Nutrition Interventions during Pregnancy and Maternal, Newborn and Child Health Outcomes
The global and serious problem of maternal undernutrition is often overlooked as an important cause of poor maternal, newborn and child health (MNCH) outcomes such as intrauterine growth restriction, preterm birth, and maternal and infant morbidity and mortality. The damaging effects of maternal undernutrition on both the mother and her child are especially prominent in low and middle income countries, where women often enter pregnancy undernourished and their nutritional deficiencies escalate during pregnancy. In consultation with Gates Foundation staff, researchers at Emory and Tulane Universities developed a project to assess the scientific evidence supporting nutrition interventions during pregnancy aimed at improving MNCH outcomes. Our project goal is to improve MNCH outcomes through better designed policies and programmes that enhance nutrition throughout the life cycle, with a particular focus on nutrition during pregnancy. The project has two objectives: (I) to assess the scientific evidence supporting nutrition interventions aimed at improving MNCH outcomes and to identify critical knowledge gaps and research needs, and (II) to examine the status of implementation and bottlenecks to expansion of coverage of effective interventions and to recommend how to overcome these supply and demand barriers. Articles in this supplement will focus on Objective I, with the main goals of: (1) presenting the scientific evidence linking nutrition interventions throughout the life cycle, and particularly during pregnancy, with MNCH outcomes; (2) identifying any critical knowledge gaps; (3) identifying priority research needed to bolster the evidence for promising new interventions; and (4) reviewing existing WHO recommendations and policies for improving maternal nutrition. A second supplement, summarising findings from Objective II, will be published in the Food and Nutrition Bulletin. Finally, a publication summarising the entire work, both the gaps in research and an agenda for programme implementation, will be prepared.
Selection of topics and outcomes
Review topics were selected based on several factors: (1) nutrients or interventions where biological plausibility of an influence on MNCH outcomes exists, or nutrients that are known to be biologically important in pregnancy (e.g. folate, iodine); (2) existing evidence from randomised controlled trials (RCTs) or non-experimental studies shows a potential influence of an intervention on MNCH outcomes (e.g. iron, calcium); (3) recommendations from our external Advisory Board, a group of experts in the field of maternal and child health and nutrition; and (4) knowledge of reviews already in progress. Several reviews were already being conducted by other groups when we selected our topics, and we therefore did not want to duplicate efforts. These topics, which are not included in our collection of reviews, include the impact of staple food fortification and of micronutrient powders on MNCH outcomes. Although the project generally focused on direct nutrition interventions, we included important non-nutritional interventions that potentially influence nutritional status and that may be included within the same programmatic structure as nutrition interventions (e.g. antihelminthics in pregnancy, inter-pregnancy interval and age at first childbirth). A range of interventions, such as conditional cash transfers, provision of anti-malarial medication in pregnancy and others, also potentially influence MNCH outcomes but were not included in order to limit the scope of the reviews.
Each review aimed to assess the following broad range of MNCH outcomes: maternal morbidity (e.g. pre-eclampsia, anaemia) and mortality, preterm birth, intrauterine growth restriction or small for gestational age, low birthweight, infant and child growth, and infant and child morbidity and mortality. For most interventions, data were not available for many of the listed outcomes; however, authors included each outcome in their initial literature search.
Description of articles
The 20 articles published in this supplement review the following topics: the biological mechanisms for nutritional regulation of maternal health and fetal development; supplementation with nutrients during pregnancy (iron + folic acid, zinc, calcium, iodine, vitamin A, vitamin D, vitamins B6, B12 and C, multiple micronutrients, iron versus multiple micronutrients, n-3 long-chain PUFA, and balanced protein energy supplementation); antihelminthics in pregnancy; interventions aimed at behaviour change such as nutrition education and counselling during pregnancy and household food production strategies; factors such as inter-pregnancy interval and age at first childbirth, intergenerational influences on child growth, and nutrition in adolescence and preconceptionally. The final article in the supplement describes current global policy and programme guidance on maternal nutrition.
