http://orcid.org/0000-0001-8996-7995
Figures
Abstract
Background
Anemia is a major public health concern among women of reproductive age leading to high maternal mortality in low- and middle-income countries. Of the prior studies conducted in Pakistan, most focused on large urban areas and did not explore the determinants of anemia among women of reproductive age (WRA) across socio-demographic, dietary, reproductive, and biological domains. Thus, we aimed to study the prevalence and determinants of anemia among WRA in rural Pakistan.
Methods
We conducted a cross-sectional study in the Thatta district of Pakistan from September 2018 to January 2019 and enrolled 150 non-pregnant, married women. Data collectors administered a structured questionnaire to collect sociodemographic, reproductive and dietary data from women, who also provided stool and blood samples. We classified all WRA as anemic if their hemoglobin was <12.0 g/dl. We performed logistic regression analysis to calculate adjusted odds ratios (aOR) and their respective 95% CIs to assess the determinants of anemia.
Results
In our study, 61.3% of the enrolled women were anemic. In the multivariable analysis, we found that factors such as serum iron levels of less than 50 μg/dl (aOR: 7.17; 95% CI (2.94, 17.47)), history of breastfeeding (aOR: 2.43; 95% CI (1.04, 5.72)), living in a katcha house (aOR: 6.61; 95% CI (2.21, 19.87)), no consumption of meat (aOR: 4.18; 95% CI (1.66, 9.96)) were significantly associated with anemia among WRA. A history of more than one abortion (aOR: 0.06; 95% CI (0.01, 0.33) appeared protective for its association with anemia.
Conclusion
Our findings demonstrate a high burden of anemia and its complex determinants among WRA in rural Pakistan. A combination of nutritional and educational strategies should be designed to encourage rural women to consume iron-rich foods in their diet with an access to adequate food. Breastfeeding women should be encouraged to consume extra calories with sufficient intake of the food to continue exclusive breastfeeding and reserve the iron stores through amenorrhea to prevent themselves from becoming anemic.
Citation: Ali SA, Abbasi Z, Shahid B, Moin G, Hambidge KM, Krebs NF, et al. (2020) Prevalence and determinants of anemia among women of reproductive age in Thatta Pakistan: Findings from a cross-sectional study. PLoS ONE 15(9): e0239320. https://doi.org/10.1371/journal.pone.0239320
Editor: Sabine Rohrmann, University of Zurich, SWITZERLAND
Received: May 31, 2020; Accepted: September 4, 2020; Published: September 24, 2020
Copyright: © 2020 Ali et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: The primary and only source of funding for this study was the University of Colorado-Denver USA | Anschutz Medical Campus. Neither my instituion (Aga Khan University), nor any other organization provided any funding for this cross-sectional study The funders had no role in study design, data collection and analysis, except they reviewed and approved the protocol of the study and provided feedback and comments on the final manuscript and approved the manuscript for publication. No author received salary for this cross-sectional study. The authors “Sumera Aziz Ali and Zahid Abbasi” received a proportion of their salary for the main and primary Women First study, but not for this cross-sectional study. The primary Women First study was funded by University of Colorado- Denver USA | Anschutz Medical Campus. We received funding for this study from the University of Colorado-Denver USA | Anschutz Medical Campus.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Anemia is one of the major public health issues among women of reproductive age (WRA), as it leads to high maternal morbidity and mortality and adverse pregnancy outcomes [1]. Anemia occurs when hemoglobin (Hb) concentration falls below an established cut-off value (<12.0 g/dl) among WRA, consequently impairing the capacity of the blood to transport oxygen to the body [2–4]. About half a billion of WRA are anemic worldwide, with a higher burden of anemia in low and middle-income countries (LMICs), affecting nearly two-thirds of WRA [5, 6]. More specifically, 41.9% of WRA are anemic in South-East Asia [6]. This is in contrast to the burden of anemia in developed countries, such as in Europe where 2–5% of WRA are classified as anemic [7]. Like other LMICs, there is a high burden of anemia in Pakistan [8]. Studies have shown that anemia affects 41.7% to 77.0% of WRA in Pakistan [8, 9]. Anemia is more prevalent in the rural areas of Pakistan, where it is often severe and linked to adverse health consequences such as postpartum hemorrhage, preterm delivery or stillbirth and low birth weight babies [10]. Furthermore, a study conducted by Hambidge et al 2019 found that anemia is an important and potentially modifiable risk factor on birth outcomes among rural Pakistani women [11]. More specifically, this study demonstrated that maternal anemia at baseline modified the treatment effect for birth outcomes such as birth weight, weight to length ratio as the improvements in these outcomes were significantly greater for Arm 1 (commencing nutrition intervention prior to conception) vs. Arm 2 (commencing same nutrition intervention early in gestation) [11].
Although few studies have investigated the determinants of anemia in Pakistan, certain gaps need to be addressed [12–14]. First, existing studies mainly focused on large urban areas of Pakistan rather than rural areas where the burden of anemia is higher. In addition, there are differences in the socio-demographic, cultural, economic, and dietary factors between rural and urban areas, which do not accurately characterize anemia in a rural setting [15]. Moreover, rural areas have limited sources of income with poor access to health facilities and other amenities such as electricity, modes of communication and access to health care facilities [16, 17]. Therefore, studies of urban populations may not be relevant in the rural context of Pakistan. Second, anemia is a complex and multifactorial phenomenon that should be understood holistically. Current studies in urban settings provide important insights about sociodemographic and reproductive determinants, but leave key biological and dietary determinants unaddressed and therefore, poorly understood [18–20]. Lastly, investigators have not utilized objective measurements of biological factors such as iron, ferritin, B12, and folic acid in the serum, hookworm in the stool, and a malarial parasite in the blood. Rather, their results are often based on self-reported data, which are subjective and may have produced biased estimates of associations between biological determinants and anemia among WRA.
