Sage Journals: Discover world-class research

Abstract

Background:

This study assessed the trend of minimum acceptable dietary practices among children aged 6 to 23 months in India in the past decade.

Methods:

Data collected in the National Family Health Survey during 2005 to 2006 (NFHS-3) and 2015 to 2016 (NFHS-4) were used. The sample size for this study was 11 727 children for NFHS-3 and 61 158 children for NFHS-4. Bivariate and multivariate analyses were done to identify the predictors of feeding practices. We analyzed data of last-born singleton children aged 6 to 23 months who were living with their mother (ever-married women aged 15-49 years).

Results:

In India, a low proportion (10%) of children aged 6 to 23 months received a minimum acceptable diet, and its estimate remained the same in the past 10 years. Older mothers, educated mothers, antenatal care, high economic status, and place of residence were the most consistent predictors of minimum acceptable dietary intake. Over the past decade, however, the minimum acceptable dietary intake has declined among older children (odds ratio [OR] = 0.57, 95% confidence interval [CI], 0.47-0.70; P < .001), children with overweight mothers (OR = 0.40, 95% CI, 0.31-0·50; P <.001), and children of well-off families (OR = 0.29, 95% CI, 0.23-0.37; P < .001). The odds for minimum acceptable dietary intake were significantly higher among children of high (4+) birth order (OR = 1.74, 95% CI, 1.41-2.15; P < .001), rural areas, and children living in the central, eastern, and north-eastern region.

Conclusion:

Results indicate the need for further investigations of poor dietary intake among children living in well-off families and in the northern region. This study suggests policymakers and program managers to address contextual barriers against minimum acceptable dietary intake among children and to reconsider existing strategies in India.

Keywords

India infant and young child feeding dietary diversity dietary adequacy

Introduction

The World Health Organization (WHO) and UNICEF have developed the Global Strategy for Infant and Young Child Feeding (IYCF), which recognizes appropriate infant feeding practices that are crucial to improve nutritional status and decrease infant mortality in all countries.¹ However, for people living in low-resource setting—across the globe—affording a diverse diet remains a challenge; main staple foods, such as cereals and starches, usually dominate their daily diet.²

Dietary inadequacy is a proximate and critical indicator of malnutrition and is associated with illness, poor health, and mortality in infants and young children aged less than 2 years.³ Malnutrition is a major concern in India. On the one hand, 38% of Indian children—younger than 5 years—are stunted, 29% are wasted, and 36% are underweight, while on the other hand, 2% are overweight.^4
-6 Nutrition-related factors are responsible for about 35% of child deaths and 11% of the total global burden of disease.⁷ Other than at individual level, factors at household and community levels such as household food security, income, occupation, and educational status of caregivers, especially of mothers, have an immediate bearing on the child’s nutritional intake.^8

-11

Data from the most recent round of the National Family Health Survey (NFHS) conducted in 2015 to 2016 reveal that the majority of mothers in India are not following the recommended level of IYCF practices, including the timely introduction of solid food,^6,12 which might result in a slow rate of improvement in prevalence of stunting from 52% in 1993 to 38% in 2016.^6,13 Other, population-based studies around the world showed that the highest risk of nutritional deficiency and growth retardation occurs in children between the ages of 3 and 15 months because of poor breast-feeding and complementary feeding practices.¹⁴ Complementary foods are often inadequate in nutritional quality, or they are given too early or too late, in too small amounts, or not frequently enough.¹² It is estimated that 6% of under-5 deaths could be prevented through the achievement of universal coverage of improved dietary practices, especially complementary feeding.¹⁵

Children need age-specific adequate nutrition of acceptable quality and quantity to prevent malnutrition.¹² So far, nutrition programs in India have extensively focused on improving feeding practices and access to supplementary nutrition of pregnant and lactating mothers and children younger than 6 years, but the adequacy of nutritional intake remains a concern¹⁶; therefore, periodic evaluation of dietary intake among children is necessary.

Very few studies have estimated the percentage of adequately fed children, aged 6 to 23 months, based on IYCF indicators and identified the determinants of appropriate feeding.^17

-20 Children of this age-group undergo a transition from breast-feeding to complementary feeding, and Indian children, of poor families in particular, often experience the growth failure due to a combination of frequent illness and an inadequate or imbalanced diet.²¹ An estimation of the levels and determinants of dietary adequacy for children of this age-group could provide valuable information for health policymakers and program managers.^22
-24

To understand the factors associated with dietary intake among children aged 6 to 23 months, the unit-level data of the NFHS enable researchers to estimate feeding practices in children aged 6 to 23 months using a nationally representative sample of children of that age-group. Therefore, this study aimed to estimate the prevalence of minimum acceptable diet (MAD) and to identify its associated factors in India for young children aged 6 to 23 months.

Methods

Data Sources

The present analysis used the data of the latest 2 rounds of the NFHS: NFHS-3 and NFHS-4 conducted in 2005 to 2006 and 2015 to 2016, respectively, which are the valid sources of information for determining complementary feeding indicators from a nationally representative sample of households.^6,25 In NFHS-3, 109 041 sampled households were visited, and 124 385 women aged 15 to 49 years were interviewed. In NFHS-4, 601 509 sampled households were visited, and 699 686 women aged 15 to 49 years were interviewed.

The analysis was restricted to the last-born singleton children who were alive, aged 6 to 23 months at the time of the survey, and who were living with the mother (ever-married women aged 15-49 years). We excluded the children with multiple births from the analysis because we assumed that the raising twins or triplet will put more stress in the family and that might change the feeding practice. The final unweighted sample size for this study was 11 727 children for NFHS-3 and 61 158 children for NFHS-4. The reason behind the difference in the sample size between NFHS-3 and NFHS-4 is that NFHS-4 assessed various indicators at the district level, while NFHS-3 estimated the indicators only up to the state level. The data from women interview provided information regarding the background characteristics of the mother and child. The household data provided information on socioeconomic characteristics of the household of the child.

Outcome Variable

Intake of MAD

We considered the intake of MAD of children aged 6 to 23 months as the dependent variable in this study. We used the MAD, recommended by the WHO (2010), to assess the intake of MAD for the children. The survey collected data on dietary intake from the mother of the child by a 24-hour recall.

A composite indicator—the proportion of children aged 6 to 23 months who received an MAD in addition to breast milk—was calculated from 2 separate indicators: children receiving minimum dietary diversity and children receiving minimum meal frequency during the previous day. The definitions of these 2 indicators are as follows:

Children receiving minimum dietary diversity: the proportion of children 6 to 23 months of age who received foods from 4 or more food groups.

