156 Chapter 7 Subgroup analyses – WHAT-hormone subgroup In the WHAT-hormone subgroup the ovulatory window did not show an increased migraine attack incidence compared to all other cycle days (OR: 0·68, 95% CI: 0·38-1·16, p=0·17) (Table 5). Table 5. Results of the mixed logistic regression model in the WHAT-hormone subgroup with the first day of a migraine attack as the dependent variable and the ovulatory window and perimenstrual window as fixed effects. Data from 58 menstrual cycles from 31 patients with migraine are shown. Women (n) Migraine attacks (n) Crude OR (95% CI) Adjusted ORa (95% CI) Outside ovulatory & menstrual window 31 79 1.00 (ref) 1.00 (ref) Ovulatory window 31 16 0.68 (0.38 – 1.15) 0.68 (0.38 – 1.16) Menstrual window 31 41 1.92 (1.28 – 2.86)** 1.92 (1.28 – 2.86)** a Adjusted ORs account for within-subject variability by including the participant as random effect. OR = odds ratio; CI = confidence interval; levels of significance = *≤0.05, **≤0.005, ***≤0.001 The case-crossover design yielded an OR of 0·56 (95% CI: 0·33-0·96, p=0·03) , indicating a lower odds of migraine attacks in the ovulatory window compared to all other cycle days. When days within the perimenstrual window were removed from the analysis to solely investigate the effect of ovulation compared to non-perimenstrual days, the OR attenuated to 0·66 (95% CI: 0·38-1·15, p=0·28) (Table e-1). In the SCCS model, migraine attack incidence was also not significantly higher during the ovulatory window when compared to the follicular phase (OR: 0·60, 95% CI: 0·34 – 1·03, p=0·07), or when compared to to the luteal phase (OR: 1·00, 95% CI: 0·52- 1·94, p=0·99) (Table e-2). Subgroup analyses – menstrual migraine group An additional subgroup analyses was conducted for all women who met the criteria for menstrual migraine.20 Of the 554 participants, 234 women (42%) met the criteria for menstrual migraine and 45 women (8%) could not be diagnosed due to less than three cycles of e-diary data. Within the menstrual migraine subgroup, the ovulatory window was also not associated with an increased migraine attack incidence (Table e-3). In the menstrual migraine subgroup the mixed model yielded an OR of 0·95 (95% CI: 0·87 - 1·05, p=0·32) compared to all other cycle days, which is similar to the results found in the entire cohort (OR: 0·95, 95% CI: 0·88-1·02, p=0·13).
RkJQdWJsaXNoZXIy MjY0ODMw