There is a need to investigate the underlying determinants of anemia comprehensively across the range of socio-demographic, reproductive, dietary, and biological domains in rural Pakistan, which have not been previously studied in this context. Understanding these determinants holistically will enable the development of local strategies and targeted interventions to address the burden of anemia among WRA in rural Pakistan. Hence, the overarching objectives of this study are to measure the prevalence and determinants of anemia among WRA in rural Pakistan. We hypothesized that poor, less educated, older, and multiparous women who consume iron-deficient diet with abnormal biomarkers will be more anemic than younger, nulliparous, well-privileged women who consume iron-rich diet.
Materials and methods
Study population and sample
We conducted a cross-sectional study in the district of Thatta, Pakistan from September 2018 to January 2019. We examined the associations of sociodemographic, reproductive, dietary, and biological determinants of anemia in 150 WRA participating in the Women First (WF) preconception maternal nutrition trial [21]. At the time of enrollment in the primary WF trial, women were eligible if they had 0–5 children and planned to conceive in the future after being enrolled in the WF primary study. We used a subset of women from the WF database (385 women) who provided biological samples in the primary trial, who had conceived during their enrollment in the WF primary study, and had delivered their baby [21]. For this cross-sectional study, we randomly selected study participants using the WF database as a sampling frame. Eligibility was contingent on being a resident of the Thatta district, not being pregnant at the time of survey, agreeing to participate, and intending to provide biological samples after enrollment in the cross-sectional study.
Data collection tool (questionnaire)
We used a structured and validated questionnaire to collect data on socio-demographic, reproductive, anthropometric, dietary, and biological determinants from WRA. We identified possible determinants of anemia based on the literature pertaining to the risk of development of anemia among women in LMICs [19]. To develop a questionnaire for the proposed cross-sectional study, we chose a series of validated questions from different studies that have been conducted in Thatta Pakistan. For example, the Pakistan Demographic Health Survey 2012–2013 (PDHS-2012-2013) was one of such sources to choose the relevant validated questions [15]. We selected questions related to socio-demographic characteristics from the PDHS-2012-2013 survey and adapted the questions wherever necessary [15]. Secondly, a maternal and newborn health registry (MNHR) is placed in Thatta district [20], whereby data is being collected on the socio- demographic, and reproductive characteristics of WRA in a similar setting and same population, where we undertook the cross-sectional study [18]. The MNHR has validated the questions in a similar population that also helped me to use several relevant questions for the current cross-sectional study [18]. Lastly, an individual randomized nutritional trial (Woman First) was conducted from 2012 to 2018 in Thatta district where MNHR is based [19, 21]. The trial used 20 validated questionnaires to explore socio-demographic, reproductive, dietary factors, anthropometric measurements, and biological factors among WRA in the same population [19]. Since all these questionnaires were relevant to the proposed study; therefore, we choose the most appropriate questions from these validated questionnaires and adapted these wherever required to develop and finalize the questionnaire for the cross-sectional study. After developing and finalizing the questionnaire from three main sources, we translated the questionnaire into the local language of district Thatta (Sindhi) followed by back-translation. To check the flow of questions, we pre-tested questionnaire on the WRA other than the study setting of Thatta but having similar characteristics to the study sample. The questionnaire consisted of different sections and collected socio-demographic, reproductive, dietary, and anthropometric data.
Study procedures & data collection
After assessing the eligibility criteria and receiving written informed consent, trained study staff, who were familiar with the local language (Sindhi), administered a structured and validated questionnaire to the enrolled women in their households. The study staff asked women about various socio-demographic, reproductive and dietary measures including maternal age, parity, education, working status, water supply, sanitation, age at the time of marriage, history of any previous adverse pregnancy outcome, date of the last delivery, consumption of fruits, vegetables and meat in their diet, and use of smokeless tobacco. The study staff also measured weight (in kilograms), height (in meters), waist circumference (in centimeters (cm)), hip circumference (cm), and mid-upper arm circumference (cm). Using a digital weighing scale, the study staff measured the weight of WRA while barefoot and wearing light clothes to the nearest 0.1 kg. The same staff measured the height of a woman barefoot to the nearest of 0.1 cm using a stadiometer. Before the measurement, the staff calibrated scales to zero. The staff measured mid-upper arm circumference on the upper right arm at the midpoint of the acromion process and the tip of the olecranon (precision 1mm). Similarly, the research staff measured waist circumference just above the iliac crest in the horizontal plane, and hip circumference at the point yielding the maximum circumference over the buttocks, all using a standard measuring tape to the nearest 1mm. We calculated body mass index (BMI) as a ratio of weight (in kilograms) to the square of height (in meters) and the study staff recorded BMI as a continuous variable. We categorized BMI as underweight (less than 18.5 kg/m2), normal (18.5–24.9 kg/m2), overweight (25.0–29.9 kg/m2), and obese (≥30.0 kg/m2) [22]. For regression analysis, we merged overweight and obese categories into a single category as overweight/obese due to low numbers in the obese category.
A trained and professional phlebotomist collected biological samples including blood and stool samples from the enrolled women. The phlebotomist collected around 12–15 mL of venous blood by following the standard procedures [19]. The phlebotomist transferred all biological samples from Thatta collection points to the main laboratory of the Aga Khan University Hospital in Karachi, Pakistan by maintaining the cold chain. The trained laboratory staff and microbiologist processed the blood samples to measure complete blood count (CBC), serum iron and ferritin levels, serum B-12 and folic acid levels. In addition, they used the blood samples to measure the inflammatory markers including erythrocyte sedimentation rate (ESR) by the Westergren method, and C-Reactive protein (CRP) using a high sensitivity ELISA assay [23, 24]. Moreover, they used an immunochromatographic test (ICT) technique to identify the malarial parasite in the blood and examined blood for hemoglobinopathies including thalassemia minor and major by hemoglobin (Hb) electrophoresis test [25, 26]. Furthermore, they used stool samples to examine different worms in the stool to primarily assess hookworm infestation. They identified parasites and eggs in the stool samples, using a microscope with conventional techniques such as direct wet smear stain (Lugol stain) [27].
