Anaemia affects a large proportion of the population in both industrialised and developing countries, particularly vulnerable groups such as children, menstruating, lactating and pregnant women [1] . It is a widespread public health problem worldwide [2] . Anaemia contributes significantly to the global burden of diseases [3] [4] such as maternal mortality and the increase in the risk of adverse neonatal outcomes during pregnancy [5] . Globally, 30.1% of women of childbearing age were anaemic in 2019, with large geographical variations. Given the magnitude of anaemia [6] , the 65th World Health Assembly (WHA) approved a global nutrition goal in May 2012. The objective was to reduce anaemia among women of childbearing age by 50% by 2025, particularly in West and Central African countries. Gestational anaemia is a major public health problem worldwide, particularly in resource-limited regions where it is estimated that one in two pregnant women is anaemic [6] [7] . Previous studies showed that iron deficiency is the most common cause. Pregnant women suffer from iron depletion [8] . A pregnant woman’s iron requirements increase from 0.8 mg/day in the first trimester to 7.5 mg/day in the third trimester [9] . Adequate iron status during pregnancy implies body iron reserves of at least 1.5 mg/day during the third trimester. Also, serum ferritin concentrations above 70 g/L at the time of conception, but only 15% - 20% of women have reserves of this magnitude [10] .

Anaemia is defined as a condition in which the number of healthy red blood cells does not meet the body’s physiological requirements for oxygen delivery to vital organs [11] . It can lead to a reduction in work productivity, cognitive and behavioural development [11] [12] . The causes of anaemia in pregnancy are multifactorial, interdependent and context-specific. In developing countries, these include nutritional deficiencies (iron, folates, vitamins A and B12), infections (malaria, hookworms, tuberculosis and HIV) and heredity (sickle cell disease, β-thalassaemia) [13] [14] . Such factors causing anaemia during pregnancy vary according to the geographical location and socio-economic status of the community [15] . In malaria-endemic areas, Plasmodium sp. is a major determinant of anaemia during pregnancy with key prevention strategies including Intermittent Preventive Treatment (IPT) and the use of Long-Lasting Insecticidal Nets (LLIN) [16] [17] .

In Burkina Faso, the National Survey of Iodine and Anaemia carried out in 2014 showed anaemia prevalence of 61.9% and 72.5% respectively in women of childbearing age and pregnant women [18] . The prevalence of anaemia among women of childbearing age in the Centre Region was 57.7% [18] . The WHO recommends that pregnant women should receive daily iron and folic acid supplementation as part of antenatal care (ANC) to prevent anaemia during pregnancy. Despite this national prevention programme, the burden of anaemia in pregnancy remains high, suggesting that other contributing factors may be overlooked. Although more than half of women of childbearing age are anaemic in the central region of Burkina, there are little studies that assessed the associated factors of anaemia in pregnant women.

The aim of this study was to determine the prevalence of anaemia and to identify its determinants among pregnant women attending antenatal clinics in the Central Region of Burkina Faso during the peak of malaria and the harvesting period. The results of this study will help the Government and its partners to undertake appropriate prevention strategies, and will also provide a database for researchers.

