Introduction

Anemia affects millions of women of reproductive age worldwide and is characterized by low levels of healthy red blood cells and hemoglobin, which reduces the ability of the blood to transport oxygen to body tissues [1,2]. Because of the physiological state of menstruation, pregnancy, and lactation, it is the most common nutritional problem disproportionately affecting women of reproductive age [3]. During each stage of a woman’s reproductive cycle, there is a varying risk of exposure to anemia, but its presence throughout all stages is associated with poor outcomes for the mother, newborn, and children [4–6]. Pregnant women who suffer from anemia during the preconception period are at an increased risk of experiencing spontaneous abortions, giving birth to infants with low birth weights, and restricting fetal growth. Furthermore, these children may be more likely to develop long-term disorders, such as autism and intellectual disabilities [7,8].

The global incidence of anemia is 30% among non-pregnant women and 36% among pregnant women [9]. Approximately 32% of childbearing women in low- and middle-income countries (LMICs) suffer from anemia. More than 50% of these countries have anemia levels between 20% and 39.9% in this age group [10]. It is most prominent in SSA countries with a prevalence of 40.5%. In Africa, the prevalence of anemia varies widely from 13% in Rwanda to 62% in Mali [11]. According to data from 2015-16Tanzania Demographic and Health Survey, anemia is alarmingly prevalent among Tanzanian women of reproductive age (45%) [2]. Iron deficiency is a serious dietary problem that accounts for 50% of instances of anemia worldwide. It is the primary contributor to anemia. Nevertheless, deficiencies in vitamin A, vitamin B12, and folic acid can also result in nutritional deficiency anemia. Acute and chronic inflammations, parasite infections, and hereditary or acquired diseases affecting hemoglobin synthesis and the creation or survival of red blood cells can also be causes of anemia [3].

Anemia in during pregnancy is linked to several unfavorable health consequences, such as an elevated risk of maternal and infant mortality, preterm delivery, low birth weight, and impaired cognitive development in offspring. These devastating impacts make anemia in pregnant mothers to be one of the global health priority areas at the global level, especially in resource-limited areas [12,13].

Evidences revealed that anemia is attributed to a complicated interplay of factors, including dietary deficiencies, parasitic infections, and socioeconomic status. Particularly iron deficiency, infectious diseases like malaria and hookworm infestation, limited access to healthcare and prenatal care, cultural practices such as early marriage and teenage pregnancy, as well as socioeconomic factors like poverty and lack of education [2,14].

To reduce anemia in women of reproductive age several interventions have been implemented such as, increasing access to iron supplements, promoting iron-rich diets, strengthening antenatal care, and educating the public about reproductive health. Despite these efforts, anemia rates among non-pregnant and pregnant women have declined by only 1% and 5% over the last two decades, respectively [9]. Furthermore, existing evidence shows that none of the LMICs, including Tanzania, are on track to achieve the nutrition target of 50% anemia reduction by 2030, indicating that the average annual rate of decline was below the level needed [15]. Thus, for effective intervention strategies and policy development in Tanzania, it is essential to generate up-to-date national, sub-regional, and regional estimates of anemia and to identify individual- and community-level factors[16]. As a result of, understanding the determinants of anemia among reproductive-age women in Tanzania is crucial for developing targeted interventions and policies to address this public health challenge.

To date, limited evidence exists in Tanzania regarding anemia and its determinants, with most studies conducted in a single setting lacking multilevel modeling to consider the hierarchical structure of the data from the Demographic and Health Survey (DHS). Moreover, the data were not weighted to compensate for the non-proportional distribution of the sample size and proper standard error estimation. Additionally, using a multilevel analysis approach can help to account for the complex interplay of individual-, household-, and community-level factors that may contribute to anemia prevalence, providing a more comprehensive understanding of the issue. Therefore, this study aimed to conduct a multi-level analysis of data from the Tanzanian Demographic and Health Survey (TDHS) to identify the individual, household, and community-level factors associated with anemia status among reproductive-age women.