Primary outcome
The primary outcome of this study was anemia and we used hemoglobin levels (grams per deciliter) in the blood to classify a woman as anemic. More specifically, we used the World Health Organization (WHO) cut-off of a hemoglobin level <12.0 g/dL to classify a non-pregnant woman aged ≥15 years as anemic [28–32]. For the current analysis, we created a binary variable of anemia and no-anemia.
Potential determinants of anemia
Independent variables in this study consisted of various socio-demographic, reproductive, dietary, and biological determinants. The socio-demographic determinants included woman’s age (in years), literacy status (illiterate / literate), residence (urban/ rural), type of house (pucca house: made of concrete, plastered walls, and roof / katcha house: made of wood, clay, and stalks), religion (Hinduism/Islam), current working status (are you working currently to earn money: Yes/No), having electricity (Yes/No), type of family system (nuclear/joint family).
Reproductive determinants included age at the time of marriage (in years), gravidity (total number of pregnancies), parity (number of children born after 20 weeks of gestation), history of adverse pregnancy outcomes including stillbirths and abortions (Yes/No), breastfeeding status at the time of the survey (Yes/No), current and ever use of family planning method (Yes/No), and menstrual cycle (regular/irregular).
Dietary determinants included intake of fruits and vegetables in past one month (Yes / No), consumption of meat (Yes/ No), and history of chewing smokeless tobacco (Yes/No). In addition, we measured biological determinants such as serum iron (normal range: 50–170 μg/dl), serum ferritin (normal range: 10 to 150 ng/ml), serum folic acid (normal range: ≥ 2.5–20 ng/ml), serum CRP (normal range: 0–0.5mg/dl), ESR (normal range: 0–20 mm/ 1st hr), thalassemia minor or major (yes/no), malarial parasite (positive/negative), and parasites in the stool (seen/unseen).
Statistical methods
To describe the characteristics of the study population, we reported frequencies and proportions. We used a Chi-squared test or Fisher’s exact test to assess the frequency distribution and the relationship between determinants and anemia status. Similarly, we performed bivariate logistic regression analyses to determine the individual effect of each significant determinant on anemia status before the multivariable logistic regression analysis. We used multicollinearity diagnostics such as variance inflation factor (VIF) and tolerance (reciprocal of VIF) to assess multicollinearity between independent variables [33]. Any VIFs of more than 4 were investigated, while VIFs greater than 10 or tolerance less than 0.1 indicated serious multicollinearity requiring correction by dropping one of the variables [33, 34].
Finally, we conducted a multivariable logistic regression analysis to determine the effect of each determinant on anemia by including all predictors and covariates in a single multivariable model using a purposeful selection method [35]. We used a p-value cut-off of <0.05 to determine if a variable should be included in the final regression model. We presented the results of regression analyses with crude/unadjusted odds ratios (OR) and adjusted odds ratios (aOR) with 95% confidence intervals (CIs). We used SPSS 19.0 to analyze the data.
Ethical considerations
Ethical approval for this study was obtained from the Aga Khan University Ethical Review Committee (2018–0262–342). All study participants provided written informed consent. Voluntary participation and the right to ask any questions and decline participation at any time were emphasized during data collection.
Results
Characteristics of women of reproductive age
We found that more than a quarter of the women (28.0%) were younger than 25 years old and 40.0% were 25–29 years old as shown in Table 1. Moreover, 82.0% of the women were not able to read or write, a similar proportion of the women (81.3%) were residents of rural areas, and 44.7% reported to be living in a katcha house. More than half of the study participants (55.3%) did not report working to earn money, and around three-fourths (73.3%) of them were living in a joint family system (Table 1). Around three-fourths of the study participants (71.3%) had given birth to at least three children, and 8.7% had a history of more than one abortion. Slightly more than a third of women (36.0%) reported not breastfeeding their children at the time of the study, and 46.0% had a normal BMI of 18.5-25kg/m2. Regarding the women’s diet, we found that 41.3% of the women had never consumed fruits or vegetables in the past month, and 68.7% had never eaten meat in the past month (Table 1).
We found that 61.3% of women were anemic in the study, and a similar proportion of the women (62.0%) had low serum iron levels (Table 2). We found that 42.7% of the women had lower serum ferritin values; however, only 6.0% of the women had low folic acid levels relative to the normal range of > 2.6ng/ml (Table 2). All of the women had serum B12 levels within the normal range (>150 pg/ml). Around 10.0% of the women had higher values of CRP than the normal range of 0–0.5mg/dl, and around 50.0% of the women had higher values for ESR than the normal range of 0–20 mm/ 1st hr. We found that only 8.7% of the women had any kind of underlying hemoglobinopathy, and 16.0% of the women’s bloods were positive for the malarial parasite and 15.2% of the women had parasites in their stool as shown in Table 2.
Sociodemographic, reproductive, and dietary characteristics of women of reproductive age by anemia status
We found no differences between anemic and non-anemic women by age, educational level, parity, working status, place of residence, and type of family system (Table 1). However, we found that a significantly higher proportion of anemic women (52.5%) were living in a katcha house when compared with 32.8% of non-anemic women (p-value: 0.02). Similarly, 68.5% of anemic and 56.9% of non-anemic women reported breastfeeding their children. A significantly higher proportion of anemic women (76.1%) had never consumed meat in the past month compared with non-anemic women (56.9%; p-value: 0.01). About half of anemic women (46.7%) were underweight, while 39.7% of the non-anemic women were underweight (Table 1).
