Gaps in the implementation and uptake of maternal nutrition interventions in antenatal care services in Bangladesh, Burkina Faso, Ethiopia and India

Abstract Antenatal care (ANC) is the largest health platform globally for delivering maternal nutrition interventions (MNIs) to pregnant women. Yet, large missed opportunities remain in nutrition service delivery. This paper examines how well evidence‐based MNIs were incorporated in national policies and programs in Bangladesh, Burkina Faso, Ethiopia and India. We compared the nutrition content of ANC protocols against global recommendations. We used survey data to elucidate the coverage of micronutrient supplementation, weight gain monitoring, dietary and breastfeeding counselling. We reviewed literature, formative research and program assessments to identify barriers and enabling factors of service provision and maternal nutrition practices. Nutrition information in national policies and protocols was often fragmented, incomplete and did not consistently follow global recommendations. Nationally representative data on MNIs in ANC was inadequate, except for iron and folic acid supplementation. Coverage data from subnational surveys showed similar patterns of strengths and weaknesses. MNI coverage was consistently lower than ANC coverage with the lowest coverage of weight gain monitoring and variable coverage of dietary and breastfeeding counselling. Key common factors associated with coverage were micronutrient supply disruptions; suboptimal counselling on maternal diet, weight gain, and breastfeeding; and limited or no record keeping. Adherence of women to micronutrient supplementation and dietary recommendations was low and associated with late and too few ANC contacts, poor maternal knowledge and self‐efficacy, and insufficient family and community support. Models of comprehensive nutrition protocols and health systems that deliver maternal nutrition services in ANC are urgently needed along with national data systems to track progress.


| INTRODUCTION
Maternal nutrition is associated with women's health and well-being, newborn survival, and child growth and development . Despite improvements in our understanding of the problem, poor nutrition among women remains a significant global concern (Victora et al., 2021). The prevalence of low body mass index (BMI) among women has declined globally, but is still high in many low-and middle-income countries (LMICs), particularly in South Asia (at 24%) and parts of sub-Saharan Africa (at 20%) (Victora et al., 2021). The prevalence of anaemia is unacceptably high, affecting 30% of women of reproductive age (equivalent to over half a billion women aged 15-49 years), and 36.5% of pregnant women (WHO, 2021). Presence of both undernutrition and overweight and obesity in women is becoming more prevalent in some countries in Asia and sub-Saharan Africa (Popkin et al., 2020).
Evidence reported in the Lancet nutrition series in 2008 (Bhutta et al., 2008), 2013  and reiterated in 2021  suggests that a package of essential nutrition interventions including micronutrient supplementation and nutrition counselling, delivered with high quality through antenatal care (ANC), can contribute to improving maternal nutrition. Findings from 81 LMICs estimate that if health systems could effectively deliver a subset of evidence-based interventions to mothers and their newborns who are already seeking care, there would be a 28% decline in maternal and neonatal deaths, and 22% decrease in stillbirths (Chou et al., 2019). Despite this evidence, the coverage of nutrition interventions is far lower than the reach of their associated health service platforms . In India (UP state), only half of pregnant women who attended ANC reported consuming iron and folic acid (IFA), and substantially fewer women initiated breastfeeding within one hour after delivery, as compared with the proportion of women who delivered in health facilities (Nguyen et al., 2021).
Reducing missed opportunities for nutrition in health services offers high potential payoffs for improving maternal and infant nutrition (Heidkamp et al., 2021).
Understanding why, where and when the gaps occur in delivering nutrition interventions through ANC is an important next step for reducing missed opportunities. Before the publication of the WHO ANC guidelines in 2016, nutrition interventions received little priority in ANC services. Thus, there is scarce evidence on the content, quality, and factors associated with maternal nutrition interventions (MNIs) during ANC (Benova et al., 2018;Torlesse et al., 2021), other than IFA supplementation (Kavle & Landry, 2018;Sanghvi et al., 2010;Sununtnasuk et al., 2016). This paper provides details on the package of four MNIs as part of ANC in two sub-Saharan and two South Asian countries (Bangladesh, Burkina Faso, Ethiopia and India). Specifically, we (1) elucidate the content of national policies and protocols on micronutrient supplementation, weight gain monitoring, and counselling on maternal diet and breastfeeding, (2) examine service delivery gaps in MNIs during ANC, and (3) identify factors associated with delivery of MNIs and maternal nutrition practices.

