Exploring the interplay of maternal anemia and BMI in determining low birth weight among rural Indian children: evidence from NFHS-5

Abstract

Background

Low birth weight (LBW) is a serious global health issue that is indicative of an increased risk of newborn illness and mortality. LBW refers to a newborn weighing under 2,500 g (5.5 pounds) at birth, primarily due to either poor fetal growth, premature labor, or both. It is more prevalent in low and middle-income countries than in high-income countries.

Methods

The entire study has been completed with the help of secondary data collected from the NFHS-5 of India, a cross-sectional national representative survey conducted from 2019 to 2021. In this study, the low birth weight of children is considered as an outcome variable. Based on previous research, various socio-economic and demographic variables were selected in this study. This study used Pearson’s chi-square statistics and multivariable binary logistic regression to identify the association between low birth weight of children and maternal anemia and BMI (Body Mass Index). And descriptive statistics is also performed to analyze the data.

Results

The result shows that there is significant association between the maternal anemia and BMI with the occurrence of low birth weight. It is found that the occurrence of low birth weight is higher among severely anemic women and underweight women. Additionally, the prevalence of LBW is high among those women who use cigarettes and tobacco, and those children of multiparity, are female children, and are from central, western, and northern regions.

Conclusion

So, it can be said that maternal anaemia and BMI is correlated with the prevalence of low birth weight among children. Hence, implementing appropriate interventions for pregnancy care and ensuring well-planned nutrition can significantly alleviate the challenges associated with LBW in India.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s41043-026-01381-z.

Background

Low birth weight (LBW) is a serious global health issue that is indicative of an increased risk of newborn illness and mortality [1]. Low birth weight refers to a newborn weighing under 2,500 g (5.5 pounds) at birth, primarily due to either poor fetal growth, premature labor, or both [1]. It is more prevalent in low and middle-income countries than in high-income countries [1]. Globally, the incidence of low birth weight affects approximately 20 million newborns, with a prevalence ranging from 15% to 20% out of all live births [2]. Sustainable Development Goals (SDG) 3.2 has targeted to reduce under-five mortality rate (U5MR) to 25 or less per 1,000 live births and neonatal mortality rate (NMR) to 12 or less per 1,000 live births worldwide [3]. Despite efforts to improve maternal and child health, the prevalence of LBW is still alarmingly high in India. According to the NFHS-5 report, 18.2% of babies were born with LBW, which is consistent with the estimation from the prior round [4]. The likelihood of infant mortality is significantly greater among newborns with low birth weight compared to those born with a higher birth weight [5].

Identifying the root causes of LBW is crucial for improving neonatal health outcomes and lowering infant death rates. LBW is frequently associated with various factors such as inadequate antenatal care, nutritional deficiencies during pregnancy, and different socio-economic conditions [6]. Lack of nutrition during pregnancy is the most significant environmental determinant contributing to adverse pregnancy outcomes [7]. Body mass index (BMI) and anaemia are the two key indicators of maternal nutrition, which can significantly affect the health of the mother and the fetus [8]. Indian women face significant challenges regarding their nutritional and reproductive health [9]. According to the NFHS-5 report, 18.7% of women in India are underweight, reflecting a 4% decrease compared to the previous round. Additionally, 24% of women are classified as overweight, marking an 8.5% increase from the previous estimation [4]. Furthermore, the prevalence of anaemia among pregnant women stands at 57%, representing a 4% increase over the prior estimate in India [4]. Adolescence is a time of considerable changes in body composition and shape, with 15% of adult height and 50% of adult body weight acquired physiologically [10]. For this reason, it is crucial to maintain enough nutrition throughout this time.

Earlier studies revealed that young girls who became pregnant within two years of menarche, had short birth intervals, and were malnourished were more likely to deliver babies with LBW [7, 11]. Another study conducted in Bangladesh showed that LBW was more common among women who suffered from maternal anemia and had a low BMI [12]. Women who were malnourished, with low BMI and short height were more likely to had unfavorable delivery outcomes, particularly for LBW [13]. Another study found that maternal anaemia and low BMI are linked to undernutrition, which subsequently leads to the delivery of LBW babies [14]. Previous research in Ghana identified that maternal anemia in first trimester and third trimester of gestational age and low BMI are the crucial factors for LBW babies [15, 16]. Findings from Ethiopian studies also revealed maternal anemia and BMI are the two important determinants of LBW. The results found that women with normal hemoglobin levels and normal BMI were less likely to have LBW babies [17–19]. The same evidence was found in research carried out in Nigeria [20], Nepal [21], and Bangladesh [22].

