household level by mode of delivery over time. The analysis is performed using SPSS version 22 and STATA statistical package version 12.

Source link

Stay in Touch

To follow the best weight loss journeys, success stories and inspirational interviews with the industry's top coaches and specialists. Start changing your life today!


Related Articles