Key points
• Delivering nutrition services to all pregnant women is essential for maternal and child health outcomes but remains a challenge.
• Nutrition gaps in antenatal care (ANC) include lack of specificity in national guidelines and protocols, bottlenecks in micronutrient supplies, low ANC provider knowledge and skills, inadequate supervision to reinforce counseling, and not engaging families to encourage key practices.
• National protocols for ANC should be more specific for the four nutrition interventions (micronutrient supplements, weight gain monitoring, counseling on diets and counseling on breastfeeding) and assign accountability for coverage and quality.
• Country models are needed for improving provision and utilisation of nutrition interventions through ANC that are based on comprehensive policy frameworks.

| METHODS
We used mixed methods for the study. First, we conducted a desk review of maternal nutrition program and policy documents including ANC guidelines and protocols to identify nutrition contents in each country and compared these contents with the WHO ANC recommendations (WHO, 2016). Then we conducted secondary data analysis of national and subnational surveys to determine coverage of services delivered by ANC providers and nutrition practices among pregnant women. Finally, we reviewed data from multiple sources (peer-reviewed publications, formative research reports and notes of stakeholder dialogues) to identify barriers and enabling factors that helped to explain the nutrition service delivery and maternal nutrition practices.  (EPHI & ICF, 2019;IIPS, 2017;INDS & ICF, 2012;NIPORT, 2020), and availability of recent information on the four priority MNIs (Kim, Ouédraogo, et al., 2020;Nguyen et al., 2015Nguyen et al., , 2018 (Table S1). The interventions were based on the WHO ANC guidelines (WHO, 2016) and endorsed by the Ministries of Health, including micronutrient supplementation, weight gain monitoring, dietary counselling, and counselling on breastfeeding.
Based on the WHO building blocks for health systems strengthening (WHO, 2010), we expected that maternal nutrition policies and protocols combined with the availability of supplies, skilled providers, supervision, monitoring, and utilisation of ANC services would result in a high level of service delivery and utilisation of nutrition interventions by pregnant women. Building on our previous work on infant and young child feeding interventions (Menon, Nguyen, Saha, Khaled, Kennedy, et al., 2016;Sanghvi et al., 2016), we expected that maternal nutrition practices would likely be influenced by individual and household factors including knowledge, self-efficacy, beliefs, access, and family support. Additionally, important community-level factors might influence the extent to which pregnant women adopt recommended practices include food taboos or misconceptions on diets (Lakshmi, 2013;Zerfu et al., 2016) and strong gendered social norms with low priority given to women's nutrition (ROSHNI, 2019).

| Policies and protocols
The Maternal Health and Nutrition Departments of the Ministries of Health provided the latest government policies, guidelines, protocols and directives for nutrition interventions in ANC. We consulted staff from the Ministries of Health to clarify terminology, overlapping or conflicting or missing content in policies and protocols. For information on the modes of service delivery for each intervention, we consulted government staff and other stakeholders working on ANC at the central and district levels.
We used a detailed checklist (Table 1) to compare the content of policies and protocols to global recommendations. For micronutrient supplementation, we reviewed the guidelines related to the types of micronutrients (IFA and/or calcium), the dose, the frequency of taking supplements (once or twice daily), and duration of supplementation. For weight gain monitoring and counselling, we reviewed guidelines on weight gain calculation and counselling. For dietary counselling, we reviewed content related to quantity and diversity of the diet during pregnancy. Lastly, we reviewed the counselling guidelines for preparing pregnant women for initiation of breastfeeding after birth and exclusive breastfeeding. For each service listed above, we also documented the modes of service delivery including types of service providers, locations of services (health facility or health service outreach sites in the community or home visits), timing and frequency of services, and ways to organise services (interpersonal or group, or community events). The information was organised by intervention for each country. Additional information was available from subnational surveys conducted by the International Food Policy Research Institute (IFPRI) as baselines for Alive & Thrive's MNI evaluations in the four countries (Kim, Ouédraogo, et al., 2020;Nguyen et al., 2015Nguyen et al., , 2018. These surveys used standardised methods and were comparable across countries.