Previous studies conducted in India had also shown that women’s nutritional status was significantly associated with adverse delivery outcomes like LBW [23–28]. Women who suffered from persistent undernutrition in their youth continue to experience nutritional deficiencies into adulthood, leading to a higher risk of delivering newborns with LBW [29]. Another study carried out in rural areas of western Maharashtra found a significant association between maternal anaemia, height, and weight, and an increased likelihood of delivering LBW babies [30]. So, the reviewed literature highlighted that maternal anaemia and BMI are the most critical factors contributing to the occurrence of LBW babies globally. Additionally, the prevalence of LBW, along with the percentage of women experiencing anemia and low BMI, is notably higher in rural areas of India, exceeding the national average [4]. Despite the existence of several studies examining the relationship between maternal anaemia, BMI, and low birth weight (LBW) in India, limited research has focused exclusively on rural populations. Furthermore, while previous NFHS-based studies have established the independent effects of maternal BMI and anaemia, their combined influence has received relatively little attention, particularly in rural India. Given the disproportionately higher burden of nutritional deprivation and LBW in rural areas, it is important to examine the role of maternal nutritional status in shaping birth outcomes. Therefore, this study aims to investigate the interplay between maternal BMI and anaemia using the most recent NFHS-5 (2019–21) data. It is hypothesised that maternal nutritional disadvantage, reflected through both low BMI and anaemia, has a compounded effect on the risk of low birth weight among rural Indian children. The findings are expected to provide context-specific evidence to inform targeted policy interventions for reducing LBW in rural India.

Methodology

Study design and sample

The study-required data were collected from the NFHS-5 (National Family Health Survey) of India, a cross-sectional national representative survey conducted during 2019–2021. The NFHS-5 (2019–2021) was conducted under the MoHFW (Ministry of Health and Family Welfare) of India, with coordination by IIPS (International Institute of Population Science), Mumbai. NFHS-5 was conducted across the nation using a two-stage stratified cluster sampling technique, which covers 28 states and 8 union territories. This survey collected the information from 6,36,699 households with a 98% of response rate, whereas men (aged 15–54 years) respondents were 1,01,839 and women (aged 15–49 years) respondents were 7,24,115, with a 92% and 97% of response rates, respectively. NFHS-5 provided information about various indicators of maternal and child health up to the district level (707) across all states and union territories. The NFHS-5 report of India provided more details about the sampling methodology [4].

Study participants

The total number of children (aged 0 to 59 months) who have been born in the last five years is 232,920. Out of this, 141,224 children were excluded. Among them, 8702 children died, 45,884 children were excluded because they lived in urban areas, excluding those children whose mothers had previous birth histories (n = 43,172), 1st of multiple, 2nd of multiple, and 3rd of multiple children were excluded (n = 1023), and also excluded those children for whom birth weight data were not collected (n = 12,287) (Fig. 1). Children whose mothers’ anaemia and BMI data were not collected or were missing (n = 4,041) were excluded. Lastly drops those children where covariate data was missing (n = 26,115), i.e. caste (n = 6,644), age at marriage (n = 939), religion (n = 39), glucose level (n = 28), sign of pregnancy complication (n = 5,434), ANC visits (n = 1,249), TT injections (n = 406), IFA tablets (n = 11,378). Finally, 91,696 children were selected for this study (Fig. 1).

Fig. 1.

Flow diagram showing sample selection in this study from NFHS-5, 2019-21, India

The included sample contains observations with comprehensive information on birth weight, mother BMI, anaemia status, and relevant covariates, making it suitable for accurate statistical analysis. In contrast, excluded cases are primarily characterised by missing data and incomplete records. Exclusion of dead children and those with missing data regarding birth weight underestimates LBW prevalence. Although it’s important for analytical accuracy, these exclusions may introduce selection bias.

Outcome variable

In this study, the low birth weight of children is considered as an outcome variable. Low birth weight is defined by WHO (World Health Organization) as a weight at birth of less than 2.5 kg or 2500 g (1,4). NFHS-5 collected the birth weight using the following questions: Was (name of the child) weighed at birth? How much did (name of the child) weight? The information was collected in two ways: either from mothers’ recall or from written records of health cards. For analysis, children are assigned code ‘0’ if their birth weight is 2.5 kg or more and code ‘1’ if their birth weight is less than 2.5 kg.

Predictor variables

Maternal anaemia and BMI are the key predictor variables in the study. Anaemia is defined by the haemoglobin concentrations of blood, and haemoglobin concentration was measured onsite with a battery-operated portable HemoCute Hb 201 + analyser. Also, in enumeration area altitudes above 1000 m, haemoglobin levels were adjusted. Therefore, blood samples for anaemia testing were gathered by a microcuvette from finger pricks. According to the WHO standard non-pregnant women are considered mildly anaemic if their haemoglobin level is 11.0–11.9 g/deciliter which is indicated by code ‘1’, moderately anaemic if haemoglobin is level 8.0–10.9 g/dl assigned here by code ‘2’, severely anaemic if haemoglobin level < 8.0 g/dl which indicated by code ‘3’ and not anaemic if haemoglobin level > 12.0 g/dl that assigned by code ‘4’.

Maternal BMI (Body Mass Index) has derived from a person’s weight in kilograms, divided by height in metres squared (kg/m²). The height is measured by the Seca 213 Stadiometer and the weight is measured by the Seca 874 Digital Scale.

After calculating BMI, the value is classified into three groups based on WHO standards. According to the WHO, women are considered underweight if their BMI value is less than 18.5 kg/m², Normal and overweight if their BMI value is 18.5–24.9 kg/m² and greater than 25.0 kg/m² respectively. For empirical analysis, underweight women are assigned code ‘1’, code ‘2’ for normal women, and code ‘3’ for overweight women.

