NFHS-6: Correlation Analysis & Policy Actionables

The National Family Health Survey (NFHS) is a large-scale, multi-round survey conducted across India to gather vital data on population, health, and nutrition. It is steered by the Ministry of Health and Family Welfare (MoHFW) and coordinated by the International Institute for Population Sciences (IIPS) in Mumbai.

This report analyzes the relationships between 101 health, social, and economic indicators across 35 states and Union Territories in India using data extracted from the National Family Health Survey (NFHS-6, 2023-2024). The objective is to identify correlated parameters to help target policy interventions that yield the highest multiplier effects.

Key Actionable Policy Insights

By analyzing cross-category correlations, four major actionable pathways for state interventions have been identified: 

1. Technology & Digital Inclusion as a Driver of Child Nutrition

Women's Mobile Phone Ownership (Indicator 93) vs Child Underweight (Indicator 72) | r = -0.806 (NFHS-6) Insight: Mobile phone ownership is the strongest single proxy and driver of household empowerment. Access to mobile phones correlates with dramatically lower rates of undernourished children. Policy Actionable: Leverage digital platforms (like Poshan Tracker, WhatsApp, SMS) to deliver nutrition/health advisory directly to mothers. Subsidizing mobile connectivity for rural women is a high-multiplier health intervention.

2. Female Education is the Single Best Contraceptive & Shield against Early Marriage

Women with 10+ Years of Schooling (Indicator 12) vs Marriage before 18 (Indicator 16) | r = -0.754 (NFHS-6) Insight: Higher female secondary education strongly suppresses early marriage. Consequently, early marriage correlates with teenage pregnancy (r = 0.904) and stunting (r = 0.530). Policy Actionable: Direct resources to ensure girls complete secondary education (10+ years) through stipends, bicycle distribution, and safe schools. This creates a cascade effect of lowering fertility rates and improving child health.

3. Addressing Domestic Violence directly improves Child Underweight Rates

Spousal Violence Experience (Indicator 95) vs Child Underweight (Indicator 72) | r = 0.575 (NFHS-6) Insight: Domestic violence is strongly linked to child malnutrition. States with higher domestic violence have significantly higher child underweight rates, likely due to poor maternal mental health and household neglect. Policy Actionable: Integrate domestic violence screening and support services into maternal and child health clinics (such as during antenatal check-ups). Protecting mothers directly protects children.

 

4. The Obesity-Caesarean Link: A Maternal Care Warning

Births Delivered by Caesarean Section (Indicator 38) vs Female Obesity (Indicator 76) | r = 0.807 (NFHS-6) Insight: There is a massive correlation between women's obesity and C-section rates. This is true cross-sectionally and dynamically—states with the fastest-growing obesity rates are seeing the fastest-rising C-section rates (change correlation r = 0.743). Policy Actionable: Promote lifestyle, physical activity, and nutritional counseling in prenatal programs. Manage non-communicable disease (NCD) risks during pregnancy to curb the rise of medically unnecessary C-sections.

Visualizing Key Clusters (Heatmaps)

To ensure legibility, indicator texts have been mapped to their standard indices and concise descriptors. The generated heatmaps are embedded below with adjusted aspect ratios and margins to ensure clear text representation:

Cluster 1: Maternal Care & Education vs Child Malnutrition

This heatmap displays how maternal schooling and the quality of prenatal/postnatal care correlate with stunting, wasting, and underweight rates. Female schooling (r = -0.557) is key protector.

 

Cluster 2: Women's Empowerment vs Social/Health Outcomes

This heatmap displays the relationships between indicators of female agency (schooling, HH decision making, bank account usage, internet and mobile phone access) and critical outcomes like early marriage (r = -0.621 with mobile) and child underweight (r = -0.806 with mobile).

 

Cluster 3: Child Vaccinations vs Disease & Diet

This heatmap looks at how child immunizations relate to diarrhoea, respiratory infections, and dietary adequacy. Early postnatal care is strongly correlated with establishing vaccine adherence (r = 0.784 with measles-containing vaccine).

Top Cross-Category Correlations

Below are the tables showing the strongest non-trivial correlations between indicators belonging to different categories.

Cross-Sectional Relationships (NFHS-6)

r Value	Indicator 1	Indicator 2
-0.874	75. Men whose BMI is below normal (%)	93. Women having a mobile phone that they use (%)
-0.855	74. Women whose BMI is below normal (%)	93. Women having a mobile phone that they use (%)
+0.839	72. Children under 5 who are underweight (%)	74. Women whose BMI is below normal (%)
+0.807	38. Births delivered by caesarean section (%)	76. Women who are overweight or obese (%)
-0.806	72. Children under 5 who are underweight (%)	93. Women having a mobile phone that they use (%)
-0.754	12. Women with 10 or more years of schooling (%)	16. Women married before age 18 (%)
+0.575	95. Ever-experienced spousal violence (%)	72. Children under 5 who are underweight (%)

Summary of Key Interventions

• Secondary Education: Ensure girls continue in school past the 10th grade to solve early marriage, high fertility, and childhood stunting.
   
• Mobile Technology: Ensure mobile phone ownership and access for women to create a direct channel for digital health and nutritional guidance.
  
  
• Integrated Clinics: Co-locate family planning, maternal mental health/domestic violence counseling, and child nutrition clinics so that progress in one category feeds into the others.
  
  
• Obesity Prevention: Screen and target gestational weight gain and adult obesity to check the rising incidence of C-sections.

Note: This is entirely a correlation exercise, and it's understood that the there might be no causation linkages between the two parameters. For going ahead with the key interventions, a study can be done to examine causation linkages or other existing studies can be relied upon. For the purpose of these recommendations, based on the anecdotal evidences and other studies available, the above key interventions are being suggested.