Magnitude and Determinates of Anemia among adolescent Girls in Africa: A Multilevel, Multicounty Analysis of 24 Countries

doi:10.21203/rs.3.rs-5004469/v1

Download PDF

Research Article

Magnitude and Determinates of Anemia among adolescent Girls in Africa: A Multilevel, Multicounty Analysis of 24 Countries

https://doi.org/10.21203/rs.3.rs-5004469/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Introduction: Anemia in Africa is a pressing public health issue with far-reaching consequences. Adolescents, especially girls, are more vulnerable to developing anemia due to rapid physical growth and development, menstruation and poor diets. Identifying the determinants of anemia among adolescent girls in Africa is crucial for developing appropriate interventions, yet evidence remains scarce.

Objective: This study aims to investigate the magnitude and determinants of anemia among adolescent girls in Africa.

Methods: Data from the recent Demographic and Health Surveys (DHSs) of 24 African countries was used, focusing on 38,966 adolescent girls aged 15–19 years. The chi-squared test was utilized for bivariate analysis, and the relationship between predictor variables and anemia was evaluated using bivariate and multivariable binary logistic regression models.

Results: The overall magnitude of anemia among adolescent girls was 43.6% (95% CI: 42.97, 44.21). The research has identified several determinants associated with anemia in adolescent girls, including education status, wealth status, contraceptive use, nutritional status, pregnancy status, and net use. These factors provide valuable insight into potential avenues for intervention and prevention efforts. The multivariable analysis indicated that education status (primary education AOR = 0.72; 95% CI = 0.50–0.80 and secondary education AOR = 0.76; 95% CI = 0.66–0.86), wealth status (being rich AOR = 0.86; 95% CI = 0.76–0.98), contraceptive use (AOR = 0.46; 95% CI = 0.40–0.53), and nutritional status were inversely associated with adolescent anemia. In contrast, pregnancy status (AOR = 1.13; 95% CI = 1.01–1.27) and treated net use (AOR = 1.26; 95% CI = 1.15–1.39) were positively associated with anemia among adolescent girls.

Conclusion:

This study serves as a catalyst for informed action and underscores the urgent need for comprehensive interventions aimed at addressing the multifaceted determinants of anemia among adolescent girls in Africa. By targeting these key factors, public health initiatives can make significant strides towards improving the health and well-being of young women across the continent.

To prevent adolescent anemia, it is recommended to encourage girls to pursue education, prevent adolescent marriage and pregnancy promote contraceptive use among married or sexually active girls, and educate on the correct use of treated nets.

Anemia

Adolescent girls

Africa

Health and Nutrition

Anemia is a condition in which the number of red blood cells or hemoglobin is lower than normal, reducing the blood's capacity to carry oxygen. It can result from several factors: nutrient deficiencies or poor nutrient absorption, infections (e.g., malaria, parasitic infections, tuberculosis, HIV), inflammation, chronic diseases, gynecological and obstetric conditions, and inherited red blood cell disorders. The most common cause of anemia is iron deficiency, which can occur due to inadequate dietary intake, poor iron absorption, or increased needs during growth spurts and menstruation[1].

Approximately one-quarter of the global population is estimated to be anemic, with the prevalence notably higher among women, expectant mothers, adolescent girls, and children under the age of five. Anemia is the third-leading cause of years lived with disability (YLDs) globally[2]. Between 1990 and 2021, the evidence indicates a global trend towards less severe anemia. The most significant reductions were observed in adult males; however progress has been slower among women of reproductive age including adolescent girls aged 15–19 and children under the age of five. In 2021, sub-Saharan Africa and South Asia are facing the greatest burden [3].

Adolescence is a period of transition from childhood to adulthood, characterized by rapid physical growth and increased nutritional needs. Among adolescents aged 10–19, especially in low- and middle-income countries, anemia is a significant issue. During this stage, the body undergoes rapid growth and development, increasing nutritional demands, including for iron. Iron deficiency anemia adversely affects the educational and economic well-being of adolescents. It has been linked to stunting, wasting, being underweight, poor cognitive function, reduced physical activity, and attention deficit hyperactivity disorder in adolescents [4, 5]

The onset of menstruation further raises the risk of anemia due to additional iron loss among adolescent girls. Adolescent girls are vulnerable especially to iron deficiency due to accelerated increase in requirements for iron, poor dietary intake of iron, menstrual losses, infection, norm of early marriage, and adolescent pregnancy. Adolescents with anemia are more likely to experience anemia during pregnancy, which can lead to various adverse effects on both pregnancy and childbirth [6, 7].

Adolescent girls in Africa are particularly vulnerable to anemia due to a combination of factors. Inadequate diets and poor absorption of iron, compounded by high rates of infectious diseases like malaria and parasitic infections, poor hygiene management, contribute significantly. Limited access to healthcare and education, along with poverty and gender inequality, exacerbate the issue. Rapid growth during adolescence increases nutritional needs, which are difficult to meet in regions with limited resources [8–12].

Studies have identified factors significantly associated with anemia among adolescent girls including the respondent's age, marital status, educational level, nutritional status and wealth status [7, 13, 14]. Additionally, early marriage and pregnancy, which are prevalent in many African countries, are linked to a range of social, physical, and health problems, including anemia among adolescents. Early marriage often leads to early pregnancy, which increases the risk of anemia due to higher iron requirements and inadequate nutritional intake. Moreover, adolescents who marry early may face limited access to healthcare, education, and economic opportunities, further exacerbating their vulnerability to anemia. These factors contribute to a cycle of poor health and socio-economic disadvantage that can have long-lasting effects on their overall well-being and future prospects [15, 16].

