Patients with migraine are right about their perception of temperature as a trigger: time series analysis of headache diary data

Abstract

Background: Researches to date on the association between headache and weather have yielded inconsistent results. Only a limited number of studies have examined the clinical significance of self-reported weather sensitivity. This study aimed to identify the difference in the association of headache with temperature between migraine patients with and without temperature sensitivity.

Methods: 66 migraine patients (75.8 % female; mean age 43.3 ± 12.9 years) provided their 1-year headache diaries from 2007 to a headache clinic in Taipei, Taiwan. 34 patients (51.5 %) reported sensitivity to temperature change but 32 (48.5 %) did not. Time series of daily headache incidence was modeled and stratified by temperature sensitivity. Empirical mode decomposition was used to identify temporal weather patterns that were correlated to headache incidence, and regression analysis was used to examine the amount of variance in headache incidence that could be explained by temperature in different seasons.

Results: Among all migraine patients, temperature change accounted for 16.5 % of variance in headache incidence in winter and 9.6 % in summer. In winter, the explained variance increased to 29.2 % among patients with temperature sensitivity, but was not significant among those without temperature sensitivity. Overall, temperature change explained 27.0 % of the variance of the mild headache incidence but only 4.8 % of the incidence of moderate to severe headache during winter.

Conclusions: This diary-based study provides evidence to link the perception of temperature sensitivity and headache incidence in migraine patients. Those who reported temperature sensitivity are more likely to have headache increase during the winter, particular for mild headaches.

Fig. 1

Flow chart of time series analysis of headache and temperature data. Using empirical mode decomposition method, temperature time series was decomposed into an overall trend and various modes of fluctuations, termed intrinsic mode functions (IMFs). Each IMF represents a specific mode of temperature fluctuations. These IMFs were correlated with headache incidence data by multiple regression analysis to derive the explained variance in headache incidence by temperature