Anemia in Pregnant Women and Children Aged 6 to 59 Months Living in Mozambique and Portugal: An Overview of Systematic Reviews

Introduction: Globally, anemia is still a public health issue faced by people in low and high-income countries. This study gives an overview of published scientific articles related to the prevalence, nutritional indicators, and social determinants of anemia in pregnant women and children aged 6 to 59 months living in Mozambique and Portugal. Methods: We performed a review of scientific literature in April 2021, searching for published indexed articles in the last 15 years (2003–2018) in electronic databases. Subsequently, quality assessment, data extraction, and content analysis were performed. Results: We have identified 20 relevant publications. Unsurprisingly, anemia plays a relevant role in disability and life imbalances for these subgroups in Mozambique compared with Portugal. For both countries, data on anemia and iron deficiency in pregnant women and children aged 6 to 59 months old are either outdated or remain unclear. Similarly, studies on social determinants and anemia are also still scarce. Conclusions: A gap of information on anemia, other nutritional indicators, and social determinants in pregnant women and children between 6 and 59 months of age living in Mozambique and Portugal is highly observed. More research is crucial to help achieve the goals established by the Sustainable Development Goals.


Introduction
Anemia is a condition in which a person's hemoglobin level is less than normal (below 11 g/dL for pregnant women and children aged 6-59 months) [1,2]. Globally, it is still a public health issue faced by people in both low-and high-income countries and is a concern for adolescent girls, women of reproductive age (particularly during pregnancy), and children in the first years of life [1,3]. It impairs health and well-being in women and increases the risk of maternal and neonatal adverse outcomes [4,5]. Physiologic anemia is the most common cause of anemia in the neonatal period, being more pronounced in preterm infants compared with term infants, accounting that half of the cases are due to iron deficiency. In women, it may develop because of prenatal, perinatal (at delivery), or Table 1. Search terms and operators used for the identification of publications in the electronic databases.

Electronic Databases
Terms The selection of papers was restricted to articles with results of studies focused on anemia in pregnant women and children aged 6 to 59 months living in Mozambique or Portugal published between the years 2003 and 2018, in English, Portuguese or Spanish languages and that were available on the selected electronic databases. This review was limited to 15 years (2003-2018) as we were primarily interested in looking at the scientific evidence published in a similar time that major demographic health national surveys and population-based studies involving anemia data were held in Mozambique and Portugal [9,[16][17][18][19]. Therefore, we expected to help to refine the development of subsequent analysis involving the mentioned surveys and anemia data. We analyzed the prevalence of anemia according to the World Health Organization's criteria [20]. Individual studies, technical reports, theses, technical specifications and standards, noncommercial translations, technical and commercial documentation, and official documents not published were excluded from this review. Articles without abstract and where the full text could not be obtained, as well as articles with duplicate data (data from the same study published in different journals) were also excluded. Table 2 summarizes the eligibility criteria used for the inclusion and exclusion of articles. Two investigators (RMC and JBC) performed a double-blinded independent screening of title/abstract, selection, and full-text review. The final selection process was blinded and independent using the feature "Blind-On" available on Rayyan-intelligent systematic review software (http://rayyan.qcri.org, accessed on 29 April 2021) [21], which helps to reduce the influence by the decisions of the researchers during the article selection stage (hiding individual authors' decisions about included studies), thus removing a significant possible source of bias [21,22]. The verification of duplicates was performed and articles without abstract and/or without full text, which did not address anemia, which did not include the countries of interest (Mozambique and/or Portugal), which did not address the target groups, and written in languages (other than Portuguese, English or Spanish) were excluded. Findings were subsequently verified and synthesized. Discussion and consensus were used to solve any emerged discrepancies between the two investigators (RMC and JBC).

