Prevalence and Determinants of Anaemia among Reproductive-aged Women in Ethiopia: A Nationally Representative Cross-sectional Study

Anaemia in reproductive-aged women is a worldwide health problem. This study was aimed to assess prevalence and determinants of anaemia among reproductive-aged women in Ethiopia. Data for the study were obtained from 2016 Ethiopian demographic and health survey data, which is a national representative cross-sectional data. A Multivariable logistic regression model was applied to identify determinants of anaemia among reproductive-aged women. A total of 14460 women who aged 15 to 49 years were included in the study. Prevalence of anaemia of among reproductive-aged women was 27.08% (95% CI: 22.88, 31.08%). Women living in Afar (AOR=2.439; 95% CI: 2.006, 2.968), Amhara (AOR=1.269; 95% CI: 1.035, 1.556), Somalia (AOR=2.592; 95% CI: 2.142, 3.133), Benshangul-Gumuz (AOR=2.019; 95% CI: 1.666, 2.447), Original Research Article Mekebo et al.; JPRI, 33(59B): 687-698, 2021; Article no.JPRI.78218 688 Gambela (AOR=2.465; 95% CI: 2.026, 2.998) were associated with high risk of anaemia. Women with 1 or 2 children (AOR=1.272; 95% CI: 1.103, 1.466), 3 or 4 children (AOR=1.277; 95% CI: 1.059, 1.539) and 5 or more (AOR=1.420; 95% CI: 1.213, 1.662) were associated with high risk of anaemia. Further, pregnant women (AOR=1.408; 95% CI: 1.263, 1.570) were associated with high risk of anaemia. Hence, concerned bodies need to pay special attention to women regarding anaemia based on place of residence and the region.


INTRODUCTION
Anaemia is defined as a condition in which concentration of red blood cells/hemoglobin is lower than the normal resulting in reduced oxygen-carrying capacity to meet physiologic needs of the body [1]. A non-pregnant and pregnant women are considered as anaemic if hemoglobin levels are lower than 120 gram/liter and lower than 110 gram/liter respectively [2]. Anaemia is one of the global widespread public health and nutritional problems affecting both developing and developed countries and occurs at all stages of life cycle prominently in young, pregnant women and other women in child bearing age [3]. It has significant adverse health consequences such as child mortality, maternal mortality, increased risk of adverse pregnancy outcomes, impaired neuro-cognitive and physical development of children and reduced work capacity, and adverse impacts on socioeconomic development [4,5]. Some of the symptoms that are resulted from impaired tissue oxygen delivery include weakness, fatigue, and difficulty in concentration [6].
Globally, it is thought that most commonly anaemia is caused by deficiency of iron. In addition, deficiencies in nutrition like folate, vitamin B12 and vitamin A, parasitic infections, acute and chronic inflammation and inherited or acquired disorder which affects synthesis of hemoglobin, production of red blood cell or survival of red blood cell can also cause anaemia [2].
The prevalence of anaemia in developed countries is estimated to be 9% and in that of developing countries 43% [7]. It is estimated that 42% of the pregnant women and 30% in nonpregnant women who are 15-49 years old are anaemic globally. It is also estimated that anaemia contributes to more than 115000 maternal deaths and also 591000 prenatal deaths globally annually [8]. Anaemia in reproductive-aged women is a worldwide health problem. The prevalence of anaemia is highest in Low Income countries predominantly in Africa. In Africa 57.1% of the pregnant and 47.5% of nonpregnant women are anaemic [3].
Despite Ethiopian ministry of health and its stakeholders are doing their best to decrease prevalence of anaemia in the country, the recent demographic and health survey report, 2016 EDHS showed an increase in the prevalence of anaemia among women aged 15-49 years as compared to 2011 EDHS report. The prevalence of anaemia among women who aged 15-49 years declined from 27% in 2005 to 17% in 2011 [9] but then increased to 24% in 2016 in Ethiopia. Moreover, the prevalence of anaemia is more among women than that of men in Ethiopia. According to 2016 EDHS report the prevalence of anaemia among women was 24% compared to 15% for men in 2016 in Ethiopia [10].
Anaemia in pregnant women causes increased risk of premature delivery and low birth weight [9]. So, it is essential to identify factors associated with anaemia among women in reproductive age to inform the planners of strategies to deal with the identified factors to reduce the chances of adverse maternal and fetal outcomes associated with anaemia, and to make the women healthy and give healthy births so that the they can freely participate and contribute their contributions in socio-economic activities for the development of the country. This study, therefore, aimed at assess the prevalence and determinants of anaemia among reproductive-aged women in Ethiopia.