Despite global efforts to mitigate anemia, progress has been slower among adolescent girls, especially in low- and middle-income countries. The current trends in anemia make it challenging to achieve the Sustainable Development Goal (SDG) of reducing anemia among women of reproductive age by 50% by 2030 [17, 18]. Addressing all forms of malnutrition is a key target of the SDGs. By effectively reducing malnutrition, the prevalence of anemia can also be lowered, thereby contributing to the successful achievement of the SDGs by 2030. To meet this ambitious goal, comprehensive strategies must be implemented including assessment the magnitude and identify the determinants associated with anemia among adolescent girls [19].

While studies have estimating the magnitude and determinants of anemia among women of reproductive age, less attention has been given to adolescent girls and multicounty analyses are rare. This study aims to provide a comprehensive understanding of the magnitude and determinants of anemia among adolescent girls in Africa by analysing data from 24 countries. These insights are crucial for developing targeted and effective policies and interventions to improve health outcomes and overall well-being for adolescent girls across the continent. This study will help identify which regions or countries need the most urgent attention and which determinants are most critical to address. A multicounty analysis will fill existing data gaps and highlight regional differences, allowing for tailored interventions that meet local needs and challenges. This comprehensive view will ensure that resources are allocated where they are needed most.

Data source and study participants

Data from the most recent demographic and health surveys (DHSs) from 24 African countries between 2016 and 2022 was used for this study. The DHS implements a stratified two-stage cluster sampling design which gives nationally representative samples. The sampling design involved randomly selected communities (clusters) at the first stage and households at the second stage. Further details on the implementation of DHS survey were available in other sources [20]. In total, 182,683 women age 15–49 years were agreed and their hemoglobin level measured. Anemia testing was performed after getting consent from the participants. Among reproductive-age women, only 38,966 were adolescents (15–19 years). Thus, in this study our analytical sample was limited to 38,966 female adolescents (15–19 years).

Measurements

Outcome variable

The outcome variable was the anemia status among adolescent girls which was dichotomized into two categories: non-anemic, coded as “0,” and anemic, coded as “1” (this category includes mild, moderate, and severe anemia). The anemia status was determined using the women’s hemoglobin levels, measured according to the cut-off points defined by the World Health Organization (WHO) [21].

Predictor variables

The variables included in this analysis was marital status, age at first marriage, educational level, employment status, place of residence, nutritional status, pregnancy status, history of childbearing/parity, contraceptive use, use of mosquito nets, and health insurance coverage. The wealth index was constructed based on household ownership of selected assets, such as televisions and materials used for constructing the house, Source of drinking water, type of toilet facility etc. Household asset factor scores were generated using principal component analysis, then standardized and categorized into quintiles (poorer, poor, middle, rich, and richer) [20] which later grouped in two three categories (poor, middle, and rich) in this analysis.

Statistical Analysis

The complexity of analyzing DHS survey datasets can be attributed to two main factors: i) the use of stratified multistage cluster sampling and, ii) the unequal probabilities of selection from target populations for sampled elements, often due to oversampling of key subgroups. To address this complexity, sampling weights were used during the estimation of proportions and their respective confidence intervals [22]. Descriptive analysis involved computing the socio-demographic characteristics of adolescent girls, presented in frequencies and percentages (see Table 1). Additionally, weighted percentages of anemia by different factors were computed (see Table 3).

Given that adolescent girls living in the same cluster may have similar anemia status, a modeling approach that accounts for the correlated nature of the data within the cluster is necessary. This involves introducing sampling clusters as random effects to accommodate cluster heterogeneity and unobserved covariates within clusters. A multilevel logistic regression model was employed to investigate the association between adolescent girls’ anemia status and demographic factors, with survey clusters treated as a random effect whilst adjusting for various covariates. Let $\:{y}_{ij}$ denote the binary outcome for adolescent girls i in cluster j, and assume $\:{y}_{ij}$ follows a Bernoulli distribution with probability of success, $\:{p}_{ij}$. Then, using the usual logit link function, a binary outcome can be associated with a linear predictor as follows:

$$\:\text{log}\left(\frac{{p}_{ij}}{1-{p}_{ij}}\right)=logit\left({p}_{ij}\right)={x}_{ij}^{{\prime\:}}\beta\:+{b}_{j},\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\left[23\right]$$

Where, $\:X$ is a vector different covariate presented in Table 1 with associated regression coefficients$\:\beta\:$. The random-effect component of the model, $\:{b}_{j},$ are mutually independent Gaussian random effects used to capture within-cluster correlation [24].

The model utilized a Bernoulli probability distribution to represent adolescent woman’s anemia level with the linear predictor assuming a specific form (Eq. 1). The random-effect component of the model captured within-cluster correlation using mutually independent Gaussian random effects with mean 0 and variance $\:{\sigma\:}_{b}^{2}$. The fixed effects parameter in the model had a conditional interpretation, reflecting the consequence of changing the value of an explanatory variable for a given value of the random effect and the other fixed effects [A]. Both descriptive and inferential data analyses were done using Stata 18.0 software [25]. In this study, for statistical significance, the considered type-I error alpha value is 0.05.

Ethical consideration

This study exclusively involved secondary analysis of existing DHS data. The data collection procedures for the DHS adhered to all relevant guidelines and regulations. Ethical approval for the surveys was obtained from the Institutional Review Boards of each respective country. For adolescent girls under the age of 18, parental or guardian permission was required prior to participation in the study. The authors obtained authorization from the DHS program to use the data: https://dhsprogram.com/data/available-datasets.cfm

Table 1 presents the socio demographic, economic and health characteristics of 38,966 adolescent girls sampled from various regions in Africa. The majority of the adolescents (60.1%) reside in rural areas. Regarding marital status, about (79.9%) adolescents were single/never married and 21.6% were been married. Among those who have been married, a larger proportion (76.8%) of adolescents married between the ages of 15–19, while 23.2% married before the age of 15.