The multi-level analysis approach is particularly relevant in this context, as it allows for the examination of the relative contributions of individual, household, and community-level factors to the prevalence of anemia. In addition, this research will provide updated and reliable data on the determinants of anemia among women of reproductive age in Tanzania in designing targeted interventions and programs to reduce the burden of anemia.

Method and materials

Study design, data source, population and sampling procedure

A nationally representative, cross-sectional study utilizing secondary data from the Tanzania Demographic and Health Surveys (TDHS) of 2022 was conducted. The TDHS is conducted every five years to collect updated health and health-related information. The 2022 TDHS sample design was carried out in two stages, with the purpose of providing national, urban, and rural estimates for Tanzania Mainland and Zanzibar. For specific indicators, such as contraceptive use, the sample design allows for estimates at the regional level for each of the 31 regions—26 regions in Tanzania Mainland and s5 regions in Zanzibar.

The 2022 TDHS employed a stratified, two-stage sample design. The first stage involved selecting sampling points (clusters) comprised of enumeration areas (EAs) delineated for the 2012 Tanzania Population and Housing Census (2012 PHC). Enumeration areas were selected using probability proportional to size (PPS) sampling method. The grouping of regions into zones allowed for larger denominators and smaller sampling errors for indicators at the zonal level. Tanzania was divided into several strata based on geographical regions and urban/rural areas. Each stratum was treated as a separate sampling domain. A total of 629 clusters were selected, 211 urban and 418 rural. In the second stage, 26 households were systematically chosen from each cluster, resulting in a total sample size of around 16,354 households. To account for the unequal probability of selection and non-response, individual weights were calculated for each respondent. For this study, we utilized the individual record (IR) file sourced from datasets within the TDHS 2022, which comprised information from 15,254 women aged 15–49 years. Hemoglobin test was conducted for the subsample of 7577 of reproductive age women. Finally, a total weighted sample size of 8,921 reproductive-age women (unweighted sample = 7,577) was considered for this analysis (Fig 1). The specifics of the sampling technique, sample size, data collection instruments, data quality control measures, and ethical considerations are all outlined in the 2022 TDHS report [17].

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Sampling procedure and sampling technique anemia status and its determinants among reproductive-age women in Tanzania using 2022 TDHS.

https://doi.org/10.1371/journal.pone.0311105.g001

Results

Study participant’s descriptive characteristics

An overall weighted sample of 8,921 women of reproductive age was used for this study. Among the study participants, the median age was 28 (IQR = 21–37) years, and the majority (38.5%) were between 15 and 24 years old. A quarter of participants (2,368 (26.5%)) came from households with the highest wealth quintile, and 5,583 (62.6%) were employed. In terms of the gender of household heads, 6,276 (70.4%) of respondents were male. The majority of our study participants, 4,370 (48.9%), had a primary education, and 4,005 (44.9%) were married. There were 3,374 (37.8%) multiparous respondents and 2,527 (28.3%) nulliparous respondents. Among the respondents, 2,035 (22.8%) were lactating, and 669 (7.5%) were pregnant. Twenty-eight percent (28.9%) of the 2,581 respondents currently use hormonal or non-hormonal contraception. A majority of 8,413 (94.3%) respondents were from households with health insurance, and only 2,581 (28.9%) use internet (Table 1).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Individual-level characteristics of reproductive age women Tanzanian demographic and health survey, 2022.

https://doi.org/10.1371/journal.pone.0311105.t001

Community-level characteristics of the study participants

Of the study participants, 5,710 (64%) were from rural areas, 2,016 (22.5%) were from the eastern administrative zone, 558 (6.2%) were from the south, and 1,334 (15.0%) were from Zanzibar. Community maternal education was high in two-thirds of 2,918 (32.7%) of the study participants, and the level of poverty was low in 5,127 (57.5%) of the community (Table 2).

