You are viewing a javascript disabled version of the site. Please enable Javascript for this site to function properly.

# Prolonged exclusive lactation and low educational level of mothers as potential risk factors for the occurrence of iron deficiency anemia among young Algerian preschool children living in poor rural area (Djelfa)

#### Abstract

BACKGROUND AND OBJECTIVES: In Algeria, iron deficiency anemia is a major public health problem. A descriptive cross-sectional study was carried out to assess the prevalence of anemia and iron deficiency anemia and to investigate the factors associated with iron deficiency anemia.

MATERIALS AND METHODS: A total of 368 clinically healthy children aged between 6 months and 5years living in a poor rural area of Algeria were assessed in this study. Biological indicators of iron status were assayed, and those of malnutrition were calculated. Socio-economic and dietary data were also collected.

RESULTS: The prevalence of iron deficiency anemia was 20.92% (95%CI: 17–25%), accounting for 86.5% of anemia cases. Nutritional components appeared as a potential cause of its occurrence. Binary logistic regression demonstrated that this elevated frequency is significantly correlated with risk factors relating to infancy(<2 years) (OR = 2.68 [95CI% : 1.47–4.97]),late introduction of weaning foods (OR = 2.51 [95%CI:1.29–5.05]), exclusive lactation(OR = 3.22 [95%CI:1.37–7.6]), low educational level of mothers (OR = 3.42 [95%CI: 1.52–7.65]) and gender (boys) (OR = 2.39 [95%CI:1.39–4.39]).

CONCLUSIONS: Anemia and iron deficiency anemia were significant health problems among the studied sample. The improvement of the socio-economic status of mothers and weaning practices should both be included in public health strategies for the control and prevention of iron deficiency anemia in our population.

## 1Introduction

Iron deficiency anemia (IDA) is considered the most widespread dietary disorder in the world [1], affecting approximately 30% of the world’s population [2]. Preschool children and pregnant women constitute the most vulnerable community because of their important physiological needs linked to growth and gestation respectively [3–6]. Iron deficiency is the major cause of anemia [7], during which irreversible alterations of neurocognitive functions and growth retardation may occur in children [8].

Generally, the prevalence of anemia and IDA among preschool children is estimated at 47.4% and 27% worldwide, respectively, with African countries and Southeast Asia recording the highest frequencies [9].

In 2011, according to WHO estimations, more than three out of ten Algerian preschool children were affected by anemia [10, 11]. The nutritional component, particularly the low bioavailability of iron in the Algerian diet, remains the main cause of this disorder [12]. There are several disparities in terms of infrastructure and care access between urban and poor rural sectors in Algeria [12], contributing to the particular vulnerability of rural populations, especially preschool children, to certain disorders such as anemia and IDA [13, 14].The dimensions of the problem have not been well defined, and few studies using reliable biochemical indicators have been performed to assess the iron status of Algerian children. Indeed, the prevalence of iron deficiency anemia and its associated factors remain largely unknown in many rural sectors of the country. We believe that the effectiveness of any strategic public health program to fight iron deficiency and IDA in preschool children must be based on knowledge of the prevalence and risk factors associated with these anomalies. Therefore, the aim of the current study was to determine the current prevalence of anemia and IDA and to investigate the possible socio-economic and nutritional risk factors associated with IDA among preschool children living in a poor rural area ofAlgeria.

## 2Materials and methods

### 2.1Location and population of the study

Our study was undertaken between June 2013 and January 2015 at the pediatric ward of the health center of the municipality of El Idrissia (southwest of Algiers). Sheep farming is the main economic activity in this area. A total of 368 children, aged between 6 months and 5 years, all seen for a vaccination pattern, were selected for the survey. On the basis of a 42.5% prevalence of anemia in Algerian preschool children [9], the prevalence of IDA (P) is estimated at approximately 40%. For sample size calculation, power was determined in the study as equal to 0.8 and alpha (α) was equal to 0.05 (as in a previous study [15]). Our estimation for the sample size was as follows:

N=Z2xPx(1-p)D2;N=(1.96)2x0.4x(1-0.4)0.052=368

where N is the required sample size, Z is the standard deviation with a 95% confidence interval, P is the estimated prevalence, and D is the 5% absolute precision.