Nearly half of the adolescents (48.9%) had attained secondary education, while 35.7% had completed primary education. Around 14.4% has no education, and a smaller fraction had higher education (1.0%). In terms of wealth distribution, about 45.6% were classified as rich, 34.1% as poor, and 20.3% as middle-income. About 67.9% of the adolescents were not working, while 32.1% were engaged in some form of employment (Table 1).

About 6.1% adolescents were pregnant during the data collection period. A large proportion of the adolescents (89.6%) do not use contraceptives. About 15.3% adolescents had one child, and 2.7% have two or more children.

Most adolescents (88.5%) do not have health insurance. The majority of adolescents had a normal weight (72.0%), while 17.6% were underweight, 8.4% were overweight, and 2.1% were obese. Most adolescents (65.7%) do not use mosquito nets, while 34.0% used treated nets, and 1.8% use untreated nets.

Table 1

Sampled adolescents characteristics in Africa (N = 38,966)
Predictors/factors			Frequency	Proportion
Residence
	Urban		15544	39.9
	Rural		23422	60.1
Marital status
	Never married/Single		30668	79.9
	Ever Married		8298	21.62
Age at first Marriage (n = 8,174)
	< 15		1927	23.2
	15–19		6371	76.8
Education status
		No education	5610	14.4
		Primary	13924	35.7
		Secondary	19047	48.9
		Higher	385	1.0
Wealth status
	Poor		13300	34.1
	Middle		7910	20.3
	Rich		17756	45.6
Currently pregnant
	No		36602	93.9
	Yes		2364	6.1
Use of contraceptive
	No		34922	89.6
	Yes		4044	10.4
Health insurance
	No		34497	88.5
	Yes		4468	11.5
Currently working
	No		26448	67.9
	Yes		12500	32.1
Parity
	No		31939	82.0
	1		5977	15.3
	2+		1050	2.7
Body mass index
	Underweight		6315	17.6
	Normal weight		25845	72.0
	Overweight		3009	8.4
	Obese		746	2.1
Type of mosquito net
	No net		25222	65.7
	treated net		13058	34.03
	Untreated net		686	1.78
Anemia status
	Not anemic		21980	56.4
	Anemic		16986	43.6

Table 2 presents the percentage of anemia among adolescent girls in various African countries. Several countries show alarmingly high rates of anemia among adolescent girls. For instance, Gabon has the highest prevalence at 72.0%, followed by Mali with 65.4%, Ivory Coast with 62.2%, Nigeria with 60.5%, Burkina Faso with 57.6%, Liberia with 55.1%, Mozambique with 54.9%, and Togo with 54.7%.

The total data aggregated from all countries show that out of 38,966 adolescent girls 16, 986 (43.6% 95% CI:42.97_, 44.21) adolescent girls were anemic. The prevalence of anemia ranged from 14.7% in Rwanda to 72% in Gabon (Table 2).

Table 2

Magnitude of Anemia among Adolescent Girls in Africa
Country	Not Anemic n(%)	Anemic n(%)
Burkina Faso	835 (42.4)	1136 (57.6)
Burundi	1235 (64.1)	692 (35.9)
Democratic Republic of the Congo	1239 (59.9)	830 (40.1)
Ivory Coast	586 (37.8)	966 (62.2)
Cameroon	966 (58.3)	692 (41.7)
Ethiopia	2574 (80.1)	641 (19.9)
Gabon	287 (28.0)	738 (72.0)
Ghana	778 (56.2)	608 (43.8)
Gambia	739 (56.5)	568 (43.5)
Equatorial Guinea	683 (52.9)	609 (47.1)
Liberia	398 (44.9)	488 (55.1)
Lesotho	564 (75.9)	179 (24.1)
Mali	341 (34.6)	645 (65.4)
Malawi	1126 (64.7)	616 (35.3)
Mozambique	1386 (45.1)	1688 (54.9)
Nigeria	1088 (39.5)	1665 (60.5)
Niger	461 (54.0)	393 (46.0)
Rwanda	1403 (85.3)	242 (14.7)
Sierra Leone	805 (50.6)	787 (49.4)
Togo	400 (45.3)	484 (54.7)
Tanzania	877 (55.0)	718 (45.0)
Uganda	932 (67.1)	457 (32.9)
South Africa	312 (66.1)	160 (33.9)
Zambia	1963 (66.6)	986 (33.4)
Total	21980 (56.4)	16986 (43.6)

Figure 1 illustrates the prevalence of anemia among adolescent girls across various African countries. Countries with a prevalence higher than the overall average (43.6%) are marked in red, while those below the average are marked in green. The highest prevalence was observed in Central and West African countries, including Gabon, Mali, Ivory Coast, and Nigeria. In contrast, lower prevalence was found in East and East-Central African countries, as well as in Southern African countries, including Rwanda, Ethiopia, Lesotho, and South Africa (Fig. 1).

Table 3 presents a chi-square analysis of the distribution of anemia among African adolescent girls by various predictor variables. The prevalence of anemia is slightly higher among urban residents, with 44.4% being anemic compared to 43.2% of rural residents. Adolescents who had ever been married show a higher prevalence of anemia (48.7%) compared to their never married/single counterparts (42.3%). The prevalence of anemia is higher among girls who married before the age of 15 (50.7%) compared to those who married between 15–19 years (48.2%).

