Study design and participants
A community based longitudinal (controlled before-after) effectiveness study was conducted in 10 upzilas/subdistricts of 5 districts of Bangladesh to evaluate the intervention provided by the WFP on the VGD beneficiaries and compared the outcomes at two-time points; baseline and end line periods. The sampling was done among the VGD beneficiaries who received either fortified rice in 5 Upazilas namely Kaligonj, Sarankhola, Tungipara, Dacope and Shyamnagar in the FFR group and unfortified rice rations in the non-FFR group. The 5 FFR Upazilas were selected by the World Food Programme from 5 districts in different geographic locations across the country. The non-FFR Upazilas were selected from the same districts with similar socio-economic backgrounds. A systematic random sampling method was employed to enrol the required number of participants for the baseline and end-line surveys from the total list of VGD women in both the FFR and non-FFR Upazilas. Participants for the FFR group were drawn from the total list of approximately 15,000 VGD beneficiaries from 40 unions under the 5 Upazilas. Similarly, participants for the non-FFR group were selected from enlisted approximately 15,000 VGD women from 53 unions of the 5 Upazilas. During the endline evaluation, a similar sampling approach was employed, and participants were allocated to FFR and a non-FFR group from the same sampling frame. However, the participants of the baseline and end line surveys were different. Baseline data collection was commenced from December 2014 to April 2015. After the baseline data collection, the rice distribution was not immediately initiated. Due to delays in fortified rice production, the onset of the intervention was also deferred for around 12 months. After 12 months of FFR/non-FFR consumption, the endline data was collected from December 2016 to April 2017.
The micronutrient composition of fortified rice
The production of fortified rice in this project took place in two steps- i)the production of fortified rice kernels, which were made from cheaper rice flour mixed with micronutrients, reconstituted via hot extrusion technology, and ii) the homogeneous blending of fortified rice with un-fortified rice, usually at a 1:100 ratio. The estimated cost implication is an additional 3–5 per cent at the retail level when compared with un-fortified rice. The micronutrient fortificants used to fortify the rice includes Vitamin A, Vitamin B1, Vitamin B12, folic acid, iron, and zinc.
Inclusion/Exclusion criteria
Predefined inclusion and exclusion criteria were employed during the enrolment of the VGD beneficiaries. The inclusion criteria include (i) women aged 15–49 years old, (ii) possession of VGD programme card and (iii) provision of a written consent agreement with their household head to participate in the study. The exclusion criteria were (i) known or suspected chronic or congenital disease, (ii) pregnancy and (iii) reported severe anaemia. The severe cases were suggested to visit the government health facility.
Each participating woman was asked about recent illnesses in the previous two weeks. Diarrhoea was defined as three or more abnormally loose or liquid stools without blood in the last 24 hours or any number of stools with blood (dysentery). Questions relating to menstrual problems, including the absence of periods, painful periods, heavy periods and irregular periods, were also included.
Collection, preparation, transport and storage of biological samples
Trained phlebotomists collected peripheral blood samples from survey respondents during both the baseline and endline evaluations. After taking consent and maintaining aseptic precautions, about 5 mL of venous blood was collected and aliquoted into appropriate tubes with or without anticoagulant. Samples were transported to the laboratory at icddr,b in Dhaka, Bangladesh, twice a week to assess concentrations of zinc, C-reactive protein (CRP) and haemoglobin. As most of the study sites were hard to reach, it was not possible to transfer the samples immediately to the icddr,b laboratory. The samples were first preserved in -20°C then transferred to icddr,b laboratory maintaining to the cold. All samples were transported to a nearby temporary laboratory set up at a clinical setting for temporary storage, before being transferred to icddr,b. Haemoglobin concentration in whole blood was measured using the cyanmethaemoglobin method. Serum samples were stored at -20°C until serum zinc was estimated by atomic absorption spectrophotometry and serum CRP was determined via an immunoturbidometric method using a Roche automated clinical chemistry analyser Hitachi 902.
Statistical analysis
Stata software (version 13; Stata Corporation, College Station, TX, USA) was used for all univariate and bivariate analyses. Frequencies and percentages (for categorical variables) or means and standard deviations (for continuous variables) were calculated for descriptive statistics. The Student’s t-test and the chi-square test were applied to compare means and to explore the associations between categorical variables respectively. P-value <0.05 was considered significant for all tests. The difference-in-difference (DID) analysis was performed to measure the effect of the fortified rice distribution programme on anaemia and serum zinc concentration. Multivariable logistic regression was done by using the stepwise backward method to determine the factors significantly associated with anaemia and zinc deficiency.