Other covariates

Based on previous research, other covariates were selected in this study. Other covariates are sex of the child, maternal age, any sign of pregnancy complication, age at marriage, birth order, education level, cigarettes and tobacco use, media exposure, religion, caste, number of ANC visits, number of TT injections taken, number of IFA tablets consumed, received full ANC, wealth index and region. Detailed information about other covariates is given in Table 1.

Table 1.

Description of other covariates included in this study, NFHS-5, 2019–2021

Variables	Description and codes
Sex of child	Coded as 1 = Male, 2 = Female
Maternal age	Mother’s age is categorised into 4 classes and coded as 1 = 15–19 years, 2 = 20–29 years, 3 = 30–39 years, 4 = 40–49 years.
Pregnancy complications	Whether women had any signs of complications during pregnancy. Coded as 0 = No, 1 = Yes
Blood glucose level	Women’s blood glucose levels are categorised into 3 groups and coded as 1 = Normal (< 141 mg/dl), 2 = High (141–160 mg/dl), 3 = Very high (> 160 mg/dl)
Marriage	Age at marriage is classified into 2 groups and coded as 1 = < 18 years, 2 = ≥ 18 years
Birth order	Coded as 1 = First, 2 = Second, 3 = Third, 4 = Four or more
Education level	Mother’s educational levels are classified into 4 categories and coded as 1 = No education, 2 = Primary, 3 = Secondary, 4 = Higher
Use cigarettes and tobacco.	Whether women ever used cigarettes and tobacco, and coded as 0 = Yes, 1 = No
Media exposure	Media exposure has been measured by frequently watching TV, listening to the radio, and reading newspapers and magazines. Then classified into 3 groups and coded as 1 = Low, 2 = Partial, 3 = High
Religion	Religion has been categorized into 4 groups and coded as 1 = Hindu, 2 = Muslim, 3 = Christian, 4 = Others
Caste	Caste has been classified into 4 categories and coded as 1 = Scheduled Caste, 2 = Scheduled Tribe, 3 = Other Backward Classes, 4 = Others
Number of ANC visits	During pregnancy, how many times visit the antenatal care and coded as 0 = < 4, 1 = ≥ 4
Number of TT injections taken	During pregnancy, how many tetanus injections were taken and coded as 0 = < 2, 1 = ≥ 2
Number of IFA tablets consumed	During pregnancy, how many IFA tablets were consumed and coded as 0 = < 100, 1 = ≥ 100
Received full ANC	Whether women during pregnancy receive at least 4 ANC visits, two TT injections taken, and 100 IFA tablets consumed as defined full ANC and coded as 0 = No, 1 = Yes
Wealth Index	Economic condition (measures based on housing assets, flooring materials, sanitation facilities, and other goods, i.e. TVs to bicycles, or cars) of respondents grouped into 5 classes and coded as 1 = Poorest, 2 = Poorer, 3 = Middle, 4 = Richer, 5 = Richest
Region	The whole geographical region of India has been divided into 6 regions and coded as 1 = South, 2 = Central, 3 = East, 4 = Northeast, 5 = West, 6 = North

Statistical analysis

This study used Pearson’s chi-square statistics and multivariable binary logistic regression to identify the association between low birth weight of children and maternal anemia and BMI. Descriptive statistics is used to estimate the frequency and percentage distribution of respondents included in this study. This distribution is shown by various socio-demographic characteristics and two key predictor variables. Here is also the percentage distribution of outcome variables (low birth weight of children) by key predictors (maternal anemia and BMI) and other covariates. Sample weight is used for percentage calculation. Pearson’s chi-square statistic is used to test differences in low birth weight among children using selected explanatory variables, and the results are considered significant if the p-value is < 0.05. A multivariable binary logistic regression model is used to estimate the crude odds ratio (COR) and adjusted odds ratio (AOR), with 95% confidence intervals (CIs); results are considered significant at p < 0.05. To investigate the combined impact of maternal BMI and anemia status on low birth weight, a composite variable was developed. Adjusted odds ratios (AORs) with 95% confidence intervals were calculated using multivariable logistic regression, with normal BMI and non-anaemic mothers serving as the reference group. Additionally, the interaction between BMI and anaemia was also tested and is reported in the supplementary material (Supplementary Table 1). All statistical analyses are performed using Stata version 12.8 (StataCorp LP, College Station, TX, USA), and ArcMap version 10.8 is used for mapping.

Results

Characteristics of the respondents

Children born with low birth weight (LBW) were 14,584 (16.8%) out of 91,696 (Table 2). Based on BMI values, underweight and overweight mothers were 18,351 (21%) and 15,154 (16.9%), respectively. Nearly one in every three women was moderately anaemic (31.2%), and 2.2% of women were severely anaemic. Male and female child was 54.1% and 45.9%, respectively. Women aged less than 20 years and between 40 years were 3.4% and 1.5%, respectively. Nearly four in every five women had any sign of pregnancy complications, and 2.9% of women had high glucose levels. Nearly 35% of women were married before they reached 18 years, and 14.4% of women were more educated. High media exposure women were nearly 6%, and three in every five women had visited 4 or more ANC. Nearly one in every two women consumed ≥ 100 IFA tablets, and nearly one in every three women had received full ANC.