The criteria for inclusion of children in this study were as follows: apparently healthy (confirmed by clinical examination), lack of medical and surgical history, no blood transfusion during the 12 months preceding the survey day and non-administration of any treatment with iron, folic acid or vitamin B12.

### 2.2Ethics and data collection

The study was approved by the scientific council of the National High School of Agronomy of Algiers. The study was validated by the pediatric ward of the health center of the municipality of El Idrissia and conforms to the guidelines of the World Medical Association Declaration of Helsinki (2000). Verbal parental consent was obtained, witnessed and formally recorded for all study participants prior to enrollment. A questionnaire was developed to collect data on several risk factors associated with increased frequency of iron deficiency anemia in children in accordance with previously conducted surveys. The investigated factors were age, gender, level of maternal education, parental income, spacing between pregnancy of the selected child and the preceding offspring, mother’s age, family size, birth weight, dietary data such as the type and duration of lactation, weaning age and the nature of daily consumed foods. The definitions of exclusive breastfeeding and age of introduction of weaning foods are in accordance with World Health Organization (WHO) definitions [16]. The type and frequency of daily consumed foods were investigated using a five-step multiple-pass five dietary repeated 24 hour recall method [17], where in mothers were asked to remember all foods eaten by their children during 3 non-successive days (one of the three days recorded included one weekend day) following the day of vaccination. The average of all three 24 h dietary recalls was used as a representation of individual intake. The mean of dietary iron intake was calculated using Diet Analysis Program, 1995 (Lifestyles Technologies, Inc., Northbridge Point, Valencia, California) supplemented by the African [18] and Tunisian food composition table [19] and then compared to the recommended daily allowance (RDA) according to the infant’sage [20].

### 2.3Anthropometric data

For each child, we conducted measurements of weight and height, which were used to calculate their BMI. Weight measurements were performed using an electronic scale (SECA881, GMBH, Germany) with a maximum range of 200 Kg. Height was measured using a mobile measuring mat for children under 24 months (SECA210, GMBH, and Germany) and a stadiometer for older children. Each measurement was carried out in accordance with the standards and recommendations of the anthropometric measurements guide [21]. The data obtained were entered and subsequently converted into z-scores using the software Anthropo Plus OMS (V.1.0.4). Underweight was considered present if the child had a z-score (weight for age) less than 2 standard deviations from that of the reference population (WZA< -2SD), while stunting was considered present only if the z-score of height for age was less than 2 standard deviations (HZA< -2SD). Overweight was considered present if the z-score of BMI for age was more than 2 standard deviations away from the reference population (ZBMIA> +2SD) [22]. For corpulence, we considered the BMI to be normal if it was between the 3rd and the 97th percentile [22]. The weight at birth was deemed low if it was less than or equal to 2.5 kg [23].

### 2.4Sampling, hematology and inflammatory data

All children included in this study underwent an assessment of their iron status and characterization of their inflammatory status. To this end, approximately 9 to 10 mL of venous blood was collected in the morning (around 10 am) by a qualified laboratory technician at the polyclinic of El Idrissia. This sample was obtained at the crease of the elbow of each child and divided equally between two 5 mL tubes: one contained ethylene diamine tetra acetic acid (EDTA), whilst the other was dry. The EDTA-containing tube was used to determine the blood count using a hematological counter (ERMA PCE-210, Erma Inc. Japan) at the laboratory of the polyclinic of El Idrissia. The latter tube was used to obtain serum by centrifugation for 5 min at 3200 rpm. The recovered serum was stored at –20°C and subsequently used for the determination of serum iron biochemical parameters (serum iron [SI] using a colorimetric method [24], serum transferrin [Trsf] and serum ferritin [Ft] were assayed by immunoturbidimetric testing [25], and inflammation (C-reactive protein [CRP] was assayed by agglutination [26]) using a Cobas Integra 400 Plus biochemistry automaton). Using the serum iron and transferrin data, we calculated the total capacity for the binding of iron by transferrin (TIBC) and the saturation coefficient (TS) using the following formulas:

TIBC(μmolL)=Tranferrin(gL)x25,1[27];TS(%)=Sarumiron(μmolL)TIBC(μmolL)x100[28].

