Atmospheric pressure, temperature, and online search activity for epilepsy: A daily time-series study

Meteorological conditions have been investigated as potential short-term correlates of epileptic seizures, but findings remain inconsistent. This study examined whether daily meteorological variation was associated with population-level online information-seeking for epilepsy.

A daily time-series study was conducted in Lombardy, Italy, from September 1, 2021, to September 2, 2026. The primary outcome was a reconstructed daily Google Trends series for the Topic Epilessia (epilepsy). Daily mean sea-level pressure and 2-m temperature derived from ERA5 were the primary exposures. Distributed lag nonlinear models evaluated cumulative associations over lags 0-7 days, adjusting for long-term and seasonal trends, day of week, and national holidays. Newey-West standard errors accounted for residual autocorrelation.

Among 1822 days with complete exposure histories, lower mean sea-level pressure (5th percentile, 1003.5 hPa) versus the median (1016.8 hPa) was associated with higher cumulative epilepsy-related search activity (difference, 8.98 units; 95% confidence interval [CI], 1.66-16.31; p = 0.016). No clear associations were observed at higher pressure levels or at low or high temperature. The association remained evident in several prespecified sensitivity analyses, including a log-transformed outcome, but was attenuated in the positive-only analysis and was not replicated using the simple search term epilessia. Secondary analyses showed no clear associations for relative humidity or precipitation, but an inverse association with high wind speed was found.

Lower mean sea-level pressure was associated with increased population-level epilepsy-related Google search activity. The observed association corresponded to a 13.3-hPa reduction in atmospheric pressure, a difference commonly encountered during the passage of low-pressure weather systems. Although the direction of the association is broadly consistent with some studies reporting increased seizure occurrence during periods of lower atmospheric pressure, Google Trends reflects information-seeking behavior rather than clinical outcomes. Therefore, these findings should be regarded as hypothesis-generating and should not be interpreted as evidence that atmospheric pressure influences seizure occurrence.