Introduction

Childhood undernutrition remains a persistent public health challenge in India. Despite sustained efforts to reduce its prevalence. According to NFHS-5, nearly 35.5% of Indian children under five are stunted, 32.1% are underweight, and 19.3% are wasted [1]. Malnutrition are known to have long-term effects on cognitive development, immunity, school performance, and economic productivity [2]. According to WHO (2008), “Dietary diversity, defined as the number of different food groups consumed over a reference period”, has emerged as a key indicator of diet quality and micronutrient adequacy in young children [3].A child meets the Minimum Dietary Diversity (MDD) requirement when he or she consumes meals from at least five of the eight food groups within the last 24 hours. A lack of diversity is frequently an indicator for poor household food security, poverty, and low nutritional understanding [4].

Several studies have demonstrated a strong association between dietary diversity and improved nutritional outcomes, including lower odds of stunting and underweight [5, 6].

Despite numerous national-level study on childhood undernutrition and dietary diversity in India, limited studies have explored how these relations vary within high-burden states like Uttar Pradesh [1], especially among different socioeconomic strata. Most existing studies address undernutrition as a uniform condition, failing to distinguish between its various forms like stunting, wasting, being underweight, and having multiple deficiencies at the same time and how dietary diversity affects each individually. Also, the interaction among wealth index and minimum dietary diversity as a moderator of nutritional outcomes is understudied.
Therefore, to develop targeted interventions to better influence effective nutrition policies, there is a critical need for state-specific, detailed study that investigates these complicated associations.

Objective: To estimate the association between dietary diversity and various forms of undernutrition among children aged 6–59 months in Uttar Pradesh, and to evaluate the moderating effect of household socioeconomic status on this relationship.

Materials and Methods

Study Design and Data Source

“This study employs a cross-sectional analytical design using data from the Fifth National Family Health Survey (NFHS-5, 2019–21), a large-scale, nationally representative household survey conducted across India. The NFHS-5 follows a multistage, stratified cluster sampling technique. The current analysis focuses exclusively on children aged 6–59 months in Uttar Pradesh, using the child recode (KR) and household datasets provided by the Demographic and Health Surveys (DHS) program.

Figure 1: Flow Diagram of Study Methodology

Study Population

Children aged 6–59 months with valid anthropometric data and complete information on dietary diversity and key covariates were included. Cases with biologically implausible Z-scores (|Z| > 6 SD)/ missing values on outcome or explanatory variables were excluded. The final analytical sample comprises 29,196 children. The Workflow of this study was mentioned in Figure 1.

Dependent/Outcome Variable: Nutritional Status

The dependent variable was nutritional status, categorized into six mutually exclusive groups derived from anthropometric Z-scores based on WHO (2006) growth standards: Normal (Reference category), Stunted only: Height-for-age Z-score (HAZ) < −2 SD, Underweight only: Weight-for-age Z-score (WAZ) < −2 SD, Wasted only: Weight-for-height Z-score (WHZ) < −2 SD, Multiple undernutrition: Any combination of two or more of the above, Overweight: WHZ > +2 SD. These categories were computed using the following definitions:

Zi=(Xi−μ)/σ

Where Zi is the z-score for child I, Xi ​is the observed anthropometric measure (e.g., height, weight), μ and σ are the median and standard deviation of the reference population.

Key Explanatory Variable:

i) Minimum Dietary Diversity (MDD) was assessed based on the WHO IYCF guidelines [7]. The mother reported the types of foods consumed by the child in the last 24 hours. Eight food groups were considered: i) Breast milk ii) Grains, roots, and tubers iii) Legumes and nuts iv) Dairy products v) Flesh foods vi) Eggs vii) Vitamin A-rich fruits and vegetables viii) Other fruits and vegetables

Children consuming ≥5 out of 8 food groups were considered to meet the MDD threshold:

Where is F_ij a binary indicator of consumption of food group j by child i.

ii) Confounding factors: The model adjusted for key covariates known to influence child nutrition, grouped into: Child-level: age (categorized), Sex (Male/Female), birth order (1/2/>2), diarrhea (Yes/No), Maternal education level, Household wealth index, type of sanitation (Improved Toilet Facility), Antenatal care (ANC).”