Children receiving minimum meal frequency: the proportion of breastfed and non-breastfed children 6 to 23 months of age, who received solid, semisolid, or soft foods (but also including milk feeds for non-breastfed children) the minimum number of times or more. (Minimum is defined as 2 times for breastfed infants 6-8 months, 3 times for breastfed children 9-23 months, and 4 times for non-breastfed children 6-23 months).

This study assessed the intake of MAD using the food groups given in Box 1; the data on dietary intake of these food groups were collected in both NFHS-3 and NFHS-4. Since in NFHS-3, the dietary intake data were collected only among breast-fed children; therefore, in this study, we estimated MAD only among breast-fed children to make the estimations comparable form both the survey rounds. A child receiving MAD in the last 24 hours, appropriate for their age, was coded as “1” and otherwise was coded as “0.” Indicators for assessing infant and young child feeding practices, asked in both rounds of the NFHS (NFHS-3 and NFHS-4), are shown in Box 1.²⁶

Box 1.

Type of Foods Classified in Food Groups, Asked in Both Rounds of NFHS (NFHS-3 and NFHS-4)

Food Groups	Food Items
Starchy staples	Any bread, roti, chapati, rice, noodles, biscuits, idli, or any other foods made from grains Any other solid, semisolid, or soft food Any white potatoes, white yams, manioc, cassava, or any other foods made from roots Infant formula Any commercially fortified baby food, for example, Cerelac or Farex
Legumes and nuts	Any foods made from beans, peas, lentils, or nuts
Dairy products	Milk such as tinned, powdered, or fresh animal milk Yogurt Drink anything from a bottle with a nipple
Flesh foods	Any liver, kidney, heart or other organ meat Any chickens, duck, or other birds Any other meat Any fresh or dried fish or shellfish
Eggs	Eggs
Vitamin A–rich fruits and vegetables	Any pumpkin, carrots, squash, or sweet potatoes that are yellow or orange inside Any dark green, leafy vegetables Any ripe mangoes, papayas, cantaloupe, or jackfruit
Other fruits and vegetables	Any other fruits or vegetables Juice or juice drinks

Explanatory Variables

This study considered 22 explanatory variables based on a review of the literature. Box 2 shows each of these explanatory variables and the categories used in this analysis.

Box 2.

List of Explanatory Variables and Their Categories Used in this Study

	Name of the Variable	Categories
1.	Sex of the child	Male, female
2.	Age of the child in months	6-11, 12-17, 18-23
3.	Birth order of the child	1, 2-3, 4+
5.	Child has anemia	Yes, no
6.	Mother’s age in years	15-24, 25-29, 30-49
7.	Mother’s education	No education, primary, secondary, and above
8.	Mother’s exposure to mass media at least once a week	No exposure, at least 1 media, 2 or 3 media
9.	Mother’s daily consumption of animal source foods	Yes, no
10.	Mother’s daily consumption of pulses/nuts/seeds	Yes, no
11.	Mother’s daily consumption of fruits and vegetables	Yes, no
13.	Mother’s body mass index	Underweight, normal, overweight
14.	Mother received complete antenatal care (ANC)	Yes, no
15.	Place of delivery	Institutional, home
16.	Type of residence	Urban, rural
17.	Religion of the household	Hindu, non-Hindu
18.	Cast/tribe of household	Scheduled caste/tribe, other
19.	Wealth index of the household	Poorest, poorer, middle, richer, richest
20.	Interaction with health worker during the past 3 months	Yes, no
21.	Received benefits from integrated child development scheme in the past 12 months	Yes, no
22.	Geographical region of India	North, central, east, northeast, west, south

Statistical Analysis

The outcome indicators were expressed as dichotomous variables, and each variable was coded as “0” for not meeting the indicator criterion and as “1” for meeting the criterion (1 for receiving MAD and “0” for not receiving MAD). Each outcome indicator was examined against a set of explanatory variables (individual, parental, household, health care, and community-level characteristics) to assess the prevalence of the outcome indicator among the women of each of the categories under explanatory variables.

Multiple logistic regression was conducted using stepwise backward elimination of variables. The stepwise regression prevents “overfitting” of the model. We preferred the backward method because the forward method produces “suppressor effects.” The backward regression analysis started with the full least squares model containing all the predictors. The analysis then iteratively removed the least useful predictor, one-at-a-time, until when all remaining variables had a statistically significant (P value <.05) contribution to the model. This analytical approach keeps most of the predictors in the model, which are significantly associated with the outcome variable.

Three logistic regression models were calculated. Models 1 and 2 identified the determinants of MAD in the 2005 to 2006 and 2015 to 2016, respectively. Model 3 identified the determinants of intake of MAD, considering the changes in the determinant variables from 2005-2006 to 2015-2016. Adding interaction terms to a regression model can greatly expand understanding of the relationships among the variables in the model and allows more hypotheses to be tested. The effect of changes in determinants on dietary intake was calculated by creating interaction variables for each of the determinants with a time factor. In that variable, the responses from NFHS-3 were coded as “0” and from NFHS-4 as “1.” Our time interaction regression models take the form of:

Y = β_{0} + δ_{0} * t + β_{1} * x_{1} + \dots + β_{n} * x_{n} + δ_{1} (x_{1} * t) + \dots + δ_{n} (x_{n} * t) + ∊

where

Y is the MAD prevalence rate, t is the time, β₀ is a vector of coefficients, β_n is a vector of explanatory variables, δ_n is a vector of interaction terms of time factor t, and ∊ is an error term.

Adjusted odds ratios (AORs) showed the strength and direction of the associations between independent variables and MAD. Appropriate sampling weights were applied to maximize the representativeness of the sample. All data analyses were done using SPSS 18.0 software.

Ethical consideration

This article is based on secondary analysis of the NFHS data, available in the public domain with all identifying information removed. The NFHS received ethical clearance from the International Institute for Population Science’s Ethical Review Board from where this survey was conducted. The survey personnel obtained informed consent from each respondent before asking questions.