Educational attainment shows a strong inverse relationship with anemia prevalence. Adolescents with no education had the highest prevalence of anemia (54.2%), followed by those with primary education (40.2%), secondary education (43.2%), and higher education (37.3%). This indicates that higher education levels are protective against anemia. Wealthier adolescents had a lower prevalence of anemia. The prevalence was highest among poor girls (46.4%), followed by those from middle-income households (44.7%), and was lowest among rich girls (41.0%). Employment status does not significantly affect anemia prevalence, with both working (43.7%) and non-working (43.7%) girls showing the same prevalence rates (Table 3).

Pregnant adolescents had a significantly higher prevalence of anemia (54.9%) compared to non-pregnant girls (43.0%). Adolescents who used contraceptives have a lower prevalence of anemia (38.3%) compared to those who do not use contraceptives (44.3%). The number of children a girl has given birth to is associated with increased prevalence of anemia. Adolescent girls with no children have a prevalence of 42.8%, those with one child have a prevalence of 47.3%, and those with two or more children have the highest prevalence at 49.0%.

The prevalence of anemia was lower among girls with health insurance (39.1%) compared to those without it (44.3%). The prevalence of anemia was highest among adolescents using untreated mosquito nets (49.0%), followed by those using treated nets (45.1%), and those not using any net (42.8%). Higher BMI was associated with a lower prevalence of anemia. Underweight girls have an anemia prevalence of 45.2%, normal weight girls have 45.0%, overweight girls have 41.5%, and obese girls have the lowest prevalence at 40.9%.

Table 3

A chi-square analysis of magnitude of anemia by predictor variables among African adolescent girls
Factors			Not anemic frequency (%)	Anemic frequency (%)
Residence
	Urban		8581(55.6)	6852(44.4)
	Rural		13359(56.8)	10174(43.2)
Marital status
	Never married/Single		17687 (57.7)	12981(42.3)
	Ever Married		4253 (51.3)	4045 (48.7)
Age at first Marriage (n = 8,298)
	< 15		959 (49.3)	985 (50.7)
	15–19		3294 (51.8)	3060 (48.2)
Education status
		No education	2587(45.8)	3059(54.2)
		Primary	8252(59.8)	5556(40.2)
		Secondary	10891(56.8)	8286(43.2 )
		Higher	210(62.7)	125(37.3)
Wealth status
	Poor		7632 (53.6)	6601(46.4)
	Middle		4336 (55.3)	3498 (44.7)
	Rich		9972 (59.0)	6927 (41.0)
Currently pregnant
	No		20864 (57.0)	15714 (43.0)
	Yes		1076 (45.1 )	1312 (54.9)
Use of contraceptive
	No		19427 (55.7)	15467 (44.3)
	Yes		2513 (61.7)	1559 (38.3)
Health insurance
	No		19165 (55.7)	15240 (44.3)
	Yes		2775 (60.9)	1785 (39.1)
Currently working/Occupation
	No		14938 (56.3)	11591 (43.7)
	Yes		6991 (56.3)	5428 (43.7)
Parity
	No		18199 (57.2)	13637 (42.8)
	1		3180 (57.2)	2851 (47.3)
	2+		561 (51.0)	538 (49.0)
Body Mass Index (BMI)
	Underweight		3499 (54.8)	2889 (45.2)
	Normal weight		14190 (55.0)	11592 (45.0)
	Overweight		1780 (58.5))	1263 (41.5)
	Obese		414 (59.1)	287 (40.9)
Type of mosquito net
	No net		14418 (57.2)	10804 (42.8)
	Treated net		7172 (54.9)	5886 (45.1)
	Not treated net		350 (51.0)	336 (49.0)

Note: all predictor variables except employment status were statistically significant with anemia.

Table 4 presents the binary logistic regression analysis of various factors associated with anemia among adolescent girls in Africa. The results are shown in terms of Crude Odds Ratios (COR) and Adjusted Odds Ratios (AOR), along with their 95% Confidence Intervals (CI).

Adolescent girls living in rural areas had a slightly lower odds of anemia compared to those in urban areas (COR: 0.950, 95% CI: 0.91–0.99). However, this association is not significant after adjusting for other factors (AOR: 0.88, 95% CI: 0.78–1.01). Adolescent girls who had ever married showed higher odds of anemia compared to their never-married/single counterparts (COR: 1.298, 95% CI: 1.34–1.36). However, this variable was not included in the multivariable analysis due to convergence issues. Adolescents who married at ages 15–19 show slightly lower odds of anemia compared to those married before age 15 (COR: 0.90, 95% CI: 0.81-1.00), but this difference is not significant after adjustment of other factors (AOR: 0.99, 95% CI: 0.88–1.11).

Education appears protective against anemia. Compared to adolescents with no education, those with primary (AOR: 0.72, 95% CI: 0.5–0.80), secondary (AOR: 0.76, 95% CI: 0.66–0.86), and higher education (AOR: 0.61, 95% CI: 0.21–1.75) had lower odds of anemia. Adolescents from rich households have significantly lower odds of anemia compared to those from poor households (AOR: 0.86, 95% CI: 0.76–0.98).

Pregnancy was associated with higher odds of anemia (AOR: 1.13, 95% CI: 1.01–1.27) compared to those who were not pregnant. Using contraceptives linked to significantly lower odds of anemia (AOR: 0.46, 95% CI: 0.40–0.53). Adolescents who had given birth to one child have higher odds of anemia compared to those with no children (COR: 1.20, 95% CI: 1.13–1.26). Similarly, adolescents with two or more children also have higher odds of anemia (COR: 1.28, 95% CI: 1.13–1.44). However, when adjusted for other factors, these associations are no longer statistically significant.