Anemia was defined as a hemoglobin (Hb) threshold ≤11 g/dL and the following thresholds were used to define its intensity: severe anemia (Hb < 7 g/dL), moderate anemia (7≤Hb≤9.9 g/dL), and mild anemia (10≤Hb < 11 g/dL) [29]. Microcytosis was defined as a mean corpuscular volume rate (MCV) <75 fL [30], while mean corpuscular hemoglobin (MCH) <27 pg defined hypochromia [31]. The total depletion of iron stores was defined as ferritin levels less than 12 μg/L (Ft < 12 μg/L) and less than 30 μg/L when inflammation was considered present [32], while a low coefficient of transferrin saturation (TS < 16%) [33, 34] with completely depleted iron stores Ft < 12 μg/L was used to identify iron deficiency. In addition, iron deficiency anemia was considered present if the hemoglobin level in children with iron deficiency was low (Ft < 12 μg/L+ST<16% +Hb<11 g/dL). Inflammation was considered present when CRP levels were higher or equal to 5 mg/L [35].

### 2.5Statistical analysis

Statistical analysis was performed using software (SPSS, version 19 Statistics, IBM, USA). Normal distribution of the variables was determined using the Kolmogorov-Smirnov test. Normally distributed data are given in means and standard deviations, whereas non-normally distributed data are depicted in medians and interquartile ranges. For data comparison of two independent samples, we employed the Mann-Whitney test and t-test for non-normally distributed and normally distributed data, respectively. The Chi-square test was used to identify the relationship between independent variables. In addition, the degree of association between quantitative variables was analyzed with the nonparametric correlation Spearman rank test. Finally, we used binary logistic regression to predict the risk factors associated with an increase in the occurrence of iron deficiency anemia with a 95% confidence interval (CI = 95%) and a significance level of p≤0.05.

## 3Results

### 3.1Descriptive and socio-demographic study of the sample

##### Table 3

Prevalence of anemia and IDA according to age and gender

 Anemia P- Value† IDA P-Value† N % (CI 95%) N % (CI 95%) Total (N = 368) 89 24.18 (19,79–24.58) 77 20.92 (17–25%) Gender Boys (N = 192) 62 32.29 (25.62–38.77) 0.0001** 52 27.08 (21–33.03) 0.002** Girls (N = 176) 27 15.34 (9.96–20.72) 25 14.2 (9.01–19) Age <24 months (N = 147) 54 36.73 (28.85–44.62) 0.0001** 43 29.25 (21–36.02) 0.003** ≥2 4 months (N = 217) 35 16.12 (10.99–20.69) 34 15.66 (11–21.1)

**:Highly significant. :P value of Chi-squared test. IDA: Iron deficiency anemia. N = Number. CI: Confidence interval.

##### Table 4

Correlative study of biological parameters of iron status, mean iron intake and socio-economic data

 IS FT TRSF Hb TS MCV MCH Iron intake‡ PI MI Family Size IS 1 0.51** –0.49** 0.73** 0.842** 0.719** 0.656** 0.234** 0.042 0.114* –0.56* FT 1 –0.62** 0.535** 0.43** 0.543** 0.5** 0.138* –0.01 0.09 –0.108* TRSF 1 –0.627** –0.799** –0.64** –0.61** 0.226** –0.004 –0.095 0.07 HB 1 0.828** 0.857** 0.785** 0.226** 0.038 0.083 –0.034 TS 1 0.676** 0.715** 0.307** 0.046 0.085 –0.07 MCV 1 0.867** 0.243** –0.019 0.098 –0.008 MCH 1 0.24** –0.007 0.054 –0.11* Iron Intake‡ 1 0.029 0.105* –0.088 PI 1 0.219* –0.036 ME 1 –0.118* Family Size 1

*Significant (p < 0.05); **highly significant (p < 0.01). : correlative test using Spearman correlation; : % of calculated mean of daily iron intake compared to RDA (recommended dietary allowance) [20]; IS: iron serum; Hb: hemoglobin (g/dL); TS: transferrin saturation (%); MCV: mean corpuscular volume (fL); MCH: mean corpuscular hemoglobin (pg); FT: Ferritin level (μg/dL); TRSF: Transferrin level (g/L). ME: mother’s educational level (number of academic years); PI: parental income.

