Boston migraine and contraception study
Awarded 2010
Large Research Grants
Deborah Bartz, MD, MPH
Planned Parenthood League of Massachusetts

Background: Over one quarter of reproductive-age women experience migraine, a condition disabling over 12 million American women annually. Hormonal contraception (HC), used by 20 million women, may cause or aggravate migraine, leading to contraceptive discontinuation. However, scientific evidence linking HC and migraine is poor. Objectives: To estimate incidence of headache due to hormonal contraception use in the first three months after method initiation. Methods: In this prospective observational cohort, women initiating estrogen-progesterone, progesterone-only, and nonhormonal contraception provided baseline headache information. Follow-up data on daily headaches and contraceptive adherence was obtained prospectively through the use of an 84-day online headache diary. This daily diary data was collected using an online system based on electronic mail and text messaging technology. Wilcoxen Rank Sum test was used to compare total headache days over 84 days in subjects using estrogen-progesterone or progesterone-only compared to nonhormonal users. We will estimate the prevalence, persistence, clinical impact and attributable risk of migraine due to hormonal contraception. Results: Fifty-seven women in each treatment arm (N=171) were enrolled and 153 (89%) provided complete dairy data sets over three months. Twenty-nine (19%) reported experiencing headaches on a regular basis at enrollment. Twelve (8%) women reported a history of symptoms of migraine without aura and 19 (12%) reported a history of symptoms of migraine with aura at enrollment. The median total headache days over the three months, and associated lower and upper quartiles, for all participants was 7 days [3:14]. The median headache days over three months in nonhormonal contraceptive users was 6 days [3:9] compared to 6 days [2:14] in users of estrogen-progesterone (p=0.57) and compared to 8 days [3:9] in users of progesterone-only contraception (p=0.08). Conclusion: When compared to a proper control group, hormonal contraceptive users experience the same incidence of headache disease as nonhormonal contraceptive users.