Table 2.

Descriptive statistics of the study participants (n = 91696) & distribution of low-birth-weight children among rural women (15–49 years) by explanatory variables in India, 2019-21

Variables	Frequency (n)	Percentage (%)	Low birth weight (%)	95% CI	p-value
Birth weight (grams)
≥ 2500	77,112	83.1	-	-	-
< 2500	14,584	16.8
Body mass index
Normal	58,191	61.9	15.3	15.10-15.68	< 0.001
Underweight	18,351	21	20	19.51–20.67
Overweight	15,154	16.9	12.8	12.31–13.37
Maternal anemia
Not anemic	36,939	39.7	15	14.65–15.38	< 0.001
Mild	24,221	26.8	16.1	15.64–16.57
Moderate	28,437	31.2	16.6	16.17–17.04
Severe	2099	2.2	20	18.40-21.83
Sex of child
Male	49,381	54.1	14.7	14.46–15.08	< 0.001
Female	42,315	45.9	17.2	16.87–17.59
Women’s age (Years)
15–19	2664	3.4	21.7	20.25–23.38	< 0.001
20–29	63,537	71.6	16.3	16.06–16.64
30–39	23,699	23.4	14.1	13.75–14.64
40–49	1796	1.5	14.2	12.71–15.95
Any sign of pregnancy complications
No	18,182	20.9	15.4	14.97–16.03	< 0.001
Yes	73,514	79	16	15.74–16.27
Glucose level
Normal	87,755	95.5	15.9	15.69–16.17	< 0.001
High	2654	2.9	15	13.76–16.48
Very high	1287	1.4	15.8	13.96–17.95
Age at marriage
< 18 years	29,454	35.1	16.8	16.40-17.26	< 0.001
≥ 18 years	62,242	64.8	15.4	15.19–15.75
Birth order
1	32,257	35.3	17.5	17.18–18.01	< 0.001
2	33,801	36.6	15	14.69–15.47
3	15,343	16.5	14.9	14.40-15.53
4 or more	11,295	11.4	14.7	14.15–15.46
Education level
No education	17,423	18.6	18	17.50-18.65	< 0.001
Primary	11,268	11.9	17.7	17.02–18.43
Secondary	50,859	54.9	15.6	15.31–15.94
Higher	12,146	14.4	12.3	11.73–12.90
Use cigarettes and tobacco
No	5332	3.4	15.8	14.86–16.81	< 0.001
Yes	86,364	96.5	15.9	15.67–16.16
Media exposure
Low	24,522	26.8	18.5	18.03-19.00	< 0.001
Partial	61,290	66.9	15	14.76–15.33
High	5884	6.2	14	13.16–14.93
Religion
Hindu	73,919	85.7	16.6	16.37–16.91	< 0.001
Muslim	7813	9.6	15.1	14.34–15.93
Christian	6140	2.1	8.6	7.94–9.34
Others	3824	2.4	15	13.96–16.23
Caste
SC	19,655	21.4	17.9	17.44–18.51	< 0.001
ST	20,513	22.3	14.2	13.73–14.69
OBC	36,962	40.3	15.8	15.51–16.25
Others	14,566	15.8	15.5	15.01–16.19
Number of ANC visits
< 4	33,539	36.7	17	16.62–17.42	< 0.001
≥ 4	58,157	63.2	15.2	14.97–15.56
Number of TT injections taken
< 2	13,655	14.7	17.2	16.59–17.85	< 0.001
≥ 2	78,041	85.2	15.6	15.42–15.93
Number of IFA tablets consumed
< 100	46,666	51.1	16.7	16.42–17.10	< 0.001
≥ 100	45,030	48.8	15	14.69–15.35
Received full ANC
No	62,550	67.8	16.1	15.09–17.24	< 0.001
Yes	29,146	32.2	15.9	15.65–16.14
Wealth quintile
Poorest	24,529	25.3	18.6	18.13–19.10	< 0.001
Poorer	24,139	25.6	16.2	15.78–16.71
Middle	20,156	22.8	14.6	14.14–15.12
Richer	14,728	17.2	13.7	13.24–14.35
Richest	8144	8.9	13.7	13.05–14.54
Region
South	12,036	17.4	14.1	13.49–14.73	< 0.001
Central	25,088	29.2	18.3	17.89–18.84
East	18,192	25.5	16.2	15.71–16.78
Northeast	11,687	3.3	10.3	9.81–10.92
West	7960	11	17.9	17.10-18.79
North	16,733	13.2	16	15.53–16.64

