ABSTRACT
Following a thesis of methodological naturalism, the desire for valid and reliable comparison across epidemiological and treatment studies, and the need to facilitate research into aetiological mechanisms, has precipitated many attempts to establish a consensus definition of PMS, both by individual researchers, and by committees or working parties meeting specifically for this purpose (see Bancroft, 1993). Equally, as was noted earlier, the existing disparity between definitions of PMS used by different researchers has been put forward by many as the reason for the confusing and conflicting results in this field. This disparity concerns factors such as the temporal definition of the “premenstrual period”; disagreement over the number and type of symptoms necessary for diagnosis; disagreement over the time period after menstruation in which symptoms must cease, or the degree of symptomatology or disruption to life necessary for diagnosis; and the inclusion or not of symptoms that are exacerbated by rather than being unique to the premenstrual period.