Data Analysis: Frequencies, percentages, and cross-tabulations were used for descriptive statistics. To assess association between predictors and nutritional status categories Chi-square tests were utilized.

Multinomial Logistic Regression: To estimate adjusted associations, Multinomial Logistic Regression (MLR) was applied with "Normal" nutrition status as the reference outcome. The general form of the MLR model is:

Where P(Y=k) is the probability of child being in nutrition category k, β_0k​ is the intercept for category k, Xi​ are predictor variables and β_ik​ are the coefficients (log-odds) for predictors in category k. The Relative Risk Ratio (RRR) for each predictor was computed by exponentiation the β-coefficients i.e. RRR=exp( β). To explore effect modification, an interaction term between Wealth Index and MDD was introduced:

Model significance was assessed at p < 0.05. Analyses were performed using R studio software.

Results

The average MDD prevalence in Uttar Pradesh is 4.2%, with considerable district-to-district variations. The regions with the lowest MDD prevalence reported in Firozabad (0.2%), Sambhal (0.99%), and Hathras (1.2%) while districts like Sonbhadra (6.8%), Balrampur (7.6%), and Bahraich (9.2%) have greater MDD prevalence (Figure 2).

Figure 2: Spatial distribution of MDD among children aged 6–59 months by district in Uttar Pradesh, India

Figure 3 shows the distribution of child nutritional status varied significantly across demographic, socioeconomic, and health-related characteristics in Uttar Pradesh. Out of 29,196 children, stunting was most prevalent among children aged 36–59 months (43.6%), whereas wasting and underweight were more common in younger age groups, particularly 6–23 months (p < 0.001). Boys had a consistently higher prevalence of stunting (53.1%), underweight (55.8%), and overweight (60.2%) than girls (p < 0.001).

Household wealth was strongly associated with nutritional outcomes. Children from poorer households showed the highest burden of multiple undernutrition (54.0%) and stunting (48.4%), whereas overweight was disproportionately higher among children from wealthier families (50.4%) (p < 0.001).

Maternal education established a clear protective factor, children of mothers with secondary or higher education experienced lower levels of stunting, underweight, and wasting, and a relatively higher prevalence of overweight (p < 0.001).

Household sanitation also found as a significant determinant. Children in households with unimproved sanitation had moderate level of stunting (33.8%) and multiple forms of undernutrition compared to those with improved facilities (p < 0.001).

ANC visits was similarly predictive, children whose mothers had four or more ANC visits had better nutritional status, including a higher proportion of overweight (52.8%), while those with no ANC visits experienced high stunting and undernutrition (p < 0.001).

An insignificant association was found between recent diarrheal episodes and nutritional status (p = 0.282), though children without diarrhea tended to fare slightly better.

Birth order was another critical factor. Children of third or higher birth order bore the greatest burden of multiple undernutritions (42.1%) (p < 0.001).

Finally, dietary diversity was suboptimal across all categories, with over 90% of undernourished children failing to meet the minimum dietary diversity threshold (<5 food groups), though this association approached but did not reach statistical significance (p = 0.052).”

Note: Normal nutritional status (n=14,149), Parenthesis include column wise percentage. Further cases with missing values were excluded from analysis

Figure 3: Distribution of Nutritional Status of Under-Five Children by Background Characteristics, Uttar Pradesh, NFHS-5 (2019–21) (N=29,196)

Further, Multinomial logistic regression was performed to identify factors associated with different types of nutritional status, using normal nutrition as the reference category (Table 1). Age of the child was strongly associated with malnutrition outcomes. Children aged 6–11 months were significantly more likely to be wasted (RRR = 2.07; 95% CI: 1.70–2.52; p < .001), while the risk of stunting increased progressively between 12–35 months. Male children had higher odds of stunting (RRR = 1.20; 95% CI: 1.10–1.30; p < .001) and multiple undernutrition (RRR = 1.11; 95% CI: 1.04–1.19; p = .003) compared to females.