The study also found that three-fourths of the anemic women (75.0%) had lower serum iron levels compared to 41.4% of the non-anemic women. Similarly, around half of anemic women (51.1%) had lower serum ferritin levels when compared to 29.3% of non-anemic women as shown in Table 2. There were no differences between anemic and non-anemic women by levels of serum folic acid, CRP, ESR, hemoglobinopathies, and malarial parasite in the blood and stool parasites (Table 2).
Determinants of anemia among women of reproductive age: Findings from a bivariate analysis
We assessed and found multiple statistically significant associations between covariates and anemia in bivariate analyses. More specifically, we assessed the association between various sociodemographic, reproductive, dietary, and biological factors and anemia among WRA. We found that women with more than four children had a 1.91(OR = 1.91; [95% CI: 0.77, 4.71]) times the odds of being anemic when compared with women having less than 2 children (Table 1). Meanwhile, women living in a katcha house had a 2.24 (OR = 2.24; [95% CI:1.13, 4.44]) times the odds of being anemic when compared with their counterparts. While exploring the association between reproductive factors and anemia, we found that women who were breastfeeding their children at the time of the study had a 1.65 (OR = 1.65; [95% CI: 0.83, 3.25]) times the odds of being anemic when compared with their counterparts; however, results were not statistically significant. Furthermore, women with low and normal BMI had 4.96 (OR = 4.96; [95% CI: 1.23, 20.01]) and 6.00 (OR = 6.00; [95% CI: 1.48, 24.31]) times the odds of being anemic, respectively, when compared with overweight and obese women. We also assessed the association between women’s diet and anemia in the bivariate analysis. We found that women who never consumed fruits and vegetables in the last month had 1.26 (OR = 1.26; [95% CI: 0.64, 2.46]) and 1.42 (OR = 1.42; [95% CI: 0.72, 2.78]) times the odds of being anemic when compared to women who consumed fruits and vegetables, respectively; however, these results were statistically insignificant (Table 1). Women who never consumed meat in the last month had 2.41 times the odds of being anemic when compared women who reported consuming meat in the past one month (OR = 2.41; [95% CI:1.18, 4.88]). Regarding the association between biological factors such as serum iron levels and anemia, we found that women with low serum iron levels had 4.25 (OR = 4.25; [95% CI:2.1,8.59]) times the odds of being anemic when compared to women having normal serum iron levels as shown in Table 2. In contrast to other potential factors, we found that a history of more than one abortion was found to be protective for its association with anemia (OR = 0.16; [95% CI:0.04, 0.62]).
Determinants of anemia among women of reproductive age: Findings of multivariable analysis
The results of multivariable analyses demonstrated that low serum iron level, living in katcha house, no consumption of meat in the past month, and history of breastfeeding increased the risk of anemia (Fig 1). More specifically, significant and positive associations between low serum iron levels (aOR = 7.17; 95% CI [2.94, 17.47]), living in a katcha house (aOR = 6.61; [95% CI: 2.21, 19.87]), no consumption of meat in the past month (aOR = 4.18; [95% CI [1.66, 9.96]) and anemia persisted in the multivariable analysis. Moreover, the association between history of breastfeeding and anemia turned out to be statistically significant in the adjusted model (aOR = 2.43; [95% CI:1.04, 5.72]). A history of more than one abortion (aOR: 0.06; 95% CI (0.01, 0.33) appeared protective for its association with anemia as shown in Fig 1.
Discussion
This was one of the first studies in Pakistan that measured the burden of anemia and its determinants comprehensively across socio-demographic, reproductive, dietary, and biological domains among WRA in rural settings. Our study demonstrated a high prevalence of anemia among WRA in a rural area of Pakistan. Factors such as low serum iron levels, no consumption of meat, living in a katcha house, and history of breastfeeding were found to be risk factors for anemia among WRA. However, women with more than one abortion were found to be at lower risk for anemia. We did not find any differences between anemic and non-anemic women by malarial parasite in the blood and stool parasites. Thus, presence of a malarial parasite or stool parasite did not turn out to be important factors related to anemia in this study. The results on the prevalence and determinants of anemia among Pakistani women should be interpreted in their economic and socio-cultural context.
Our observation regarding the prevalence of anemia is consistent with other studies conducted in different rural areas of Pakistan, where the prevalence of anemia has been found to be between 41.7% to 77.0% [8, 9, 36, 37]. However, it is important to note that a direct comparison of our findings with earlier studies conducted in Pakistan may not be possible because of the differences in the number and composition of the study subjects enrolled, study settings and socio-cultural circumstances.
The association of low serum iron levels with a higher risk of anemia in this study is compatible with the other studies conducted in urban areas of Pakistan and other low-income countries, such as Vietnam and Ethiopia [31, 38–40]. Although studies conducted in urban areas of Pakistan or Ethiopia did not measure serum iron levels in the blood, their results demonstrate that women not consuming iron-rich foods or iron supplements and those with food insecurity are at higher risk of anemia than women who consume iron-rich foods or take iron supplements [31, 38–40]. The consistently positive association between low serum iron levels and anemia may be due to poor diet quality and poor intake of iron-rich foods in our study population. This was further evident from our study’s findings where two-fifths of the women reported no consumption of vegetables and fruits in the prior month and more than two-thirds of the women reported no consumption of meat in the prior month. Thus, inadequate intake of iron, predominantly caused by reduced access to heme iron, which is primarily found in meat and is highly bioavailable, can contribute to iron deficiency [41–43]. Furthermore, Lander et al. [44] reported that four-fifths of rural Pakistani women in the same district (Thatta) were found to consume a diet with inadequate diversity, which might make WRA susceptible to anemia in Thatta district [44]. Dietary intake data from a similar study also indicated rural Pakistani women had the lowest mean intake of calories, protein and key micronutrients [44]. In addition to inadequate dietary diversity, the use of smokeless tobacco, e.g. “gutkha” is common among rural Pakistani women and has been shown to reduce appetite and cause inflammation and thus may have affected the intake of iron-rich foods in our study population [45].