Source of Data
This study was based on a nationally representative cross-sectional survey, 2016 EDHS (2016 Ethiopian demographic and health survey) which was implemented by the Central Statistical Agency (CSA) from January 18 to June 27, 2016 in Ethiopia.

Study Area
This study was conducted in Ethiopia. Ethiopia is one of the countries in Africa continent and located in the Horn of Africa. It is bordered by six African countries: to the north and northeast by Eritrea, to the east by Djibouti and Somalia, to the west by Sudan and by South Sudan, and to the south by Kenya. And, it is the second most populous nation in African continent.

Sample Size
In this study, we used a sample of 14460 women aged 15 to 49 years.

Dependent variable
Dependent variable was anaemic status of women at a time survey.

Independent variables
Independent variables included in this study were region, place of residence, marital status, level of education, wealth index, smoking cigarette, age, body mass index, parity, pregnancy status, and contraceptive methods.

Data Analysis
Data analysis was done using SPSS version 25. Multivariable logistic regression model was employed to identify the determinants of anaemia among women.

Prevalence of anaemia among women
A total of 14460 women of which 3916 (27.08%) anaemic were included in this study ( Table 1).

Background characteristics of women
Of total of 14460 women included, about twothird (67.52%) of the women were living in rural while remaining 32.48% of them were living in urban at a time of the survey. Regarding geographic region, 10.89% of them were living in Tigray region while remaining 7.05%, 11.65%, 12.46%, 8.70%, 7.07%, 12.15%, 6.80%, 5.17%, 11.13% and 6.58% of them were living respectively in Affar region, Amhara region, Oromia region, Somalia region, Somalia region, Benshangul-Gumuz region, SNNPR, Gambela region, Harari region, Addis Ababa, and Dire Dawa at time of survey ( Table 2).
Regarding age, more than one-fifth (21.89%) of the women were in the age group of 15-19 years, about one-sixth (18.16%) of the them were in the age group of 20-24 years, about one-sixth (18.06%) of them were in the age group of 25-29 years, about one-seventh (14.29%) of them were in the age group of 30-34 years, about one-eighth (13.07%) of them were in the age group of 35-39 years, 8.25% of them were in the age group of 40-44 years and remaining 6.28% of them were in age group of 45-49 years at a time of the survey. Regarding marital status, majority (63.64%) of them were married or living together, about one-fourth (26.29%) of them were single, 2.88% of them were widowed while the remaining 7.19% of them were divorced/no longer living together/separated at a time of the survey. Regarding parity, majority (33.29%) of the women had no child, about one-fourth(25.66) of them had 5 or more children, 23.62% of the them had 1 or 2 children while the remaining 17.43% of them had 3 or 4 children at a time of the survey ( Table 2).