Household wealth emerged as a strong predictor. Children from poor households had significantly higher risk of stunting (RRR = 1.59; 95% CI: 1.40–1.80; p < .001), underweight (RRR = 1.57; p = .004), and multiple undernutrition (RRR = 1.90; p < .001) compared to those from wealthier households. Similarly, low maternal education was consistently associated with higher risk of stunting and multiple nutritional deficits. For example, children whose mothers had no formal education had 2.09 times greater risk of multiple undernutrition (95% CI: 1.84–2.39; p < .001) compared to those whose mothers had higher education. Interestingly, Minimum Dietary Diversity (MDD) on its own was not significantly associated with any nutritional outcome after adjusting for socioeconomic, demographic, and maternal variables (p > .05 for all categories).

Figure 4 shows the association between interaction term of Wealth Index and MDD with nutrition varied by socioeconomic context. A significant interaction effect was observed for several outcomes. Among poor children who did not meet the MDD threshold, the risk of being stunted was 39% higher (RRR = 1.39; 95% CI: 1.08–1.79; p = .012), and the risk of being underweight was more than twofold higher (RRR = 2.43; 95% CI: 1.11–5.35; p = .027), compared to rich children with adequate diet. Most notably, the risk of multiple undernutrition was nearly three times higher among poor children with low dietary diversity (RRR = 2.89; 95% CI: 2.22–3.75; p < .001) and remained elevated even among those with higher diversity (RRR = 2.36; p < .001). In contrast, among children from rich households, dietary diversity was not significantly associated with nutritional outcomes, and in some cases, low MDD even appeared protective (e.g., RRR = 0.72 for stunting; p = .014).

Figure 4: Interaction between Dietary Diversity and Wealth index

Discussion

The study shows the evidence that malnutrition in Uttar Pradesh is shaped by a complex interplay of demographic, socioeconomic, and maternal factors and diet diversity with differential effects across types of undernutrition. This study estimated a low average MDD of 4.2% among children aged 6–59 months in Uttar Pradesh, with substantial variation across districts. The finding is consistent with study by Gunnal et al., who found that among younger children aged 6 to 23 months, the central region of India, including Uttar Pradesh, had a high prevalence of minimum diet diversity failure (MDDF) of 84.6%. According to both studies, Uttar Pradesh is a crucial region for inadequate dietary diversity [8].

Age came about to be a key determinant of child nutritional outcomes. Stunting was more common in older children (36–59 months), but wasting and underweight were more common in younger children (6–23 months), which is in line with the theory that infections and early food deficiencies cause acute undernutrition to appear earlier [9]. Male children were more prone toward stunting and multiple under-nutrition, aligned with prior Indian studies attributed this disparity to gendered feeding practices and biological susceptibility [10, 11].

All types of undernutrition were significantly predicted by household wealth, but multidimensional undernutrition was almost twice as common in households with low incomes (RRR = 1.90; p < .001). Notably, overweight was concentrated among children from wealthier households, highlighting India’s double burden of malnutrition [12]. Similarly, low maternal education was significantly associated with higher risk of all forms of undernutrition, supporting the well-known connection between maternal knowledge and child feeding practices [13].

Remarkably, better cleanliness by itself predicted a lower incidence of wasting and multiple undernutritions but was not substantially linked to stunting or underweight. This implies that the prevention of acute nutritional shocks may be more strongly impacted by sanitation [14].

Better nutritional outcomes were positively correlated with the frequency of antenatal care (ANC) visits. Prenatal contact may be an invaluable opportunity for maternal nutrition counseling, IFA supplementation, and health behavior promotion, as children whose mothers attended four or more ANC visits were more likely to be overweight and well-nourished [15]. On the other hand, there was a gradient effect in birth order, with children of higher birth order having far higher risks of multiple undernutritions, probably as a result of home resource dilution [16].