| Nutrition services delivered in ANC and maternal nutrition practices
Indicators for coverage of ANC and nutrition service delivery used self-reported data based on interviews with women who received services during pregnancy for their last child in the past five years (national surveys) or in the past 6 months (subnational surveys). We first extracted data on the following indicators: percentage of women who attended any ANC from a skilled provider, completed four or more ANC visits, and started ANC in the first trimester of pregnancy. We then compiled information related to MNIs including percentage of women who consumed any IFA and received counselling on: IFA supplements, weight gain, diet, and breastfeeding. with either nutrition service delivery or maternal nutrition practices related to at least one of the four interventions. The information was organised by author, location, study design and approach, intervention(s) addressed, and key findings.

| Ethical considerations
Ethical approval was not required for this study as anonymized data used for secondary data analysis were from existing national and subnational surveys, which all had prior ethical clearance from international and national institutional review boards.

| National policies and guidelines on MNIs
Information in national protocols and policies on key nutrition interventions was often not available in a consolidated form but fragmented and scattered across several documents (

Content of protocols and programs
Breastfeeding counselling

Protocols
Counselling on the importance of early initiation within 1 h and how to place the newborn on the chest with the support of a health provider, preventing and managing common difficulties Counselling on the importance of six months of exclusive breastfeeding, skills (position, attachment, manual expression) and how to address common breastfeeding difficulties Source documents for Table 2

| Coverage of ANC services and delivery of nutrition interventions
At the national level, coverage of at least one ANC visit is high (from 74% in Ethiopia to 95% in Burkina Faso), however, the proportion of pregnant women in the four countries who sought ANC in the first trimester or completed at least 4 ANC visits was lower ( Figure 1). Data is available for coverage of IFA supplementation, but not for coverage of counselling on IFA, weight gain, diet, and breastfeeding. Subnational data showed gaps between ANC coverage and nutrition service delivery, indicating missed opportunities for delivering nutrition interventions to pregnant women during ANC visits (Figure 2).

| Dietary counselling
Coverage of dietary counselling also varied by country according to subnational surveys, ranging from 21% in India to 60% in Ethiopia.
More women received ANC than those who received counselling on dietary diversity; these missed opportunities ranged from 39 pp in India to 71 pp in Burkina Faso.

| Breastfeeding counselling
According to subnational surveys, low proportions of pregnant women were counselled on early initiation of breastfeeding: 12% in Burkina Faso and 9% in India. These proportions are higher in Ethiopia (50%) and Bangladesh (54%), but still suboptimal. Access to free supplies of IFA and calcium was enabling, side effects were a concern for some Maternal diets were influenced by misperception that dietary diversity meant only costly foods, gender inequities and social norms (women expected to sacrifice their own wellbeing for the good of the family), husbands were willing and able to help, and family budgets were amenable to change with counselling (benefits for the child emphasised, specifying affordable foods)

| Factors associated with delivery and utilisation of MNIs
Nguyen et al. (2017) (Nguyen et al., 2017. These enabling factors were also identified in formative research studies in Burkina Faso (Kere, 2020) and Ethiopia (Siekmans et al., 2018). Among barriers, bottlenecks in the supply chain affected women's ability to procure IFA and complete the protocol of 180 IFA tablets. Limited knowledge of pregnant women was found to be a common barrier including not knowing the importance of completing six months or 180 tablets during pregnancy, not understanding why healthy PW should consume the tablets in Ethiopia (Siekmans et al., 2018). Side effects and forgetfulness were barriers in Bangladesh (Schuler, 2015) and India (CMS, 2015

| Dietary counselling and dietary intake
Key factors associated with better dietary practices of pregnant women were maternal knowledge, confidence, self-efficacy and beliefs (Nguyen et al., 2017

| Breastfeeding counselling
Early start of ANC and frequent ANC visits by pregnant women were found to be associated with better breastfeeding practices. Mothers' perception that "milk doesn't form that early" led to lack of confidence in early initiation according to qualitative studies in Burkina Faso (Ky-Zerbo et al., 2019). In addition to pregnant women, family members needed to be guided by ANC providers before the birth of the child in understanding why and how women should be supported to initiate on time. Obtaining family support can improve the effectiveness of counselling, when family members help to remind the PW and reinforce the practices recommended by service providers (Ghosh et al., 2020).