Anaemia during pregnancy can result from both non-nutritional factors such as haemorrhage, chronic diseases and nutritional factors such as iron, vitamin and mineral deficiencies [21] . In this study, the prevalence of anaemia among pregnant women was 59.12%. This prevalence was higher than the global average reported by the WHO, which was 37% for anaemia in pregnant women aged 15 - 49 years [22] . It is close to those reported by the Demographic and Health Survey (56.6%) and by the study of Meda et al. (66%) in Burkina Faso [23] [24] . Other studies on anaemia during pregnancy have reported variable prevalence rates. The prevalence of anaemia in Turkey was 38.2%, while in Algeria it was 74% [25] [26] . These variations could be due to differences in the socio-demographic, nutritional factors and economic characteristics of the study populations in the countries concerned. On the other hand, public health measures and improved antenatal care services could account for these trends. Anaemia in this study was predominantly mild (47.88%) or moderate (51.06%). Only one woman had severe anaemia (1.06%). This woman was prim gravida, under 30 years of age, a housewife and had a negative malaria status. These characteristics corroborate the overall results and the associated factors obtained. This prevalence was lower than that found in Mali for the severe form (31.5%) [27] . Some authors estimate that the prevalence of anaemia and its degree of severity increases with the age of pregnancy, reaching 76.57% in the 3rd trimester [26] . This difference could be explained by the fact that in this study, the women enrolled were at their 1st antenatal consultation and were mostly in their 1st trimester of pregnancy. Pregnancy is the main factor associated with severe anaemia in women, with a risk of 3 to 4 times higher in pregnant women [28] . In this study, more than half of pregnant women (67%) suffered from normocytic anaemia and one woman in three had microcytic anaemia (30.90%). These proportions are different from those in the study by Demmouche and Moulessehoul in Algeria, who found that 54.9% of women had microcytic anaemia and that 45% had normocytic anaemia [26] . Adebo et al. noted that all anaemic pregnant women were microcytic and normochromic [29] . The occurrence of microcytic anaemia could be related to iron deficiency. In the case of normocytic anaemia, it could be related to the haemodilution usually observed during pregnancy [30] . Macrocytic anaemia could be the result of folate or vitamin B12 deficiency [28] .

Anaemia was higher in women with no education. This finding is consistent with the results of other publications [31] [32] . The study also showed that women with no or low levels of education had a low dietary diversity score. They were 2 times more likely to have a low DDS and 5 times more likely to have a borderline FCS. This could be due to the fact that illiterate women may be economically unstable [33] and may not meet their nutritional needs (foods rich in vitamin A, iron and folic acid, etc.) during pregnancy, leading to anaemia. Two studies in the sub-region reported variable results depending on the level of education [29] [34] . Therefore, the level of education can improve awareness of good nutrition during pregnancy and throughout life [35] . Thus, improved nutrition reduces vulnerability to anaemia in pregnant women. Similarly, the present study found an association between women age and anaemia during pregnancy. The rate of anaemia was higher in women aged between 20 and 34. They were 3 times more likely to suffer from anaemia than pregnant women aged over 34. This result is consistent with studies carried out in Nigeria [36] , Uganda [37] and Saudi Arabia [38] . Various pieces of evidence have shown that early marriage is associated with low economic status, dropping out education and pregnancy complications (including anaemia) [39] [40] . Some studies have shown a higher risk in women with two or more children [25] [41] [42] . In this study, the results did not confirm the correlation between parity and abnormal haemoglobin concentrations. Prim gravidas and secondi gravidas were more affected. However, they are in agreement with two previous studies in Malawi and Nigeria [43] [44] . These results may be explained by the fact that multigestational women, thanks to the experience of their previous pregnancies, are more likely to adhere to good practice advices from health workers for the prevention of anaemia during pregnancy. Conversely, a study in Ethiopia showed that the risk of anaemia increased with parity [35] . This finding was similar to the results of the Jordanian study where a statistically significant increase (p < 0.01) was found between the overall prevalence of anaemia in multiparous women (64.0%) compared to primigravida women (49.3%) [45] . In this study, the prevalence of anaemia was higher in pregnant women with an inter-gestational interval of more than 2 years. Birth spacing was not found to be protective against anaemia. These results are in line with those of a study in Cameroon [30] . Variations could be attributed to differences in diet and differences in populations. This study found that subjects with low serum iron levels were more likely to suffer from anaemia (71.4%) than those with normal serum iron levels (57.9%). The prevalence of normosiderimic anaemia was lower than that reported in Benin (75%) in pregnant women [29] . The difference was not statistically significant. Women with a high transferrin level were 2 times more affected than those with a normal transferrin level. The difference was statistically significant (OR = 2.291, 95% CI: 1.107 - 4.742, p = 0.024). Pregnant women with anaemia and high transferrin levels showed iron depletion, implying iron deficiency anaemia. With regard to malaria infection, no statistically significant association with anaemia was found. This result is in agreement with those of other authors who also found no correlation between the occurrence of anaemia in pregnant women and malaria infection [29] [46] . This could be linked to the fact that the population observes protective measures against malaria through the daily and appropriate use of LLIN.

