Menu
Menu
Menu
Menu
Background: Anemia is a public health problem in many developing countries. It affects a sizable proportion of women of reproductive age. Anemia increases the risk of morbidity and mortality from infectious diseases, and can lead to poor fetal outcomes, and low productivity. This study examined the trends and determinants of anemia among women of reproductive age in Uganda. Methods: This study analyzed data from the Uganda Demographic and Health Surveys conducted in 2006, 2011, and 2016. The study was based on 10,956 weighted cases of women age 15-49. Bivariate analysis and multiple logistic regression analysis examined the association between the outcome variable and the determinants. Potential determinants of anemia in women were selected based on literature. Results: The results of the analysis show that anemia decreased in Uganda between 2006 and 2016, but with an increase between 2011 and 2016. The overall prevalence of anemia among women was 50, 23, and 32% respectively in 2006, 2011, and 2016. Women who were pregnant at the time of the survey had higher odds of being anemic across the surveys (OR 2.00, 95% CI 1.49-2.67; OR 1.47, 95% CI 1.02-2.10; OR 1.33, 95% CI 1.07-1.65). Women in households with nonimproved sources of drinking water also had higher odds for anemia (OR 1.32, 95% CI 1.09-1.61) in 2016. Wealth index, region and age were also significantly associated with anemia in women. Conclusion: In order to reduce anemia in women, there is need to target pregnant women during antenatal and postpartum visits, and ensure that nutrition education during such visits is supported. There is also need to ensure sustainable household access to safe water. This should be combined with interventions aimed at enhancing household wealth.
The study used datasets from the 2006, 2011, and 2016 Uganda Demographic and Health Survey (UDHS). The UDHS is a nationally representative population-based household survey, conducted every 5 years. The UDHS uses a stratified two-stage cluster sampling procedure. In the first stage, clusters are selected from sampling frames using the most recent census. Households are selected from each cluster at the second stage. The UDHS captures information in such areas as births to women age 15–49, women’s demographic and socioeconomic characteristics, household characteristics, maternal and child health and nutrition, access to health facilities and involvement in household decision making using questionnaires. It further includes testing of height and weight of women and children, and testing for anemia, malaria and Vitamin A deficiency [20, 21, 25]. In this study we only considered women whose blood sample had been drawn for testing, who had a test result for the anemia level, and who were usual members of the household in which they were surveyed. These criteria resulted in 10,956 weighted cases of women age 15–49 years for the three survey years. The total sample included 2672, 2539 and 5745 women in 2006, 2011 and 2016 survey years respectively. The dependent variable—woman’s anemia status—was recoded from the anemia level variable in the DHS datasets. Anemia level was determined from the result of hemoglobin level from blood testing. During the survey, blood specimens were collected for eligible women who voluntarily consented to be tested. This was done by obtaining a blood sample from a drop of blood taken from a finger prick. Hemoglobin analysis was carried out on site using a portable Hemocue analyser. Results were provided both verbally and in writing and all severe cases were referred for follow-up care. Anemia is marked by low levels of hemoglobin in the blood. For the analysis, all nonpregnant women age 15–49 who had less than 11.0 g of hemoglobin per deciliter (g/dl) were coded as anemic. Among pregnant women, those with hemoglobin values less than 12.0 g per deciliter were considered anemic. Nonpregnant women with hemoglobin values below 4.0 g per deciliter and those above 18.0 g per deciliter (g/dl) were regarded as implausible. Also, hemoglobin values below 3.0 g per deciliter and those above 17.0 g per deciliter (g/dl) in pregnant women were regarded as implausible. All implausible cases were excluded from the analysis. For the analysis, woman’s anemia status was recoded into a binary outcome variable. All women whose anemia level was severe, moderate, or mild, were recoded as yes, and nonanemic cases were recoded as no. For the analysis, covariates were selected based on literature [4, 26, 27]. Covariates were grouped into three categories: community, household and individual variables. Community-level variables included place of residence and region. The region variable for the 2011 and 2016 UDHS was recoded as in the 2006 UDHS, for comparability. It was categorized as Kampala, Central 1, Central 2, East Central, Eastern, Northern, West Nile, Western, and South Western. Household-level variables included wealth index, sex of the household head, type of toilet facility, source of drinking water, and number of children in the household. Wealth index is a composite measure of a household’s living standards. It is calculated using data on a household’s ownership of assets, household construction materials, and water and sanitation facilities [28]. Wealth index was coded as 1 poorest, 2 poorer, 3 middle, 4 richer, and 5 richest. Sex of the household head was coded as 1 male and 2 female. Type of toilet facility was recoded as 1 improved toilet, 2 shared toilet, 3 nonimproved toilet, and 4 no facility. Source of drinking water was grouped into improved and nonimproved as in the DHS grouping. Individual-level variables for women included age, educational attainment, involvement in decision-making, ever giving birth, access to health services and pregnancy status. Age was coded into seven 5-year groups: 15–19, 20–24, 25–29, 30–34, 35–39, 40–44, and 45–49, for better illustration of results [20]. Educational attainment was coded as 0 no education, 1 primary, 2 secondary, and 3 higher. Involvement in decision-making was generated from three variables: Women who were involved in decision-making individually or jointly with their partner regarding spending of their income, their own health care, and household purchases were recoded as 1 involved, otherwise 0 not involved. All missing cases were recoded as 9. Women who reported having given birth were recoded as 1 yes, otherwise 0 no. Access to health care was recoded as 1 yes, otherwise 0 no, depending on whether distance to facility was reported as a big problem in accessing health care or not. Pregnancy status was coded as 1 yes for women who reported that they were pregnant at the time of the survey, and 0 no for women who were not pregnant or not sure of their status. Only women who had plausible results of the blood hemoglobin levels were included in the analyses. Data were weighted using the women’s individual sample weight to adjust for nonresponse and disproportionate selection. The svy command was used to account for complex survey design. The independent variables were tested for multicollinearity using the pairwise correlation coefficient and only variables with a relationship below 0.5 cutoff were included in the analysis [29]. Bivariate analyses were conducted to examine association between the dependent variable (anemia) and the explanatory variables. Pearson’s chi-squared (χ2) tests were used to examine the significant differences between anemia and the explanatory variables. Statistical significance using p-values was set at p < 0.05. Multivariate logistic regressions were used to examine the relationship between anemia status and the determinant variables. The results are presented for four models: Model 1 for 2006; Model 2 for 2011; Model 3 for 2016; and Model 4 for pooled data for the 3 survey years. Adjusted odds ratios and 95% confidence intervals are presented. All analyses were conducted using Stata version 15, and results are reported for the UDHS survey years 2006, 2011, and 2016.
N/A