Quality Assessment and Data Extraction
To assess the quality of the publications, 22 criteria of the STROBE -Strengthening the Reporting of Observational Studies in Epidemiology checklists were applied [23][24][25]. These criteria involve a series of characteristics that relate to the title, abstract, introduction, statistical methods, bias, results, and discussion sections of the articles [24] and were used independently by two members of the team (RMC and JB) for the evaluation. An excel spreadsheet was elaborated for the verification and calculation of the percentage of methodological quality (Supplementary Files S1 and S2). We also performed the critical appraisal of the publications using the 'AMSTAR 2 -A Measurement Tool to Assess systematic Reviews software (https://amstar.ca/index.php, accessed on 7 May 2021) [26,27].
With the completion of the quality assessment, we performed (RMC and JBC) to the extraction of relevant data, the content analysis, and the summarization of the information. The extracted data included the following information: author(s), year of publication, mentions of Mozambique and/or Portugal, target groups, Mozambique or Portugal specific findings, and period of analysis/analysis performed (Table 2 and Supplementary File S3). The description of relevant findings by thematic categories was also included (Supplementary File S6). We contacted the corresponding author in cases we were unable to access the article's Supplementary Files.
We performed a content analysis to identify the type of information published and specific key findings to Mozambique and Portugal. Table 3, Supplementary Files S3 and S6 summarize the general and specific country findings of the 20 publications. Regarding the AMSTAR 2 critical domains, 60.0% performed a comprehensive literature search, 95.0% performed data selection and extraction in duplicate, 65.0% used satisfactory techniques for assessing the risk of bias, 71.0% discussed or investigated publication bias, and 95.0% reported conflicts of interest ( Figure 3; Supplementary Files S2.1 and S2.2).   2013; Proportion of pregnant women with anemia whose blood hemoglobin would increase to at least 110 g/L was calculated. The proportion of severe anemia that would increase to at least 70 g/L was also calculated.
Iron deficiency contribute substantially to maternal deaths and maternal iron deficiency is associated with babies with low weight (<2500 g) at birth; anaemia (haemoglobin <110 g/L), which might be attributable to low consumption or absorption in the diet or to blood loss, such as from intestinal worms, is highly prevalent during pregnancy. In Africa, the prevalence of iron deficiency anaemia (haemoglobin <110 g/L) was of 20.2% (18.6-21.7) and 20.3% (18.3-22.4) in children < 5 years and pregnant women, respectively. In Europe, the prevalence of iron deficiency anaemia (haemoglobin < 110 g/L) was of 12.1% (7.8-16.2) and 16.2% (12.6-19.7) in children < 5 years and pregnant women, respectively.

Anemia and Iron Deficiency Anemia (IDA) Prevalence
Global and regional anemia and iron deficiency anemia prevalence data were reported in five studies, but only two studies reported Mozambique and Portugal-specific data [30,33,40,42,43]. In 2013, the prevalence of iron deficiency cases (by thousands) was of about 5193.9 (4985.7-5442.7) and of about 1205.7 (1180.2 to 1227.5), in Mozambique and Portugal, respectively [40]. The absence of postpartum anemia in pregnant women with vaginal deliveries who were admitted in labor was also reported in one study trial conducted in Portugal during the years 2006 and 2007 [38]. A review reported that in 2005, the prevalence of anemia was severe (≥40.0%) among Mozambican children of preschool age (0-5 years) [30]. Nonetheless, for both countries, data on anemia and iron deficiency in pregnant women and children aged 6 to 59 months old are either outdated or remain unclear as the findings hereby reported were not disaggregated into these specific subgroups.

Years Lived with Disability (YLDs) and Disability-Adjusted Life-Years (DALYs)
Eight studies provided data on anemia, iron deficiency, years lived with disability (YLDs), and disability-adjusted life-years (DALYs) [39][40][41][42][43][44][45][46]. It was documented that iron deficiency anemia was among the ten causes of years lived with a disability (YLD) during the year 2013 in Mozambique, but not for Portugal [40]. For the same year, Mozambique's YLDs iron deficiency anemia was of about 199.3 (133.8 to 290.1) and of 29.4 (19.2 to 43.1) in Portugal. Between 2015 to 2016 anemia was among the top ten leading causes of disability in Mozambique (ratio of observed YLD to YLD on basis of the social demographic index was 0.71 and 0.98, respectively) but not in Portugal. For the same period, iron deficiency anemia was also among the leading Level 4 causes of age-standardized YLD rates for females in Mozambique but not in Portugal [42][43][44]. Concerning DALYs, one study reported that in 2013 children undernutrition and iron deficiency were ranked as 2nd and 9th risk factors among those ten risk factors of attributable DALYs for males and females in Mozambique. Iron deficiency wasn't among the ten leading risk factors in terms of DALYs neither in Mozambique nor Portugal, for both sexes combined [41]. Nonetheless, one study reported that in the same year, iron deficiency was one of the top ten causes of DALYs for adolescent girls aged 10-19 years old [37]. Overall, anemia and iron deficiency anemia play a crucial role among children and adolescent girls living in Mozambique-a country still facing a high burden of malnutrition [32].

