PROTOCOL: Effects of nutritional interventions during pregnancy on birth, child health, and development outcomes: A systematic review of evidence from low‐ and middle‐income countries

Optimal nutrition plays a crucial role before, during and after pregnancy (Alfaradhi & Ozanne, 2011; Black et al., 2013; Ota, Hori, Mori, Tobe-Gai, & Farrar, 2015). Poor maternal nutritional status is a risk factor for serious foetal complications and the outcomes for the neonate, including intrauterine growth restriction, stillbirth, low birth weight, preterm birth, increased risk of neonatal infections, neonatal hypothermia and neonatal death (Ahmed, Hossain, & Sanin, 2012; Black et al., 2013). Moreover, women who are undernourished at the time of conception have a higher risk of obstructed labour, preeclampsia, anaemia and mortality when compared with healthy women (Christian, Mullany, Hurley, Katz, & Black, 2015; Zerfu, Umeta, & Baye, 2016). The prevalence of maternal malnutrition is higher in low- and middle-income countries (LMICs) when compared with high-income countries (Black et al., 2013). Malnutrition refers to a group of nutritional disorders that include micronutrient deficiencies, undernutrition and overweight/obesity. Maternal under-nutrition is typically defined by a body mass index (BMI) less than 18.5 kg/m2, while overweight is classified as BMI ≥25 kg/m2 and obesity as BMI ≥30 kg/m2. The double burden of malnutrition is the coexistence of undernutrition and overweight and obesity, which has also been found to be highly prevalent in LMICs (Kimani-Murage et al., 2015) due to diets that chronically lack diversity and infections and/or chronic disease that could contribute to deficiencies by directly inhibiting nutrient absorption. The prevalence of maternal undernutrition ranges from 10 to 19% in LMICs, with variation by region and by country (Black et al., 2013). In addition, >10% of women aged 15–45 living in LMICs have heights (i.e., maternal stunting defined as maternal height <145 cm) that are considerably below the average (Black et al., 2013). The prevalence of low BMI in adult women is >20% in Sub-Saharan Africa and South-Central and Southeastern Asia (Black et al., 2013). Some individual countries fare worse than others. For example, in India, the prevalence of undernutrition among women of reproductive age (WRA) reaches almost 40% (Black et al., 2013). In 2014, about 1.9 billion adult people worldwide were found to be overweight, a prevalence that surpassed that of underweight, which constituted about 462 million people. In addition, >600 million were reported to be obese (World Health Organization, 2017). The prevalence of obesity is higher in the Americas and the Caribbean when compared to Africa, but overall, rates of overweight and obesity are rising globally, a situation that mimics that in high-income countries and may be reflective of changing food environments (Black et al., 2013; WHO, 2017). Both maternal undernutrition and overnutrition can have adverse effects before, during and after pregnancy (Kimani-Murage et al., 2015). Here add something about maternal undernutrition being related to low birthweight. Maternal undernutrition throughout pregnancy has also been associated with long-term health issues for the infant, such as obesity, diabetes mellitus, hypertension and cognitive dysfunction (Crispi, Miranda, & Gratacós, 2018; Maršál, 2016). In addition, low birth weight has been associated with increased risk of death from coronary heart disease and stroke in adulthood (Crispi et al., 2018). Malnutrition or inadequate dietary intake during pregnancy can expose the foetus to a harsh environment, which forces the foetus to adapt. However, this adaptation can lead to permanent changes in function and structure that can later lead to chronic diseases in adult life (Crispi et al., 2018; Maršál, 2018). Maternal obesity has also been associated with a higher risk of stillbirth and congenital abnormalities (Alfaradhi & Ozanne, 2011; Stothard, Tennant, Bell, & Rankin, 2009). In addition, obesity during pregnancy is associated with an increased risk of foetal macrosomia (Catalano & DeMouzon, 2015), which could lead to obstructed labour, and preterm birth, which is a major risk factor for infant mortality (Meehan, Beck, Mair-Jenkins, Leonardi-Bee, & Puleston, 2014). This review will focus on macronutrient supplementation during pregnancy. Micronutrient supplementation is being evaluated in a separate Campbell review of this series. Several macronutrient supplementation interventions have been proposed to address maternal malnutrition especially in LMICs including balanced energy supplementation, food provision and distribution and dietary intervention to prevent maternal obesity (Bhutta et al., 2013; Imdad & Bhutta, 2012). In LMICs, diets often lack foods rich in macronutrients and micronutrients that are typically found in meat, poultry and fish (Gibson & Hotz, 2018). Therefore, it is important to increase the availability of macronutrients and micronutrients by promoting and introducing diverse crops, integrating farming systems with small livestock, promoting fish farming and promoting better food storage (Gibson & Hotz, 2018). In addition, this intervention includes supplementation, which is designed to supply pregnant women in LMICs with multiple micronutrients (Allen, De Benoist, Dary, & Hurrell, 2006; Gibson & Hotz, 2018; Zerfu et al., 2016). Such interventions have found to be positively related to a reduced risk of maternal anaemia, preterm birth and low birth weight in a single study in Ethiopia (Zerfu et al., 2016). A