Identifying symptom clusters during the menopausal transition: observations from the Seattle Midlife Women's Health Study.

OBJECTIVES: First, to explore the variability in how symptoms clustered together over the late reproductive stage, early and late menopausal transition (MT) stages, and early postmenopause; second, to determine whether the symptom factor structure of the preceding MT stage would predict the symptom factor structure of the MT stage immediately following.
METHODS: The sample consisted of a subset of Seattle Midlife Women's Health Study participants who were in late reproductive or early or late menopausal transition stages or early postmenopause and provided self-reported data on symptoms experienced between 1990 and 2005. Principal components analysis was used to determine how symptoms clustered together across the stages. Variables predicting the symptom factor structure were analyzed by multiple regression.
RESULTS: Principal components analysis with varimax rotation revealed different factor structures for each of the four stages. The three-factor solution in the late reproductive stage explained a total of 54.9% of the variance. The four-factor solutions in the early and late menopausal transition stages and postmenopause explained a total of 56.5%, 59.3%, and 60.7%, respectively.
CONCLUSIONS: This analysis revealed similar factor structures across the four stages in that each stage revealed a mood component, a vasomotor component, and a pain component. However, the symptoms differed somewhat in how they grouped from stage to stage. Regression analysis demonstrated that a relationship exists between the symptom factor structures across stages. Controlling for demographic and lifestyle variables, it was revealed that the symptom clusters at the early and late menopausal transition stages and early postmenopause were best predicted by the symptom factor structure of the previous stage.