Bivariate analysis result
The prevalence of anaemia among women was highest in Somalia region (58.35%) followed by Affar region (45.65%) and it was lowest in Addis Ababa city administrative (15.79%) followed by Amhara region (17.58%) ( Table 2).
Regarding place of residence, prevalence of anaemia among women in rural area (30.61%) was higher than in urban (19.76%). The prevalence of anaemia among women was highest for those who aged 30-34 years (29.93%) followed by age group 35-39 years (28.83%) and it was lowest in those women who aged 15-19 years (23.73%) followed by age group 45-49 years (24.35%) (Table2).
The prevalence of anaemia among women decreased with increased level of education. It was highest among those women who were not educated (33.94%) and it was lowest among those women whose level of education was higher than the secondary education (15.19%). Likewise, it decreased with increased body mass index. It was highest among those women whose body mass index was less than 18.5 (32.11%) and lowest for those women whose body mass index was 25.0 and above (21.35%) ( Table 2).
The prevalence of anaemia among women was highest for those women whose total number of children ever born were 5 or more (34.36%) followed by those women whose total number of children ever born were 3 or 4 (29.29%) and lowest for those women who had no child (12.75%). Similarly, it was higher for those women who were pregnant (37.58%) than those who were not pregnant (26.26%) ( Table 2).
The prevalence of anaemia among women was also higher for women who were not using contraceptive methods (29.36%) than those who were using contraceptive methods (18.79%). Similarly, it was higher for those women who were smoking cigarette (29.27%) than those who were not smoking (27.06%) ( Table 2).
All independent variables except smoking cigarette were significant in bivariate analysis. Those significant independent variables (with pvalue < 0.25) in bivariate analysis were included in the multivariable logistic regression analysis.

Inferential Statistics Results
The multivariable logistic regression analysis result revealed that region, place of residence, age, wealth index, body mass index, parity, and pregnancy status were significantly associated with anaemia among women ( Table 3).  (Table 3).
Furthermore, the odds of being anaemic for the richer women was 0.277 (AOR: 0.277, 95% CI for AOR: 0.172, 0.447) times less than for the poorest women. The odds of being anaemic for the richest women was 0.158 (AOR: 0.158, 95% CI for AOR: 0.073, 0.343) times less than for the poorest women. Likewise, the odds of being anaemic for the women whose body mass index is between 18.5 and 24.5 was 0.272 (AOR: 0.272, 95% CI for AOR: 0.144, 0.512) times less than for those whose body mass index is less than 18.5. Similarly, the odds of being anaemic for the women whose body mass index is 25.0 and above was 0.420 (AOR: 0.420, 95% CI for AOR: 0.294, 0.600) times less than for those whose body mass index was lower than 18.5 (Table 3).
Also, pregnancy status was found to be predictor factor of anaemia among women. The odds of being anaemic for the pregnant women is 1.408 (AOR: 1.408, 95% CI for AOR: 1.263, 1.570) times higher than for non-pregnant women. The odds of being anaemic for women who ever bore 1 or 2 children was 1.272 (AOR: 1.272, 95% CI for AOR: 1.103, 1.466) times higher than for those who had no child. The odds of being anaemic for women who ever bore 3 or 4 children was 1.277 (AOR: 1.277, 95% CI for AOR: 1.059, 1.539) times higher than for those who had no child. Similarly, the odds of being anaemic for women who ever bore 5 or more children was 1.420 (AOR: 1.420, 95% CI for AOR: 1.213, 1.662) times higher than for those who had no child (Table 3).

Goodness of fit of the Model
The model goodness of fit was checked using Hosmer and Lemeshow test. The Hosmer and Lemeshow test result showed p-value = 0.120, which implies good fit for the model.