Results

Profile of the Children

Table 1 shows the sample distribution of children aged 6 to 23 months by selected background characteristics during NFHS-3 and NFHS-4. The results of the t test show that the proportion of children by sex, age, and birth order was similar during both rounds. Prevalence of anemia among children had significantly declined for the past 10 years. Mothers’ educational level improved in the past decade; a higher percentage of mothers had secondary and above education in NFHS-4 compared to NFHS-3. Mothers’ exposure to mass media (reading the newspaper, listening to the radio, or watching television) increased since NFHS-3. Daily consumption of animal source foods among mothers increased from NFHS-3 to NFHS-4, while at the same period, daily consumption of pulses/nuts/seeds or fruits and vegetables declined. The proportion of underweight mothers declined since 2005 to 2006, whereas the proportion of overweight mothers increased. The proportion of mothers who received antenatal care (ANC), had institutional delivery, met health workers in past 3 months, and received benefits from ICDS in the last 12 months preceding the survey, increased from 2005-2006 to 2015-2016. The proportion of women living in urban areas increased from NFHS-3 to NFHS-4. More than half of the children aged 6 to 23 years were from central and eastern regions of the country in both rounds of the NFHS.

Table 1.

Sample Distribution of Children Aged 6 to 23 Months by Selected Background Characteristics, 2015 to 2016, India.^a

Characteristics	NFHS-3, 2005-2006		NFHS-4, 2015-2016		Tests of Independence (χ² Value)
Characteristics	Percentage	Number (Unweighted)	Percentage	Number (Unweighted)	Tests of Independence (χ² Value)
Sex of child
Male	53.1	6177	52.8	32 321	0.26
Female	46.9	5550	47.2	28 837
Age of child in months
6-11	38.1	4578	38.7	23 497	1.75
12-17	34.0	4021	33.7	20 776
18-23	27.9	3128	27.6	16 885
Birth order
First	30.2	3707	36.7	21 863	1126.1^f
2-3	44.0	5262	49.3	29 619
4 or more	25.7	2758	14.0	9676
Anemia among children
Yes	82.6	7470	69.6	38 648	791.9^f
No	17.4	2038	30.4	17 621
Mother’s age, years
15-24	52.2	5508	45.7	25 993	218.7^f
25-29	28.8	3661	34.9	21 425
30-49	19.0	2558	19.5	13 740
Mother’s education
No education	49.1	4688	28.3	17 825	2348.9^f
Primary	14.2	1684	13.8	8633
Secondary and above	36.7	5355	57.8	34 700
Mother’s exposure to mass media at least once a week
No exposure	47.9	4558	35.1	22 762	989.1^f
At least 1 media	30.9	3916	44.9	26 567
Two or 3 media	21.2	3253	20.0	11 829
Mother’s daily consumption of animal source foods
Yes	37.8	4698	44.5	24 297	198.3^f
No	62.2	7029	55.5	36 861
Mother’s daily consumption of pulses/nuts/seeds
Yes	51.4	5638	44.5	25 800	208.1^f
No	48.6	6089	55.5	35 358
Mother’s daily consumption of fruits and vegetables
Yes	63.0	7536	49.8	30 882	761.8^f
No	37.0	4191	50.2	30 276
Anemia among mothers
Yes	62.5	6263	57.7	34 180	101.0^f
No	37.5	4377	42.3	26 033
Mother’s body mass index (BMI)
Underweight	44.9	4273	30.4	17 315	1217.5^f
Normal	50.5	6201	58.6	36 724
Overweight	4.7	789	11.0	6337
Mother received complete antenatal care^b
Yes	9.2	1368	16.9	9228	500.8^f
No	90.8	10359	83.1	51 930
Place of delivery
Home	61.8	6481	18.1	12 906	10734.0^f
Institution	38.2	5246	81.9	48 252
Place of residence
Urban	23.1	4073	26.2	13 738	53.6^f
Rural	76.9	7654	73.8	47 420
Religion of household
Hindu	78.3	8117	79.0	44 515	3.0
Non-Hindu	21.7	3610	21.0	16 643
Cast/tribe of household
Scheduled caste/tribe	31.6	4088	33.3	24 254	14.8^e
Other	68.4	7639	66.7	36 904
Wealth index
Poorest	26.4	2191	26.0	16 374	9.3
Poorer	23.3	2287	22.5	14 546
Middle	19.7	2422	20.5	12 377
Richer	17.6	2575	17.4	9934
Richest	13.0	2252	13.6	7927
In the past 3 months met with health workers^c
Yes	37.6	3671	51.2	29 761	812.0^f
No	62.4	8056	48.8	31 397
Received benefits from ICDS in the past 12 months
Yes	29.6	3371	64.3	39 153	5352.2^f
No	70.4	8356	35.7	22 005
Geographical regions of India
North	12.2	2017	12.8	11 274	49.8^d
Central	28.7	2565	26.4	17 264
East	27.7	2020	28.1	13 880
North-east	4.0	2283	3.8	8991
West	12.9	1288	12.5	4227
South	14.6	1554	16.5	5522
Total	100.0	11727	100.0	61 158

Abbreviation: ICDS, Integrated Child Development Scheme.

^aData on 2219, 464, and 1087 cases in NFHS-3 and 4889, 782, and 945 cases in NFHS-4 were missing in anemia among children, BMI, and anemia among women, respectively, excluded from the analysis.

^bANC in the first trimester, at least 4 antenatal visits, at least 1 tetanus toxoid (TT) injection, and iron-folic acid tablets or syrup is taken for 100 days or more.

^cHealth workers include Auxiliary Nurse Midwife (ANM), Lady Health Visitor (LHV), Anganwadi Worker (AWW), or Accredited Social Health Activist (ASHA).

^dSignificant at 10% level.

^eSignificant at 5% level.

^fSignificant at 1% level.

Food Consumed Among Children

The consumption of different food groups among breast-feeding children aged 6 to 23 months during NFHS-3 and 2015 to 2016 is presented in Table 2. The prevalence of breast-feeding among children aged 6 to 23 months slightly declined from NFHS-3 (86.9%) to NFHS-4 (84.2%; not given in the table). The most common food items were starchy staples (grains/roots, or tubers), followed by dairy products. The consumption of a starchy staple diet (rich in energy-dense carbohydrates) significantly decreased from 2005-2006 (83%) to 2015-2016 (72%). The consumption of dairy products, which are a rich source of animal proteins, also decreased from 2005-2006 (49%) to 2015-2016 (46%). However, the consumption of vitamin A–rich fruits and green leafy vegetables and other fruits and vegetables significantly increased in the past 10 years as observed 32% and 19% in NFHS-3 and 38% and 32% in NFHS-4, respectively. A similar proportion of children (1 in every 10 children) consumed nonvegetarian sources of protein, such as flesh foods or eggs, compared to other food groups in 2015 to 2016. However, the consumption of nonvegetarian foods significantly increased among children from 2005-2006 to 2015-2016. The percentage of children consuming minimum meal frequency significantly declined from 45% in 2005 to 2006 to 31% in 2015 to 2016. One of the major reasons for low MAD among these children could be due to the decline in minimum meal frequency.