Having health insurance shows no significant association with anemia (AOR: 1.03, 95% CI: 0.81–1.30). Regarding to nutritional status, adolescents who were obese had significantly lower odds of anemia compared to those who were underweight (AOR: 0.57, 95% CI: 0.38–0.86). Using a treated mosquito net was associated with higher odds of anemia (AOR: 1.26, 95% CI: 1.15–1.39).

The multivariable analysis revealed that 1.14% of the variation in anemia among adolescent girls was due to differences at the cluster/community level (ICC = 0.014, p < 0.05). This suggests that about 1.14% of anemia cases may be attributable to other unobserved community-level determinants.

Table 4

Bivariate and multivariable analysis to determine anemia among adolescent girls in Africa.
Factors			COR (95% CI)	AOR (95% CI)
Residence
	Urban		1
	Rural		0.950 (0.91–0.99)**	0.88(0.78–1.01)
Marital status
	Never married/Single		1
	Ever Married		1.298(1.34–1.36)**
Age at first Marriage (n = 8,298)
	< 15		1	1
	15–19		0.90(0.81–0.99)*	0.99(0.88–1.11)
Education status
		No education	1	1
		Primary	0.57(0.53–0.60)**	0.72(0.5–0.80)*
		Secondary	0.64(0.60–0.68)**	0.76(0.66–0.86)*
		Higher	0.50(0.40–0.64)**	0.61(0.21–1.75)
Wealth status
	Poor		1	1
	Middle		0.93(0.88–0.99)**	0.95(0.85–1.08)
	Rich		0.80(0.77–0.84)**	0.86(0.76–0.98)
Currently pregnant
	No		1	1
	Yes		1.617(1.49–1.76)**	1.13(1.00-1.27)*
Use of contraceptive
	No		1	1
	Yes		0.78(0.72–0.83)**	0.46(0.40–0.53)*
Health insurance
	No		1	1
	Yes		0.81(0.76–0.86)**	1.03(0.81–1.30)
Currently working/Occupation
	No		1	1
	Yes		0.99(0.96–1.04)
Parity
	No		1	1
	1		1.20(1.13–1.26)**	1.01(0.91–1.12)
	2+		1.28(1.13–1.44)**	1.05(0.89–1.21)
Body mass index
	Underweight		1	1
	Normal weight		0.99(0.93–1.04)	1.04(0.91–1.21)
	Overweight		0.85(0.78–0.94)**	0.91(0.75–1.12)
	Obese		0.84(0.71–0.98)**	0.57(0.38–0.86)*
Type of mosquito net
	No net		1	1
	Treated net		1.09(1.05–1.14)**	1.26(1.15–1.39)*
	Not treated net		1.28(1.10–1.50)**	1.31(0.9–1.76)
	ICC		-	0.011

Note: **: P < 0.05. *: P < 0.2, boldface with *: P < 0.05

This study aimed to investigate the magnitude and determinants of anemia among adolescent girls in Africa. The analysis utilized both bivariate and multivariable logistic regression to identify relationship between various socio-demographic, economic and health-related factors with the likelihood of anemia. The study investigated that the overall prevalence of anemia among adolescent girls in Africa was 43.6% (95% CI: 42.97, 44.21_). According to the World Health Organization (WHO), anemia is categorized based on prevalence cut-off values that indicate its public health significance. If the prevalence of anemia is below 5%, it is considered not to be a public health problem. A prevalence of 5–19% is classified as a mild public health problem, while a prevalence of 20–39% is considered a moderate public health problem. When the prevalence reaches 40% or higher, anemia is considered a severe public health problem [26]. Given that the prevalence of anemia among adolescent girls in Africa exceeds this threshold at 43.6%, it is classified as a severe public health problem in the region.

The pooled magnitude of anemia in this study was higher than the studies conducted in Indonesia

(14.3%) [27], India (20%)[12], Nepal (31%)[28], Ethiopia (23%) [8], Kenya (28.9%)[29], but lower than in Pakistan (47.9%) [30], Tanzania (53.3%) [9] and Ghana (50.3%) [31]. Even within the same study, the magnitude of anemia among adolescent girls varied significantly across different African countries. The prevalence ranged from 14.7% in Rwanda to 72% in Gabon. This wide range in anemia prevalence reflects differences in factors such as religion, cultural and dietary practices, access to iron-rich foods, socioeconomic conditions, health system and the effectiveness of public health interventions in these countries [32–34].

Furthermore, out of the 24 African countries included in this study, 16 of them (66.7%) had a prevalence of anemia above 40% among adolescent girls is alarming. This indicates that in the majority of African countries, the magnitude of anemia among this population is a substantial public health problem that requires urgent attention. Addressing this public health challenge will require a comprehensive, multi-faceted approach tailored to the specific needs and contexts of each country. Interventions may include improving dietary diversity and nutrient intake, implementing supplementation programs, provide health education and enhancing access to healthcare services to adolescent girls [35, 36].

