Health SDGs are at risk from climate change: Evidence from India

PLOS One | https://doi.org/10.1371/journal.pone.0335529 November 26, 2025 8 / 17 provide the VIF values). Sampling weights from the NFHS were applied to all the regressions. The analysis is pri- marily focused on individual level determinants and avoids household level variables such as sanitation coverage and drinking water sources. There are also estimation concerns related to some of the variables that are otherwise often considered to have an influence on child health. For instance, household wealth is not used as it tends to be correlated with income, and the latter is already used in computing vulnerability. Information on variables related to dietary diversity and immunization is usually available for a sub-sample (for example, ages 6–23 months) or is heter- ogeneously specified (e.g., on diet) and inclusion of these would have led to reduction of the observations available for the analysis. The coefficients for the climatic vulnerability variable indicate the log-odds of observing the associated health out- come in highly vulnerable districts as compared to the other districts. For example, in stunting (estimation 1, Table 3), the climatic vulnerability coefficient is 0.14, indicating a positive association, where individuals in highly vulnerable Table 2. Descriptive data on explanatory variables from NFHS. Explanatory Variables Level Percentage Distribution (%) Explanatory Variables for Stunting, Wasting, being Underweight and Non-Institutional deliveries (number of observations = 154547) Age of Child Less than 1 year (less than 12 months) 20.34

1 years (12–23 month) 19.38 2 years (24–35 months) 19.60 3 years (36–47 months) 19.91 4 years (48–59 months) 20.77 Sex of Child Female children 48.21 Male children 51.79 Birth Order of the child* Birth order 1 37.16 Birth order between 2–3 48.53 Birth order greater than 3 14.32 Mother’s Education Up to primary 38.22 Higher than primary 61.78 Mother had received prenatal care from a doctor during pregnancy Not Received 46.31 Received 56.39 Number of antenatal visits during pregnancy Less than four visits 46.03 Four or more visits 53.97 Explanatory variables for ‘Problems with Access to Healthcare’ (number of observations = 447348) Marital Status Not Married/others 28.05 Married 71.95wh Educational attainment Up to primary 38.83 Higher than primary 61.17 Nature of the problem in getting a person to accompany the woman to a health facility No problem 45.85 Small problem 34.21 Big problem 19.94 Nature of the problem in accessing trans- portation for the travel to a health facility No problem 36.75 Small problem 35.39 Big problem 27.87 Notes: 1. * The birth order of the child is the chronological position of the child in the birth schedule for a mother 2. The KR file (for children) is used for data related to stunting, wasting, underweight, and non-institutional deliveries, while the IR file (for individual women) was used for data related to problem in access to healthcare. Consequently, the number of observations and the descriptive statistics on variables differs for the two estimations. https://doi.org/10.1371/journal.pone.0335529.t002