Data from RCTs were not available for all topics, especially those where RCTs would not be possible (e.g. age at first childbirth). For some interventions where RCTs may be possible, such as antihelminthics in pregnancy, surprisingly, very few trials have been conducted. In cases where data from RCTs were lacking, authors reviewed evidence from high quality observational studies. Although RCTs are considered the ‘gold standard’ in epidemiological research addressing efficacy, well-designed and appropriately analysed observational studies can provide valuable evidence where evidence from RCTs is absent.
For each review article, authors abstracted individual study data from all trials and non-experimental studies that met the inclusion criteria. Data were abstracted into a standardised Excel data abstraction sheet created specifically for this project, and in many cases, >50% of the data were double-abstracted to ensure accuracy. The data abstraction sheets, which contain 49 potential variables, capture important information about each study, for example, sample size, study context, definition of the outcome, information about study quality (e.g. blinding), and effect estimates (Table S1).
The effect of interventions during pregnancy or of factors such as inter-pregnancy interval on MNCH outcomes was assessed by meta-analysis, when possible. Finally, the topic of intergenerational influences on child growth covered a broad subject matter and was written as a review.
Papers were written both by Emory authors involved in the project and by topic experts external to the project. Each article was reviewed by at least 2 peer reviewers, and papers written by Emory authors were reviewed by 2 or more peers unassociated with the project. We thank the external reviewers for providing rigorous reviews of the included papers. We assessed the quality of evidence of the influence of each intervention on the aforementioned MNCH outcomes based on a modified version of the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system.1,2 Dr Christa Fischer-Walker, from The Johns Hopkins Bloomberg School of Public Health, visited Emory and trained the Emory study team on GRADE methods used by the Child Health Epidemiology Reference Group (CHERG).3 We thank Dr Fischer-Walker for her contribution. Barbara Abu-Zeid, an Emory University reference librarian, conducted a training session for the Emory study team on how to conduct a systematic review, and we thank her for her guidance.
The collection of reviews published in this supplement aimed to (1) review the scientific evidence on the influence of nutrition interventions during pregnancy on MNCH outcomes and (2) identify research gaps in each subject area. In a future publication we will summarise in detail the gaps in research (Objective I) and programme implementation (Objective II) regarding maternal nutrition during pregnancy. There is much in research that needs to be addressed. Overall, the role of nutrition interventions during pregnancy in improving MNCH outcomes merits future study. For example, scientific evidence linking low maternal pre-pregnancy body mass index (underweight) with MNCH outcomes is lacking. Several interventions provided during pregnancy have beneficial effects on MNCH outcomes, but are currently not widely adopted. For example, although calcium supplementation in pregnancy decreases the risk of several poor maternal health outcomes, it is not a widely recommended intervention in public health, especially in low and middle income countries. Several other nutritional interventions have been proven efficacious in improving select MNCH outcomes, for example protein energy supplementation and iron + folic acid supplementation during pregnancy; however, variable programmatic experiences have lead to questionable effectiveness of these interventions. Finally, the safety of interventions such as multiple micronutrient supplementation and vitamin C + E supplementation in pregnancy merits future study. Ramakrishnan et al. report in their review that newborns born to women who began multiple micronutrient supplementation after the first trimester of pregnancy experienced a higher risk of neonatal mortality. Likewise, Dror and Allen report an increased risk of pregnancy-related hypertension in women supplemented with vitamins C + E during pregnancy. Whether these findings occurred by chance, or whether supplementation with these micronutrients in pregnancy truly poses an increased risk of poor maternal and infant health outcomes remains unclear.
The articles published in this supplement provide a current review of interventions that benefit MNCH outcomes, those where evidence of benefit is lacking, and gaps in knowledge. These articles uniquely contribute to the current knowledge base because we (1) included a broad range of interventions and factors that may influence MNCH outcomes, (2) included a variety of MNCH outcomes, and (3) took the less conventional approach of including data from well-designed observational studies where data from RCTs were lacking.
This publication is based on research funded by the Bill & Melinda Gates Foundation. The findings and conclusions contained within are those of the authors and do not necessarily reflect positions or policies of the Bill & Melinda Gates Foundation.
Conflicts of interest
The authors declare no conflicts of interest.