Prevention of Low Birthweight Infants Among Pregnant Women in Rural Bangladesh: A Cluster Randomized Controlled Trial

The low birthweight (LBW) prevalence in Bangladesh varies between 22-50%, which positioned Bangladesh 4th globally among the countries having highest burden of LBW. LBW imposes greater risk of mortality and morbidities among children under five and creates a long-term negative impact on development and wellbeing in adolescence and adulthood. Therefore, the objective of the research was to investigate the effects of ‘balanced plate nutrition education’ (nutrition education on balanced diet with practical demonstration of balanced plate) in combination with engagement of family decision makers, to reduce incidences of LBW infants among pregnant women by increasing birthweight. The specific objectives included developing the ‘balanced plate nutrition education’ intervention for rural pregnant women and measuring the effect of the intervention on birthweight of infants and incidences of LBW, and exploring the barriers of and household coping strategies related to compliance. It also examined the association of household food insecurity and size of infants at birth. Both quantitative and qualitative methods were applied for this research. For the quantitative analysis, primary data were derived from a randomized controlled trial conducted in rural Bangladesh (Chapter 4) and secondary data from Bangladesh Demographic and Health Survey 2011 (Chapter 6). Statistical analyses were performed using multivariable linear and binomial regression with log link function. We adjusted for the clustered randomisation using generalised estimating equations (GEE). We constructed survey-weighted logistic regression models for BDHS data to account for different sampling probabilities and different response rate. A qualitative study was conducted in the trial area among the balanced plate nutrition education intervention recipients. We conducted in-depth interviews (n=10) with mothers of infants (0-6 months), focus group discussions (n=2) with their husbands and older women in the family and key informant interviews (n=4) with Shasthya Kormi (community health workers) of BRAC (an NGO in Bangladesh, formerly known as Bangladesh Rural Advancement Committee). Interviews were audio-recorded during collection, later transcribed in Bangla, and translated into English. The transcripts were manually coded and analysed using the thematic approach. Chapter 4 demonstrated that pregnant women who received balanced plate nutrition education had heavier infants compared to those received standard nutrition education. The incidence of low birthweight (LBW) was also lower among women in the balanced plate group compared to the latter. The mean birthweight increased by 125.3 g (95% confidence interval (CI) 5·7, 244·9; p=0·04) and the risk of LBW was reduced by 54% (relative risk (RR) 0·46; 95% CI 0·28, 0·78; p=0·004) in the intervention compared to the comparison group. The effect of intervention was greater among adolescent mothers in terms of birthweight and incidence of LBW than the non-adolescent mothers mean difference 297·3 g; 95% CI 85·0, 509·6; p=0·006 and RR 0·31; 95% CI 0·12, 0·77; p=0·01). Chapter 5 exhibited that accessing animal source food was the greatest barrier in practicing balanced diet. Perceived gap in understanding appropriate portion size and importance of diversified food for a pregnant woman were the other reported barriers. Mothers-in-law’s authority and control over pregnant women’s diet led to intrahousehold food mal-distribution with less nutritious food share for them. Active engagement of the family decision makers (husbands and elderly women in the family) in the nutrition counseling and demonstration session created an agreement on balanced diet for pregnant women. Husbands were inclined to finance more for purchasing nutritious foods such as cheap fishes, milk and fruits. Women were selfmotivated to increase consumption of vegetables and animal source foods with right proportion of rice. Mothers-in-law’s approval and husbands’ voluntary contribution enabled women to practice the balanced diet. Chapter 6 showed an inverse association between household food insecurity and perceived birth size of infants in Bangladesh. Infants from food insecure household were 36% more at risk of being small at birth compared to infants born in food secure households, which was aggravated by less utilization of antenatal care and first birth. There was an obvious regional variation of prevalence of smaller infants in geographically hard to reach areas; women in Sylhet and Chittagong districts were more vulnerable to give birth to smaller infants than mothers living in Barisal. In conclusion, balanced plate nutrition education in pregnancy impacted on the birthweight of infants and incidence of LBW in rural Bangladesh. Practical demonstration of making balanced plate in combination with family engagement can create an enabling environment for pregnant women to adopt a balanced diet with self-motivation. Household food insecurity is major driver in determining fetal growth and subsequent size of infants at birth. The insights from this research will help to design nutrition behaviour change communications for pregnant women and target household with greatest need to improve perinatal nutrition for better child survival, growth, development and productivity in Bangladesh and other LMICs