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

PLOS One | https://doi.org/10.1371/journal.pone.0335529 November 26, 2025 10 / 17 vulnerability, and are statistically significant, though the strength of the association is lower at a coefficient value of 0.009, indicating a small increase in the log-odds. All associations are statistically significant at 1% significance level (p values<0.01). The analysis also reveals that there is variation in the extent of the association of climatic vulnerability and health outcomes. It is to be noted that these results hold after introducing other determinants of health outcomes in the estimation. The signs and significance levels obtained on other explanatory variables confirm to expectations in almost all the cases. For instance, the negative and significant association between mothers/women’s education and adverse health outcomes are well-established. As the level of education increases, the likelihood of an adverse health outcome reduces in all cases, with the coefficients being negative (and statistically significant) for stunting (−0.36), wasting (−0.18), underweight (−0.38), non-institutional deliveries (−0.78) and facing problems in accessing healthcare (−0.13). Similarly, receiving prenatal care from a doctor significantly reduces the log-odds of an individual being stunted, wasted, underweight and having non- institutional deliveries. Climatic vulnerability clearly plays a role in determining health outcomes, with high vulnerability levels being associated with more adverse outcomes.

A sensitivity analysis was conducted by using alternative groupings of vulnerability for each health outcome (dependent variable). Regressions were run using all 5 categories, and with alternative groupings across vulnerability categories. Very similar results are obtained for outcomes on stunting, wasting, underweight, and non-institutional deliveries. The signs of all explanatory variables remain the same (correctly signed) and significant with the outcome variables demonstrating an inverse relationship with climatic vulnerability, which intensifies with higher magnitudes of climatic vulnerability. For the access to health care outcome, the transportation availability variable was found to be significantly and highly correlated (chi square test) with the climatic vulnerability variable and the availability of a person to accompany the woman in an alternative grouping, which could be taken care of by dropping the transportation availability variable. The results from the alternative regressions confirm that our results are robust and reliable. (S7 Table, S8 Table and S9 Table present results from the sensitivity analysis). Fig 3 provides a graphical representation of the results, using the odds ratios from the estimations. The dots repre- sent the odds ratios and the red lines show the corresponding 95% confidence intervals. The odds ratio for the outcome variables are greater than 1. This indicates that individuals living in highly vulnerable districts have a higher likelihood Table 4. Regression results for problems with access to healthcare.

SDG Health Outcomes (Dependent Variables) Problems with Access to Healthcare Explanatory Variables Climatic Vulnerability 0.009** (0.00) Woman’s Education −0.13** (0.00) Woman’s age −0.00** (0.00) Current marital status 0.02**(0.00) Problem in getting a person to accompany the woman to a health facility 1.99**(0.00) Problem in accessing transportation for the travel to a health facility 2.40**(0.00) Constant 0.10**(0.00) LR chi2 1.50e + 11** Observations 447348 Notes: Standard errors are provided in parentheses. The significance levels are marked as follows: * denotes 5% and ** denotes 1% level of significance (* p < 0.05, ** p < 0.01) https://doi.org/10.1371/journal.pone.0335529.t004