Low coverage of effective nutrition interventions in many high-burden countries, due to service provision and demand factors, result in poor uptake of recommended practices and nutrition outcomes. We examined the factors that influence maternal nutrition and early breastfeeding practices and determined the extent that the key factors could improve these practices in two regions in Burkina Faso. We used household survey data among pregnant (n = 920) and recently delivered women (n = 1840). Multivariable regression analyses were conducted to identify the determinants of a diverse diet and iron-folic acid (IFA) supplement consumption, weight monitoring during pregnancy and early initiation of breastfeeding (EIBF). Population attributable risk analysis was used to estimate how much the outcomes can be improved under optimal conditions of interventions that address the modifiable determinants. During pregnancy, 21% of women achieved minimum diet diversity (MDD-W), 70% consumed 90+ IFA tablets and 65% were weighed 4+ times; EIBF was 40%. Nutrition knowledge was associated with MDD-W (odds ratio [OR]: 3.2), 90+ IFA (OR: 1.5) and EIBF (OR: 1.9). Positive social norms and family support were associated with 90+ IFA (OR: 1.5). Early and 4+ ANC visits were associated with 90+ IFA (OR: 1.5 and 10) and 4+ weight monitoring (OR: 6.2). Nutrition counselling was associated with 90+ IFA (OR: 2.5) and EIBF (OR: 1.5). Under optimal programme conditions, 41% of women would achieve MDD-W, 93% would consume 90+ IFA, 93% would be weighed 4+ times and 57% would practice EIBF. Strengthening the delivery and uptake of interventions targeted at these modifiable factors has the potential to improve maternal nutrition practices.
Alive & Thrive (A&T) is an initiative that supports the scaling up of nutrition interventions to save lives, prevent illnesses and contribute to healthy growth and development through improved maternal nutrition and infant and young child feeding practices in several countries. In Burkina Faso, A&T developed a set of interventions aimed at strengthening maternal nutrition services integrated into ANC provided through the government health system (Sanghvi et al., 2022). A&T aimed to test the feasibility of improving the provision and uptake of maternal nutrition interventions such as counselling on dietary diversity, adequate food intake, consumption of IFA supplements, adequate weight gain and early breastfeeding practices and community mobilization. This study used data from the baseline household survey conducted as part of the evaluation of A&T maternal nutrition interventions in two regions—Hauts‐Bassins and Boucle du Mouhoun, in Burkina Faso (registered at ClinicalTrials.Gov: {“type”:”clinical-trial”,”attrs”:{“text”:”NCT04155437″,”term_id”:”NCT04155437″}}NCT04155437). These two regions were selected with the Government of Burkina Faso, based on regional‐level engagement and ownership, presence of a cadre of community health workers, size of region and level of security. The survey was carried out in 80 health centre catchment areas in four health districts (Boromo, Toma, Dandé and Léna). Three villages were randomly selected within each health centre catchment area, and a census was conducted within each village; from the two census lists (for pregnant women and recently delivered women), women were selected by simple random sampling until the required sample sizes were reached. The sample included 920 pregnant women and 1840 recently delivered women with children under 6 months of age, to determine current dietary practices during pregnancy and service exposure throughout the last pregnancy and maternal nutrition and early breastfeeding practices respectively. Structured questionnaires were administrated face‐to‐face using computer‐assisted personal interviewing at the respondents’ homes by survey teams trained and supervised under AFRICSanté (Agence de Formation, de Recherche et d’Expertise en Santé pour l’Afrique) in November–December 2019. Four primary outcomes related to maternal nutrition practices were constructed: (1) minimum dietary diversity during pregnancy (five or more food groups); (2) consumption of at least 90 IFA tablets during the last pregnancy; (3) weight monitoring at least 4 times during the last pregnancy and (4) EIBF. A fifth outcome of exclusive breastfeeding (EBF) was also analyzed and presented in Supporting Information. Maternal dietary diversity during pregnancy was assessed among pregnant women using the individual report of foods consumed over a 24‐h recall period. These foods were grouped into 10 categories based on the minimum dietary diversity guidelines for women (MDD‐W) (FAO and FHI 360, 2016). A diet diversity score was calculated as the number of food groups consumed out of 10 total food groups, and the cut‐off of at least 5 food groups per day was used to define MDD‐W to achieve micronutrient needs. For the other outcomes, recently delivered women were asked and probed about the number of IFA tablets they consumed and the number of times they were weighed during their last pregnancy. For breastfeeding, women were asked how many hours/days after birth they started breastfeeding their child, and EIBF was defined as within 1 h after birth. EBF was assessed using a report of any foods or liquids fed to the child over a 24‐h recall period and defined as feeding the child no food or liquids other than breast milk in the past 24 h (WHO & UNICEF, 2021). The independent variables were identified based on a conceptual framework applied in a previous study where the predicted effects of the potentially modifiable determinants were assessed on IFA supplement consumption and dietary diversity outcomes (Nguyen et al., 2017). These determinants included maternal, household and health service factors. At the