We found a positive association between poor consumption of meat and anemia in our study population and this finding is consistent with other studies, which reported that poor intake of meat is a risk factor of anemia among WRA [46, 47]. In rural Vietnam, the consumption of meat of at least three times a week is more common in non-anemic women than in anemic women [48]. Likewise, anemic women in one of the observational studies reported lower intake of red meat than non-anemic women [40]. This finding can be explained by the fact that poor intake of meat can lead to low levels of iron in the body, thus resulting in anemia. Overall, two types of dietary iron have been found; one is heme iron that comes mainly from animal flesh foods and the other is non-heme iron, which is the only type found in plant-based foods, such as grains and vegetables [49, 50]. In humans, the heme form is the most bioavailable, with an estimated absorption rate of 15%–35% when compared to absorption rate of 2% to 20% for non-heme iron [50, 51].
As expected, women living in a katcha house were more likely to be anemic than their counterparts. Living in a katcha house is one of the proxy indicators of poor socioeconomic status, which is further related to the intake of a poor diet [28]. This relationship between poor socio-economic status and anemia is consistent with other studies conducted in different parts of Asia, Africa, and urban areas of Pakistan [28, 37, 39, 52]. For instance, a study conducted in northern Pakistan revealed that women with a poor socioeconomic status were more likely to be anemic than women with a higher socioeconomic status [37]. Likewise, a study conducted in Bangladesh found that rich women were less likely to be anemic when compared to poor women [28]. Collectively, these findings can be explained by the fact that a poor woman’s purchasing power is meager; therefore, she cannot afford iron-rich food, such as meat and certain iron-rich fruits and vegetables, and is therefore consumes a non-nutritious diet consequently resulting in anemia [53, 54]. Moreover, due to food insecurity, these poor women also tend to consume smokeless tobacco and non-nutritious substances such as fuller’s earth, clay, and ice to reduce their appetite, and consequently, are more susceptible to anemia due to a lack of adequate nutrition [55]. Generally, a woman’s diet also reflects her socioeconomic status. Therefore, living in katcha homes and consuming low meat may all reflect a more marginal living circumstances of a woman, including more risk of poor sanitation and poorer health status (more inflammation as reflected by higher values of inflammatory markers and poorer iron absorption). Thus, diet may well be part of the problem, but it is complicated and interconnected, and it is likely more than just not eating an adequate iron rich diet.
A history of breastfeeding was found to be associated with anemia among WRA, and this finding is consistent with findings from other low-income countries [39, 56]. For example, a study conducted in Ethiopia found that lactating women were at higher risk of anemia when compared with non-lactating women [39]. There is evidence that a lactating woman needs to consume more calories with a more diversified diet than non-lactating women [57]. Studies from Pakistan have reported that women during pregnancy and lactation tend to avoid some foods (due to misconceptions and beliefs) such as beef, eggs, brinjal, fish and citrus fruits as these are considered hot and could have ill effects on their babies [58, 59]. Since the amenorrhea associated with full exclusive breastfeeding is generally assumed a protective factor for iron status; insufficient intake of adequate diet could lead to insufficient milk production and thus shorter exclusive breastfeeding and resumption of menses, which eventually leads to worse iron status and anemia [60].
Contrary to findings reported in the literature [61–63], a history of more than one abortion was found to be protective against anemia in our study; however, this finding should be interpreted with caution. There could be different plausible explanations for this inverse finding. Women with more than one abortion were not able to complete the nine months of pregnancy duration, thus preventing them from physiological anemia caused by pregnancy [64]. This means that these women avoided anemia that might have occurred had they completed their full-term pregnancy, as pregnancies generally result in a reduction in the concentration of red blood cells [65]. Secondly, our study measure regarding the history of abortion was binary without any further details about the abortions. For example, we did not ask about the type of abortion from these women [spontaneous vs induced]. In addition, we did not ask women about how long they had experienced bleeding after abortion, nor did we investigate the volume of blood loss or any clots in the blood during the abortion. This lack of information limits our ability to interpret these results. Lastly, it could be possible that women may not have correctly reported their history of abortion due to the sensitive nature of the issue and stigma associated with abortion, or they might have considered a spontaneous abortion as normal menstrual bleeding. If this reporting were differential across anemic and non-anemic women, it could have led to a higher number of women with less than one abortion among anemic than non-anemic women. Due to these potential limitations for this inverse finding, it is hard to make any conclusion from this protective association between abortion and anemia, and this needs to be explored further with sufficient details about the history of abortion.
Strengths and limitations
This is one of the first studies in rural Pakistan that investigated potential determinants of anemia comprehensively across sociodemographic, reproductive, dietary and biological domains. In addition, the random sampling of women suggests that these data are representative of the overall population in the district. In addition, unlike other studies conducted in a few urban areas, our data regarding biological determinants of anemia were based on objective measurements rather than relying on self-reported data from women. However, there are some inherent limitations. First, the cross-sectional nature of the survey does not allow temporal precedence between various socio-demographic, dietary, reproductive, and biological determinants, and anemia to be unambiguously determined. However, unlike other cross-sectional studies reported in the literature, our study explored the determinants of anemia holistically across socio-demographic, reproductive, dietary, and biological domains. Secondly, our dietary history questionnaire queried only what was eaten in the prior month and may not have fully represented the daily or seasonal dietary patterns of the women. Although the measured dietary history did not represent daily or seasonal dietary patterns, it did indicate monthly dietary patterns of WRA among rural women. A third limitation is that due to the sensitive nature of some questions such as the use of contraception and the history of abortion, women might not have reported the information accurately. However, we tried to overcome this limitation by asking the sensitive questions in privacy and assured participants about the confidentiality of their responses. Lastly, the small sample size of the study may limit our findings to be generalized to the larger group of rural Pakistani women, thus more longitudinal and robust tepidemiological studies with larger sample size are required in the future.
