This report replicates the analysis of zero-food prevalence among children aged 6–23 months across Indian states, using a simulated dataset modeled after the Comprehensive National Nutrition Survey (CNNS) 2016–2018. The analysis estimates the proportion of children who consumed no solid or semi-solid food in the past 24 hours, highlighting regional differences in child feeding practices.
# Simulate CNNS-like data for demonstration (replace with actual CNNS data)
set.seed(123)
states <- c("Uttar Pradesh", "Kerala", "Tamil Nadu", "Gujarat", "Bihar",
"Maharashtra", "Punjab", "Assam", "West Bengal", "Karnataka")
cnns_data <- data.frame(
child_id = 1:2000,
age_months = sample(6:23, 2000, replace = TRUE),
total_food_groups_consumed = sample(0:7, 2000, replace = TRUE,
prob = c(0.1, 0.15, 0.2, 0.2, 0.15, 0.1, 0.05, 0.05)),
state = sample(states, 2000, replace = TRUE)
)
# Filter children aged 6-23 months (already filtered in simulation)
cnns_filtered <- cnns_data %>%
filter(age_months >= 6 & age_months <= 23)
# Create zero-food indicator
cnns_filtered <- cnns_filtered %>%
mutate(zero_food = ifelse(total_food_groups_consumed == 0, 1, 0))
state_prevalence <- cnns_filtered %>%
group_by(state) %>%
summarise(
n = n(),
zero_food_count = sum(zero_food),
prevalence = mean(zero_food)
) %>%
arrange(desc(prevalence))
knitr::kable(state_prevalence, digits = 2, caption = "Zero-Food Prevalence by State")
| state | n | zero_food_count | prevalence |
|---|---|---|---|
| West Bengal | 204 | 29 | 0.14 |
| Karnataka | 184 | 23 | 0.12 |
| Kerala | 209 | 26 | 0.12 |
| Assam | 194 | 24 | 0.12 |
| Tamil Nadu | 196 | 22 | 0.11 |
| Gujarat | 194 | 19 | 0.10 |
| Bihar | 203 | 18 | 0.09 |
| Uttar Pradesh | 221 | 17 | 0.08 |
| Maharashtra | 185 | 14 | 0.08 |
| Punjab | 210 | 15 | 0.07 |
ggplot(state_prevalence, aes(x = reorder(state, prevalence), y = prevalence)) +
geom_col(fill = "steelblue") +
coord_flip() +
labs(title = "Zero-Food Prevalence Among Children Aged 6–23 Months by State",
x = "State",
y = "Prevalence (Proportion)") +
theme_minimal()
This analysis shows variation in zero-food prevalence across Indian states using simulated CNNS-style data. Real CNNS data should be used to obtain accurate values. The current simulation demonstrates how to calculate and visualize prevalence using real-world methods.
CNNS provides valuable insights into child nutrition in India. With access to actual CNNS microdata, this analysis can be refined to guide real policy actions on improving infant and young child feeding practices.