p-value: Significance level of chi-square statistic

Table 2 represents the percentage distribution of low-birth-weight children among rural women (15–49 years) by body mass index, maternal anaemia, and socio-economic characteristics. The women experiencing underweight were more likely to have low birth weight children (20%) than normal women. Overweight women were less likely to have LBW children (12.8%) compared to normal women. Severe anaemic (20%) mothers were more likely to have LBW children compared to non-anaemic mothers. Among female children (17.2%), LBW were more than among male children. The decreasing trend of LBW children was observed with increasing women’s age, in which the age group 15–19 years (21.7%) had more LBW children compared to women aged 40 years or older. LBW children were higher among the women who married before 18 years (16.8%). The prevalence of LBW was decreasing with increasing birth order. The percentage of LBW decreased with increased maternal educational level and family wealth status; among the higher educated women (12.3%), the prevalence of LBW was less compared to no formally educated women (18%). The proportion of LBW children was higher among the poorest women compared to those who were the richest (18.6% vs. 13.7%). The prevalence of LBW was higher among the women who had low media exposure compared to those who had high media exposure (18.5% vs. 14%). The percentage of LBW children was high among those women who received < 4 ANC visits (17%), took < 2 TT injections (17.2%), consumed < 100 IFA tablets (16.7%) and did not receive full ANC (16.1%). The prevalence of LBW children was common among women who belonged to the scheduled caste (17.9%) and the Hindu (16.6%) religion. The prevalence of LBW children was also unequally distributed across the geographical region. A higher percentage of LBW babies was observed in the central region (18.3%), followed by the west region (17.9%), east region (16.2%), and north region (16%).

Maps 1a, b, c and d, represent the district-level prevalence patterns of low birth weight, maternal anaemia, and underweight and overweight, respectively, in the rural Indian context. A higher prevalence of LBW exceeding 20% is mostly observed in Punjab, Haryana, West Bengal, western Uttar Pradesh, and northwestern Madhya Pradesh. A lower prevalence of LBW (below 11.6%) is observed in most parts of Rajasthan, Gujarat, Ladakh, Himachal Pradesh, Arunachal Pradesh, Manipur, Mizoram, and Nagaland. Higher prevalence of maternal anaemia (above 58.2%) is observed in the eastern part of India e.g. West Bengal, Bihar, Jharkhand, and Odisha. It is also observed in Assam, Tripura, Ladakh, Haryana, Western Punjab, Eastern Gujarat and Andhra Pradesh. Lower prevalence of maternal anaemia (below 40.9%) is mostly found in Arunachal Pradesh, Nagaland, Manipur, Mizoram, western Rajasthan, eastern Uttarakhand and Kerala. The prevalence of underweight is found to be high (above 25%) in most part of Bihar and Jharkhand. Lower prevalence of underweight (less than 12.9%) is observed in most part of Arunachal Pradesh, Nagaland, Manipur, Meghalaya, Mizoram, Jammu and Kashmir, Ladakh, Himachal Pradesh, Punjab and Kerala. Higher Prevalence of overweight (above 29.2%) is found in most part of Punjab, Sikkim, Kerala and Tamil Nadu. On the other hand, lower prevalence (below 14.8%) is mostly observed in Rajasthan, Madhya Pradesh, Chhattisgarh, Bihar, Jharkhand, Maharashtra, Meghalaya, Assam and Nagaland.

Map. 1.

District-level patterns of (a) prevalence of low birth weight, (b) prevalence of maternal anaemia, (c) prevalence of underweight among rural women, (d) prevalence of overweight among rural women in India, 2019-21

Table 3 represents the association between LBW children with maternal anaemia and maternal body mass index and other covariates by odds ratio from the multivariable binary logistic regression model in rural India. It was found that among the underweight mothers (AOR = 1.26; 95% CI = 1.20–1.31), the prevalence of LBW babies is likely more than that of normal mothers. Women who were overweight (AOR = 0.89; 95% CI = 0.84–0.94) among them, the chances of LBW babies were likely less compared to normal women. Chances of LBW babies among the severely anaemic mothers (AOR = 1.29; 95% CI = 1.15–1.44) were likely higher compared to mothers who were not anaemic. The occurrences of LBW babies were more among female children (AOR = 1.20; 95% CI = 1.16–1.25) than male children, and as the women’s age increased, the chances of LBW babies were likely reduced. Among women aged 30–39 years (AOR = 0.80; 95% CI = 0.72–0.90) with LBW babies was likely lower than that of women aged 15–19 years. Birth order and the mother’s educational level were negatively associated with LBW babies. Among the higher educated women (AOR = 0.70; 95% CI = 0.65–0.76) risk of LBW children was less compared to those women who had not received formal education. The risk of LBW babies was likely more among those women who used cigarettes and tobacco (AOR = 1.15; 95% CI = 1.06–1.24). Occurrences of LBW children were possibly less among those women who were highly exposed (AOR = 0.90; 95% CI = 0.82–0.98) to mass media than those women who had low media exposure. Women who belonged to the scheduled tribe (AOR = 0.80; 95% CI = 0.75–0.85) and OBC (AOR = 0.90; 95% CI = 0.85–0.94) had a lower risk of LBW babies than women who belonged to the scheduled caste. Risk of LBW children was likely low among women who received ≥ 4 ANC visits (AOR = 0.91; 95% CI = 0.87–0.96), took ≥ 2 TT injections (AOR = 0.87; 95% CI = 0.83–0.92) and consumed ≥ 100 IFA tablets (AOR = 0.92; 95% CI = 0.88–0.98). The Wealth Index was negatively associated with LBW babies. The chances of LBW children among the richest women (AOR = 0.77; 95% CI = 0.70–0.84) were likely much less than those of the poorest women. As compared to the south region, the probability of LBW babies among women who live in the west region (AOR = 1.18; 95% CI = 1.09–1.28) was higher, followed by the central region (AOR = 1.15; 95% CI = 1.07–1.22).

