Factors Associated with Migraine Attack Duration in Adults: A Cross-Sectional Study

Abstract

Abstract Introduction Migraine is a common neurological disorder with attacks lasting 4–72 h. Although duration is a key diagnostic criterion, the clinical and demographic factors associated with its variability are not well defined. Identifying temporal phenotypes may improve diagnostic accuracy and inform individualized management. Methods In this cross-sectional observational study, we evaluated 694 patients fulfilling ICHD-3 migraine criteria. Patients were categorized by attack duration into short (4–12 h), medium (13–24 h), and long (>24 h) groups. Sociodemographic features, migraine characteristics, associated symptoms, and disability measures were compared across groups. Multinomial logistic regression was performed using the short-duration group as the reference category to identify variables independently associated with medium- and long-duration attacks. Results Of 694 participants (84% female; mean age 36.66 ± 12.20 years), 33.3% (n = 231) experienced short, 28.2% (n = 196) medium, and 38.5% (n = 267) long attacks. Long-duration attacks were associated with older age (39.98 ± 11.36 vs. 34.09 ± 12.62 years, p < 0.001), higher female ratio (89.1% vs. 78.8%, p = 0.007), and lower educational level (p = 0.008). Patients with prolonged attacks had higher allodynia scores (median ASC 6 vs. 4, p < 0.001), greater pain intensity (VAS 9 vs. 8, p < 0.001), and more frequent nausea and vomiting (p = 0.002). Illness duration was the longest in the long-attack group (median 120 vs. 60 months, p < 0.001). Other clinical characteristics, including aura, localization, and chronic migraine rates, showed no significant differences. In multivariable analysis, older age and higher pain intensity remained independently associated with long-duration attacks, while lower educational level and higher pain intensity were independently associated with medium-duration attacks. Conclusion Migraine attack duration is associated with distinct demographic and clinical characteristics independent of attack frequency. In multivariable analysis, older age and higher pain intensity remained independently associated with long-duration attacks, while lower educational level and higher pain intensity were independently associated with medium-duration attacks. These findings suggest that attack duration could represent a meaningful clinical phenotype; therefore, frequency-based classification approaches might overlook patients with infrequent but prolonged and disabling attacks.