balanced energy protein (BEP) supplement is a macronutrient food-based supplement where proteins provide <25% of total energy content (Imdad, 2012). BEP supplements, therefore, come in several forms. For example, a study from India provided supplements that consisted of dehusked sesame cake, jaggery and oil containing 30 g of proteins and 417 kcal energy for undernourished pregnant women (Girija, Geervani, & Rao, 1984). In another study from The Gambia, undernourished pregnant women were given daily supplements of high energy biscuits made with roasted nuts, rice flour, sugar and groundnut oil as supplements that contained 4,250 kJ energy, 22 g of proteins, 56 g fat and vitamins and minerals (Ceesay et al., 1997). Two previous reviews have demonstrated the positive association of BEP interventions with pregnancy outcomes, such as a reduced risk of stillbirth and small for gestational age babies and increase of birth weight (Imdad, 2012; Ota et al., 2015). Food distribution programmes provide low-income and undernourished pregnant and nonpregnant women and children with access to supplemental nutritious foods and often nutrition education (Baqui et al., 2008; Heaver, 2002; Kapil, Chaturvedi, & Nayar, 1992). These programmes are typically run by local or international organisations. For example, India has the Integrated Nutrition and Health programme, which is a non-governmental organisation-based programme that is implemented together through CARE-India and the Indian government (Baqui et al., 2008). This programme educates pregnant women alongside the provision of healthcare services and supplementary nutrition, with the aim of increasing knowledge about maternal and newborn care. The long-term goal of reducing neonatal mortality (Baqui et al., 2008; Kapil, 2002). India also has the Tamil Nadu Integrated Nutrition Programme (TINP), which is implemented by the state government of Tamil Nadu and supported by the World Bank. TINP aims to reduce maternal and child malnutrition through the use of a Community Nutrition Centre that makes supplementary nutrition available to pregnant women and children in villages (Heaver, 2002). In Bangladesh, the nutrition-focused Maternal, Neonatal, and Child Health programme supports pregnant women by providing several cross-cutting services such as counselling on nutrition and health, micronutrient supplementation and weight-gain monitoring (Nguyen et al., 2017). As noted above, obesity during pregnancy is associated with a host of maternal and foetal complications such as pre-eclampsia, caesarian birth, macrosomia and congenital malformations (Dodd, Crowther, & Robinson, 2008; Muktabhant, Lawrie, Lumbiganon, & Laopaiboon, 2015). Several behavioural interventions, including dietary control and exercise, have been found to be positively related to a reduced risk of macrosomia, caesarean delivery and gestational weight gain (GWG; Catalano & DeMouzon, 2015; Dodd et al., 2008; Guelinckx, Devlieger, Mullie, & Vansant, 2010; Muktabhant et al., 2015; Renault et al., 2014). Interventions can vary, and could include light to moderate-intensity exercise, strength training, stretching and relaxation exercises to prevent excessive weight gain (Nascimento, Surita, Parpinelli, Siani, & Pinto e Silva, 2011) or combined dietary control and exercise interventions whereby diet counselling and advice is paired with exercise. However, in this review we will only focus on dietary interventions to prevent maternal obesity. BEP supplementation is used to help undernourished women achieve the recommended daily energy intake (Bhutta et al., 2013; Imdad, 2012). Currently, there is strong evidence to support the benefits of BEP supplementation when compared with both high protein energy supplements and isocaloric supplements (Imdad, 2012; Ota et al., 2015). Evidence from a Cochrane review has linked BEP supplementation to a reduction in stillbirths, small for gestational age births and improvement in birth weight (Ota et al., 2015). However, no significant impact on preterm birth or neonatal death was observed (Ota et al., 2015). Food distribution programmes directly provide nutritious foods or supplements to vulnerable populations, including pregnant women. There is some evidence to support the targeting of programmes to pregnant women through the subsequent improvement in birth weight and reduction of infant mortality among infants of recipient mothers (Frith, Naved, Persson, & Frongillo, 2015). Often, programmes will provide pregnant women with healthy foods along with access to additional services, such as nutrition counselling. Counselling sessions may include information on the components of a healthy diet, the importance and consequences of poor nutrition, and food demonstrations, which provide women with the tools and knowledge necessary to maintain good antenatal health (Nguyen et al., 2017). Other interventions use community platforms, such as community health centres, to provide services such as immunisation, promotion of maternal and neonatal care, and distribution of food supplements. These strategies have have been shown to reduce neonatal deaths and improve maternal anaemia (Baqui et al., 2008; Leroy, Olney, & Ruel, 2016). Lifestyle interventions that include dietary control, exercise and behavioural change have been associated with a reduced risk of excessive GWG and macrosomia and decreased risk of adverse pregnancy outcomes (Catalano & DeMouzon, 2015; Dodd et al., 2008; Guelinckx et al., 2010; Muktabhant et al., 2015; Renault et al., 2014). Moreover, lifestyle interventions for maternal obesity can be implemented