Table 3.

Binary logistic regression models for low birth weight of children among rural women (15–49 years) in India, 2019-21

Variables	Unadjusted		Adjusted
	Odds ratio	95% CI	Odds ratio	95% CI
Body mass index (BMI)
Normal®	1.00		1.00
Underweight	1.38***	1.32–1.44	1.26***	1.20–1.31
Overweight	0.80***	0.76–0.85	0.89***	0.84–0.94
Maternal anaemia
Not anemic®	1.00		1.00
Mild	1.08***	1.03–1.13	1.04	0.99–1.09
Moderate	1.12***	1.08–1.17	1.05*	1.01–1.10
Severe	1.42***	1.27–1.58	1.29***	1.15–1.44
Sex of child
Male®	1.00		1.00
Female	1.20***	1.16–1.24	1.20***	1.16–1.25
Women’s age (Years)
15–19®	1.00		1.00
20–29	0.70***	0.63–0.77	0.82***	0.74–0.91
30–39	0.59***	0.53–0.65	0.80***	0.72–0.90
40–49	0.59***	0.50–0.70	0.82	0.69-098
Any sign of pregnancy complications
No®	1.00		1.00
Yes	1.03	0.99–1.08	1.02	0.98–1.07
Glucose level
Normal®	1.00		1.00
High	0.93	0.84–1.04	1.00	0.89–1.11
Very high	0.99	0.85–1.15	1.09	0.96–1.27
Age at marriage
< 18 years®	1.00		1.00
≥ 18 years	0.90***	0.87–0.93	1.00	0.96–1.04
Birth order
1®	1.00		1.00
2	0.83	0.79–0.86	0.79***	0.76–0.83
3	0.82	0.78–0.86	0.72***	0.68–0.76
4 or more	0.81***	0.76–0.86	0.66***	0.61–0.71
Education level
No education® 1.00			1.00
Primary	0.97	0.91–1.03	1.03	0.96–1.09
Secondary	0.83***	0.80–0.87	0.90***	0.85–0.95
Higher	0.63***	0.59–0.68	0.70***	0.65–0.76
Use cigarettes and tobacco
No®	1.00		1.00
Yes	1.00	0.93–1.08	1.15***	1.06–1.24
Media exposure
Low®	1.00		1.00
Partial	0.78***	0.75–0.81	0.90***	0.86–0.94
High	0.71***	0.66–0.77	0.90*	0.82–0.98
Religion
Hindu®	1.00		1.00
Muslim	0.89***	0.83–0.95	0.90**	0.84–0.97
Christian	0.47***	0.43–0.51	0.72***	0.65–0.80
Others	0.88	0.81–0.97	1.04	0.94–1.15
Caste
SC®	1.00		1.00
ST	0.75***	0.71–0.79	0.80***	0.75–0.85
OBC	0.86***	0.82–0.90	0.90***	0.85–0.94
Others	0.84***	0.79–0.89	0.97	0.91–1.03
Number of ANC visits
< 4®	1.00		1.00
≥ 4	0.87***	0.84–0.91	0.91***	0.87–0.96
Number of TT injections taken
< 2®	1.00		1.00
≥ 2	0.89***	0.85–0.93	0.87***	0.83–0.92
Number of IFA tablets consumed
< 100®	1.00		1.00
≥ 100	0.87***	0.84–0.90	0.92**	0.88–0.98
Received full ANC
Yes®	1.00		1.00
No	0.98	0.90–1.06	0.97	0.90–1.04
Wealth index
Poorest®	1.00		1.00
Poorer	0.84***	0.80–0.88	0.88***	0.84–0.93
Middle	0.74***	0.71–0.78	0.79***	0.75–0.84
Richer	0.69***	0.66–0.74	0.75***	0.70–0.80
Richest	0.69***	0.65–0.75	0.77***	0.70–0.84
Region
South®	1.00		1.00
Central	1.37***	1.28–1.45	1.15***	1.07–1.22
East	1.18***	1.10–1.26	0.92**	0.86–0.99
North East	0.70***	0.65–0.76	0.68***	0.62–0.74
West	1.33***	1.23–1.43	1.18***	1.09–1.28
North	1.16***	1.09–1.24	1.09**	1.01–1.17

^®Reference category; *p < 0.05; **p < 0.01; ***p < 0.001; CI: Confidence Interval, n = 91,696

The interaction between maternal BMI and anaemia was examined; however, the interaction terms were not statistically significant, indicating no evidence that the association between BMI and low birth weight differed according to anaemia status. Nevertheless, analyses of combined BMI–anaemia categories revealed differences in the odds of low birth weight across exposure groups. Please see the Supplementary Table 1: Statistical interaction effects of maternal BMI and anaemia on low birth weight in rural India, 2019-21.

