The impact of low temperatures on mortality in Vojvodina (2000-2020): a quantitative analysis of cold spells

This study investigates the impact of low temperatures and cold spells on mortality in the Autonomous Province of Vojvodina, Serbia, during the 2000-2020 winter seasons. While previous research in the region has predominantly focused on heat-related mortality, this study provides a comprehensive quantitative analysis of cold-related mortality using daily data and a multi-method statistical approach. Cold days were defined as those with an average temperature below 0 °C, and cold spells were defined as periods with at least three consecutive days with minimum temperatures at or below - 10 °C. The results confirm a statistically significant increase in mortality during both cold days and cold spells. Nonlinear models, particularly LOESS and quadratic regression, identified two inflexion points in mortality response: at - 3.5 °C (onset of increase) and - 7 °C (marked escalation), with the highest mortality observed below - 13 °C, though based on fewer observations. The analysis of temporal lags revealed that mortality tends to increase between 2 and 9 days following cold exposure, with the most pronounced effects occurring during cold spells. Compared to Poisson regression, which failed to capture these associations, nonlinear methods more accurately reflect cold-related mortality dynamics. The study also proposes that - 7 °C is the most reliable operational threshold for early warning systems, given its consistency across models and statistical significance. These findings contribute to understanding the health effects of cold weather in Eastern Europe and may inform the development of targeted public health interventions.