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Community-level characteristics of reproductive age women Tanzanian demographic and health survey, 2022.

https://doi.org/10.1371/journal.pone.0311105.t002

Random effect and model fitness analysis results

In this study, a single-level binary Poisson regression model was compared to a multilevel robust Poisson regression model without predictors. The deviance (-2 Likelihood Ratio (LR) test) was used to determine which model was more appropriate, and the result was statistically significant (p<0.05), suggesting that the multilevel robust Poisson regression model with robust variance was a better fit than the single-level robust Poisson regression model. Four random effects models were fitted, and the model with the lowest deviation was selected. Furthermore, an intra-class correlation coefficient (ICC) of 0.1034 was observed in the null model, suggesting that 10.34% of anemia was influenced by unobserved factors at community levels. The Akaike’s Information Criteria (AIC) is the smallest in model 4 (AIC = 13742.0) as compared to other models. Thus, model 4 is the best-fitting model. Therefore, all interpretations and reports were made based on this model (Table 3).

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Model summary of multilevel robust Poisson regression analysis of factors associated with anemia among reproductive age women Tanzanian demographic and health survey, 2022.

https://doi.org/10.1371/journal.pone.0311105.t003

Determinants of anaemia among reproductive age women

In the multivariable multilevel mixed-effect Poisson regression model, poorest household wealth index, being employed, being pregnant, hormonal contraceptive use, underweight, alcohol consumption, and women from (eastern, central, southern highland and Zanzibar) administrative zones and high community poverty level were predictors of anemia among reproductive-age women. It was found that anemia prevalence was 1.07 higher among unemployed reproductive-age women compared with employed women [APR = 1.07, 95% CI: 1.01, 1.16]. Reproductive-age women from the poorest household wealth index had 1.17 times [AOR = 1.17, 95% CI: 1.12, 1.57] higher prevalence of anemia as compared to reproductive-age women from the richest household wealth index. Women with a body mass index of <18.5 kg/m2 (underweight) had 1.15 times higher prevalence of anemia than women who had BMI ≥18.5 kg/m2 [APR = 1.15, 95% CI: 1.09, 1.82]. Women who are using hormonal contraceptives were 28% less likely to be anemic compared to those who do not using [APR = 0.72, 95% CI: 0.64, 0.81]. Being a pregnant had 1.21 times higher prevalence of anemia as compared to non-pregnant women [APR = 1.21, 95% CI: 1.1, 1.34]. Women of reproductive age who did not drink alcohol had 1.29 times higher prevalence of anemia than women who did [AOR = 1.29; 95% CI: 1.08, 1.52].

The prevalence of anemia was 1.16 times higher among reproductive age women from eastern administrative zones [APR = 1.16, 95% CI: 1.05, 1.28] and 1.36times higher women from Zanzibar administrative zones of Tanzania [APR = 1.36, 95% CI: 1.21, 1.51]. Whereas being anemic was 27% less likely among women from southern highland administrative zones [APR = 0.73, 95% CI: 0.62, 0.86] and 12% less likely among women from central administrative zone [APR = 0.78, 95% CI: 0.66, 0.93] as compared to reproductive age women from lake administrative zone and in southern [AOR = 0.73, 95% CI: 0.62, 0.82]. Being anemic was 1.13 times more likely among reproductive-age women from high community poverty levels as compared to reproductive-age women from low community poverty levels [APR = 1.13, 95% CI: 1.01, 1.25] (Table 4).

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Multi-variable multilevel mixed-effects robust Poisson regression analysis of factors associated with anemia among reproductive age women Tanzanian demographic and health survey, 2022.

https://doi.org/10.1371/journal.pone.0311105.t004

Discussion

Anemia is a global health problem, particularly prevalent among women of reproductive age in low-income countries. Tanzania, like many other low-income countries, grapples with a high prevalence of anemia among its population, posing adverse consequences for individuals and society at large. This research aimed to shed light on the prevalence, contributing factors, and implications of anemia in Tanzania.

