Other researchers disagree
There are other doctors / researchers who do not agree with the hypotheses or conclusions of this study.
Dr. Stephen Silberstein, director of the Jefferson Headache Center in Philadelphia, didn't mince words:
"This study doesn't prove anything. It's not that I don't believe the results, it's that the results are not believable."3
Dr. Ellen Drexler of Maimonides Medical Center in Brooklyn, New York, questions the conclusions and the premise that female Migraineurs have lower estrogen levels. She commented:
"Migraine brains are more sensitive to many exogenous and endogenous factors, of which falling estrogen levels are an important one for many female Migraineurs,.... However, female Migraineurs are not known to have consistently lower levels of estrogen than are non-Migraineurs... genetic factors may also be playing a role, as well as exogenous factors such as medication use, smoking and alcohol use, which may vary between Migraineurs and non-Migraineurs..."3
Summary and comments
Although the authors of this study make bold statements that Migraine lowers the risk of breast cancer by 30%, other researchers are far from certain and far from accepting either the hypothesis or the conclusion.
It is not clear how accurate their statistics regarding the number of Migraineurs in the study are. They relied upon self-reporting by the study participants. In other words, they determined whether or not a participant was a Migraineur based only on what the patient said. Migraine is often misdiagnosed and often goes undiagnosed. Thus, since no validation was performed regarding participants' Migraine status, there is no way of knowing how many study participants actually were Migraineurs.
The researchers put too much emphasis on the role of estrogen fluctuation among their participants who supposedly had Migraine. Yes, estrogen fluctuations around menses, pregnancy, and menopause are triggers for some women. However, this does not hold true for all women. For some women, those estrogen fluctuations are not a trigger. Again, this makes it difficult to consider the data valid.