Table 4 shows the combined effects of Maternal BMI and anaemia on low birth weight in rural India, 2019–2021. The weighted prevalence of low birth weight was highest among underweight anaemic mothers (21%), followed by underweight non-anaemic mothers (20.1%), compared with 16.1% among mothers with normal BMI and no anaemia (reference category). Consequently, the gross differential in low birth weight prevalence was positive among anaemic mothers with underweight and normal BMI, indicating a higher prevalence relative to the reference group. In contrast, overweight non-anaemic (13.6%) and overweight anaemic mothers (14%) exhibited lower prevalence levels, resulting in negative gross differential values.

Table 4.

Combined exposure categories of maternal BMI and anaemia on low birth weight in rural India, 2019-21

Maternal BMI × anaemia status	LBW (weighted %)	Gross differential (%)	Crude odd ratio	95% CI	Adjusted odds ratio (AOR)	95% CI
Normal BMI + Not anaemic (Ref.)	16.1	-	1.00	-	1.00	—
Underweight + Not anaemic	20.1	4	1.41***	1.31–1.52	1.28***	1.19–1.37
Underweight + Anaemic	21	4.9	1.50***	1.42–1.59	1.34***	1.27–1.42
Normal BMI + Anaemic	16.6	0.5	1.11***	1.06–1.16	1.07**	1.02–1.12
Overweight + Not anaemic	13.6	-2.5	0.83***	0.77–0.90	0.91*	0.84–0.99
Overweight + Anaemic	14	-2.1	0.88**	0.82–0.95	0.94	0.87–1.01

Gross differential (%) represents the absolute difference in the prevalence of low birth weight between each maternal BMI–anaemia category and the reference category (Normal BMI + Not anaemic); Adjusted for maternal age, age at marriage, glucose level, education, wealth index, caste, religion, media exposure, region, birth order, ANC visits, iron–folic acid intake, tetanus toxoid injection, full ANC visits, use of cigarettes and tobacco, any sign of pregnancy complication, and sex of child; Ref. Reference category; *p < 0.05; **p < 0.01; ***p < 0.001; CI: Confidence Interval

The regression results further show that underweight mothers, particularly those who were anemic, had far greater risks of low birth weight than mothers with normal BMI and no anemia. Anaemia alone also increased the risk among mothers with normal BMI. In addition to this, overweight women without anaemia showed lower odds of LBW compared with women of normal BMI and anaemia. However, this finding represents an observed association rather than evidence of a protective effect or causal relationship.

Discussion

The present study shows the impact of maternal anaemia and BMI on low birth weight among the rural children of India with the help of a nationally representative survey, i.e. NFHS-5 (2019–2021). This study also shows the impact of different socio-demographic, maternal anthropometric, and economic determinants of utilization of maternal health care services on the low birth weight among the most recent child. One of the main causes of newborn illness and mortality is low birth weight. Low birth weight is a growing concern in many nations, including India, as it can lead to major health problems in adulthood [1]. Several factors affect the likelihood of giving birth to a child weighing less than 2,500 g. Maternal anthropometry is one such aspect connected to the outcome of pregnancies [27, 31]. Birth weight is influenced by a number of variables, including the genetic makeup of the mothers, sociocultural, socioeconomic, and behavioural aspects, as well as the pre-pregnancy body mass index (BMI), gestational weight gain (GWG), and others [23, 25, 28, 32, 33]. Low birth weight (LBW) is more common in Asia than in other regions, primarily due to undernutrition of the mother both before and during the pregnancy [24, 34]. The dietary needs for early and late pregnancy differ qualitatively; early pregnancy requires micronutrients and proteins to support organogenesis, placental development, and fetal cell differentiation, whereas later pregnancy requires calories and other nutrients to meet the demands of rapid fetal growth, maternal tissue expansion, and energy storage for lactation [26, 31, 35, 36]. It has been demonstrated that micronutrient deficits during pregnancy have detrimental effects on the growing fetus. In Asia, India has the greatest rate of maternal fatalities and the highest frequency of anaemia, with nearly half of pregnant women still suffering from varied degrees of the condition [12, 37, 38]. Compliance with iron supplements, cultural perspectives on pregnant eating, and the problem of nutrition supplementation and fortification are of particular relevance [10, 13, 14, 29]. Maternal anaemia and maternal BMI are crucial elements of maternal anthropometry that directly affect the pregnancy outcomes and delivery outcomes, as well as newborn health outcomes, i.e. type of birth (preterm birth and term birth) and birth weight (normal and low) [30, 34, 36, 38]. Regardless of stature, the current investigation verified a robust correlation between low birth weight (LBW) and maternal underweight, as underweight women had a 1.38-fold higher likelihood of having LBW kids than the group as a whole. Many previous studies showed that there is a high chance of giving birth to anaemic newborn babies whose mothers are also anaemic during delivery [22, 27, 39–41]. In the present study, it is also found that the occurrence of low birth weight is more (1.42 times) among the severe anaemic women than the mild and moderate anaemic women. Many previous studies also found that the occurrence of low birth weight is more common among women with severe anaemia and low BMI worldwide [5, 12, 17, 20, 22].