Table 2.

Food Group Consumption Among Breast-Feeding Children Aged 6 to 23 Months in India and its Pattern Between NFHS-3 and NFHS-4.

Type of Food Group	NFHS-3 (2005-2006), n = 11 727 (Unweighted)		NFHS-4 (2015-2016), n = 61 158 (Unweighted)		Difference^a
Type of Food Group	%	95% CI	%	95% CI	Difference^a
Starchy staples (grains roots and tubers)	82.9	82.3-83.6	72.3	71.9-72.7	−10.6^e
Legumes and nuts	13.4	12.8-14.0	12.6	12.4-12.9	−0.8^c
Dairy products (milk yogurt cheese)	48.6	47.8-49.5	45.7	45.3-46.1	−2.9^e
Flesh food^b	7.2	6.8-7.7	9.0	8.8-9.3	1.8^e
Eggs	4.7	4.4-5.1	13.3	13.0-13.6	8.6^e
Vitamin A–rich fruits and vegetables	32.4	31.6-33.2	38.2	37.8-38.5	5.8^e
Other fruits and vegetables	18.7	18.0-19.3	32.3	31.9-32.7	13.6^e
Minimum dietary diversity	14.5	13.6-15.3	21.6	21.0-22.2	7.1^e
Minimum meal frequency	44.6	43.8-45.5	31.2	30.8-31.5	−13.4^e
Minimum acceptable diet	9.8	9.3-10.4	9.5	9.2-9.7	−0.3

Abbreviation: CI, confidence interval.

^aNFHS-4 – NFHS-3.

^bMeat/fish/poultry & liver/organ meats.

^cSignificant at 10% level (t test).

^dSignificant at 5% level (t test).

^eSignificant at 1% level (t test).

Association of MAD

Table 3 presents the association of MAD among children aged 6 to 23 months with selected characteristics in India. A direct relation of MAD with increasing age of the children is observed during NFHS-3 and NFHS-4. However, in the past 10 years, the proportion of younger children, aged 6 to 11 months, who received the MAD increased from 4% (NFHS-3) to 6% (NFHS-4), but it decreased among children of older age (12-23 months) during the same period. The percentage of first birth order children receiving MAD declined in the past decade, while no major change was observed among children with anemia during the same period. Consumption of MAD among children of mothers of different age groups did not change much from NFHS-3 to NFHS-4.

Table 3.

Descriptive Statistics of Minimum Acceptable Diet Among Children Aged 6 to 23 Months by Selected Background Characteristics.^a

Characteristics	NFHS-3 (2005-2060), n = 11 727 (Unweighted)		NFHS-4 (2015-2016), n = 61 158 (Unweighted)		Difference^b
Characteristics	% (n)	95% CI	% (n)	95% CI	Difference^b
Sex of child
Male	10.1	9.4-10.8	9.5	9.2-9.8	−0.6
Female	9.6	8.9-10.3	9.5	9.2-9.8	−0.1
Age of child in months
6-11	4.2	3.6-4.8	5.9	5.6-6.2	1.7^g
12-17	11.7	10.8-12.6	10.2	9.8-10.6	−1.5^f
18-23	15.3	14.1-16.5	13.6	13.1-14.1	−1.7^f
Birth order
First	11.7	10.7-12.7	9.5	9.1-9.9	−2.2^g
2-3	10.8	10.0-11.6	10.0	9.7-10.3	−0.8^e
4 or more	6.0	5.2-6.8	7.6	7.0-8.2	1.6^f
Anemia among children
Yes	9.3	8.7-9.9	9.3	9.0-9.6	0.0
No	14.3	12.7-15.9	10.8	10.3-11.3	−3.5^g
Mother’s age, years
15-24	9.3	8.6-10.0	9.1	8.8-9.4	−0.2
25-29	10.9	9.9-11.9	9.7	9.3-10.1	−1.2^e
30-49	9.6	8.4-10.8	9.8	9.3-10.3	0.1
Mother’s education
No education	5.5	4.9-6.1	7.1	6.7-7.5	1.6^g
Primary	10.7	9.3-12.1	8.7	8.1-9.3	−2.0^f
Secondary and above	15.3	14.3-16.3	10.8	10.5-11.1	−4.5^g
Mother’s exposure to mass media at least once a week
No exposure	5.9	5.3-6.5	7.0	6.7-7.3	1.1^f
At least 1 media	10.6	9.7-11.5	10.3	9.9-10.7	−0.3
Two or 3 media	17.6	16.2-19.0	12.0	11.4-12.6	−5.7^g
Mother’s daily consumption of animal source foods
Yes	14.4	13.4-15.4	11.1	10.7-11.5	−3.3^g
No	7.1	6.5-7.7	8.1	7.8-8.4	1.1^f
Mother’s daily consumption of pulses/nuts/seeds
Yes	10.8	10.1-11.5	9.9	9.5-10.3	−0.8^e
No	8.9	8.2-9.6	9.1	8.8-9.4	0.2
Mother’s daily consumption of fruits and vegetables
Yes	11.2	10.5-11.9	11.8	11.4-12.2	0.6
No	7.5	6.8-8.2	7.2	6.9-7.5	−0.3
Anemia among mothers
Yes	8.8	8.2-9.4	9.7	9.3-10.0	0.9^e
No	11.9	11.0-12.8	9.3	8.9-9.7	−2.6^f
Mother’s body mass index (BMI)
Underweight	7.7	7.0-8.4	8.7	8.3-9.1	1.1^e
Normal	10.8	10.0-11.6	9.6	9.3-9.9	−1.3^f
Overweight	23.7	20.3-27.1	11.5	10.7-12.3	−12.2^g
Mother received complete antenatal care^c
Yes	23.0	20.6-25.4	13.2	12.5-13.9	−9.8^g
No	8.5	8.0-9.0	8.7	8.4-9.0	0.2
Place of delivery
Home	6.5	6.0-7.0	8.2	7.7-8.7	1.7^g
Institution	15.3	14.3-16.3	9.7	9.4-10.0	−5.5^g
Place of residence
Urban	14.9	13.6-16.2	11.0	10.5-11.5	−3.9^g
Rural	8.3	7.8-8.8	8.9	8.6-9.2	0.6^e
Religion of household
Hindu	9.4	8.8-10.0	8.9	8.6-9.2	−0.5
Other	11.4	10.2-12.6	11.6	11.0-12.2	0.2
Cast/tribe of household
Scheduled caste/tribes	7.3	6.5-8.1	9.3	8.9-9.7	2.0^g
Other	11.0	10.4-11.6	9.6	9.3-9.9	−1.5^g
Wealth index
Poorest	4.3	3.6-5.0	7.6	7.2-8.0	3.3^g
Poorer	7.7	6.8-8.6	9.0	8.5-9.5	1.3^e
Middle	9.1	8.0-10.2	9.9	9.4-10.4	0.7
Richer	13.6	12.2-15.0	10.3	9.7-10.9	−3.3^g
Richest	20.9	19.0-22.8	12.2	11.5-12.9	−8.7^g
In the past 3 months met with health workers^d
Yes	10.8	9.9-11.7	10.3	10.0-10.6	−0.4
No	9.3	8.7-9.9	8.6	8.3-8.9	−0.7^e
Received benefits from ICDS in the past 12 months
Yes	10.2	9.2-11.2	10.2	9.9-10.5	0.0
No	9.7	9.1-10.3	8.1	7.7-8.5	−1.6^g
Geographical regions of India
North	9.5	8.1-10.9	7.2	6.6-7.8	−2.3^f
Central	6.7	5.9-7.5	7.1	6.7-7.5	0.4
East	11.1	10.1-12.1	11.4	10.9-11.9	0.4
North-east	10.2	7.6-12.8	11.9	10.6-13.2	1.7
West	7.7	6.4-9.0	6.7	6.1-7.3	−1.0
South	15.7	14.1-17.3	13.1	12.4-13.8	−2.6^f
Total	9.8	9.3-10.3	9.5	9.3-9.7	−0.4