| DISCUSSION
Although maternal nutrition is considered essential for achieving sustainable development goals (Victora et al., 2021) and ANC is the primary service platform for delivering MNIs, coverage of MNIs provided through ANC is low and many missed opportunities remain.
Our paper examines why, where and when the gaps occur and highlights common strengths and weaknesses in the policies and programs related to MNIs delivered through ANC. Our findings also provide critical insights on barriers and enabling factors at individual, family, community and health system levels that are associated with either the delivery of nutrition interventions in ANC or related maternal nutrition practices followed by pregnant women or both.
Policies and protocols examined in our study at the global and national levels had several gaps, including the lack of specificity on: (1) counselling on adherence, record keeping and use of data for micronutrition supplementation; (2) adapting recommended weight gain ranges to local contexts and counselling on optimal weight gain; (3) determining contextualized dietary counselling content and quality; and (4) content and modality for breastfeeding counselling.
Key MNIs were mentioned in national ANC guidelines. However, the content of guidelines and protocols were not consolidated and easily accessible, sometimes lacked specificity and did not con- Our study identified substantial data gaps on coverage and quality of maternal nutrition services, which is consistent with previous findings in the Asia region (Torlesse et al., 2021). National surveys and routine monitoring data systems lacked process indicators needed for tracking nutrition interventions in ANC. Our findings align with global concerns on limited process monitoring indicators (Choufani et al., 2020;Gillespie et al., 2019;Heidkamp et al., 2021). For micronutrients, we found that the number of doses distributed per woman and counseling provided are not routinely recorded. Most women are weighed during ANC visits according to survey data and formative research, but weight gain and weight gain counseling are not recorded. Other indicators such as counseling on maternal diet, breastfeeding counseling and obtaining the support of family members, are also missing. The lack of data highlights an apparent lack of priority and accountability for maternal nutrition in health services. Efficient planning and targeting of scarce resources cannot be achieved without data. Previous evidence has shown that ongoing cycles of problem-solving based on data and adjustments made in program delivery can improve worker performance (Bootwala, 2020). Accountability is essential to ensure that global and country actions are delivered with quality and equity to make a difference.
Our study highlights multiple barriers and enabling factors that influence the utilisation of MNIs; addressing these should be a priority of the health system. At scale programs face supply-side health system level challenges such as micronutrient stock-outs that can be anticipated and pre-empted. Additional challenges at the system level include gaps in information systems and performance of vital tasks, such as counseling and community engagement for encouraging and facilitating utilisation of the key interventions.
Demand-related factors common across countries include late and infrequent care-seeking for ANC, and poor adherence associated with gaps in knowledge, beliefs, self-efficacy and family or community support. We found considerable variation in completion of at least four ANC visits, and this influenced the coverage and utilisation of nutrition interventions (Nguyen et al., 2017. Several nutrition interventions require sustained action at the household level, starting early in pregnancy and can be enabled or impeded by family members and social norms (Martin et al., 2020). Combining individual counseling with community-based activities for involving husbands and other family members of pregnant women has shown impact on improving dietary diversity and household food security (Frongillo et al., 2019). Adherence to recommended maternal nutrition practices requires timely care-seeking to obtain adequate supplies and reinforcement; repeated counseling throughout pregnancy to problem-solve, build skills and confidence of PW; and active community-level engagement for obtaining family support.
Our study is among the first to document current gaps in the content of a package of nutrition interventions within ANC protocols and existing health systems, thus providing critical information for filling missed opportunities in ANC. Using mixed methods including review of policies, programs and relevant literature, compiled data from national and subnational levels, and findings of formative research and stakeholder workshops, we offer rich evidence and suggest solutions to address bottlenecks in delivering nutrition interventions through ANC. We suggest how to begin to facilitate scale up of four proven interventions that provide substantial payoffs (WHO, 2016). Limitations of this study include the lack of nationally representative data on coverage and variations in sources of information.

| CONCLUSION
Missed opportunities to address maternal malnutrition through health services remain widespread despite rising ANC coverage and evidence of negative impacts on maternal and child health outcomes.
National policies acknowledge the need for implementing four priority nutrition interventions but do not provide specific guidance on or accountability for delivering them. Our review provides insights on where, when and why health system gaps are likely to occur, and key barriers and enabling factors associated with coverage and utilisation of MNIs. Attention is needed to strengthen micronutrient supply chains, establishment and use of information systems for accountability and performance improvement, enhance ANC provider knowledge and skills to deliver tailored counseling, and develop strategies for effectively engaging with families. ANC and nutrition departments within ministries of health need to fill policy gaps and develop country models for improving the provision and utilisation of nutrition interventions through ANC.