Malaria and HIV/AIDS
Two study trials that included 5469 pregnant women conducted between 1987-2013 in Mozambique, showed that mefloquine was more efficacious than sulfadoxine-pyrimethamine in HIV-uninfected women or daily cotrimoxazole prophylaxis in HIV-infected pregnant women for prevention of malaria infection and was associated with lower risk of maternal anemia (RR 0.84, 95% CI 0.76 to 0.94), no adverse effects on pregnancy outcomes (stillbirths and abortions), and no effects on low birth weight and prematurity. The high proportion of mefloquine-related adverse events constitutes an important barrier to its effectiveness for malaria preventive treatment in pregnant women [28]. Two other trials carried between 2004-2008 in Mozambique found a large effect in reducing the risk of cord blood anemia (RR 0.49, 95% CI 0.30 to 0.80), and increase in mean cord packed cell volume (MD 1.01%, 95% CI 0.05 to 1.97) [29]. Nonetheless, the possible effects or benefits of routine malaria chemoprevention, hookworm and HIV/AIDS prevention and treatment strategies and hookworms on reducing the risk of maternal anemia and negative pregnancy outcomes are still not well documented, in particular in provinces of Mozambique with the highest burden of malnutrition and food insecurity. Unsurprisingly, for Portugal comparatively to Mozambique, which has been facing a burden of malaria and HIV/AIDS for several years, data remain unclear.