Adequate utilization of antenatal care services can significantly reduce the chances of LBW among rural women. Prior studies conducted in India reported that inadequate utilization of ANC services is a significant predictor of LBW. Women who had not taken 4 or more ANC services were more likely to have LBW babies [26, 28]. Regular health check-ups and practicing healthy habits help women identify different complications, including LBW. Moreover, receiving ≥ 2 TT injections and consumption of ≥ 100 IFA tablets were also associated with low odds of LBW children. Another study conducted in Ghana revealed that women who never consumed 100 IFA tablets had 3 times higher odds of having LBW babies [42]. Different socio-demographic and economic factors are also significantly associated with LBW of children. It is found that the women who had completed their higher education, belonged to the richer wealth quintile, were fully exposed to mass media, and had not engaged in different smoking activities, had a lower occurrence of low birth weight than the rest of the women [10, 27, 28, 37, 39, 40]. This study’s regional heterogeneity in LBW suggests that the burden was not evenly dispersed throughout rural India. The increased frequency in central, western, and northern areas may be explained by interregional variations in maternal nutrition, education, poverty, and access to high-quality prenatal care. According to earlier research conducted in India and other low- and middle-income nations, maternal undernutrition and anemia can biologically affect placental function and fetal development, raising the risk of LBW [12, 17–19, 23–28].

The findings of this study remain relevant in light of the recently released NFHS-6 (2023–24) fact sheets. NFHS-6 indicates that 19.7% of women aged 15–49 years are underweight and 30.7% are overweight or obese, highlighting the continued burden of maternal nutritional challenges in India. Although maternal healthcare utilization has improved, with higher coverage of antenatal care (65.2%) and iron–folic acid supplementation for 100 or more days (54.9%) than in NFHS-5 [43], maternal undernutrition persists. These findings support the continued importance of maternal BMI and anaemia as determinants of adverse birth outcomes, including low birth weight. Furthermore, as unit-level NFHS-6 data are not yet publicly available, NFHS-5 remains the most recent nationally representative dataset suitable for examining the combined effects of maternal BMI and anaemia on low birth weight in rural India.

There are certain limitations associated with this study. Firstly, its cross-sectional design prevents the establishment of causal relationships between the outcome variable and various predictors. Secondly, information on birth weight was obtained either from written records or recalled by the mother, which introduces the potential for recall bias. Thirdly, Maternal BMI and haemoglobin were measured at the time of survey rather than during pregnancy, and some women may have been pregnant at the time of haemoglobin measurement, which may lead to misclassification of anaemia status. Fourthly, in contrast, excluded cases are primarily characterised by missing data and incomplete records. Exclusion of dead children and those with missing data regarding birth weight underestimates LBW prevalence. Although it’s important for analytical accuracy, these exclusions may introduce selection bias. Furthermore, important determinants of birth weight, including gestational age (preterm birth) and maternal height, were unavailable and could not be included in the analysis. As these factors are known to influence fetal growth and birth outcomes, residual confounding cannot be ruled out. Therefore, the observed associations between maternal BMI, anaemia, and LBW should be interpreted with caution. To gain a clearer understanding of the factors contributing to the high prevalence of low birth weight (LBW) among rural Indian children, future research should employ longitudinal data. Nevertheless, the study highlights important maternal factors such as BMI and anaemia status, which are significantly linked to LBW. These findings can support policymakers in developing targeted interventions to help reduce the incidence of LBW in India.

Conclusion

This study concludes that the rate of LBW remains high and the predominance has been on the rise for India over the past decade. Women who are underweight had expanded chances of having LBW babies; however, the chances were measurably noteworthy only for rural India. The rise in the burden of LBW in India serves as a sign of poor/inadequate usage of national well-being arrangements to achieve the SDGs focused on improving child well-being results (SDG 3). This merits extraordinary investigative consideration, digging into the basic causes of the rise in LBW predominance, and calls for utilizing a stronger policy plan to reverse the situation. A commonly proposed technique to anticipate maternal and childbirth-related complications is to require early safeguard by giving vital care for pregnant moms through antenatal care services. Coordination and the arrangement of supplements/nutritious nourishment programs could benefit the mothers as well as newborn babies.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

LBW

Low Birth Weight

BMI

Body Mass Index

NFHS

National Family Health Survey

Author contributions

Mithun Sarkar: Methodology; Software; Writing - original draft. Koyel Majumder: Conceptualization; Data curation; Investigation; Writing - original draft, Writing - review & editing. Bikash Barman: Data curation; Formal analysis. Pradip Chouhan: Conceptualization; Supervision; Writing - review & editing.

Funding

This research did not receive any sort of grant from any agency, non-profit organization, or commercial entity.

Data availability

The data has been derived from DHS (Demographic Health Survey), which is publicly available. The data set is accessible through this link via DHS. ( [https://dhsprogram.com/data/available-datasets.cfm](https:/dhsprogram.com/data/available-datasets.cfm) ).

Declarations

Ethics approval and consent to participate

The NFHS-5 survey was approved by the relevant ethics committees, and written informed consent was obtained from all participants. This study is a secondary analysis of anonymized NFHS-5 data obtained from the DHS Program with appropriate permission.

Competing interests

The authors declare no competing interests.

Human Ethics and Consent to Participate declarations

Not applicable.