Our multivariable analysis found that adolescents’ educational level, wealth status, being pregnant, contraceptive use, nutritional status and use of mosquito net were statistically significant with anemia. Education emerged as a crucial protective factor against anemia. Adolescent girls with primary and secondary education levels showed significantly lower odds of anemia compared to those with no education which was supported by other studies done in sub-Sahara Africa [13], in west and central Africa [37] and Ethiopia [15]. This finding indicates that education is crucial in enhancing health literacy, nutritional awareness, and overall health practices, which can considerably lower the risk of anemia. Education likely equips adolescents with the knowledge and resources needed to make health decisions, leading to improved nutritional choices and proactive health-seeking behaviors [38, 39]. Unsurprisingly and similar with other study findings [40, 41], our study found that, adolescents from rich households had also lower odds of anemia compared to those from poor households. This suggests that economic resources improve access to nutrition and healthcare services, both of which are essential for preventing anemia. Families with higher incomes can afford more diverse, nutritious diets, healthcare services, and supplements, all of which help to prevent anemia.

Expectedly, and align with previous studies [42, 43], this study found that, Pregnancy was significantly associated with higher odds of anemia. This association underscores the heightened nutritional demands placed on pregnant adolescents, without effective intervention, can increase the risk of anemia [44]. The physiological changes during pregnancy, such as a surge in blood volume and a substantial increase in iron requirements, make adolescent girls particularly susceptible to anemia. Given their growing bodies and developmental needs, the nutritional demands are even more acute. Anemia during adolescent pregnancy can lead to several adverse outcomes, including increased risks of preterm delivery, low birth weight, preeclampsia, unsafe abortion, mental-health related effects and maternal and child morbidity and mortality [45]. Therefore, it is crucial to implement targeted interventions to prevent adolescent pregnancy and, if it occurs, to provide adequate nutritional support and prenatal care for pregnant adolescents.

This study revealed that the use of contraceptives was significantly associated with lower odds of anemia which was supported by other studies [46, 47]. This could be justified by the fact that adolescents who use contraceptives can prevent complications related to early pregnancy and childbirth, which may reduce the prevalence of anemia caused by recurrent blood loss. Another plausible justification is that the use of contraceptives can minimize menstrual bleeding and decrease their susceptibility to anemia [48].

Adolescent girls who were obese showed significantly lower odds of anemia compared to those who were underweight. Our result was consistent with the study conducted in Pakistan [49], in Bangladesh [50] and in Ethiopia [15]. This suggests that excessive body weight, potentially linked to better micronutrient status, may confer some protection against anemia. On the other hand, underweighted adolescents have more chances to be deficient in micronutrients which may result in an increased risk of anemia. However, it's important to note that while obesity might reduce the risk of anemia, it is not an indicator of overall health and poses other health risks. Nevertheless, studies done in Korea [51] and in Morocco[52] reported a higher probability of anemia among obese adolescents. The possible explanation might be obesity is often associated with chronic inflammation, which can affect iron metabolism. Inflammatory cytokines can inhibit iron absorption and utilization, leading to anemia, despite adequate or elevated iron stores. Obese individuals may have diets that are high in calories but low in essential nutrients, including iron [53]. Furthermore, obese individuals have a higher blood volume, which may dilute the concentration of hemoglobin and ferritin, potentially masking the presence of anemia when only hemoglobin levels are measured.

The relationship between obesity and anemia among adolescents is multifaceted. It underscores the importance of considering various factors, including micronutrient status, dietary quality, and inflammation, when evaluating the risk of anemia. Further research is needed to clarify these associations and inform public health strategies to improve the nutritional status of adolescent girls.

Contrary to other studies suggesting a link between the use of treated mosquito nets and reduced anemia in children under five years old [54, 55], this findings show that the use of treated mosquito nets among adolescents (15–19 years old) was associated with higher odds of anemia. This unexpected result indicates that while mosquito nets are effective in preventing malaria, their usage alone may not directly reduce the risk of anemia among adolescents. The observed association could be influenced by other unmeasured factors, such as net usage patterns among adolescents. For instance, adolescent girls experience menstruation, which leads to blood loss and makes them more susceptible to iron-deficiency anemia. Additionally, adolescent girls have increased nutritional needs to support their growth and reproductive health. Furthermore, adolescents might not use mosquito nets as consistently or correctly as needed, possibly due to changes in lifestyle, increased independence, or a lack of perceived risk, thereby reducing the nets' effectiveness in preventing malaria and, consequently, anemia. This study suggests developing targeted education programs to ensure adolescents use mosquito nets correctly. This includes training on proper hanging techniques, consistent usage, and care and maintenance of the nets to maximize their effectiveness.

This study has some limitations to consider. First, due to the limited variables available in this secondary data, we did not include all predictors associated with adolescent anemia, such as dietary patterns and food security. Second, the use of data from a cross-sectional study limits our ability to infer causality between the predictors and the outcome variable (anemia). Despite these limitations, our study included important proxy variables for food security, such as wealth status, as well as other sociodemographic and economic factors. Using a continentally representative large sample, the results from this analysis may be generalized to African adolescent girls aged 15–19 years.

The study has shown that anemia among adolescent girls in Africa poses a significant public health concern, with a prevalence of 43.6%. Factors such as higher levels of education and wealth, as well as the use of contraceptives and obesity, were found to decrease the likelihood of being anemic. Conversely, being pregnant and using treated mosquito nets were associated with increased odds of anemia among adolescent girls. It is recommended that efforts be made to promote school attendance for girls, encourage contraceptive use among married or sexually active adolescents, prevent adolescent marriage and pregnancy, educate on the proper use of mosquito nets, and provide nutritional support in order to effectively combat adolescent anemia.

Competing Interests

The authors have declared that no competing interests exist.

Human Ethics and Consent to Participate declarations
not applicable.