After controlling for confounders, our analysis revealed no statistically significant correlation between Minimum Dietary variety (MDD) and nutritional status, despite the increasing focus on dietary variety as a global IYCF indicator. Still, more than 90% of undernourished children ate less than five food groups, indicating a general lack of adequate nutrition.

A unique finding revealed from the wealth-MDD interaction analysis which showed MDD was protective only in wealthier households. While meeting the MDD criteria, poor children were more likely to suffer from undernutrition, particularly multiple undernutrition (RRR = 2.36; p <.001). This study indicates only dietary diversity not be enough to avoid malnutrition in settings with limited resources, where overall food intake, meal frequency, and underlying food insecurity could weigh down the benefits of dietary quality. But in wealthy families, MDD seems to offer neutral or even protective benefits, especially when it comes to stunting (RRR = 0.72; p =.014).

These results are consistent with recent research warning against using MDD alone as a stand-in for adequate nutrition in susceptible groups [17]. Instead, a multi-sectoral approach like targeting food security, maternal literacy, sanitation, and antenatal care is crucial for addressing multiple forms of child malnutrition. The study also supports the argument made by de Onis et al. that dietary indicators should be explaining within the socioeconomic context [4]. The integration of MDD into India’s nutritional surveillance must therefore consider economic disparities, food environments, and regional dietary patterns.

Strengths and Limitations: A key advantage of this study is to use of nationally representative, biomarker-based anthropometric data from NFHS-5. Unbiased sub-group analysis is made possible by the high sample size. However, this study depends on a 24-hour dietary recall, which might limit the finding.

Conclusion

Undernutrition among children in Uttar Pradesh is primarily shaped by socio-economic disadvantages, including poverty/low maternal education, rather than dietary diversity alone. While low dietary diversity is prevalent among undernourished children, its impact on nutritional status is especially prominent in poorer households, as the significant interaction found between wealth and minimum dietary diversity. Children from disadvantaged backgrounds who do not meet minimum dietary diversity thresholds face a significantly higher risk of multiple forms of undernutrition.

Policy recommendation: The government ongoing programs like Integrated Child Development Services (ICDS)/Poshan Abhiyan have made important advancement by providing supplementary nutrition and promoting health services. However, these programs alone are insufficient to address the multifaceted nature of undernutrition. The findings justify and reinforce the need for these policies to develop by emphasizing not only the quantity but also the quality/diversity of diets, promote maternal education and reduce nutritional disparities in Uttar Pradesh.

Acknowledgements:

We sincerely acknowledge Mahatma Jyotiba Phule Rohilkhand University, Bareilly, U.P. for providing the academic support and guidance and resources extended throughout the research.