Discussion
In this overview, we have identified 20 relevant publications. Regarding specific results on anemia in both countries, few were linked to anemia and iron deficiency prevalence and those related to the causes of disability and life imbalances were commonly presented [30,37,[40][41][42][43][44][45].
In 2013, Mozambique showed slightly lower iron deficiency prevalence rates and YLDs iron deficiency anemia values compared to neighboring countries such as Tanzania ((with a prevalence of iron deficiency cases of 55,599,526.2 (9365.7-9731.6); and the YLDs iron deficiency anemia values of 346.2 (229.1-498.1)). Likewise, the same data was valid for Portugal in comparison with the neighboring country, Spain ((with a prevalence of iron deficiency of 5907.3 (5756.9-6026.4); and the YLDs due to iron deficiency anemia values of about 145.2 (96.2-213.4)) [40]. Concerning DALYs, in 2013, similarly to Mozambique children's undernutrition was ranked also as 2nd risk factor for other countries of Eastern Sub-Saharan Africa such as Djibouti, Kenya, Malawi, Tanzania, Uganda, and Zambia. Nonetheless, differently from Mozambique, iron deficiency was ranked not as the 9th risk factor but the 7th for Somalia and as the 10th for Burundi, Comoros, Djibouti, Malawi, and Tanzania [41]. Undoubtedly, anemia and iron deficiency anemia continue persisting as an unsolved risk factor/cause of DALYs throughout the past years in Mozambique and possibly overshadowing the in-country fight against maternal and child morbidity and mortality.
Evidence indicates that anemia plays a relevant role in disability and life imbalances in pregnant women and children under five years of age in Mozambique compared with Portugal [40,42,43]. These findings are consistent with previous studies in Portugal that showed a low prevalence of maternal anemia in the first half of pregnancy. However, careful attention needs to be paid as these studies were conducted with Portuguese women with the 20th week of pregnancy, whereas anemia prevalence often tends usually to be higher in late pregnancy, in women without iron supplementation [48]. Our findings related to Mozambique agree well with previous reports that suggest that we are faced with a population of mothers with low iron stores, conditioning the hemoglobin values of infants in the first months of life [49]. Adding to this are many contextual cultural and socio-economic factors that may contribute to high levels of anemia throughout infancy in Mozambican children.
A growing body of literature showed that adolescents are at risk of iron deficiency because of their high iron requirements during the growth spurt period, particularly in girls for whom the start of menstruation leads to iron losses [50][51][52][53]. Concerning adolescent women, iron deficiency anemia is still a public health problem in both countries, since the data indicate that it is a risk factor for disadjusted life years [37,41]. The results found in Portuguese adolescent girls, also coincide with those reported in the last statistics of Portugal that showed prevalence rates of 9.0% in adolescent girls aged 12-15 years and 16.0% in adolescent girls aged 16-19 years [54]. Reports from previous years from the HELENA Study showed that the overall proportion of iron depletion among adolescents was 17.6%, being higher in girls compared to boys. By geographical location, rates were higher in Eastern Europe (Pecs, 23%) followed by Northern Europe (Stockholm, 19.0%), Western Europe (17.0% in Ghent and 19.0% in Lille), Central Europe (16.0% in Dortmund and 19.0% in Vienna) and 15.0% in Southern Europe (14.0% in Athens, 17.0% in Heraklion, 19.0% in Rome and 10.0% in Zaragoza) [51]. Other authors reported a prevalence of anemia of 2.6% in Portuguese adolescents living in the city of Porto, being these rates higher in girls (4.1%) compared to boys (1.0%) [50,55]. Nonetheless, these results only allow for statistical comparison of differences in iron status between selected European cities but not between European regions [51]. On the other hand, our findings are similar to those found in Spain, which shows that 15.0% of Spanish adolescents had iron deficiency anemia in a similar period, thus pointing more specifically to iron deficiency anemia as a clinical challenge in the daily practice of medicine at all levels of care [56]. There is evidence to suggest that dietary iron intake may be poor because of inadequate intake during adolescence or due to poor iron intake since infancy. Other factors such as a change in dietary habits by peer influence, eating disorders (refusal to eat, excessive weight-loss diets, and skipping meals), dependence on food that can be prepared rapidly and simply (fast food) can also play a crucial role on iron deficiency [50]. Our findings suggest the need for specific attention to adequate dietary intake from infancy up to adolescence particularly to adolescent girls ensuring that their dietary iron intake is adequate to their requirements, agreeing with those results previously reported by De Andrade et al. [50]. Findings here reported related to anemia in Mozambican preschoolers under five years of age are also of huge concern, as other studies conducted in Sub-Saharan Africa showed that anemia persists through later ages of life, being also a burden in elder preschoolers aged 7-15 years [30,57].
The global burden of diseases in 2015 and 2016 classifies Mozambique as a low social demographic index country (SDI) and Portugal as a high-middle social demographic index country (SDI) [42,43,46]. In the year 2016, iron deficiency anemia was one of the leading causes of YLDs in low-middle-SDI and low-SDI quintiles. Pooled analysis that included among other African countries, neighbor-countries of Mozambique with a similar SDI (such as Tanzania, Eswatini, and Madagascar), showed that the risk of anemia among women living in the lowest wealth quintiles was 25.0% higher than among those in the highest wealth quintile. Women with no education were more likely to be anemic than were those with greater than secondary education. Patterning of anemia by socioeconomic status was also noted for children: a child living in a household in the lowest wealth quintile was 21.0% more likely to be anemic than were those in the highest wealth quintile [30]. Conditional on demographic and socioeconomic factors, the mother's anemia status was among the strongest predictors of anemia in children [30,34]. Our review shows that age and gender were the social determinants more analyzed comparatively to indicators such as the years of schooling or wealth index [37,40,42,43]. Previous studies highlighted the relevance of social determinants analysis to better understand the causal association of social determinants and the occurrence of anemia [58]. Despite the relevance of this issue has been studied in other countries hereby mentioned, our review shows that, limited research evidence on anemia's social and structural determinants prevails for Mozambique and Portugal.
A gap of information on anemia, other nutritional indicators (in addition to iron deficiency), and social determinants in pregnant women and children between 6 and 59 months of age living in Mozambique and Portugal are highly observed in this study overview. Little literature related to anemia; our focus subgroup was also found at Index Medical Portuguese Journals-a Portuguese national database with non-open access. Anemia prevalence among Portuguese pregnant women and Mozambican children under 59 months old was a topic approached in some of the publications found, nonetheless, the majority of studies had its focus on supplementation with iron during pregnancy, preschoolers' food habits, and case studies interlinked with anemia during infancy [48,49,[59][60][61][62][63][64]. More research is required to gather scientific-based evidence that can contribute to improving strategies that allow us to achieve the goals established by the World Assembly of WHO of reducing anemia in women of reproductive age, low birth weight, and under-five mortality. Undoubtedly efforts to attain these goals are interlinked to achieving the following Sustainable Development Goals (SDGs): "2.2 End all forms of malnutrition", "3.1. Reduce global maternal mortality ratio to less than 70 per 100 000 live births by 2030" and "3.2. End preventable deaths of newborns and children under five years of age, with all countries aiming to reduce neonatal mortality to at least as low as 12 per 1000 live births and under-5 mortality to at least as low as 25 per 1000 live births, by 2030" [65].
An opportunity acknowledged by the Universal Health Coverage (UHC) 2030 s platform is that the inclusion of universal health coverage in the SDGs can be a coherent approach to health, allowing the acceleration of equitable and sustainable progress toward universal health coverage and health systems strengthening at global and country levels [66]. Even though both countries which were the focus of our study present huge contextual, geographic, and socioeconomic differences, Mozambique despite some progress (made in previous years) had a remarkably low UHC service coverage index and life expectancy (of 46.0; 61 years) compared with Portugal (of 82.0; 80 years) by the year 2017. External health expenditure was also strikingly higher in Mozambique (62.92) compared with Portugal (0.09) by the year 2018 [66,67]. With this in mind, the best performances in health indicators are clearly observed for Portugal compared with Mozambique where major health (triple burden diseases, infectious illnesses), lack of water, lack of good sanitation, lack of access to health facilities problems, as well as food insecurity still prevails [68]. Nevertheless, Costa et al. showed that throughout Eastern and Southern European countries, including Portugal, population health inequalities prevail across metropolitan areas, generally, with municipalities presenting worse health determinants value scores. Despite geographic disparities in the distribution of value scores of health outcomes between municipalities may not be as evident as expected in Portugal, more specifically in Lisbon, health inequalities still prevail and need to be tackled [69]. As stated by Malta et al. [68], Mozambique and Portugal are both Community of Portuguese Language Countries (CPLP), and the strengthening of network collaborations focused on research, and more specifically on anemia can be of added value to give more visibility to this health issue and to improve health policies and to reinforce strategies aimed to reduce anemia between pregnant women and children under five years living in Mozambique and Portugal.

