Maternal mortality and morbidity remain a challenge to promoting global health and achieving health equity. In 2017, an estimated 295,000 women died in pregnancy and childbirth, 94% of which were in low- and middle-income countries (LMICs) (1). While global maternal mortality fell by 42.9% between 1990 and 2015, accelerated progress, particularly in low-income countries, is needed to achieve the SDG goal of 70 deaths per 100,000 live births by 2030 (2). The leading causes of maternal death are largely preventable and include obstetric hemorrhage, hypertensive disorders in pregnancy, sepsis or infection, embolism, obstructed labor, and abortion-related complications (3). Contributors to these causes include individual level risk factors, such as infection, poor nutrition, education, and obstetric history, but also broader social, economic, environmental and health systems determinants (4).

Pregnancy increases susceptibility for various infections, including influenza, malaria, and reproductive tract infections (5). In malaria-endemic regions of Africa, one in four pregnant women are estimated to have a malarial infection at the time of delivery, resulting in high risks for adverse birth outcomes and potential for transmission to the child (6). While the prevalence of helminthic infections varies by region, it can be a significant contributor to maternal iron-deficiency anemia (7). Other infections, such as sexually transmitted diseases, bacterial vaginosis, and asymptomatic bacteriuria are also common in pregnancy, with higher rates in LMICs (8, 9). Inflammation and infection in pregnancy can impact fetal development and play a role in the etiology of adverse birth outcomes, including prematurity (10).

Maternal undernutrition represents one of the most prevalent burdens among pregnant women in LMICs (11). Underweight, defined as body mass index (BMI) less than 18.5 kg/m², affected 14.2% of women aged 20–49 years in LMICs in 2015, with prevalence as high as 42% in India in 2014. Anemia in pregnancy, defined as hemoglobin less than 110 g/L, affected 40.1% of pregnancies in 2016. The prevalence of other maternal micronutrient deficiencies is not well documented, but evidence of improved maternal micronutrient status and birth outcomes following supplementation suggests deficiencies are common (11). Maternal undernutrition, including inadequate gestational weight gain, has direct impacts on the growth and development of the fetus, resulting in babies being born too soon and/or too small (12, 13). In LMICs, maternal nutritional and infection-related risk factors often coexist, resulting in synergistic effects, further underscoring the need for integrative antenatal care services (14).

Interventions administered in pregnancy provide the opportunity to lift children out of the intergenerational effects of poor health. As such, many evidence-based interventions and recommendations during pregnancy are targeted at reducing adverse birth outcomes, such as low birth weight (LBW), preterm birth (PTB), small for gestational age (SGA), stillbirth, and neonatal mortality (15). In trials generating the evidence of benefit, outcomes in the newborn are examined primarily, whereas those in the mother are considered secondary and are measured and reported with less frequency (15). Understanding the physiological impact of interventions on the mother can elucidate the mechanisms that promote better fetal development and what factors moderate its efficacy. Furthermore, full knowledge of intervention effects is necessary to ensure they are, at minimum, safe for pregnant women and, ideally, contribute to improvements in maternal health. The objective of this review is to present the evidence on maternal outcomes of those interventions administered in pregnancy where the primary aim is improving birth outcomes.

We conducted a literature review of peer-reviewed systematic reviews (SRs), meta-analyses (MAs), and reviews of reviews of select interventions (n = 27) during pregnancy that focused on infection prevention and management, and nutritional support in LMIC settings with a focus on maternal health and nutrition outcomes. The selected interventions were informed by a parallel review by Ashorn et al. focusing on birth outcomes (Ashorn, P., personal communication). We did not include antiretroviral therapy for HIV-1 in pregnancy as it has been tested for maternal infection and mother-to-child transmission outcomes.

Outcomes of interest included maternal mortality and morbidity such as hemorrhage, hypertensive disorders, anemia, infections, micronutrient status indicators, adverse events, side effects, and others. Unspecified outcomes were also considered if found in the literature. We did not apply any restrictions to outcome definitions and report the definitions used by the authors when provided.