According to this study, Tanzania faces a significant burden of anemia, with a prevalence rate of 42.02% among reproductive-age women. The finding from this study was consistent with the studies from Sub-Saharan Africa (40.5%) [11], low-middle income countries (41.58%) [21] and Gambia (44.28%) [22]. However, the finding from this study is higher than the prevalence reported in Ethiopia (37.4%) [23], loa peoples of democratic republic (39.2%) [24], Philippines (9.0%) [25], Rwanda (19.2%) [13], Mali (38%) [26], east Africa (34.85%) [27] and systematic review in SSA (32.1%) [28]. On the other hand the finding from the current study was lower than the study conducted in Nepal 64.32% [29], Senegal 55.2% [12], Mali (63.5%) [30], Pakistan 61.3% [31] and India (65.5%) [32]. The potential reason may be due to differences in socioeconomic status among countries such as chronic malnutrition, limited access to nutritious food, poor sanitation and hygiene, inadequate healthcare infrastructure and lack of education and awareness [33]. Addressing these issues through strategies such as improving food access, promoting hygiene practices, and enhancing healthcare infrastructure can reduce anemia in Tanzania.

In this study, determinants of anemia were also identified. In the final model, we found that Poorest Wealth index, Being employed, BMI <18.5 kg/m², hormonal contraceptive use, pregnancy status, alcohol consumption, administrative zones and high community poverty level were significantly associated with anemia among reproductive age women.

Women from families with poorest household wealth index had increased the prevalence ratio of anemia by 7% as compared to women from richest households. This is in line with findings reported from Ethiopia east Africa [27], Ethiopia [23], India [32], SSA [11] and low and middle income [34]. The possible reason could be that low household wealth is connected to food insecurity, which prevents individuals from being well-nourished and increases their risk of micro nutrient deficiency including iron deficiency anemia. Additionally, the widespread occurrence of anemia may be due to the fact that households with lower wealth are less likely to have sufficient purchases of nutrient-rich food and access to healthcare services for women when they are ill.

This study showed that women who were unemployed are more likely to develop anemia than those who are employed. Similar finding were reported from a multilevel analysis in east Africa [27], Mauritania, Nigeria [35] and primary study in Senegal [12]. This may be explained by the fact that the woman’s financial independence gives her more control over the decisions she makes regarding her follow-up care. An excellent degree of socioeconomic well-being acts as a buffer against anemia. Thus, anemia is a health issue associated with the nations’ socioeconomic status. A higher socioeconomic status would improve nutritional status and act as a preventive factor against anemia.

This study also revealed being underweight was significantly associated with increased the likelihood of anemia among reproductive age women. This is in line with studies conducted in Indonesia [36], Ethiopia [23], SSA [11], and [13] Rwanda. Women who are underweight are more likely to develop anemia due to several factors. One reason is that they may not be consuming enough iron-rich foods in their diets. Another reason is that they may have lower total body iron stores due to lower muscle mass, which can result in lower ferritin levels [37]. Ferritin is a protein that stores iron in the body and helps to regulate its release. Having lower ferritin levels means that less iron is available to produce hemoglobin and maintain healthy red blood cells [38,39]. Therefore, it is crucial for underweight women to maintain a healthy weight and consume a balanced diet rich in iron and other essential nutrients to prevent anemia and promote overall health.

The current study reveal that the prevalence of anemia were reduced by 28% among women current users of hormonal contraceptives; these findings are consistent with those of other studies in Tanzania [39], Rwanda [13], twenty four SSA countries [40] and population based study in SSA [41]. This could be explained by hormonal contraception methods like the pill, patch, ring, and injection use synthetic hormones (estrogen and progestin) to prevent pregnancy. These hormones help regulate menstrual cycles, decrease the amount of blood lost during periods, and prevent ovulation altogether. As a result, women who use hormonal contraception methods are less likely to experience heavy menstrual bleeding and prolonged periods, which can lead to iron deficiency anemia [42,43].

