Exposure to elevated temperatures and risk of preterm birth in Valencia, Spain.

Abstract Background Prematurity is the second-leading cause of death in children under the age of 5 worldwide. It is predicted that the future climate will have more intense, longer lasting and frequent extreme heat episodes, and so the temperature effect on the risk of preterm birth is generating considerable interest in the public health field. Our aim was to explore the potential short-term effects of elevated temperatures on the risk of preterm birth in Valencia (Spain). Methods All singleton natural births born in the metropolitan area of Valencia during the warm season (May–September, 2006–2010) were included ( N =20,148). We applied time-series quasi-Poisson generalized additive models to evaluate the risk of preterm birth at different maximum apparent and minimum temperature values (50th, 90th and 99th percentiles of the warm season) up to 3 weeks before delivery (reference: overall annual median value). In addition, three temperature-interval-specific estimates were obtained for changes between each of these temperature values. We took into account the pregnancies at risk adjusted by the gestational age distribution of the set in each day. We used distributed-lag non-linear models with a flexible function in the shape of the relationship and lag structure. Results Risk of preterm birth increased up to 20% when maximum apparent temperature exceeded the 90th percentile two days before delivery and 5% when minimum temperature rose to the 90th percentile in the last week. Differences between interval-specific risk estimates across lags were observed. Conclusion Exposure to elevated temperatures was associated with an increased risk of preterm birth in the following three weeks.