Introduction
Iron deficiency is one of the most leading causes of disability in developing as well as developed countries. About 90,000 maternal and neonatal deaths were due to iron deficiency anaemia in developing regions in 2013 (WHO, 2014). According to the World Health Organization, in 2015 more than 50% of people were suffering from anaemia in the South East Asia region, and about half of cases were due to iron deficiency (WHO, 2015). Though the primary cause of anaemia is iron deficiency, among pre-school children it is also caused by other factors such as vitamin A and vitamin B12 deficiencies, and hookworm and malaria infections (Pasricha et al., Reference Pasricha, Black, Muthayya, Shet, Bhat and Nagaraj2010; Dey et al., Reference Dey, Gosawmi and Dey2013). Iron deficiency anaemia has many harmful effects among pre-school children. For example, it reduces learning capacity, decreases attentiveness and causes low intelligence (Katzman et al., Reference Katzman, Novack and Pearson1972; Pablo et al., Reference Pablo, Windom and Pearson1985; Behrman et al., Reference Behrman, Kleigman and Nelson1994; Agaoglu et al., Reference Agaoglu, Torun, Unuvar, Sefil and Demir2007; Zhao et al., Reference Zhao, Zhang, Peng, Li, Yang, Liu and Wang2012). Anaemia among children has also been indirectly linked to the developmental stage of a region, as is evident from region-wise data (WHO, 2008).
Among the many risk factors for death, iron deficiency anaemia used to be regarded as one of the ‘Top Ten Risk Factors’ (Dubey, Reference Dubey, Sachdev and Choudhury1994). Iron deficiency anaemia is still very common in South Asian countries, including India, Bangladesh and Pakistan. It causes many serious problems, especially among pregnant and lactating women, children and adolescents (Lokeshwar et al., Reference Lokeshwar, Mehta, Mehta, Shelke and Babar2011). During the last decade, India was the largest contributor to child anaemia among developing countries (WHO, 2008; Pasricha et al., Reference Pasricha, Black, Muthayya, Shet, Bhat and Nagaraj2010). In 2007, 69.5% of 6- to 59-month-old children in India were anaemic, of which 26.3% were mildly anaemic, 40.2% moderately anaemic and 2.9% severely anaemic (IIPS & ICF, 2007). The present paper aimed to assess the anaemia status of this age group of children using data from the most recent National Family Health Survey (IIPS & ICF, 2017), carried out in 2015–16. The objectives of the study were to assess: (i) the distribution of anaemia prevalence by child age group, (ii) the prevalence of child anaemia by zone and state and (iii) the relation of child anaemia prevalence with social, demographic and economic variables, including maternal nutritional status and low birth weight.
Methods
Data
Unit-level cross-sectional data were obtained from the National Family Health Survey (NFHS-4), conducted in India by the International Institute for Population Sciences in 2015–16. Data for 1,37,347 children aged 6–59 months living in 29 states and seven Union Territories were analysed. The child’s anaemia status was the dependent variable. This was measured through the level of haemoglobin in the blood. Consent was obtained from a parent, or an adult member of the household responsible for the child, for collection of a blood sample from the youngest 6- to 59-month-old child in the household. A drop of blood was taken from a finger prick (or sometimes a heel prick) and collected in a micro-cuvette. Haemoglobin analysis was conducted on-site with a battery-operated portable Hemo CueHb 201+ analyser (IIPS and ICF, 2017). Levels of anaemia severity were as follows: mild (10.0–10.9 g/dl), moderate (7.0–9.9 g/dl) and severe (<7.0 g/dl).
The independent variables were type of place of residence (urban or rural), education of mother (illiterate, primary [(classes I–IV], secondary [classes V–XII] and higher than class XII), religious group (Hindu, Muslim, Christian and Other) and wealth index. The supplied NFHS-4 data contained the wealth index, which was calculated from the ownership scores for consumer goods such as TV, car and refrigerator. These scores were then combined to group households into quintiles, namely ‘Poorest’, ‘Poorer’, ‘Middle’, ‘Richer’ and ‘Richest’, including approximately 20% of the sample population in each group. These five categories were further modified and then categorized into three groups (Poor, Medium and Rich) by pooling ‘Poorest’ and ‘Poorer’ households into the ‘Poor’ category, and ‘Richer’ and ‘Richest’ households into the ‘Rich’ category. The middle category was kept unchanged.