using a combination of dietary control and physical activity (Renault et al., 2014) or diet and exercise and behavioural change alone (Muktabhant et al., 2015; Nascimento et al., 2011). Dodd et al. (2014) used a comprehensive antennal dietary and lifestyle counselling intervention for pregnant women in Australia. The intervention included exercise, home visits that provided dietary advice and behavioural strategies delivered by a registered dietician (Dodd et al., 2014). Several reviews have been published that examine the impact of the interventions described above (Bhutta et al., 2013; Gibson & Hotz, 2018; Imdad & Bhutta, 2012; Muktabhant et al., 2015; Ota et al., 2015; Zerfu et al., 2016). However, most of these reviews focused on the efficacy of these interventions using randomised trials and did not address the question of effectiveness of large-scale nutrition programmes. Studies of effectiveness are needed to understand whether an intervention will be impactful in a real-world setting. Additional studies have been published recently (Devi et al., 2017; Dwarkanath et al., 2016; Huseinovic et al., 2017; Saville et al., 2018), indicating a need to update the systematic review evidence. Dietary interventions alone to prevent maternal obesity during pregnancy have not been reviewed previously. Therefore this review will make a first attempt to study its evidence. Furthermore, previous reviews did not assess the long term effects of these interventions during childhood. Taken together, this review will incorporate the latest evidence from RCTs and nonrandomized trials, and also assess the long term effects of maternal nutritional supplementation (Figure 1). Framework on Maternal nutritional supplementation [Color figure can be viewed at wileyonlinelibrary.com] BEP supplementation Food distribution programmes Dietary interventions to prevent maternal obesity Each intervention will be assessed, analysed and reported separately. Randomised controlled trials (RCTs), where participants were randomly assigned, individually or in clusters, to intervention and comparison groups. Cross-over designs will be eligible for inclusion. Quasi-experimental designs, which include: Natural experiments: studies where non-random assignment is determined by factors that are out of the control of the investigator. One common type includes allocation based on exogenous geographical variation. Controlled before-after studies (CBA), in which measures were taken of an experimental group and a comparable control group both before and after the intervention. We also require that appropriate methods were used to control for confounding, such as statistical matching (e.g., propensity score matching, or covariate matching) or regression adjustment (e.g., difference-in-differences and instrumental variables). Regression discontinuity designs; here, allocation to intervention/control is based upon a cut-off score. Interrupted time series (ITS) studies, in which outcomes were measured in the intervention group at least three time points before the intervention and after the intervention. Two review authors will independently screen titles and abstracts of all retrieved references. We will retrieve the full-text study reports for all citations that at least one review author considers potentially relevant. Two review authors will independently screen the full text articles and identify studies for inclusion, and identify and record reasons for exclusion of ineligible studies. We will include studies irrespective of whether measured outcome data are reported in a “usable” way. We will resolve any disagreement through discussion or, if required, we will consult a third review author. We will identify and excluded duplicates and collate multiple reports of the same study so that each study, rather than each report, is the unit of interest in the review. We will record the selection process in sufficient detail to complete a Preferred. This review will include healthy pregnant women of any age living in LMICs, as defined by the World Bank. Studies where women are recruited in the preconception period are eligible, given that women are followed throughout pregnancy. In this review, we will consider women who are undernourished (inadequate nutrition) and obese women who have no other co-morbids. BEP supplementation: Defined as a food supplement where proteins provide <25% of the total energy content (Imdad, 2012). Food distribution programme: Food distribution programmes are defined by their direct provision of foods to recipients, who, in this case, are pregnant women. Eligible food distribution programmes could be locally or internationally led, and may or may not include elements of nutrition education. Dietary interventions for prevention of maternal obesity: Eligible interventions for preventing or reducing maternal obesity include dietary control only. Each intervention will be analysed separately and will not be compared to the other interventions listed here. may include of This review will include studies that have the and maternal outcomes, foetal outcomes, newborn and child Maternal outcomes BMI and newborn outcomes defined as of a foetus before it is and especially the and of defined as with no of life at or after mortality is defined and deaths mortality Child outcomes mortality and mortality and Maternal outcomes Maternal mortality defined as the death of a while pregnant or of of irrespective of as defined by authors and of anaemia outcomes outcomes birth age (World Health Child outcomes score for height for score for weight for score for weight for outcomes for