DISCUSSION
This study was aimed to assess prevalence and determinants of anaemia among women of reproductive in Ethiopia. A total of 14,460 women of which 3,916 (27.08%) were anaemic were included in this study.
In this study, the prevalence of anaemia among reproductive-aged women in Ethiopia was 27.08% (95% CI: 22.88, 31.08%) which is almost similar with earlier study conducted in Turkey [12]. Prevalence of anaemia in our study was higher than the studies conducted in Iran [13], Ethiopia [14,15], Vietnam [16], while it was lower than the other studies conducted in Lao PDR [17], India [18], Nepal [19], Cambodia [20], and Bangladesh [21]. The reason for the variation in prevalence of anaemia in this study from those mentioned studies might be because of the differences in socioeconomic status, geographical location of the study area above sea level, and study period.
Furthermore, region, place of residence, age, wealth index, body mass index, parity and pregnancy status were significant determinants of anaemia among women.
Place of residence was found to be significant determinant of anaemia among women of reproductive age in Ethiopia. It was revealed that women who live in rural areas were more likely to be anaemic than those who live in urban areas. This result agrees with findings of the study done previously in Ethiopia [22], and Lao PDR [17]. The possible reason could be that those women who live in rural area may not have adequate health services and access for information on factors that influence anaemia due to lack of facilities and services like education. The study also showed that region had association with anaemia among women of Ethiopia. In support to our study, earlier studies in Uganda [23], Myanmar [24], Rwanda [25], Pakistan [26], and Lao PDR [17] also showed that geographic location had significant association with anaemia among women of reproductive age. Women from poorest families are more likely to be anaemic as compared to those from richest families. This result is consistent with the result of the previously conducted studies in in Uganda [23], Ethiopia [27], Meghalaya [28], Rwanda [25], and Sudan [29] which revealed that women from poorest families were more likely to be anaemic than those from the richest families. The reason might be that the poorest households cannot afford good diet, and may not have good sanitation.
In this study, it was also found that age was significantly associated with anaemia among women in Ethiopia. Previously conducted study in Uganda [23] showed that age had significant associated with anaemia among women of reproductive age. Our study revealed that women aged between 20 and 39 years are more likely to be anaemic than those who were in other age groups, which is almost similar to the results obtained from studies conducted in Uganda [23], and Ethiopia [30]. The possible explanation might be that woman could have more of her lifetime births by this age group.
In line with previously conducted studies in Ethiopia [31][32][33], India [34], and Pakistan [26] our study also revealed that women with higher body mass index were less likely to be anaemic than those with lower body index. Therefore, it is recommended to give particular attention to include micronutrients initiatives as a prioritized program for those who with lower body mass. Pregnant women were more likely to be anaemic than non-pregnant ones. This result agrees with the result obtained from the study done in Uganda [23], Mynmar [24], and Ethiopia [30], Tanzania [35] which revealed that pregnant women were at higher risk of anaemia compared to non-pregnant women. This could be explained by the fact that in pregnancy period nutritional demand of woman is highest in a woman's life and pregnant women are advised to eat more diversified diets than usual and they might not get diversified enough diets.
In this study, it was also found that women with higher number of ever born children were more likely to be anaemic than those with lower number of ever born children. This result is supported by the results revealed from the earlier studies [36][37][38] which revealed that women with parity of two or more were at higher risk of anaemia compared with those with lower parity. This might be due to the fact that in pregnancy there is blood volume expansion that increases iron demand and for this more blood is produced to support the growth of the baby. When the woman's dietary needs are not met during the pregnancy, she would be at risk of anaemia, and the more a woman gets pregnant, the higher risk she will be anaemic.

CONCLUSIONS
The result of this study demonstrated that about one-fourth of women had anaemia. In the study, region, place of residence, age, wealth index, body mass index, parity, and pregnancy status were identified as significant determinants of anaemia among reproductive-aged women. Women living in Gambela, Somali, Affar, and Benshangul-Gumuz region were associated with higher risk of being anaemic. Similarly, women aged 20-39 years, women with lager number of ever born children, and pregnant women were associated with higher risk of being anaemic. On the other hand, women living in urban areas, women with higher economic status, and women with higher body mass index were associated with reduced risk of being anaemic. Hence, concerned bodies need to pay special attention to women regarding anaemia based on place of residence and region.

LIMITATIONS
Some important determinant factors were not incorporated in the analysis due to high missing values in the data used. Furthermore, because of cross-sectional nature of the data used for the study, it was not possible to determine the cause-effect relationship between anaemia and its predictors.

CONSENT
It is not applicable.

ETHICS APPROVAL
Ethical approval was not necessary as this study used the 2016 EDHS publicly available secondary data which is available on the DHS website (http:// dhsprogram.com).

AVAILABILITY OF DATA
The data used for the final analysis in this study is available from corresponding author upon reasonable request.