The assessment of the symptoms of premenstrual syndrome and their response to therapy.

The symptoms of premenstrual syndrome should be rated daily, or at frequent intervals throughout the menstrual cycle. Self-rating is usually most feasible and separate rating of differing symptom groups is important, as symptoms differ in their response to therapy. Daily scores should be analysed to assess periodicity, either by subdividing the cycle into phases or by using the least mean square method of fitting sine waves. Standardized scores enable data to be compared across cycles. In a clinical trial it is important to include an untreated cycle to assess whether the subject has premenstrual syndrome and as a baseline with which to compare treated cycles. Allowance should be made for a carry-over effect and for high placebo response. One solution is to use a change-over design balanced for carry-over effects. The criteria used to define a patient should be stated.