Conclusion
Our study confirmed that anemia is a major public health problem in women of reproductive age in rural Pakistan and a large proportion of women were found to have low levels of serum iron. Lack of dietary iron intake mainly contributes to iron deficiency among poor rural Pakistani women. These findings provide a rationale to take necessary steps aimed at alleviating iron deficiency.
There could be several policy implications for this study in the future. The government should target disadvantageous rural areas and social groups where the purchasing power, knowledge, and access to health care facilities are limited. More specifically, government of Pakistan need to develop some mechanisms to reduce the poverty at the household level by creating job opportunities for the people living in remote areas of Pakistan. One of the ways for poverty alleviation is to ensure good education in the schools both for boys and girls mainly in the rural areas, which are highly reliant on public schools. Secondly, government can provide small loans or introduce social safety-net programs to help people living below the poverty line with proper monitoring of the funds allocated for such poor households.
In addition, there is a need to design combined nutrition and educational strategies to encourage rural women to consume iron-rich foods in their diet. Proactive efforts can be made to educate women through effective campaigns for motivating them to take iron tablets according to the prescribed schedule. Moreover, breastfeeding women should be encouraged to consume extra calories during lactation to prevent themselves from becoming anemic. Similarly, ensuring poor women’s access to information on risks of eating an iron-poor diet and having a less adequate diet during lactation is also important to reduce anemia among women in rural Pakistan.
Acknowledgments
The authors would like to extend their gratitude to all the study participants, for their contribution to this research. More specifically, the authors want to acknowledge all data collectors mainly four field supervisors, Miss Abida Khowaja, Miss Sumaira Fatima, Miss Aziza Khudai and Miss Shagufta Begum who contributed to field supervision.
References
- 1. Milman N. Anemia—still a major health problem in many parts of the world! Annals of hematology. 2011;90(4):369–77. pmid:21221586
- 2.
World Health Organization. Haemoglobin concentrations for the diagnosis of anaemia and assessment of severity. 2011.
- 3.
Cappellini MD, Motta I, editors. Anemia in clinical practice—definition and classification: does hemoglobin change with aging? Seminars in hematology; 2015: Elsevier.
- 4. Beutler E, Waalen JJB. The definition of anemia: what is the lower limit of normal of the blood hemoglobin concentration? 2006;107(5):1747–50. pmid:16189263
- 5. Ouédraogo S, Koura GK, Bodeau-Livinec F, Accrombessi MM, Massougbodji A, Cot M, et al. Maternal anemia in pregnancy: assessing the effect of routine preventive measures in a malaria-endemic area. The American Journal of Tropical Medicine and Hygiene. 2013;88(2):292–300. pmid:23296448
- 6.
World Health Organization.The global prevalence of anaemia in 2011. Geneva: World Health Organization; 2015. 2017.
- 7. Milman N, Taylor CL, Merkel J, Brannon PM. Iron status in pregnant women and women of reproductive age in Europe. The American Journal of Clinical Nutrition. 2017;106(suppl_6):1655S–62S. pmid:29070543
- 8.
Ministry of National Services Regulation and Coordination (MoNHSR&C) GoP. National Nutrition Survey. 2018.
- 9. Baig-Ansari N, Badruddin SH, Karmaliani R, Harris H, Jehan I, Pasha O, et al. Anemia prevalence and risk factors in pregnant women in an urban area of Pakistan. Food and nutrition bulletin. 2008;29(2):132–9. pmid:18693477
- 10. Parks S, Hoffman MK, Goudar SS, Patel A, Saleem S, Ali SA, et al. Maternal anaemia and maternal, fetal, and neonatal outcomes in a prospective cohort study in India and Pakistan. BJOG: An International Journal of Obstetrics & Gynaecology. 2018. pmid:30554474
- 11. Hambidge KM, Bann CM, McClure EM, Westcott JE, Garcés A, Figueroa L, et al. Maternal Characteristics Affect Fetal Growth Response in the Women First Preconception Nutrition Trial. Nutrients. 2019;11(10):2534. pmid:31640153pmid:31640153
- 12. Anjum Anam, Manzoor Maleeha, Manzoor Nadia, Shakir HA. Prevalence of anemia during pregnancy in district Faisalabad, Pakistan. Punjab University Journal of Zoology. 2015;30(1):15–20.
- 13. Habib MA, Raynes-Greenow C, Soofi SB, Ali N, Nausheen S, Ahmed I, et al. Prevalence and determinants of iron deficiency anemia among non-pregnant women of reproductive age in Pakistan. 2018;27(1):195. pmid:29222899
- 14. Ullah I, Zahid Muhammad, Ismail Khan Muhammad S M. Prevalence of anemia in pregnant women in district Karak, Khyber Pakhtunkhwa, Pakistan. International Journal of Biosciences. 2013;3:77–83.
- 15.
National Institute of Population Studies. Pakistan Demographic and Health Survey 2012–13. Isamabad 2013 2017 http://www.nips.org.pk/abstract_files/Priliminary%20Report%20Final.pdf.
- 16. Ali SA, Saleem S, Sami N, Shabbar M, Ahmed M, Rozi S, et al. Geographic access to working family planning centers and unintended pregnancies among married women: a community based nested case control study. Open Journal of Epidemiology. 2016;6(1):95.
- 17. Ahmad M. Health care access and barriers for the physically disabled in rural Punjab, Pakistan. International Journal of Sociology and Social Policy. 2013.
