PMS (pre-menstrual syndrome) has proven a useful concept in many situations—in jokes, to sell products, to construct legal arguments, etc. It may surprise you, then, that the existence of PMS as a biological phenomenon is actually far from scientifically established. At a basic level, PMS can be defined as negative shifts in emotional, physical, or mental well-being that occur in the week or so before the beginning of one’s menstrual period. This implies that, if PMS is a “real” biological phenomenon, it must have its origins in reproductive physiology. In other words, natural, normal shifts in hormone concentrations or levels that exist during this “premenstrual” time must be related to changes in well-being. Though many researchers have sought such a relationship, there is very little evidence to support its existence. In fact, there isn’t even agreement on which hormones influence symptoms of PMS—estrogen, progesterone, and testosterone have been the most popular culprits, but other hormones have also been studied.
How can it be so difficult to study something so many take for granted?
To study this proposed relationship, we need, of course, to measure both hormone levels and well-being. As you might have guessed from the previous post on the variation in the timing of menstrual events, accurately determining hormone levels is no easy task! Because assessing hormone levels takes a lot of time, specialized laboratory skills, and money, many researchers skip this step. Instead, some have assumed that all women have “regular” 28 day cycles, with ovulation occurring 100% of the time, right smack dab in the middle of that cycle. As we’ve already discussed, that almost never happens. Working from this assumption of “regularity,” some researchers take only a few hormone samples, or none at all, and try to determine the timing of the follicular and luteal phases based on menstrual bleeding dates. For all of the reasons discussed in the previous post, they shouldn’t make these assumptions—there’s too much variation in the timing of these menstrual cycle events, and, without hormonal or other kinds of evidence, we can’t assume ovulation even occurred!
As if measuring hormones wasn’t enough of a problem, other authors have pointed out that the survey methods used to establish well-being are often problematic (Hawes and Oei, 1992). Of course, even if the survey works well, the timing of the survey matters when we’re trying to track changes over time, changes that are thought to be related to underlying physiological functioning. If the timing of the surveys is based on assumptions about that physiological functioning, not the actual functioning, then we have the same problem described above: survey answers might not represent the supposedly time-sensitive events we think they do—and there’s no sure way to know without carefully measuring hormones and looking for ovulatory markers.
So what do we know?
In a series of studies conducted with a large sample of European women (Sanders, Warner, Backstrom, and Bancroft, 1983; Backstrom et al., 1983), there were no hormonal differences between women who complained of PMS-like symptoms and those who did not. Interestingly, women who complained of more PMS-like symptoms had more pregnancies and miscarriages, were more likely to remain home with their children, and were more likely to have had negative reactions to oral contraceptives. Other studies have come to similar findings, which makes it difficult to believe that there is a strong hormonal influence in PMS (see Harris and Vitzthum, 2013, for review of those, too).
Limited cross-cultural evidence indicates that women outside the United States experience changes in well-being during the “premenstrual phase” (which has been very loosely defined). However, as in the U.S. literature, the prevalence and reported symptoms range widely. Interestingly, authors often find that premenstrual symptoms are influenced by things such as race and/or ethnicity, education and stress levels, susceptibility to major depression, and, tellingly, culturally-transmitted fears about menstruation.
Some authors have suggested it may be that, in some women, the degree of changes in hormone levels from one phase of the menstrual cycle to the next is more dramatic than in other women, and that this—the more-dramatic-than-average shift in hormone levels—might induce negative changes in wellbeing. This, however, has yet to be demonstrated unequivocally in reasonably healthy, normally cycling women. Furthermore, studies in which women are given unnaturally high levels of synthetic hormones should be viewed with caution, as we do not know that women’s bodies respond the same way to these chemicals as they do to the ones their own bodies produce.
At this point there is no evidence that PMS is a biologically-induced phenomenon that all, or even most, normal, healthy women experience. This is not to say that some women do not experience changes in well-being around the time of menstruation; it’s just to say that, at this time, we don’t know how common the phenomenon really is, and there’s no evidence that it is the result of normal physiological processes. Perhaps not surprisingly, one prominent sex researcher has concluded that, given the range of symptoms, inconsistencies in days of reports of symptoms, and the fact that the mental/emotional problems experienced are more likely associated with general depression or anxiety, PMS as a “clinical concept” is “unsatisfactory” (Bancroft et al., 1993).
Alternative explanations for PMS focus on the medicalization of women’s bodies and emotions, and point out that much of the PMS literature focuses heavily on the ways in which PMS sufferers break cultural norms of femininity (Martin, 1987; Gottlieb, 1988; Zita, 1988). This perspective points out the ways in which assumptions about women’s bodies, their psychologies, and their roles in society are built into research regarding PMS. For example, it is telling that some of the complaints most cited by researchers, irritability and hostility, are the exact opposite of stereotypical feminine virtues (Gottlieb ,1988). In some PMS literature, there is a great deal of emphasis not just on the female “sufferer” but also on all the ways her “hostility” and “irritability” affect her children and her husband. This cultural bias about how women should (or shouldn’t) behave “often constitutes…a collection of negative facts about women’s nature, a nature which in turn is seen as requiring medical surveillance and management, along with a ‘protective’ secondary citizenship” (Zita, 1988). Where women aren’t “functioning” according to acceptable (often, passive) cultural norms, their defective bodies are blamed.
The objective of these alternative explanations is not to belittle real women’s negative experiences or to recommend that their complaints go untreated. These explanations do, however, suggest that the treatment for chronic, severe PMS may require something more than the standard biomedical intervention. Such explanations also emphasize the need to read the literature on PMS with an especially critical eye, given the historical tensions between women and the (mostly) male physicians who have attempted to mold their bodies and minds to suit a particular set of cultural norms.