Abbreviations: CI, confidence interval; ICDS, Integrated Child Development Scheme.

^aData on 2219, 464, and 1087 cases in NFHS-3 and 4889, 782, and 945 cases in NFHS-4 were missing in anemia among children, BMI, and anemia among women, respectively, excluded from the analysis.

^bNFHS-4 – NFHS-3.

^cAt least 4 antenatal visits, at least 1 tetanus toxoid (TT) injection and iron-folic acid tablets or syrup is taken for 100 days or more.

^dAuxiliary Nurse Midwife (ANM), Lady Health Visitor (LHV), Anganwadi Worker (AWW), or Accredited Social Health Activist (ASHA).

^eSignificant at 10% level (t test).

^fSignificant at 5% level (t test).

^gSignificant at 1% level (t test).

Mothers’ education and exposure to mass media were associated with better dietary adequacy of children at both time periods. A decline in intake of MAD was observed from NFHS-3 to NFHS-4 among mothers with secondary and above education and mothers who were exposed to 2 or 3 mass media at least once a week. Mothers’ consumption of animal protein was associated with MAD intake of the children, but children’s intake of MAD significantly declined from 14% (NFHS-3) to 11% (NFHS-4) in the past 10 years.

Body mass index (BMI) of mothers also had an association with the dietary adequacy of children. A higher proportion of children of obese mothers get MAD compared to children of underweight mothers, and this trend did not change from NFHS-3 to NFHS-4. However, the percentage of children of obese mothers, who were receiving MAD, significantly declined from NFHS-3 (24%) to NFHS-4 (12%).

Antenatal care visits and institutional delivery showed direct relation with dietary adequacy of the child. Intake of MAD by children whose mothers received complete ANC and had institutional delivery was higher than whose mother did not. Also, more children living in urban areas received MAD than children from rural area. However, the dietary adequacy declined from NFHS-3 to NFHS-4 among those children who were from urban areas, had institutional delivery, and whose mother had full ANC. The wealth index showed a positive association with intake of MAD, but the gap in dietary adequacy between poorest and “richest” families reduced in the past 10 years. More children of the “poorest” families received MAD in NFHS-4 (8%) compared to NFHS-3 (4%), and this trend reversed among children of “richest” families.

A significant disparity in dietary adequacy of children was observed across the regions of India. The dietary adequacy was the lowest in Central India and the highest in Southern India throughout the past 10 years. However, in the southern and northern regions of India, the proportion of children consuming MAD significantly declined from NFHS-3 to NFHS-4.

Determinants of MAD

Table 4 summarizes the determinants of MAD in breast-feeding children aged 6 to 23 months in models 1, 2, and 3. Models 1 and 2 identified the determinants of MAD in the 2005 to 2006 and 2015 to 2016, respectively. The odds for the intake of MAD were significantly high among children aged 12 months and above (both in 2005-2006 and 2015-2016). Similarly, mother’s education; exposure to mass media; daily consumption of animal source foods, fruits, and vegetables; ANC; institutional delivery; and well-off families were positively related to MAD as observed at both time periods (2005-2006 and 2015-2016).

Table 4.

Determinants of Minimum Acceptable Diet (MAD) Among Children Aged 6 to 23 Months in 2005 to 2006 (NFHS-3) and 2015 to 2016 (NFHS-4) and Change in MAD From 2005-2006 to 2015-2016.^a-c