- 18. Goudar SS, Carlo WA, McClure EM, Pasha O, Patel A, Esamai F, et al. The maternal and newborn health registry study of the global network for women's and children's health research. International Journal of Gynaecology and Obstretics. 2012;118(3):190–3. pmid:22738806pmid:22738806
- 19. Hambidge KM, Krebs NF, Westcott JE, Garces A, Goudar SS, Kodkany BS, et al. Preconception maternal nutrition: a multi-site randomized controlled trial. BMC pregnancy childbirth. 2014;14(1):1–16. pmid:24650219pmid:24650219
- 20. Bose CL, Bauserman M, Goldenberg RL, Goudar SS, McClure EM, Pasha O, et al. The Global Network Maternal Newborn Health Registry: a multi-national, community-based registry of pregnancy outcomes. Reproductive Health. 2015;12(S2):S1. pmid:26063166pmid:26063166
- 21. Hambidge KM, Westcott JE, Garcés A, Figueroa L, Goudar SS, Dhaded SM, et al. A multicountry randomized controlled trial of comprehensive maternal nutrition supplementation initiated before conception: the Women First trial. The American journal of clinical nutrition. 2019;109(2):457–69. pmid:30721941
- 22. Waninge A, Ligthart K, Kramer J, Hoeve S, van der Schans C, Haisma H. World Health Organization, BMI classification, 2006. Measuring physical fitness.31(3):38.
- 23. Namaste SM, Aaron GJ, Varadhan R, Peerson JM, Suchdev P. Methodologic approach for the biomarkers reflecting inflammation and nutritional determinants of anemia (BRINDA) project. The American journal of clinical nutrition. 2017;106(suppl_1):333S–47S. 10.3945/ajcn.116.142273 pmid:28615254
- 24. Preet K, Anchinmane VT, Sankhe S. Evaluation of micro-ESR method with Westergren method for determination of erythrocyte sedimentation rate. International Journal of Research in Medical Sciences. 2018;6(2):628–31.
- 25.
Ayong L, Moukoko CEE, Mbacham WF. Diagnosing Malaria: Methods, Tools, and Field Applicability. Malaria Control and Elimination: Springer; 2019. p. 73–82.
- 26. Devanesan S, AlQahtani F, AlSalhi MS, Jeyaprakash K, Masilamani V. Diagnosis of thalassemia using fluorescence spectroscopy, auto-analyzer, and hemoglobin electrophoresis—A prospective study. Journal of infection public health. 2019;12(4):585–90. pmid:30808594
- 27. Amin OM. Evaluation of a new system for the fixation, concentration, and staining of intestinal parasites in fecal specimens, with critical observations on the trichrome stain. Journal of microbiological methods. 2000;39(2):127–32. pmid:10576702
- 28. Kamruzzaman M, Rabbani MG, Saw A, Sayem MA, Hossain MG. Differentials in the prevalence of anemia among non-pregnant, ever-married women in Bangladesh: multilevel logistic regression analysis of data from the 2011 Bangladesh Demographic and Health Survey. BMC women's health. 2015;15(1):54. pmid:26219633pmid:26219633
- 29. Bhandary S, Shrestha A. Prevalence of anemia among non-pregnant and pregnant women of reproductive age in Nepal. Journal of General Practice and Emergency Medicine of Nepal. 2011;2(2):21–6.
- 30. Stevens GA, Finucane MM, De-Regil LM, Paciorek CJ, Flaxman SR, Branca F, et al. Global, regional, and national trends in haemoglobin concentration and prevalence of total and severe anaemia in children and pregnant and non-pregnant women for 1995–2011: a systematic analysis of population-representative data. The Lancet Global Health. 2013;1(1):e16–e25. pmid:25103581
- 31. Habib MA, Raynes-Greenow C, Soofi SB, Ali N, Nausheen S, Ahmed I, et al. Prevalence and determinants of iron deficiency anemia among non-pregnant women of reproductive age in Pakistan. Asia Pacific journal of clinical nutrition. 2018;27(1):195. pmid:29222899
- 32. Trinh LTT, Dibley M. Anaemia in pregnant, postpartum and non pregnant women in Lak district, Daklak province of Vietnam. Asia Pacific journal of clinical nutrition. 2007;16(2). pmid:17468088
- 33.
Collinearity Diagnostics, Model Fit & Variable Contribution.
- 34. Senaviratna N, Cooray T. Diagnosing Multicollinearity of Logistic Regression Model. Asian Journal of Probability and Statistics. 2019:1–9.
- 35.
Hosmer DW Jr, Lemeshow S, Sturdivant RX. Applied logistic regression: John Wiley & Sons; 2013.
- 36. Parks S, Hoffman MK, Goudar SS, Patel A, Saleem S, Ali SA, et al. Maternal anaemia and maternal, fetal, and neonatal outcomes in a prospective cohort study in India and Pakistan. BJOG: An International Journal of Obstetrics & Gynaecology. 2019;126(6):737–43. pmid:30554474
- 37. Ullah I, Zahid M, Khan MI, Shah M. Prevalence of anemia in pregnant women in district Karak, Khyber Pakhtunkhwa, Pakistan. International Journal of Biosciences. 2013;3:77–83.
- 38. Nguyen PH, Gonzalez-Casanova I, Nguyen H, Pham H, Truong T, Nguyen S, et al. Multicausal etiology of anemia among women of reproductive age in Vietnam. European journal of clinical nutrition. 2015;69(1):107–13. pmid:25205323
- 39. Gebremedhin S, Enquselassie F. Correlates of anemia among women of reproductive age in Ethiopia: evidence from Ethiopian DHS 2005. Ethiopian Journal of Health Development. 2011;25(1):22–30.