We conducted individual searches for each intervention in four databases: PubMed, Embase, Cochrane Database of Systematic Reviews, and Scopus. Each search consisted of intervention-related terms, pregnancy terms, and study type specification, i.e., review articles. The search was limited to reviews written in English and Spanish and published from 1950 onwards and last updated on January 25th, 2021. Search terms were restricted to title, abstract, and keywords to obtain the most relevant results. An example of our search strategy is available in Supplementary Table S1.

Two authors (EZ, MR) reviewed the articles independently, using Covidence software. Titles and abstracts were screened, and relevant articles underwent full-text review. We included systematic reviews and meta-analyses that evaluated the intervention of interest and reported at least one maternal outcome. We excluded narrative or scoping reviews, clinical guidelines, and intervention trials in non-pregnant women.

The following information was extracted from the review articles: title, author, year published, review type (SR/MA), effect sizes of pooled data (for MAs) or synthesis of findings (for SRs) by outcome, number of studies and participants, intervention and comparator details, and quality assessment by review authors. We did not conduct an independent quality assessment of the included review articles but report on the published assessments.

We screened 3,983 articles, conducted 481 full text reviews, and extracted data from 102 systematic reviews or meta-analyses across all interventions (Supplementary Table S2). To avoid duplication, we present the most recent and comprehensive findings from 24 reviews across infection-related interventions (Table 1), nutritional interventions (Table 2) and other interventions (Table 3). Relevant articles were not identified for eight interventions, specifically maternal Hib vaccination, WASH interventions, IPTp plus antibiotics, the screening and treatment of STIs other than HIV and syphilis, the treatment of deep caries of periapical periodontal disease, the screening and treatment of tuberculosis, unconditional cash transfers, and the monitoring of gestational weight gain in undernourished settings. For the other 19 interventions, at least one relevant SR/MA was identified with maternal outcomes. Additional intervention details and the published risk of bias and quality of evidence are provided (Supplementary Table S3).

The goal of this scoping review was to present the evidence on the maternal effects of infection-related, nutrition and other interventions administered in pregnancy with the aim of improving birth outcomes. Findings of maternal effects were far less reported nor primary compared to birth and infant outcomes, as demonstrated by the eight interventions where no reviews reporting maternal outcomes could be identified. Despite this lack of data, several important findings can be drawn from this work.

Vaccination of pregnant women against infectious diseases has been predominately evaluated for its effect on young infants, with limited evidence on maternal morbidity effects. However, vaccine use in pregnancy has been shown to protect mothers from infection, including influenza, and more recently, SARS-CoV-2 (56). Tdap vaccination in pregnancy is considered safe for the mother and fetus and should be scaled up globally with continued surveillance of chorioamnionitis outcomes. Additional research is needed to discover more effective vaccines, to characterize the optimal timing and seasonality of vaccination in pregnancy for maternal immunogenicity and protection of the infant, and to identify strategies for increasing coverage and uptake globally (57–60).

Insecticide-treated bed nets and frequent IPTp-SP are effective interventions in reducing maternal malaria infection and transmission. In areas of high SP resistance, transitioning to DP should be considered given its effectiveness in reducing maternal malaria and anemia. Studies of cost-effectiveness will help determine whether this new drug regimen should be widely adopted. Conversely, the intermittent screen and treat approach is not yet a suitable alternative to IPTp, even in SP-resistant areas, because the lack of sensitivity of the rapid diagnostic tests (RDTs) may cause more maternal malaria infections to go untreated.

Anthelmintic treatment in pregnancy is recommended and effective in reducing anemia, but evaluations of large-scale implementation programs could shed more light on the effectiveness of anthelmintic treatment for specific parasites. Additional research on whether two doses result in better outcomes, particularly in highly endemic areas, is needed (61).

Treatment with antibiotics is effective in clearing BV in pregnancy but does not appear to reduce the risk of maternal or infant adverse events. With high rates of BV being reported in LMICs (62), more research is needed to determine whether broad screen and treat programs are effective in these settings, whether specific sub-groups benefit more from treatment, and whether earlier administration of treatment improves outcomes (63). Given the significant increased risk of side effects with antibiotic treatment, exploration of alternatives to antibiotic therapy is warranted (64).