References

1. International Institute for Population Sciences (IIPS) and ICF. (2021). National Family Health Survey (NFHS-5), 2019–21: India (Volume I). Mumbai: IIPS. https://dhsprogram.com/pubs/pdf/FR375/FR375.pdf
2. UNICEF. Stop stunting in South Asia: A common narrative on maternal and child nutrition. 2019. Available from https://www.unicef.org/rosa/sites/unicef.org.rosa/files/2019-11/StopStuntinginSouthAsia-ACommonNarrativeonMaternalandChildNutrition_UNICEF.pdf
3. World Health Organization. Indicators for assessing infant and young child feeding practices: Part 1. 2008. Available from https://apps.who.int/iris/bitstream/handle/10665/43895/9789241596664_eng.pdf
4. de Onis M, Garza C, Onyango AW, Rolland-Cachera MF and le Comité de nutrition de la Société française de pédiatrie. Les standards de croissance de l'Organisation mondiale de la santé pour les nourrissons et les jeunes enfants [WHO growth standards for infants and young children]. Archives de Pédiatrie. 2009;16(1):47–53. https://doi.org/10.1016/j.arcped.2008.10.010
5. Rajan S, Sawant R, Faas S. Dietary diversity and malnutrition among children under 5 in India: DHS Working Papers. The DHS Program. 2024. Available at https://dhsprogram.com/pubs/pdf/WP201/WP201.pdf
6. Chatterjee K, Sinha R, Kumar P. Finding sustainable solutions for childhood under-nutrition in India: An assessment of association of childhood under-nutrition with multiple factors. International Journal of Community Medicine and Public Health. 2022;9(5):2214–2224.
7. The DHS Program. Guide to DHS Statistics: DHS-7 (Version 2). 2021. pp. 11.25–11.28. Rockville, MD: ICF. Available at https://www.dhsprogram.com/pubs/pdf/DHSG1/Guide_to_DHS_Statistics_DHS-7.pdf
8. Gunnal G, Bagaria D, Roy S. Regional patterns in minimum diet diversity failure and associated factors among children aged 6–23 months in India. Natl Med J India. 2024;37:181–90. DOI: 10.25259/NMJI_241_2023
9. Victora CG, de Onis M, Hallal PC, Blössner M, Shrimpton R. Worldwide timing of growth faltering: revisiting implications for interventions. Pediatrics. 2010 Mar;125(3):e473-80. doi: 10.1542/peds.2009-1519.
10. Kanjilal B, Mazumdar PG, Mukherjee M, Rahman MH. Nutritional status of children in India: Household socio-economic condition as the contextual determinant. International Journal for Equity in Health. 2010;9(1):19. https://doi.org/10.1186/1475-9276-9-19
11. Corsi DJ, Mejía-Guevara I, Subramanian SV. Risk factors for chronic undernutrition among children in India: Estimating relative importance, population attributable risk and fractions. Social Science and Medicine. 2016;157:165–185. https://doi.org/10.1016/j.socscimed.2015.11.014
12. Nguyen PH, Scott S, Headey D, Singh N, Tran LM, Menon P, Ruel MT. The double burden of malnutrition in India: Trends and inequalities (2006-2016). PLoS One. 2021 Feb 25;16(2):e0247856. doi: 10.1371/journal.pone.0247856.
13. Tariqujjaman M, Hasan MM, Mahfuz M, Hossain M, Ahmed T. Association between Mother's Education and Infant and Young Child Feeding Practices in South Asia. Nutrients. 2022 Apr 5;14(7):1514. doi: 10.3390/nu14071514.
14. Humphrey JH. Child undernutrition, tropical enteropathy, toilets, and handwashing. The Lancet. 2009;374(9694):1032–1035. https://doi.org/10.1016/S0140-6736(09)60950-8
15. Alhassan Y, Otiso L, Okoth L et al. Four antenatal care visits by four months of pregnancy and four vital tests for pregnant mothers: impact of a community-facility health systems strengthening intervention in Migori County, Kenya. BMC Pregnancy Childbirth 2024;24:224. https://doi.org/10.1186/s12884-024-06386-2.
16. Datar A, Jacknowitz A. Birth weight effects on children's mental, motor, and physical development: Evidence from twins data. Maternal and Child Health Journal. 2009;13(6):780–794. https://doi.org/10.1007/s10995-009-0461-6
17. de Oliveira Otto MC, Anderson CAM, Dearborn JL, Ferranti EP, Mozaffarian D, Rao G, Wylie-Rosett J, Lichtenstein AH; American Heart Association. Behavioral Change for Improving Health Factors Committee of the Council on Lifestyle and Cardiometabolic Health and Council on Epidemiology and Prevention; Council on Cardiovascular and Stroke Nursing; Council on Clinical Cardiology; and Stroke Council. Dietary Diversity: Implications for Obesity Prevention in Adult Populations: A Science Advisory From the American Heart Association. Circulation. 2018 Sep 11;138(11):e160-e168.