cognitive disease disease Studies were excluded if have not reported the outcomes We will include all participants in eligible studies that outcomes of interest There will be no based on of of or of outcome outcomes are reported at multiple time of we will include outcome based on of outcomes (e.g., infants the time of is not we will the authors for the For and adulthood outcomes, we will include the outcome at the We will include studies from These countries are defined as with a of or less in and middle-income are with a and in (World 2017). We will not any for example, on the described We will also for any and for The will be in the Cochrane Controlled Health the nutrition and Health which the from Africa and so We will also the of or (e.g., World and The trials and will be for A Nutrition for Nutrition World Food Programme Nutrition Food and Health We will also the of the published included studies and systematic We will also run studies of included studies in and of and/or We will also other and individual authors who are in the of maternal For this review, the designs of interest were experimental and study designs as as studies with a control including We will also studies with at least three time points before and three time points after the intervention. The studies will include pregnant women. the we will that all articles on the same study were and appropriate combination of outcome will be to the type of interventions studies are intervention will be combined or and we will that there was no double of an outcome is reported in several we will unit in to the We will in the of and will that any of by studies and that no double of evidence when studies. time period when study type of (e.g., full-text and study country or and of interest and age and of study unit of or study individually or randomised trials and total data group to of and delivery of and their We described all the study intervention in the of included studies, we only reported the intervention that the review group randomised to of and and their of the total in intervention and comparison change at each time Other Two review authors will independently assess the risk of for each included We will resolve any by discussion or by a third review author. of participants and of outcome for each outcome outcome data outcome Other such as the of outcome and outcome outcome of the interventions during the study outcome Other of We will the outcome data for each study the data in to the effects 2014). We will use the risk for We will use the for outcomes reported on the same and the for outcomes the same outcome but measured on We will the with for all and are not we will use other available data example, and appropriate methods described in the Cochrane for of Interventions et al., 2011) to the and other available data are not sufficient to we will the study we are to the in we will in the or the data the We will also consider the and of data when outcomes as can due to small We will outcomes from studies with multiple in an appropriate to double of participants by to same In such a we will not the and only We will have a of outcomes and outcome these be combined (e.g., the cognitive be combined with Therefore, a will be separately for each Furthermore, for each we will separately study designs and We will all the for each outcome and not any from others. trials have used we will that study have their after for effects (e.g., it is whether a has for the study will be for appropriate for are not we will an of the and are in using the et al., 2011). will be combined with from We will authors to study and outcome data where (e.g., when we identify a study as an we did not a full after we the study we will such an as a outcome data are such as or and we could not these from the study we will from other available such as to the methods described in the Cochrane for of Interventions et al., 2011). will be using and of the we will also assess using the of the and we and in in this Therefore, we will attempt to any observed statistical using sufficient studies are will be to any and study studies are sufficient to a As a direct for we will the from published reports with from other may this will not the case, and for any found will be and in the text of the review. We will a of all studies for each intervention which all the in the studies in the and so and examine to were effects given the diverse interventions, and so For each we will the from the factors such as of of people interventions, and so to assess their impact on the and effectiveness of each intervention. We will of for all of the outcomes using the of and et al., 2011). These of risk of of of and risk of We will the of evidence for each outcome as or The evidence is described in studies will be as there are no serious we will the evidence for studies with a of of a and of defined as with no of life at or after mortality mortality and mortality and birth maternal pregnant women defined based on BMI Africa Asia and of supplementation: pregnancy third weight overweight obese interventions on maternal The will be using with a for We will use statistical to assess will be significant for We will assess the of and if the of intervention be in inadequate and/or that have been in We not to include Imdad, review Imdad, for this review from a from the & to the Centre for Child Health at The for to for of the systematic review. will be for this review and the review will be update after pregnant Nutrition food studies if can add to this for a so if were using it or or and

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.