Nonlinear and lagged association between ambient temperature and pre-weaning piglet mortality in Cuba
Related Experiment Video
Updated: Nov 7, 2026

Use of a Piglet Model for the Study of Anesthetic-induced Developmental Neurotoxicity (AIDN): A Translational Neuroscience Approach
Published on: June 11, 2017
O Fonseca-Rodríguez1,2, E A Curbelo-Benítez3, F A Morales-Calvo1,2
1Unitat mixta d'Investigació IRTA-UAB en Sanitat Animal, Centre de Recerca en Sanitat Animal (CReSA), Campus de la Universitat Autònoma de Barcelona (UAB), Bellaterra, Catalonia, 08193, Spain.
Abstract:
Climate change is increasing heat exposure, yet epidemiological evidence on temperature-related mortality in pre-weaning piglets remains limited, particularly in humid tropical production systems. We assessed the immediate and delayed association between daily mean ambient temperature and pre-weaning piglet mortality using records from 14 commercial farms in 12 Cuban municipalities during 2022-2024 (17,599 deaths). Farms mainly use open or semi-open naturally ventilated maternity buildings with limited active microclimate control. Municipality-level temperature and relative humidity were obtained from ERA5-Land. We applied a space-time-stratified case-crossover design with conditional quasi-Poisson regression and a distributed lag non-linear model over lags 0-7 days, adjusting for relative humidity. The cumulative temperature-mortality association was non-linear and strongly heat-dominated, with a model-derived minimum-mortality temperature (MMT) of 23.7 °C; the increase at the extreme cold tail was less precise. Effects were predominantly acute. At lag 0, the estimated RRs were elevated at both the maximum observed temperature of 31.2 °C (RR 1.49, 95% CI 1.06-2.09) and the minimum of 14.3 °C (RR 2.26, 95% CI 1.27-4.02); however, cold effects attenuated rapidly and contributed little to the overall burden. The estimated temperature-attributable fraction was 16.6%, with most of the burden associated with temperatures above the MMT (15.5%). Most heat-attributable deaths occurred on non-extreme warm days above the minimum-mortality temperature. Exposure to temperatures at or above the 99th percentile (P99) of the farm-day temperature distribution accounted for 0.7%. These findings support routine heat-risk management in farrowing environments to help reduce pre-weaning mortality and strengthen climate resilience in tropical swine systems.