Child’s present health status was measured through z-score values, i.e. z-scores of the two nutritional indices weight-for-age and height-for-age. The z-score is defined as the deviation of the value observed for an individual from the median of the reference population, divided by the standard deviation (SD) of the reference population, i.e.:
$$z {\vskip.3pt\hyphen} {\rm{score}} = {\frac{{{\rm{Observed}}\,{\rm{value}} - {\rm{Median}}\,{\rm{of}}\,{\rm{the}}\,{\rm{reference}}\,{\rm{population}}}}\over{{{\rm{SD}}\,{\rm{of}}\,{\rm{the}}\,{\rm{reference}}\,{\rm{population}}}}$$
The classifications of the z-score (following NCHS/WHO) were below normal (<–2), normal (–2 to <+2) and above average (≥+2). Child’s birth weight was classified into three categories, namely: very low (<1500 g), low (<2500 g) and normal (2500 g or above). Mother’s nutritional status was measured through BMI (WHO, 1998), and classified into four categories: underweight (BMI<18.5), normal (BMI = 18.5–24.99), overweight (BMI = 25.0–29.99) and obese (BMI≥30.0). Anaemia was grouped into anaemic (<11.0 g/dl) and non-anaemic (≥11.0 g/dl).
Analysis
The sample children were divided into six groups by age (in months) to assess variation in anaemia rates by age group. The percentages of mild, moderate and severe anaemic children by zone and state were also computed. These were summed to give a total prevalence of anaemic children. In addition, the percentages of non-anaemic children were calculated. The relationships between the anaemia status of the child and socioeconomic and demographic variables (place of residence, mother’s educational level, religion, wealth index, nutritional status of the child and mother, birth weight of the child and anaemia status of the mother) were assessed using bivariate analysis and the application of chi-squared tests.
The combined effects of the socioeconomic and demographic variables were assessed by carrying out binary logistic regression. All explanatory variables were taken as categorical. Anaemic condition was considered as the dependent variable, with anaemic children being coded ‘1’ and the rest as ‘0’. The coefficient of each explanatory variable, along with its significance, determined how the variable influenced the level of anaemia. A positive coefficient implied a positive relationship. More specifically, an estimated odds ratio (OR) >1 (which is equivalent to the value of the coefficient being >0) indicated that the probability of becoming anaemic was more in this category compared with the reference category of the explanatory variable. If it was <1, then it was opposite to that of the ‘>1’ case. An estimated OR of ‘1’ indicates that the nature of the dependent variable was not different from that of the reference category. All calculations were carried out using SPSS (Statistical Package for Social Science) version 18.0. Significance levels of p < 0.01 and p < 0.05 were considered.
Results
Table 1 shows the distribution of anaemia prevalence in the sample children by age group, zone and state. At the all-India level, the percentage of anaemic children increased from 65.9% in the age group 6–8 months to 68.1% in the age group 12–23 months, then decreased to 42.7% in the older 48- to 59-month age group. At the regional level, this pattern was more or less the same, with only slight variations by state and zone. The requirement for iron decreases after the first year of life due to a decrease in growth rate. Still, iron intake during the second year might not always be adequate (Stekel, Reference Stekel and Stekel1984; FAO/WHO, 1988). A balanced diet should contain the right amount of iron. A poor diet may contain insufficient iron, and this might be a reason for the increasing prevalence of anaemia in this age group.
Table 1. Prevalence of anaemia in 6- to 59-month-old children in India by age group, 2015–16 (NFHS-4)
a Dadra & Nagar Haveli.
b Andaman & Nicobar Islands.
Table 2 shows the distribution of anaemia prevalence among the sample children by anaemia severity, zone and state. Overall, 56.3% of the children were anaemic in 2015–16; 1.5% were severely anaemic, 27.6% were moderately anaemic and 27.2% were mildly anaemic. The North-east region was the least affected (34.6%) and the Central region the most affected (62.3%). According to the NFHS-3, in 2005–06 the percentage of anaemic children in North-east India was 55.6%. It is surprising how much this zone improved within a decade. The Central zone had 71.9% anaemic children, so this had improved by only 10 percentage points over the same period. The only two states with less than 25% anaemic children in 2015–16 were Mizoram (22.4%) and Nagaland (24.8%). Both are in the North-east zone. The states with an anaemia rate over 60% were Bihar (62.2%), Jharkhand (68.8%), Madhya Pradesh (68.4%), Uttar Pradesh (63.5%), Gujarat (63.0%), Dadra & Nagar Haveli (84.7%), Daman & Diu (70.6%), Haryana (70.5%) and Chandigarh (73.5%). The Union Territories of Dadra & Nagar Haveli and Daman & Diu had a very high prevalence of anaemia among pre-school children. The reason for this is not clear, but it might be due to an improper diet (Zhang et al., Reference Zhang, Qin, Liu, Li, Li and Varma2018), childhood illness such as diarrhoea (Tariq et al., Reference Tariq, Kamal, Reddy, Bajantri, Niazi and Matela2017), worm infestations (Seidelman et al., Reference Seidelman, Zuo, Udayakumar and Gellad2016) or lack of appropriate iron supplementation (Okam et al., Reference Okam, Koch and Tran2017).