Characteristics	Model 1 (n = 9462) NFHS-3		Model 2 (n = 56023) NFHS-4		Model 3 (n = 65451) Change From NFHS-3 to NFHS-4
Characteristics	AOR	95% CI	AOR	95% CI	AOR	95% CI
Age of child in months
6-11	1.00		1.00		1.00
12-17	3.41^h	2.83-4.12	1.83^h	1.70-1.98	0.60^h	0.49-0.73
18-23	5.06^h	4.19-6.11	2.50^h	2.32-2.70	0.57^h	0.47-0.70
Birth order
First	–		1.00		1.00
2-3	–		1.16^h	1.09-1.24	1.27^g	1.09-1.49
4 or more	–		1.10^f	0.99-1.23	1.74^h	1.41-2.15
Anemia among children
Any anemia	1.00		1.00		1.00
No anemia	1.27^g	1.08-1.49	1.09^f	1.02-1.16	0.74^h	0.63-0.87
Mother’s education
No education	1.00		1.00		1.00
Primary	1.52^h	1.23-1.87	1.12^f	1.01-1.25	0.61^h	0.49-0.76
Secondary and above	1.37^g	1.13-1.67	1.26^h	1.16-1.38	0.54^h	0.46-0.64
Mother’s exposure to mass media at least once a week
No exposure	1.00		1.00		1.00
At least 1 media	1.11	0.93-1.33	1.31^h	1.20-1.42	0.81^h	0.72-0.92
Two or 3 media	1.41^g	1.15-1.73	1.38^h	1.25-1.54	0.53^h	0.47-0.61
Mother’s daily consumption of animal source foods
No	1.00		1.00		1.00
Yes	1.37^h	1.18-1.59	1.25^h	1.17-1.34	0.64^h	0.58-0.71
Mother’s daily consumption of pulses/nuts/seeds
No			1.00		1.00
Yes	–		0.89^h	0.84-0.95	0.89^h	0.83-0.94
Mother’s daily consumption of fruits and vegetables
No	1.00		1.00
Yes	1.33^h	1.14-1.54	1.50^h	1.40-1.60	–
Mother’s body mass index (BMI)
Underweight	1.00		–		1.00
Normal	1.24^g	1.08-1.43	–		0.70^h	0.63-0.78
Overweight	1.64^h	1.26-2.15	–		0.40^h	0.31-0.50
Anemia among mothers
Any anemia			1.00		1.00
No anemia	–		0.93^f	0.87-0.99	0.66^h	0.57-0.76
Complete antenatal care^d
No	1.00		1.00		1.00
Yes	1.66^h	1.36-2.02	1.33^h	1.23-1.43	0.59^h	0.49-0.71
Place of delivery
Home	1.00				1.00
Institution	1.33^g	1.13-1.57	–		0.59^h	0.53-0.65
Place of residence
Urban			1.00		1.00
Rural	–		0.92^f	0.85-1.00	1.46^h	1.25-1.70
Religion of household
Hindu	1.00		1.00
Non-Hindu	1.19^f	1.02-1.40	1.34^h	1.25-1.44	–
Cast/tribe of household
Scheduled caste/tribe			1.00		1.00
Other	–		0.90^g	0.84-0.96	0.69^h	0.63-0.75
Wealth index
Poorest	1.00		1.00		1.00
Poorer	1.69^h	1.34-2.13	1.04	0.95-1.15	0.58^h	0.46-0.75
Middle	1.53^g	1.19-1.97	1.07	0.96-1.19	0.59^h	0.46-0.76
Richer	2.17^h	1.66-2.84	1.05	0.93-1.19	0.38^h	0.30-0.49
Richest	2.98^h	2.18-4.06	1.27^g	1.10-1.46	0.29^h	0.23-0.37
In the past 3 months met with health workers^e
No	1.00		1.00
Yes	1.35^h	1.18-1.56	1.23^h	1.15-1.31	–
Received benefits from Anganwadi center (ICDS) in the past 12 months
No	1.00		1.00		1.00
Yes	1.18^f	1.01-1.37	1.16^h	1.08-1.25	1.18^h	1.10-1.26
Regions of India
North	1.00		1.00		1.00
Central	1.13	0.89-1.44	1.28^h	1.14-1.44	1.46^g	1.13-1.89
East	1.88^h	1.48-2.38	1.94^h	1.72-2.18	1.45^g	1.14-1.84
North-east	1.49^f	1.01-2.19	1.76^h	1.48-2.09	1.55^f	1.04-2.31
West	0.65^g	0.48-0.86	0.87^f	0.76-1.00	1.29	0.95-1.75
South	1.33^f	1.03-1.72	1.73^h	1.54-1.94	1.19	0.92-1.54

Abbreviations: AOR, adjusted odds ratio; CI, confidence interval; ICDS, Integrated Child Development Scheme.

^aModel 1: Multivariate model identifying determinants of MAD for NFHS-3. Data on 2265 cases in NFHS-3 were missing and were excluded from the analysis.

^bModel 2: Multivariate model identifying determinants of MAD for NFHS-4. Data on 5135 cases in NFHS-4 were missing and were excluded from the analysis.

^cModel 3: Multivariate model identifying determinants of change in MAD form NFHS-3 to NFHS-4. Data on 7434 cases in NFHS-3 and NFHS-4 were missing and were excluded from the analysis.

^dANC in the first trimester, at least 4 antenatal visits, at least 1 tetanus toxoid (TT) injection, and iron-folic acid tablets or syrup is taken for 100 days or more.

^eAuxiliary Nurse Midwife, Lady Health Visitor, Anganwadi Worker, or Accredited Social Health Activist.

^fSignificant at 10% level.

^gSignificant at 5% level.

^hSignificant at 1% level.

Model 3 identified the determinants of change in MAD intake, considering the changes in the determinant variables from 2005-2006 to 2015-2016. The odds for the intake of MAD were 5 times higher in 2005 to 2006 (OR: 5.1; confidence interval [CI], 4.2-6.1) but significantly declined by half in 2015 to 2016 (OR: 2.5; CI, 2.3-2.7). The odds of a change in intake of MAD in the past 10 years among children older than 12 months was 0.6 (40% less likely to increase) compared to change in intake of MAD among younger children aged below 1 year (model 3). Similar decline in intake of MAD in the past10 years can be noticed with nonanemic childrenversus children with anemia, mothers who had at least primary education versus mothers with no education, mothers who were exposed to at least 2 channels of mass media versus mothers with no exposure to mass media, mothers who consumed animal source proteins daily, had complete ANC, and who lived in better-off households compared to their respective counterpart.

But, on the other hand, intake of MAD had improved in the past 10 years among children of certain characteristics. For example, a higher proportion of children of 2 or higher birth order were given MAD compared to change in intake of MAD among first birth order children (OR: 1.3 times among 2-3 birth order and 1.7 times among 4+ birth order). Similarly, intake of MAD among children of women living in rural areas increased 1.4 times compared to change in intake of MAD among children of women living in urban areas. Change in consumption of MAD among children with anemia when compared to nonanemic counterparts. Also, change in consumption of MAD among children of women who received benefits from Anganwadi centers in the past 12 months and who lived in central, eastern, north-eastern regions of India were more likely to feed MAD to their children compared to change in consumption of MAD among children of women who did not receive benefits from Anganwadi centers and who lived in northern region of India, respectively.

Discussion

We found that the prevalence of MAD (only 10%) was consistently low in the past decade in India among breast-feeding children aged 6 to 23 months. Although MAD did not significantly change from 2005-2006 to 2015-2016, however, minimum meal frequency reduced by 14 percentage points from 2005-2006 (45%) to 2015-202016 (31%), which led to exceptionally low consumption of acceptable diet. Such a low intake of MAD is a public health concern in India because an unbalanced diet could increase the prevalence of malnutrition in the country.