maternal level, knowledge scores were generated for dietary diversity, IFA, weight gain during pregnancy and breastfeeding. For knowledge of dietary diversity, women were asked to name at least five food groups, examples of locally available food rich in essential nutrients and the importance of food variety during pregnancy (eight question items). For IFA knowledge, recently delivered women were asked whether they heard of anaemia, its effects and causes, about the recommended numbers of IFA tablets per month and throughout pregnancy and benefits of IFA during pregnancy (5 question items). Knowledge of weight gain was assessed based on women’s knowledge of how much weight a pregnant woman should gain during pregnancy, where a response of 10–12 kg was scored as 1, that is, correct. For the knowledge of breastfeeding, women were asked the time after birth a baby should start breastfeeding, the reason a baby should breastfeed soon after birth, the benefits of colostrum, how long a baby should be exclusively breastfed and why and at what age a baby can receive liquid other than breast milk (16 question items). Each question item was scored as 0 or 1, and the sum represented the knowledge score. Belief, self‐efficacy and social norms related to maternal nutrition practices were measured on a 5‐point Likert scale by asking women about the extent to which they agreed or disagreed with statements. Belief and self‐efficacy statements asked women whether they believed that the recommended practices were beneficial and feasible to do, respectively; statements about social norms asked whether they perceived other women in their community were doing these practices. Each statement was given a score of 1 for strongly agree or agree and 0 for strongly disagree, disagree, neither agree nor disagree. The knowledge, belief, social norms and self‐efficacy scores were constructed using outcomes‐specific factors, scaled and ranged from 0 to 10 and then on divided into high and low categories with cut‐offs at the median, for the regression analyses. For household factors, support from husband and/or other family members was assessed by asking women whether husbands or other members helped to acquire diverse foods or IFA supplements, reminded them to consume them, monitored their weight and provided other support during pregnancy. Each statement was scored 1 for strongly agree or agree and 0 for strongly disagree, disagree, neither agree, nor disagree. The sum of scores was scaled and ranged from 0 to 10, and then divided into high and low categories with cut‐offs at the median. Measures of belief, self‐efficacy, perceived social norms and family support specifically related to breastfeeding were not collected due to study prioritization on maternal nutrition practices and limitations on length/duration of survey questionnaires. Health service factors included timing of first ANC visit; number of ANC visits; home visit by a community health agent; receipt of IFA supplement for free and receipt of counselling or provision of information on dietary diversity, IFA supplementation, weight gain and breastfeeding. Control variables in these analyses included maternal age, education as a binary variable (no schooling/koranic literacy training/not completed first grade vs. primary/secondary or higher), religion (Muslim vs. Catholic/Protestant/Traditional) and parity (0 vs. 1, 2, 3+), which are common factors that influence dietary and other maternal nutrition practices. Household wealth index was constructed using principal components analysis of variables on housing conditions and assets, and the first component was used to divide the score into terciles (low vs. middle, high) (Vyas & Kumaranayake, 2006). Household food security was measured using the FANTA/USAID Household Food Insecurity Access Scale (Coates et al., 2007) and treated as a binary variable (food secure vs mildly/moderately/severely food insecure). For analyses of breastfeeding outcome, we included delivery at a health facility and caesarean section as control variables, as these influence the capacity for early breastfeeding practices. Descriptive analysis was used to report the sample characteristics including the sociobehavioural factors. Bivariate analyses were used to test for associations between each potential determinant and the dependent variables. Multiple regression analyses were used to identify factors associated with MDD‐W during pregnancy (five or more food groups); consumption of at least 90 IFA tablets during the last pregnancy; weight monitoring at least 4 times during the last pregnancy; EIBF and EBF. Regression analyses were run adjusting for geographical clustering and control variables at maternal and household levels. Odds ratio (OR) with its 95% confidence interval was estimated for the logistic regression models. Population attributable risk is the proportion of the outcome in the population (exposed and unexposed) that is due to exposure. Thus, we used population attribute risk analysis (Newson, 2013) to estimate by how much our study outcomes can be improved under scenarios of either exposure to each modifiable determinant alone or in the combination of modifiable determinants, using the determinants identified from the regression results. From among the independent variables above, we considered modifiable factors as those that may be targeted and modified by programme interventions. Statistical significance levels at p < 0.05, p < 0.01 and p < 0.001 were used. All statistical analyses were performed using Stata version 17. Ethical approval was obtained from the Ethics Committee of Centre Muraz (Burkina Faso) and the Institutional Review Board of the International Food Policy Research Institute (USA). Written informed consent was obtained from all study participants.