The impact of climatic factors on headache patterns: a 14-year time series analysis of cluster and tension-type headaches in primary care.
Cuenca-Zaldívar Juan Nicolás JN, Corral Del Villar Carmen C, García Torres Silvia S, Araujo Zamora Rafael R et al.
Weather-related influences on pain and neurological symptoms are widely discussed in biometeorology. However, the long-term impact of climatic variables on primary care headache consultations remains unclear. This study aimed to examine whether climatic variables are associated with consultations for cluster headaches, tension-type headaches (TTH), and unspecified headaches over 14 years. Data (2010-2023) were extracted from medical records using International Classification of Primary Care, Second Edition (ICPC-2) codes. Meteorological variables (temperature, rainfall, wind direction, barometric pressure, and sunshine hours) were obtained from the State Meteorological Agency. Time-series analyses used Exponential Smoothing State Space Model with External Regressors (ETSX) and AutoRegressive Integrated Moving Average Models with External Regressors (ARIMAX) using age, sex, and meteorological factors as external regressors. Model accuracy was evaluated using the Root Mean Squared Error (RMSE), Symmetric Mean Absolute Percentage Error (SMAPE), and Mean Absolute Scaled Error (MASE). A total of 5127 headache consultations were analyzed (mean age, 45.0 ± 19.9 years; 68.9% female). ETSX models best fit the overall, unspecified, and cluster headache series, whereas the ARIMAX model provided optimal performance for TTH. Sex was the strongest predictor across all models. In TTH, female sex increased consultations (p < 0.001), whereas higher temperature (p = 0.031) and wind direction (cosine component; p = 0.027) were associated with fewer consultations. For cluster headache, male sex was associated with fewer consultations (p = 0.020), and no climatic variables showed a significant association. Meteorological variables were not independently associated with unspecified headache consultations. Climatic variables showed limited, subtype-specific associations, with only temperature and wind direction independently associated with TTH. No weather variables predicted cluster headache or unspecified consultations. Demographic factors appeared to be more strongly associated with healthcare utilization than climatic variables, although ETSX and ARIMAX models may support forecasting and resource planning.