Ambient temperature and child nutritional status of more than 6 million children in Brazil: a cohort study

Background

Evidence regarding the association between temperature and children's nutritional status is scarce. We aimed to estimate the relationship between ambient temperature and the weight and height of Brazilian children aged 12–59 months.

Methods

This large longitudinal cohort study from all 5570 Brazilian municipalities used data from the 100 Million Brazilians Cohort linked with data from the Information System on Live Births (SINASC), Food and Nutrition Surveillance System (SISVAN), and temperature data from the Brazilian Daily Weather Gridded Data (BR-DWGD). The study included children aged between 12–59 months with nutritional monitoring recorded by SISVAN between Jan 1, 2008 and Dec 31, 2017. We applied distributed lag non-linear models to investigate the cumulative influence of temperature over the year preceding each child’s measurement on weight and height (ie, lag: 0–51 weeks). Subgroup analyses were conducted to identify population groups facing greater risk of worse nutritional outcomes.

Findings

6 498 546 children aged between 12–59 months were included in the study. 3 226 520 (49·65%) participants were male, 3 272 026 (50·35%) were female, and 4 194 195 (64·54%) were Parda or Brown ethnicity. For each 1°C increase in ambient temperature above the median (26°C), the cumulative odds over the weekly lags 0–51 increased by 10·0% (odds ratio [OR] 1·10 [95% CI 1·099–1·103]) for underweight (low weight-for-age), 8·0% (OR 1·08 [1·078–1·081]) for wasting (low weight-for-height), and 8·0% (OR 1·08 [95% CI 1·078–1·080]) for stunting (low height-for-age) in Brazil. In stratified analyses, the odds of malnutrition associated with a 1°C increase in ambient temperature were highest in the North (followed by the Northeast, the two poorest regions of Brazil), in municipalities with higher deprivation index, in rural areas, and among children of Indigenous mothers.

Interpretation

Children experiencing high temperatures in Brazil are more likely to suffer from malnutrition, with worse outcomes observed among more vulnerable social groups. Our results highlight the deep environmental injustice in Brazilian society, with the influence of temperatures interacting with existing social and economic inequalities and exacerbating nutritional outcomes in the most vulnerable groups. Further studies are needed to investigate causal pathways and identify strategies capable of mitigating the effects of climate change on child nutrition.