Prior reproductive and non-reproductive depression, and depressive symptoms in menopausal transition

The menopausal transition is associated with an increased risk of depression. Prior depression is a well-established risk factor, but studies do not distinguish between prior reproductive and non-reproductive depression. We aimed to examine whether prior reproductive (i.e., premenstrual mood disorder and perinatal depression) and non-reproductive (i.e., not related to hormonal transitions) depression were differentially associated with depressive symptoms during the menopausal transition.

We analyzed cross-sectional data from menopause outpatient clinics, including a multidisciplinary clinic. Exposures included premenstrual mood disorder assessed with the Premenstrual Symptom Screening Tool, perinatal depression assessed with the Edinburgh Postnatal Depression Scale-Lifetime version, and prior non-reproductive depression recorded in medical records. Current depressive symptoms were measured with the Inventory of Depressive Symptomatology-Self Rated. We used univariable and multivariable linear regression models to examine associations; multivariable models accounted for overlap between exposures.

Among 180 participants (median age 51; 61% perimenopausal, 39% postmenopausal), premenstrual mood disorder showed the strongest association with depressive symptom severity (B = 9.0, 95% CI 5.1-12.9, p < 0.001), followed by perinatal depression (B = 7.8, 95% CI 3.4-12.1, p < 0.001) and prior non-reproductive depression (B = 4.7, 95% CI 0.7-8.7, p = 0.021). In multivariable analysis, only premenstrual mood disorder (B = 7.2, 95% CI 2.4-12.1, p = 0.0037) and perinatal depression (B = 5.7, 95% CI 1.2-10.1, p = 0.013) remained associated with depressive symptom severity.

Prior reproductive depression, but not prior non-reproductive depression, was associated with greater depressive symptom severity during the menopausal transition. Histories of premenstrual mood disorder and/or perinatal depression may help identify individuals vulnerable to depressive symptoms during this period. Prospective studies in population-based samples are needed.