In addition, the quality of the pregnant women’s diet was assessed using the dietary diversity score and no correlation was found with the occurrence of anaemia during pregnancy. However, malnourished pregnant women (91.67%) were 8 times more likely to be anaemic than pregnant women in a normal nutritional state (OR = 8.482, 95% IC = 1.06 - 67.41). In contrast, Nikiema et al. in Burkina Faso found no association between brachial circumference and anaemia [46] . This study found a significant association between anaemia and the consumption of vegetables/fruit rich in vitamin A, legumes and fruit. Consumption of iron-rich foods such as meat, dark green leafy vegetables and eggs showed no statistically significant difference. These results are in line with those reported in a recent study [47] .

The proportion of anaemic pregnant women living in peri-urban area (70%) was high compared with those living in urban area (41%). Similar results have been reported in Ethiopia [35] [48] [49] . In their study, Samuel et al. reported that peri-urban residents were 2 times more likely to be affected by anaemia than urban residents. The higher prevalence of anaemia among pregnant women in peri-urban areas was probably linked to the lack of information on adequate nutrition during pregnancy and to socio-economic factors.

The determination of biological examinations carried out in the three health centres with different technical platforms may lead to results of different rank. Calibrations or the diversity of methods and equipment used may influence the results. The dietary diversity of women estimated on the basis of a simple 24-hour recall may not reflect daily and seasonal variations although the questionnaire is the one recommended by the FAO [50] . Potential memory bias can occur during the survey. This study did not consider household food security and income levels. It should be noted that there are differences in the methodology used to calculate the dietary diversity score for women. In some studies, the DDS is determined on the basis of the 10 food groups using a 24h recall data set, with points only acquired when the food subcategories are consumed. In this study, a 9 food groups method was used.

This study shows that the prevalence of anaemia was high with the majority of women being mildly and moderately anaemic. This prevalence exceeds the WHO global critical threshold. In addition, several factors were associated with the occurrence of anaemia, such as low levels of education and low consumption of vegetables/fruits rich in vitamin A. To tackle this problem, efforts should focus on raising women’s awareness of early prenatal consultations. This would enable iron and folic acid supplementation to be introduced in good time, provide good nutritional education for a diversified and balanced diet, and also intensify anaemia education at these prenatal consultations, which could reduce the prevalence of anaemia during pregnancy.

The data are available to anyone who expresses a need by contacting me

LM and PAN designed and carried out the study. LM, KV and SM participated in data collection, analysis and interpretation. LM wrote the manuscript. PAN and FWD made critical revisions of the article. The final manuscript was approved by PAN and JS. All authors have read and approved the final manuscript as submitted.

Thanks to the managers and staff of the Saint Camille hospital in Ouagadougou and the Schiphra hospital for their excellent cooperation in conducting the study. Thanks to the Centre Regional Health Department and the staff of the Medical Centre of Saaba for their valuable collaboration. Thanks to the interviewers for their help in collecting high-quality data. Thank you to the women and their husbands for their voluntary participation in this study.

ANC: Ante-Natal Consultation; BC: Brachial Circumference; DDS: Dietary Diversity Score; ECHR: Ethics Committee for Health Research; FAO: Food and Agriculture Organization; FCS: Food Consumption Score; HOSCO: Hôpital Saint Camille de Ouagadougou; IPT: Intermittent Preventive Treatment; LLIN: Long-Lasting Insecticidal Nets; MCV: Mean Corpuscular Volume; MMS: Multiple Micronutrient Supplementation; WCA: Women of Childbearing Age; WHA: World Health Assembly.