Multivariate logistic regression analysis (Table 4: model 3) showed improvement in intake of MAD among children of higher birth order (4 or more), living in rural areas, who were benefitting from Anganwadi centers and living in central, eastern, and north-eastern regions of the country. The Anganwadi centers provide meals to preschool-going and school-going children throughout the country and especially in rural areas through the government-run Integrated Child Development Scheme (ICDS). The children who are living in urban areas are not only more likely to consume fast food but also have limited access to the Anganwadi center compared to children living in rural areas.^27,28 At the same time, local availability of the food variety and social customs related to dietary habits also cause the geographic variation in intake of MAD.²⁹

The consumption of different food groups and MAD among breast-feeding children aged 6 to 23 months showed similar results for NFHS-3 to the findings of Patel et al.¹⁹ The literature showed that the consumption of MAD varies with age and it improves as the age increases.^30,31 The present study also found that a higher proportion of older children consumed MAD compared to younger children. However, in the past decade, the consumption of MAD was alarmingly decreased, and the decline was more pronounced among older children (12-23 months) than younger children (6-11 months). Dietary adequacy among children with anemia improved in the past 10 years compared to their nonanemic counterparts. This could be due to various government programs, such as ICDS, that focus on the eradication of anemia in the country²⁷; therefore, the program ensured MAD to the targeted children.

Mothers’ education above primary level, their exposure to mass media, and their daily consumption of animal protein foods were significant predictors of MAD, and this finding corroborates to the findings of Borkotoky et al.³² The decline in the consumption of MAD among children could be due to a change in lifestyle and eating patterns over the past 10 years. Children might be consuming more in quantity but not in quality and diversity. However, we could not examine this explanation due to lack of data on the quantity of dietary intake in NFHS. The literature showed that selection of foods by children is mainly taste driven rather than based on the nutritional content.³³ Easy accessibility, rising incomes of the family, and increased consumption of junk foods could be the associated factors for poor diet diversity among children.³⁴

Several technical consultations and documents on complementary feeding recommend that it should be started when breast-milk alone is insufficient to meet all nutritional requirements.^7,30,35,36 This late introduction of complementary foods has been cited as a factor contributing to the downward trend of the growth curves seen among infants after 6 months of age.³⁴

Nonanemic mothers, maternal obesity, and household wealth had some influence on MAD as the odds of MAD among children of these mothers show a significant decline since 2005 to 2006. Nutritional counseling for mothers about infant and young child feeding options during ANC is an important continuum of care for promoting appropriate infant and young child feeding practices.^20,37

The findings further showed that children were more likely to receive MAD when their mothers receive full ANC (as per the WHO guidelines) and have institutional delivery; this finding aligned with the findings of a previous study in India.³² The literature showed that complementary feeding practices improved among the mothers who had more exposure to health professionals during their ANC visits, more likely to take appropriate actions to improve the dietary diversity of their children.³ However, contrary to existing evidence, we found that in NFHS 2015-2016, children, whose mothers received complete ANC and had institutional delivery, received significantly less MAD than those of NFHS 2005-2006. In the past 10 years, India experienced a tremendous increase in maternal health coverage indicators; institutional delivery increased by 40 percentage points and receipt of 4 antenatal checkups by 14.2 percentage points, and these increases were higher among vulnerable socioeconomic groups. For example, the institutional delivery among the lowest wealth index group increased from 13% in 2005 to 2006 to 60% in 2015 to 2016, while the increase for the highest wealth index group was from 84% to 95% during the same period.^6,26 The change in dietary intake among children aged 6 to 23 months did not coincide with an increase in coverage of maternal health services, mostly after initiation of Janai Suraksha Yojana in 2005, across the socioeconomic groups.

This study has some potential limitations. The cause and effect relationships cannot be established due to the cross-sectional design of the NFHS. It also relied on a 24-hour recall of different types of food groups and the frequency but not the quantity of the food which is vital information to estimate dietary adequacy. However, the strength of this study lies with the results from a large nationally representative sample, which can help policymakers and researchers design interventions to improve feeding practices for infant and young children in India.

Findings of the study reveal that a very low proportion of breast-feeding children aged 6 to 23 months received MAD, and this remained the same in the past 10 years in the country. Intake of MAD in India is consistently low among children of a younger age-group, children of 4 or higher birth order, children with anemia, children of mothers having no exposure to mass media, mothers who do not consume animal source foods daily, mothers with low BMI, mothers who did not receive ANC during their last pregnancy, children of urban area, poor household, of mother who did not receive any benefit from ICDS in past 12 months, and children from the northern, central, and western India. In the long term, improvements in higher levels of parental education and media can be used as an effective means of promoting meal frequency practices and nutritional counseling for mothers about infant, and young child feeding options during ANC can result in better practices of MAD in India.

Footnotes

Authors’ Note

All authors contributed to the study concept and design. NK contributed to statistical analysis; SK and AM contributed to interpretation of data; NK and AM drafted the article; NK and SK contributed to the critical revision of the article for important intellectual content; and all authors have approved the final version of the manuscript.

Authors analysed the secondary data accessible from DHS website with a request for use of research purposes only. DHS had no role in the design, analysis, or writing of this article.

Acknowledgments

The International Institute of Population Sciences, Mumbai, and ORC Macro, Maryland, USA, carried out the NFHS series for the Health Sector of the Ministry of Health and Family Welfare, India.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Arupendra Mozumdar

Ethical Standards Disclosure

This article is based on secondary analysis of the National Family Health Survey (NFHS) data available in public domain with all identifying information removed. The NFHS received ethical clearance from the International Institute for Population Science’s Ethical Review Board from where this survey was conducted. The survey personnel obtained informed consent from each respondent before asking questions.

References

Mukuria

Kothari

Abderrahim

.Infant and Young Child Feeding Updates. Calverton, MD: ORC Macro; 2006. https://dhsprogram.com/pubs/pdf/NUT1/NUT1.pdf.

Arimond

Ruel

. Dietary diversity is associated with child nutritional status: evidence from 11 demographic and health surveys. J Nutr. 2004;134(10):2579–2585.

Bhutta

Ahmed

Black

, et al. What works? Interventions for maternal and child undernutrition and survival. Lancet. 2008;371(9610):417–440.

Unicef Statistics. Joint Malnutrition Dataset from UNICEF, World Bank and WHO. UNICEF; 2018. http://data.unicef.org/nutrition/malnutrition.html. Accessed October 6, 2018.

Ramachandran

. Food & nutrition security: challenges in the new millennium. Indian J Med Res. 2013;138(3):373–382.

International Institute for Population Sciences (IIPS) and ICF. National Family Health Survey (NFHS-4), 2015–16: India. Mumbai, India: IIPS; 2017.

World Health Organization. Technical Consultation on Infant and Young Child Feeding. Fifty-Third World Health Assembly, Provisional Agenda Item 12.4. 2000. http://apps.who.int/gb/archive/pdf_files/WHA53/eaid2.pdf. Accessed October 6, 2018.