- 40. Thomson CA, Stanaway JD, Neuhouser ML, Snetselaar LG, Stefanick ML, Arendell L, et al. Nutrient intake and anemia risk in the women's health initiative observational study. Journal of the American Dietetic Association. 2011;111(4):532–41. pmid:21443985
- 41. Bhargava A, Bouis HE, Scrimshaw N. Dietary intakes and socioeconomic factors are associated with the hemoglobin concentration of Bangladeshi women. The Journal of nutrition. 2001;131(3):758–64. pmid:11238756
- 42. Monsen ER. Iron nutrition and absorption: dietary factors which impact iron bioavailability. Journal of the American Dietetic Association. 1988;88(7):786. pmid:3290310
- 43. Liabsuetrakul T. Is international or Asian criteria-based body mass index associated with maternal anaemia, low birthweight, and preterm Births among Thai population?—an observational study. Journal of health, population, nutrition. 2011;29(3):218. pmid:21766557
- 44. Lander RL, Hambidge KM, Westcott JE, Tejeda G, Diba TS, Mastiholi SC, et al. Pregnant Women in Four Low-Middle Income Countries Have a High Prevalence of Inadequate Dietary Intakes That Are Improved by Dietary Diversity. Nutrients. 2019;11(7). pmid:31295916pmid:31295916
- 45. Wabe NT, Mohammed MA. What science says about khat (Catha edulis Forsk)? Overview of chemistry, toxicology and pharmacology. Journal of Experimental & Integrative Medicine. 2012;2(1).
- 46. Al-Sayes F, Gari M, Qusti S, Bagatian N, Abuzenadah A. Prevalence of iron deficiency and iron deficiency anemia among females at university stage. Journal of Medical Laboratory Diagnosis. 2011;2(1):5–11.
- 47. Pasricha S-R, Caruana SR, Phuc TQ, Casey GJ, Jolley D, Kingsland S, et al. Anemia, iron deficiency, meat consumption, and hookworm infection in women of reproductive age in northwest Vietnam. The American journal of tropical medicine hygiene. 2008;78(3):375–81. pmid:18337329
- 48. Pasricha S-R, Caruana SR, Phuc TQ, Casey GJ, Jolley D, Kingsland S, et al. Anemia, iron deficiency, meat consumption, and hookworm infection in women of reproductive age in northwest Vietnam. The American journal of tropical medicine and hygiene. 2008;78(3):375–81. pmid:18337329
- 49. Heath A-LM, Fairweather-Tait S. Clinical implications of changes in the modern diet: iron intake, absorption and status. Best practice research Clinical haematology. 2002;15(2):225–41. pmid:12401305
- 50. Martinez-Torres C, Layrisse M. Iron absorption from veal muscle. The American journal of clinical nutrition. 1971;24(5):531–40. pmid:5578513
- 51. Harvey LJ, Armah CN, Dainty JR, Foxall RJ, Lewis DJ, Langford NJ, et al. Impact of menstrual blood loss and diet on iron deficiency among women in the UK. British Journal of Nutrition. 2005;94(4):557–64. pmid:16197581
- 52. Bentley M, Griffiths P. The burden of anemia among women in India. European journal of clinical nutrition. 2003;57(1):52–60. pmid:12548297
- 53. Ghose B, Tang S, Yaya S, Feng Z. Association between food insecurity and anemia among women of reproductive age. PeerJ. 2016;4:e1945. pmid:27168968
- 54. Harding KL, Aguayo VM, Namirembe G, Webb P. Determinants of anemia among women and children in Nepal and Pakistan: An analysis of recent national survey data. Maternal & child nutrition. 2018;14:e12478. pmid:28857410pmid:28857410
- 55. Young SL, Khalfan SS, Farag TH, Kavle JA, Ali SM, Hajji H, et al. Association of pica with anemia and gastrointestinal distress among pregnant women in Zanzibar, Tanzania. The American journal of tropical medicine and hygiene. 2010;83(1):144–51. pmid:20595493
- 56. Monárrez-Espino J, Martínez H, Greiner T. Iron deficiency anemia in Tarahumara women of reproductive-age in Northern Mexico. salud pública de méxico. 2001;43:392–401. pmid:11763687
- 57.
Medicine Io, Lactation NRCSoND, Lactation IoMSoNd, Lactation IoMSo, Resources USH, Administration S, et al. Nutrition during lactation: Henry Holt and Company; 1991.
- 58. Hassan U. Beliefs of women regarding food during lactation. Pakistan Armed Forces Medical Journal. 2013;63(2):266–70.
- 59. Niloufer Sultan Ali SIA, Noor Rahat. Womens’ beliefs and practices regarding food restrictions during pregnancy and lactation: A Hospital Based Study. Journal of Ayub Medical College Abbottabad. 2004;16(3).
- 60. Dewey KG, Cohen RJ, Brown KH, Rivera LL. Effects of exclusive breastfeeding for four versus six months on maternal nutritional status and infant motor development: results of two randomized trials in Honduras. The Journal of nutrition. 2001;131(2):262–7. pmid:11160544
- 61. Uche-Nwachi E, Odekunle A, Jacinto S, Burnett M, Clapperton M, David Y, et al. Anaemia in pregnancy: associations with parity, abortions and child spacing in primary healthcare clinic attendees in Trinidad and Tobago. African health sciences. 2010;10(1):66. pmid:20811527
- 62. Mardani M, Rezapour S, Ahmadipour S, Mohsenzadeh A, Khalkhali Rad A, Roosta S, et al. Prevalence of anemia and its risk factors among pregnant women in Khorramabad (Iran) 2010–2014. The Journal of Maternal-Fetal & Neonatal Medicine. 2017;30(7):826–9.
- 63. Bahaj AA. Prevalence of and risk factors of anemia among pregnantwomenMukalla-Hadhramout-Yemen. Tikret Journal of Pharmaceutical Sciences. 2006;2(1):71–7.
- 64. Chowdhury S, Rahman M, Moniruddin A. Anemia in pregnancy. Medicine Today. 2014;26(1):49–52.
- 65. Akinlaja OJ. Hematological changes in pregnancy-The preparation for intrapartum blood loss. International Journal of Obstretics and Gynecology 2016;4(3):00109.