Table 2. Prevalence of anaemia among 6- to 59-month-old children in India by anaemia severity, zone and state, 2015–16 (NFHS-4)
aDadra & Nagar Haveli. bAndaman & Nicobar Island.
The relationship between child anaemia prevalence and different socioeconomic, child health and maternal health variables is presented in Table 3. Rural children were more anaemic than urban children (p < 0.001). Mother’s education has a significant role in reducing anaemia among the children: 71.7% of the children of illiterate mothers were anaemic while only 49.9% of those of higher educated mother were affected by anaemia. Hindu and Muslim children were highly affected (57.9% and 57.8% respectively). Christians were the least affected (35.3%) (p < 0.001). Wealth affected the incidence of anaemia: the level among the poorest group was 60.0%, while that among the richest group was 52.4% (p < 0.001). Children’s nutritional status and birth weight both had an inverse relationship with the children’s level of anaemia (p < 0.001). In other words, the better the health status and higher the birth weight of the children, the lower the prevalence of anaemia. Similarly, the better the health status and lower the level of anaemia of the mothers, the lower the prevalence of anaemia in the children.
Table 3. Relationship between anaemia prevalence among 6- to 59-month-old children in India and household socioeconomic conditions and child and mother health factors
Table 4 shows five binary logistic regressions models assessing the differential impact of socioeconomic and demographic variables on children’s anaemia status. Model 1 included the socioeconomic conditions of the household, Model 2 the socioeconomic conditions of the household and child nutritional status, Model 3 the socioeconomic conditions of the household and mother’s health, Model 4 only included child’s nutritional status and mother’s health, and Model 5 included socioeconomic conditions of the household, mother’s health and child’s nutritional status. Model 5 included all the variables. No marked differences in the direction of influence were observed among the five models. Place of residence did not show any significance in any of the models. All other variables had significant effects in the expected direction. When all the variables were taken together in Model 5, the wealth index ceased to have a significant effect on the children’s anaemia status, though the direction remained the same as in the other models.
Table 4. Categorical logistic regression analysis of anaemia prevalence among 6- to 59-month-old children in India by household socioeconomic conditions and child and mother health factors
See text for explanation of Models 1–5. Ref.: reference category.
**p < 0.01; *p < 0.05.
Discussion
This study attempted to provide an overview of the most recent situation for the anaemic condition of 6- to 59-month-old children in India and to assess the influence of the socioeconomic conditions of the mother and children’s health status. The most common cause of anaemia is iron deficiency, but it may occur for other reasons such as malabsorption of the nutrients in the diet, chronic disease (e.g. cancers such as leukaemia, lymphoma and multiple myeloma), radiation, blood loss, drugs and medications (antibiotics and medicines for rheumatoid arthritis) and infections such as malaria and hepatitis. It may also be an inherited disease, such as thalassaemia or sickle cell disease (NIH, 2011, WHO, 2014; Wikipedia). Iron is necessary for effective immune system function. It is also needed by pathogenic organisms such as non-typhoid Salmonella. Thus, iron is needed by both the host and pathogen and therefore in some cases iron in the body may have an adverse effect (Doherty, Reference Doherty2007).
From the analysis of NFHS-4 data, the study found that the percentage of anaemia among children in India was 65.9% at 6–8 months, increased to 68.1% at 12–23 months then decreasing to 42.7% at 48–59 months. The trend was the same over different zones and states, but varied slightly in magnitude. The probable cause of this increasing then decreasing trend is that, from the 6th to 12th months, iron requirements increase with body weight as the infant’s weight triples by the end of the first year of life. For this, 30% of iron is required from food (Osorio, Reference Osorio2002). If this supplementation is not fulfilled, a child in this age group has a high chance of being anaemic. After 2 years, as growth becomes stable and the requirement for iron becomes less, the percentage of anaemic children goes down. The requirement for absorbed iron is 0.96 mg for age 3 months to 1 year, 0.61 mg between 1 and 2 years and 0.70 mg from 2 to 6 years (FAO/WHO, 1988). If these requirements are not fulfilled by food, then the child may become anaemic.