Conclusions
This review highlighted the research gap on anemia, iron deficiency among pregnant women and children living in Mozambique and Portugal.
Anemia and iron deficiency anemia continue persisting as an unsolved risk factor/cause of DALYs especially for Mozambique-highly burdened by major health burden and food insecurity challenges. For both countries, data on anemia and iron deficiency in pregnant women and children aged 6 to 59 months old reported in the studies are either outdated or remain unclear as most of the findings were not disaggregated into these specific subgroups.
There is a lack of information on the effects of malaria chemoprevention, HIV/AIDS prevention and treatment strategies, micronutrient supplementation, iron correction treatment on women of reproductive age, pregnant women and children under 59 months living in these countries, as well, on the influence of social-structural determinants into the risk of anemia due to inequalities, especially in the current context of COVID-19 pandemic.
The strengthening of international and/or Community of Portuguese Language Countries (CPLP) network collaborations for innovative research on anemia and iron deficiency among this specific population group is needed, as well as the increase in awareness of anemia as impairment for life quality and a risk factor for maternal and child morbidity and mortality. Better resource-focused strategies for nutritional education, prevention, and treatment of anemia and iron deficiency can also be drawn by using more realistic and updated socio-contextual in-country-based evidence.

Limitations
The reporting of results of studies conducted in Mozambique and Portugal may be underestimated. As it is possible that some studies may have been conducted after the period set for this review or may be found in other databases (private, difficult to access by common users, or not included in this overview).
Our review was limited to 15 years (2003-2018) due to reasons previously stated. Once we finished our review, and despite being out of our primary study focus conduct any further evaluation of its content, we performed an update search and identified only five potentially relevant studies (Supplementary File S6) [70][71][72][73][74]. Thus, we observed that publications on this specific issue are still scarce for both countries. Language may also be a source of bias as most of the journals indexed in major databases are often published in the English language. Therefore, the actual number of studies might not be representative and be only a subset of studies that are being performed in these two Portuguese Language countries. We tried to overcome this limitation by broadening our search using additionally to the English language the Portuguese and Spanish languages.