Author Contribution

Conceptualization: Fentanesh Nibret TirunehData curation: Bedilu Alamirie EjiguFormal analysis: Bedilu Alamirie Ejigu Methodology: Bedilu Alamirie Ejigu, Fentanesh Nibret TirunehValidation: Bedilu Alamirie Ejigu, Fentanesh Nibret TirunehWriting – original draft: Fentanesh Nibret Tiruneh Writing – review & editing: Fentanesh Nibret Tiruneh, Bedilu Alamirie EjiguBoth authors read and approved the final manuscript.

Data Availability

The data we used in this study were obtained from the DHS program (www.dhsprogram.com), but the `Dataset Terms of Use' do not permit us to distribute this data as per data access instructions (http://dhsprogram.com/data/Access-Instructions.cfm). Interested scholars will have access to the relevant data used in this study in the same manner as it was accessed by the authors of this study from the aforementioned website.

Funding Declaration

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

WHO. Anaemia. 2024.
Lancet T. The Lancet: New study reveals global anemia cases remain persistently high among women and children. Anemia rates decline for men. 2023.
Collaborators GA. Prevalence, years lived with disability, and trends in anaemia burden by severity and cause, 1990–2021: findings from the Global Burden of Disease Study 2021. Lancet Haematol. 2023;10(9):e713–34.
Bahrami A, et al. Anemia is associated with cognitive impairment in adolescent girls: A cross-sectional survey. Appl Neuropsychology: Child. 2020;9(2):165–71.
Soman SK, et al. Adolescent anaemia its prevalence and determinants: a cross-sectional study from south Kerala, India. Int J Community Med public health. 2017;4(8):2750–6.
Sari P, et al. Iron Deficiency Anemia and Associated Factors Among Adolescent Girls and Women in a Rural Area of Jatinangor, Indonesia. International Journal of Women's Health; 2022. pp. 1137–47.
Tura MR et al. Prevalence of anemia and its associated factors among female adolescents in Ambo Town, West Shewa, Ethiopia. J Blood Med, 2020: pp. 279–87.
Habtegiorgis SD, et al. Prevalence and associated factors of anemia among adolescent girls in Ethiopia: A systematic review and meta-analysis. PLoS ONE. 2022;17(3):e0264063.
Yusufu I, et al. Factors associated with anemia among school-going adolescents aged 10–17 years in Zanzibar, Tanzania: a cross sectional study. BMC Public Health. 2023;23(1):1814.
Nti J, et al. Variations and Determinants of Anemia among Reproductive Age Women in Five Sub-Saharan Africa Countries. Biomed Res Int. 2021;2021(1):9957160.
Bellizzi S, et al. Iron deficiency anaemia and low BMI among adolescent girls in India: The transition from 2005 to 2015. Public Health Nutr. 2021;24(7):1577–82.
Chauhan S, et al. Prevalence and predictors of anaemia among adolescents in Bihar and Uttar Pradesh, India. Sci Rep. 2022;12(1):8197.
Worku MG, et al. Multilevel analysis of determinants of anemia among young women (15–24) in sub-Sahara Africa. PLoS ONE. 2022;17(5):e0268129.
Merid MW, et al. An unacceptably high burden of anaemia and it’s predictors among young women (15–24 years) in low and middle income countries; set back to SDG progress. BMC Public Health. 2023;23(1):1292.
Tiruneh FN, et al. Associations of early marriage and early childbearing with anemia among adolescent girls in Ethiopia: a multilevel analysis of nationwide survey. Archives Public Health. 2021;79(1):91.
Saleh MA. Outcomes of Teenage Pregnancy at Benghazi Medical Center 2019–2020. Int J Sci Acad Res. 2022;3(3):358–3602.
WHO. Anaemia in women and children.
Organization WH. Global nutrition targets 2025: Stunting policy brief. World Health Organization; 2014.
Cf O. Transforming our world: the 2030 Agenda for Sustainable Development. New York, NY, USA: United Nations; 2015.
CSA.. Ethiopia Demographic and Health Survey 2016. 2016.
Organization WH. Haemoglobin concentrations for the diagnosis of anaemia and assessment of severity. World Health Organization; 2011.
Carle AC. Fitting multilevel models in complex survey data with design weights: Recommendations. BMC Med Res Methodol. 2009;9:1–13.
!!!. INVALID CITATION !!!.
Leyland AH, Groenewegen PP. Multilevel modelling for public health and health services research: health in context. Springer Nature; 2020.
StataCorp L. Stata Statistical Software: Release 18. Stata Surv data Ref Man release, 2023. 14.
WHO. Nutrition and nutrition-related health and development data (Anemia). 2008.
Sari P, et al. Anemia among adolescent girls in west java, Indonesia: related factors and consequences on the quality of life. Nutrients. 2022;14(18):3777.
Chalise B, et al. Prevalence and correlates of anemia among adolescents in Nepal: Findings from a nationally representative cross-sectional survey. PLoS ONE. 2018;13(12):e0208878.
Ekasanti I, et al. Determinants of anemia among early adolescent girls in Kendari City. Amerta Nutr. 2020;4(4):271.
Habib N, Abbasi S-U-RS, Aziz W. An analysis of societal determinant of anemia among adolescent girls in Azad Jammu and Kashmir, Pakistan. Anemia. 2020;2020(1):1628357.
Tandoh MA, Appiah AO, Edusei AK. Prevalence of anemia and undernutrition of adolescent females in selected schools in Ghana. J Nutr Metabolism. 2021;2021(1):6684839.
Alonso EB. The impact of culture, religion and traditional knowledge on food and nutrition security in developing countries. 2015.
Chouraqui J-P, et al. Religious dietary rules and their potential nutritional and health consequences. Int J Epidemiol. 2021;50(1):12–26.