Johnston

Fanzo

Cogill

. Understanding sustainable diets: a descriptive analysis of the determinants and processes that influence diets and their impact on health, food security, and environmental sustainability. Adv Nutr. 2014;5(4):418–429.

Mayen

Marques-Vidal

Paccaud

, et al. Socioeconomic determinants of dietary patterns in low- and middle-income countries: a systematic review. Am J Clin Nutr. 2014;100(6):1520–1531.

10.

Zarnowiecki

Dollman

Parletta

. Associations between predictors of children’s dietary intake and socioeconomic position: a systematic review of the literature. Obes Rev. 2014;15:375–391.

11.

McPhie

Skouteris

Daniels

, et al. Maternal correlates of maternal child feeding practices: a systematic review. Matern Child Nutr. 2014;10(1):18–43.

12.

World Health Organization (WHO). Guidance on Ending the Inappropriate Promotion of Foods for Infants and Young Children: Implementation Manual. Geneva, Switzerland: WHO; 2017: Licence: CC BY-NC-SA 3.0 IGO.

13.

International Institute for Population Sciences (IIPS). National Family Health Survey (MCH and Family Planning), India 1992–93. Bombay, India: IIPS; 1995.

14.

Shrimpton

Victora

De Onis

Lima

Blössner

Clugston

. Worldwide timing of growth faltering: implications for nutritional interventions. Pediatrics. 2001;107(5):e75.

15.

Leutter

. Meeting the challenge to improve complementary feeding. SCN News. 2003;27:4–9. https://www.popline.org/node/233102.

16.

Ministry of Women and Child Development. Report of the Working Group of Nutrition for 12th Five-Year Plan. 2012–17. 2011. http://planningcommission.gov.in/aboutus/committee/wrkgrp12/wcd/wgrep_nutition.pdf. Accessed October 6, 2018.

17.

Ruel

Menon

. Child feeding practices are associated with child nutritional status in Latin America: innovative uses of the demographic and health surveys. J Nutr. 2002;132(6):1180–1187.

18.

Aemro

Mesele

Birhanu

Atenafu

. Dietary diversity and meal frequency practices among infant and young children aged 6–23 months in Ethiopia: a secondary analysis of Ethiopian Demographic and Health Survey 2011. J Nutr Metab. 2013;Article ID: 782931. doi:10.1155/2013/782931.

19.

Patel

Pusdekar

Badhoniya

Borkar

Agho

Dibley

. Determinants of inappropriate complementary feeding practices in young children in India: secondary analysis of National Family Health Survey 2005–2006. Matern Child Nutr. 2012;8(suppl 1):28–44.

20.

Amugsi

Mittelmark

Oduro

. Association between maternal and child dietary diversity: an analysis of the Ghana Demographic and Health Survey. PLoS One 2015;10(8):e0136748.

21.

World Health Organization (WHO). Indicators for Assessing Infant and Young Child Feeding Practices. Conclusions of a Consensus Meeting Held 6–8 November 2007 in Washington, DC, USA. Part 1: Definitions. Geneva, Switzerland: WHO; 2008.

22.

Prasad

Costello

. Impact and sustainability of a ‘baby friendly’ health education intervention at a district hospital in Bihar, India. BMJ. 1995;310(6980):621–623.

23.

Aggarwal

Patwari

. Breastfeeding among urban women of low-socioeconomic status: factors influencing introduction of supplemental feeds before four months of age. Indian Pediatr. 1998;35(3):269–273.

24.

Coutinho

De Lira

De Carvalho

Ashworth

. Comparison of the effect of two systems for the promotion of exclusive breastfeeding. Lancet. 2005;366(9491):1094–1100.

25.

International Institute for Population Sciences (IIPS) and Macro International. National Family Health Survey (NFHS-3), 2005–06: India: Volume I. Mumbai, India: IIPS; 2007.

26.

World Health Organization (WHO). Indicators for Assessing Infant and Young Child Feeding Practices, Part 2, Measurement. Geneva, Switzerland: WHO; 2010.

27.

Chudasama

Patel

Kadri

Mitra

Thakkar

Oza

. Evaluation of integrated child development services program in Gujarat, India for the years 2012 to 2015. Indian J Public Health. 2016;60(2):124–130.

28.

Kotecha

. Nutritional anemia in young children with focus on Asia and India. Indian J Community Med. 2011;36(1):8–16.

29.

Longvah

Khutsoh

Meshram

, et al. Mother and child nutrition among the Chakhesang tribe in the state of Nagaland, North-East India. Matern Child Nutr. 2017;13(suppl 3):e12558.

30.

Dewey

. Guiding Principles for Complementary Feeding of the Breastfed Child. Washington, DC. Pan American Health Organization/WHO; 2003. https://www.popline.org/node/236830

31.

Paul

Muti

Chasekwa

, et al. Complementary feeding messages that target cultural barriers enhance both the use of lipid-based nutrient supplements and underlying feeding practices to improve infant diets in rural Zimbabwe. Matern Child Nutr. 2010;8:11740–18709.

32.

Borkotoky

Unisa

Gupta

. State-level dietary diversity as a contextual determinant of nutritional status of children in India: a multilevel approach. J Biosoc Sci. 2018;50(1):26–52.

33.

Maternal and Child Survival Programme. Nutrition Brief: Junk Food Consumption is a Nutrition Problem among Infants and Young Children: Evidence and Program Considerations for Low and Middle Income Countries. 2016. http://www.mcsprogram.org/wp-content/uploads/2016/09/MCSP-Nutrition-Brief.pdf.

34.

Philips

Chimmule

Engle

, et al. Does Timely Introduction of Complementary Foods Lead to Improved Nutritional Status. Analysis of Data from Maharashtra, India. 2000. https://pdfs.semanticscholar.org/3009/e5673d273a3bd6b44405f658095755589ce9.pdf.

35.

World Health Organization (WHO). Complementary Feeding of Young Children in Developing Countries – A Review of Current Scientific Knowledge. WHO/NUT 98.1. Geneva, Switzerland: WHO; 1998.

36.

World Health Organization (WHO). Development of a Global Strategy on Infant and Young Child Feeding: Report on a WHO/UNICEF Consultation for the WHO European Region Budapest. Hungary: WHO; 2001.

37.

Reddy

. Improving childhood nutrition in India. BMJ. 2011;343:d7188.

Dietary Adequacy Among Young Children in India: Improvement or Stagnation? An Investigation From the National Family Health Survey