Periodontal treatment in pregnancy ameliorates maternal health through improving dental health, but additional research is needed to understand how periodontal indices may link to fetal growth and birth outcomes, and whether timing and type of treatment plays a role. Treatment of ASB in pregnancy significantly reduces the risk of pyelonephritis, but research on effectiveness and cost-effectiveness of RDTs on maternal outcomes is lacking in LMICs. Similarly, evidence on the effectiveness of screen and treat programs for STIs and TB in pregnancy are needed, particularly in LMICs where burdens are high.

Nutrition interventions more commonly reported direct maternal effects compared to the infection-related interventions. Nutrition education may help undernourished pregnant women improve their dietary intakes, but nutritional support, i.e., provision of supplements or food, may be key to improving nutritional status, particularly in food insecure contexts. While the current WHO recommendation does not promote BEP as an individual level intervention, many women in LMICs may experience low BMI when entering pregnancy, even if 20% of the overall population is not underweight as illustrated by sub-national map of the world for low BMI among women (11), and many more may experience inadequate GWG regardless of their BMI at the start of pregnancy (65). As such, regular screening of GWG has the potential to be an effective intervention to ensure adequate weight gain and dietary intakes, especially if followed by nutritional supplementation such as BEP (12). Exploring this targeted approach to nutrition in antenatal care is pressing as populations across LMICs grow more heterogenous in nutritional status. Trials underway using expert consensus formulations of BEP will provide future evidence for the impact of fortified BEP supplementation on birth and maternal outcomes (NCT03533712, NCT03668977, NCT04012177) (66).

IFA is highly effective in reducing maternal anemia and improving iron status. The replacement of IFA with MMS should be considered based on the evidence regarding improved birth outcomes and maternal micronutrient status with MMS supplementation. Findings from individual trials suggest additional maternal health benefits with MMS supplementation, such as reduced cortisol and erythropoietin levels in the third trimester (67), and reductions in obstetric complications such as PPROM, postpartum hemorrhage, and puerperal sepsis (68).

High dose calcium is recommended for reducing the risk of hypertension and preeclampsia in pregnancy in low calcium intake settings; the cost and adherence issues related to high dosage have limited the scale-up in such settings (69). Ongoing non-inferiority trials of low vs. high dose calcium will shed light on this issue (NCT03725891, NCT03735433).

Supplementation with omega-3 fatty acids has not been strongly associated with any maternal outcomes, though additional research on hypertension-related and metabolic outcomes are indicated (70, 71). RCT evidence is needed to evaluate conditional and unconditional cash transfer programs in pregnancy on maternal outcomes, short and long-term. Implementation research is needed to evaluate how either high or low dose aspirin interventions may be integrated within antenatal care and brought to scale.

Much of the current focus in maternal and child health in LMICs is on halting the intergenerational cycles of poor health, and pregnancy and pre-pregnancy interventions are evaluated by their effects on birth outcomes, particularly gestational age and growth indicators. The likelihood that these indicators will be influenced by a particular intervention relies in large part on the maternal physiological response. Maternal health and nutritional status are often considered ‘secondary’ and are either omitted from research results or never measured in the first place, except in the context of interventions that are directly targeted to impact maternal infection, health and survival. Our review demonstrates the importance of capturing maternal outcomes to elucidate whether and by what extent direct effects on the mother impact her and may in turn be the pathway for influencing fetal and infant outcomes. We found an overall under-reporting of maternal outcomes for the selected interventions, and very few results on maternal mortality and serious morbidity. Some studies reported long-term impacts of pregnancy interventions on children, but no long-term health effects on mothers were described, despite the plausible hypothesis that such effects could accrue. There is also a gap in our understanding of the cost-benefit ratios of interventions, which exclude assessments of benefits to maternal health. Future RCTs should include and measure maternal outcomes as primary and should systematically report non-significant estimates and rare events. In summary, our scoping review of antenatal interventions may help policy makers, program implementers, and researchers identify what interventions should be prioritized for the benefit of maternal health in high burden settings.