Iron supplementation is necessary for child growth and brain development in the age group 6–24 months. Earlier studies have shown that iron-deficient children have low IQ scores, less attentiveness and lower scores on academic IQ tests (More et al., Reference More, Shivkumar, Gangane and Shende2013). Iron deficiency anaemia is caused by many factors. Also, poor brain development is associated with multiple factors such as low socioeconomic status, poverty, lack of stimulation at home, poor parental education, maternal depression, low birth weight, faulty feeding, malnutrition, parasitic infection. Many studies have shown that iron deficiency in infancy may be associated with impaired cognitive function during the school years (Lozolff et al., Reference Lozoff, Brittenham, Wolf, McClish, Kuhnert and Jimenez1987).
The prevalence of anaemia in 6- to 59-month-old children by different levels of severity gave a clearer picture. The prevalence of severe anaemia was 1.5%, and those of moderate and mild anaemia were 27.6% and 27.2%, respectively. Overall in India, 56.3% of 6- to 59-month-old children were anaemic in 2015–16. Compared with the previous national survey conducted in 2005–06 (NFHS-3), this had decreased by 13.5 percentage points. The North-east zone was found to be the least affected (34.6%) and the Central zone the most affected (62.3%) by anaemia. According to the NFHS-3, the prevalence of anaemia in children in the North-east region in 2005–06 was 55.6%. Thus, it had reduced by 20 percentage points over a 10-year period. The Central zone had 71.9% anaemic children during NFHS-3. This improved by 10 percentage points over the decade to 2015–16. This is a commendable achievement.
Different programmes have been initiated by the Government of India for the reduction of anaemia among children over recent years. The National Nutritional Anaemia Prophylaxis Programme (NNAPP), introduced by the Ministry of Health and Family Welfare, is one such programme. Vitamin A and IFA supplementation to children was done through this programme. Other important programmes include the National Nutritional Anaemia Control Programme (NNACP) and the Integrated Child Development Services (ICDS) Scheme. However, it is doubtful whether these programmes were fully implemented, because it was found that many of the intended beneficiaries were not even aware of all the programmes. The supply of Iron/Folic Acid (IFA) tablets was lower than expected. The IFA tablets given to children below 3 years of age were not easily acceptable by them, so liquid IFA was recommended for young children (Gragnolati et al., Reference Gragnolati, Shekar, Gupta, Bredenkamp and Lee2005). In 2007, the Government of India modified the policy and replaced tablets with liquid IFA. Young children aged 6–59 months then received 1 ml IFA syrup for 100 days in a year containing 20 mg elemental iron and 100 mg folic acid. It seems likely that the reduction in anaemia among children seen over the study decade was due to these Government policies.
Rural children were found to be more anaemic than urban children. The relationship between anaemia and different socioeconomic variables was profound. Mother’s education was found to have a very significant role in the reduction in child anaemia as 71.7% of the children of illiterate mothers were anaemic compared with only 49.9% of the children of higher educated mothers. Hindu and Muslim children were more highly affected (57.9% and 57.8% respectively) than Christians (35.3%). Among the children of poor households, anaemia was 60.0%, whereas among the rich households it was only 52.4%.
The binary logistic regression analysis found no marked differences in the direction of influence among the five models. There is no marked difference in anaemia status between rural and urban children. Mother’s education has a profound effect on reducing the incidence of anaemia: the higher the level of education of a mother, the lower the incidence of anaemia among children. Christian children are uniformly in a much better position. Wealth index has some effect in reducing anaemia and the effect is most conspicuous among the richest households. The health status of children has a negative effect on the level of anaemia. Mother’s nutritional status is also very much effective. Non-anaemic mothers have fewer children with anaemia than anaemic mothers.
This study shows that, besides iron supplementation, the socioeconomic conditions of households need to be improved to reduce child anaemia in India. Low birth weight and low maternal nutritional status were also found to be responsible for the high occurrences of anaemia among children in India.
Funding
This research received no specific grant from any funding agency, commercial entity or not-for-profit organization.
Conflicts of Interest
The authors have no conflicts of interest to declare.
Ethical Approval
The study data were taken from IIPS, who had received ethical clearance to conduct the survey.