##### Table 5

Univariate analysis of investigated risk factors for IDA in children

 Normal (N = 291) % (CI95%) IDA (N = 77) % (CI95%) Crude OR [CI95% ] P value Total (N = 368) 79.08 (74.9–83) 20.92 (17–25) – – Gender Boys (N = 192) 72.91 (67–79.3) 27.08 (21–33%) 2.24 (1.32–3.81) 0.002** Age<24 months (N = 147) 71.4 (64.04–78.8) 28.57 (21–36) 2.12 (1.27–3.5) 0.003** ME Low or Medium† (N = 278) 76.61(79.5–93.3) 23.38 (18–28) 1.98 (1.01–3.86) 0.04* PI Low‡ (N = 161) 78.26 (72.59–83) 21.73 (16–27) 1.12 (0,67–1.86) 0.66 Mothers’ age >30 years (N = 213) 77 (71.3–82.69) 23 (17–29) 1.35 (0.8–2.27) 0.25 Underweight (WAZ< -2SD) (N = 16) 75 (51–98.83) 25 (10–49) 1.27 (0.39–4.06) 0.68 Stunting (HAZ< -2SD) (N = 72) 76.38 (66–86.4) 23.6 (13–32) 1.14 (0.67–2.13) 0.67 Overweight (BMIZ> +2SD) (N = 27) 81.49 (74–83.2) 18.51 (17–25) 0.84 (0.31–2.32) 0.75 VC Incomplete (N = 44) 86.37 (73.6–82.6) 13.63 (3–21) 0.56 (0.22–1.38) 0.2 Exclusive lactation§ (N = 34) 55.88 (38.3–73) 44.1 (27–62) 2.61 (1.24–5.49) 0.009** Weaning age >6 months (N = 67) 64.18 (52–75) 35.82 (24–48) 2.63 (1.46–4.46) 0.001** Iron intake <50% of RDA (N = 112) 71.56 (62–80.4) 28.43 (20–37) 1.8 (1.06–3.07) 0.028* SBP<18 months (N = 182) 78.03 (71.9–84) 21.97 (16–28) 1.13 (0.68–1.87) 0.623 Birth weight <2.5 kg (N = 41) 73.17 (58–86.9) 26.8 (13–48) 1.5 (0.71–3.17) 0.279 Pregnancy Care Irregular (N = 89) 78.5 (72.7–89.2) 22 (17–26) 0.86 (0.47–1.57) 0.627 Large family size >8 members (N = 18) 72.2 (49.3–95.14) 27.7 (5–51) 1.48 (0.51–4.3) 0.46

*: Significant (p < 0.05). **: Highly significant (p < 0.01). : less than 12 academic years; : less than 7 U\$/day; §: exclusive lactation: breast feeding, both breast and/or bottle feeding, fed exclusively on infant formula and fed exclusively on whole milk powder; ME: mother’s educational level; PI: parental income; RDA: recommended dietary allowance [20]; VC: vaccination coverage; OR: Odds Ratiov; SD: standard deviation; IDA: iron deficiency anemia; SBP: spacing between pregnancy of the selected child and preceding offspring; HAZ: Height for age z-score; WAZ: weight for age z-score: BMIZ: Body mass index for age z-score; SD: standard deviation. CI: Confidence interval.

##### Table 6

Multivariate analysis of potential factors for IDA among the investigated children†,‡

 Prevalence of IDA Estimated β EXP (β) CI 95% for EXP (β) P-value Variable Constant –3.36 – – Age (<24 months) 0.98 2.68 1.45–4.97 0.002** Gender (Boys) 0.78 2.39 1.29–4.39 0.005** Low mother’s educational level (Less than 12 academic years) 1.23 3.42 1.52–7.65 0.003** Exclusive lactation§ 1.17 3.22 1.37–7.6 0.007** Late weaning age (>6 months) 0.92 2.51 1.29–5.05 0.011*

*: Significant P < 0,05; **: highly significant P < 0,01. : Variables included in the model were as follows: age, gender, mother’s educational level, exclusive lactation, weaning age and mean of daily iron intake. : Multivariate analysis using binary logistic regression. §: Exclusive lactation includes exclusive breastfeeding and/or bottle feeding. IDA: iron deficiency aneamia. β: Estimated parameter. CI: Confidence interval.