Mildon A, et al. Integrating and coordinating programs for the management of anemia across the life course. Volume 1525. Annals of the New York Academy of Sciences; 2023. pp. 160–72. 1.
Rohmatika D et al. Education and Reminder Software for Strength-ening Anemia Prevention Program in Adolescent Girls. in ISPHE 2020: Proceedings of the 5th International Seminar of Public Health and Education. 2020.
Aisah S, Ismail S, Margawati A. Animated educational video using health belief model on the knowledge of anemia prevention among female adolescents: An intervention study. Malaysian Family Physician: Official J Acad Family Physicians Malaysia. 2022;17(3):97.
Sagalova V et al. Socio-economic predictors of undernutrition and anaemia in adolescent mothers in West and Central Africa. J global health, 2021. 11.
Krause BL. Networks and knowledge: Women's empowerment, networks, and health information-seeking behavior in rural Guatemala. J Rural Stud. 2024;105:103169.
Riddle AY, et al. Associations between dimensions of empowerment and nutritional status among married adolescent girls in East Africa: a structural equation modelling study. BMC Public Health. 2023;23(1):225.
Zhu Z, et al. Anemia and associated factors among adolescent girls and boys at 10–14 years in rural western China. BMC Public Health. 2021;21:1–14.
Sales CH, et al. Prevalence and factors associated with iron deficiency and anemia among residents of urban areas of São Paulo, Brazil. Nutrients. 2021;13(6):1888.
Gueye M, et al. Neonatal complications of teenage pregnancies: prospective study about 209 Cases in Senegal. Am Jf Pediatr. 2020;6(4):504–7.
Pons-Duran C, et al. Adolescent, pregnant, and HIV-infected: risk of adverse pregnancy and perinatal outcomes in young women from Southern Mozambique. J Clin Med. 2021;10(8):1564.
Adam I, Ali AA. Anemia Dur pregnancy Nutritional Defic. 2016;978:953–51.
Lambonmung A, Acheampong CA, Langkulsen U. The effects of pregnancy: a systematic review of adolescent pregnancy in Ghana, Liberia, and Nigeria. Int J Environ Res Public Health. 2022;20(1):605.
Misunas C, et al. The Association Between Hormonal Contraceptive Use and Anemia Among Adolescent Girls and Young Women: An Analysis of Data From 51 Low-and Middle-Income Countries. J Adolesc Health. 2024;74(3):563–72.
Worku MG, Tesema GA, Teshale AB. Prevalence and determinants of anemia among young (15–24 years) women in Ethiopia: A multilevel analysis of the 2016 Ethiopian demographic and health survey data. PLoS ONE. 2020;15(10):e0241342.
Haile ZT, Teweldeberhan AK, Chertok IR. Association between oral contraceptive use and markers of iron deficiency in a cross-sectional study of Tanzanian women. Int J Gynecol Obstet. 2016;132(1):50–4.
Khan UH, et al. Prevalence of Nutritional Anaemia with Association of Body Mass Index among Karachi University students, Pakistan. Volume 71. JOURNAL OF THE PAKISTAN MEDICAL ASSOCIATION; 2021. pp. 55–8. 1.
Kamruzzaman M. Is BMI associated with anemia and hemoglobin level of women and children in Bangladesh: A study with multiple statistical approaches. PLoS ONE. 2021;16(10):e0259116.
Jeong J, et al. Association between obesity and anemia in a nationally representative sample of South Korean adolescents: a cross-sectional study. Healthcare. MDPI; 2022.
Mehdad S et al. Association between overweight and anemia in Moroccan adolescents: a cross-sectional study. Pan Afr Med J, 2022. 41(1).
Harding KL, Aguayo VM, Webb P. Hidden hunger in South Asia: a review of recent trends and persistent challenges. Public Health Nutr. 2018;21(4):785–95.
Mwaiswelo RO, et al. Malaria infection and anemia status in under-five children from Southern Tanzania where seasonal malaria chemoprevention is being implemented. PLoS ONE. 2021;16(12):e0260785.
Adugna DG, et al. Prevalence and determinants of anemia among children aged from 6 to 59 months in Liberia: a multilevel analysis of the 2019/20 Liberia demographic and health survey data. Front Pead. 2023;11:1152083.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
10 Sep, 2024
Editor assigned by journal
04 Sep, 2024
Submission checks completed at journal
04 Sep, 2024
First submitted to journal
30 Aug, 2024

You are reading this latest preprint version

Magnitude and Determinates of Anemia among adolescent Girls in Africa: A Multilevel, Multicounty Analysis of 24 Countries

Status:

Version 1

Abstract

Figures

Introduction

Method

Data source and study participants

Measurements

Outcome variable

Predictor variables

Statistical Analysis

Ethical consideration

Results

Discussion

Conclusion

Declarations

Competing Interests

Human Ethics and Consent to Participate declarations
not applicable.

Author Contribution

Data Availability

Funding Declaration

References

Additional Declarations

Status:

Version 1

Magnitude and Determinates of Anemia among adolescent Girls in Africa: A Multilevel, Multicounty Analysis of 24 Countries

Status:

Version 1

Abstract

Figures

Introduction

Method

Data source and study participants

Measurements

Outcome variable

Predictor variables

Statistical Analysis

Ethical consideration

Results

Discussion

Conclusion

Declarations

Competing Interests

Human Ethics and Consent to Participate declarations not applicable.

Author Contribution

Data Availability

Funding Declaration

References

Additional Declarations

Status:

Version 1

Human Ethics and Consent to Participate declarations
not applicable.