The prevalence of be anemic among pregnant reproductive age women were 1.21 times higher compared to non-pregnant women. This is consistent with the findings reported in previous studies Senegal [12], Gambia, Indonesia [36], Democratic Republic of Congo [24], and Ethiopia [23]. During pregnancy there are factors that can increase the chances of women developing anemia. The changes, in the body during pregnancy such as an increase in blood volume can dilute blood cells and hemoglobin levels leading to a decrease in their levels and eventually causing anemia [44]. Additionally, pregnant women need nutrients, iron to produce extra red blood cells. Insufficient intake of nutrients, through diet having pregnancies or pregnancies that are closely spaced apart and certain complications can further strain the body’s resources and raise the risk of anemia. It is crucial to screen for, monitor and manage anemia during pregnancy to ensure the well-being of both the mother and the fetus.

In this study, alcohol consumption decreases the likelihood of developing anemia among reproductive age women; in line with the study in Ethiopia [45] and Tanzanian adults[46]. This happens because of a few factors. One is that when we consume alcohol it has been found to increase the absorption of iron, from the food we eat. Additionally research has shown that alcohol can stimulate the production of erythropoietin, a hormone that tells our bone marrow to make blood cells [47]. Furthermore alcohol can also improve blood flow and circulation in our bodies, which helps deliver oxygen and nutrients to our tissues and organs [48]. It is important to note, however, that excessive alcohol consumption can have a negative impact on health and may increase the risk of developing other conditions such as liver disease, heart disease, and certain types of cancer. Therefore, moderation is key when it comes to alcohol consumption [49].

Reproductive-age women from eastern and Zanzibar administrative zones were significantly associated with higher odds of having anemia, whereas women from southern administrative zones had a lower prevalence of anemia. Several factors contribute to the variation in anemia prevalence rates across different administrative zones in Tanzania. These factors include socioeconomic factors such as income levels, poverty rates, and access to clean water, sanitation facilities, and healthcare services. Food security and dietary patterns also play a role, with variations in agricultural productivity and the availability of nutritious foods affecting anemia prevalence rates [50]. It is important to consider that these factors interact with each other, and multiple factors may contribute to the geographic variation of anemia in Tanzania.

Women living in communities with a high level of community poverty were more likely to suffer anemia than women who lived in communities with low levels of community poverty. This is consistent with another study SSA [11] and India [51]. The likelihood of anemia is heightened in areas with high community poverty due to a lack of access to nutritious food, inadequate sanitation and hygiene practices, limited healthcare, and essential services, as well as poor living conditions [52,53]. It is crucial to address community poverty and underlying socioeconomic factors in order to decrease anemia prevalence rates.

The high prevalence of anemia among reproductive-age women in Tanzania has significant implications for the health and development of the country. Addressing the root causes of anemia, such as improving access to nutritious food, implementing effective interventions, and increasing awareness of anemia prevention, is crucial for the overall health and well-being of women and children in Tanzania.

Conclusion

In conclusion, the prevalence of anemia among reproductive-age women in Tanzania is significantly higher compared to the global prevalence reported by the WHO. This finding underscores the urgent need for concerted efforts to address this public health issue in Tanzania. The high prevalence of anemia among reproductive-age women not only poses a significant threat to their health but also has severe implications for maternal and child health outcomes. Interventions targeting the improvement of nutrition, access to healthcare services, and education on anemia prevention and management should be prioritized to effectively reduce the burden of anemia among reproductive-age women in Tanzania. Additionally, further research is warranted to explore the unique factors contributing to this high prevalence and tailor interventions accordingly. By implementing these measures, we can strive to improve the health and well-being of women in Tanzania, ensuring better maternal and child health outcomes.

Acknowledgments

We would like to thank the measure DHS program for providing the datasets.