Date Published: September 26, 2017
Publisher: Public Library of Science
Author(s): Marie-Louise H. Rasmussen, Marin Strøm, Jan Wohlfahrt, Poul Videbech, Mads Melbye, Jenny E. Myers
Abstract: BackgroundSome 5%–15% of all women experience postpartum depression (PPD), which for many is their first psychiatric disorder. The purpose of this study was to estimate the incidence of postpartum affective disorder (AD), duration of treatment, and rate of subsequent postpartum AD and other affective episodes in a nationwide cohort of women with no prior psychiatric history.Methods and findingsLinking information from several Danish national registers, we constructed a cohort of 457,317 primiparous mothers with first birth (and subsequent births) from 1 January 1996 to 31 December 2013 (a total of 789,068 births) and no prior psychiatric hospital contacts and/or use of antidepressants. These women were followed from 1 January 1996 to 31 December 2014. Postpartum AD was defined as use of antidepressants and/or hospital contact for PPD within 6 months after childbirth. The main outcome measures were risk of postpartum AD, duration of treatment, and recurrence risk. We observed 4,550 (0.6%) postpartum episodes of AD. The analyses of treatment duration showed that 1 year after the initiation of treatment for their first episode, 27.9% of women were still in treatment; after 4 years, 5.4%. The recurrence risk of postpartum AD for women with a PPD hospital contact after first birth was 55.4 per 100 person-years; for women with postpartum antidepressant medication after first birth, it was 35.0 per 100 person-years. The rate of postpartum AD after second birth for women with no history of postpartum AD was 1.2 per 100 person-years. After adjusting for year of birth and mother’s age, women with PPD hospital contact after first birth had a 46.4 times higher rate (95% CI 31.5–68.4) and women with postpartum antidepressant medication after their first birth had a 26.9 times higher rate (95% CI 21.9–33.2) of a recurrent postpartum episode after their second birth compared to women with no postpartum AD history. Limitations include the use of registry data to identify cases and limited confounder control.ConclusionsIn this study, an episode of postpartum AD was observed for 0.6% of childbirths among women with no prior psychiatric history. The observed episodes were characterized by a relatively short treatment duration, yet the women had a notably high rate of later AD and recurrent episodes of postpartum AD. The recurrence risk of postpartum AD was markedly higher among women with PPD hospital contact after first birth compared to women with postpartum antidepressant medication after first birth. Our results underline the necessity of measures targeted at specific vulnerable groups, such as women who experience PPD as a first psychiatric episode.
Partial Text: Postpartum depression (PPD) is a nonpsychotic depressive episode occurring in the period following delivery of a child. Depending on, for example, the inclusion criteria and the quality of follow-up of women who have given birth, it is reported to affect 5%–15% of all women after childbirth [1,2], which makes it one of the most common postnatal complications of childbearing. Left untreated, the disorder can have long-term implications for both mother and child, including impairment of the child’s development [3–5] and increased risk of long-term maternal depression . A number of different risk factors for PPD have been identified, of which the majority are antenatal, personal, and psychosocial factors [3,7–9]. However, such factors can account for at most a third of the variance in the diagnosis of PPD [10,11], which could indicate a genetic predisposition, as suggested in some studies .
In accordance with Danish law, the use of the register-based data in the study was approved by the Danish Data Protection Agency (no. 2008-54-0472). The study is reported as per STROBE guidelines (S1 STROBE Checklist). A detailed analysis plan was not written prior to the initiation of the project. However, based on the objectives of this study, we defined all basic analyses to be undertaken in meetings with all involved parties (epidemiologists, clinician, and statisticians) prior to the receipt of the registry data and before the start of the analyses. We did not depart from the analysis plan built during these meetings but added post hoc sensitivity analyses as presented. However, in response to review comments, postpartum antidepressant treatment was divided post hoc into postpartum medication and PPD hospital contact.
In this nationwide, population-based cohort study, 0.6% of childbirths among women with no prior history of psychiatric disease resulted in a postpartum AD, defined as a prescription fill for antidepressant medication and/or hospital contact for depression during the first 6 months after birth. However, less than 1/3 of the women were still receiving treatment 1 year after treatment initiation. Compared to women with no episode of postpartum AD after their first childbirth, women with postpartum antidepressant medication and PPD hospital contact, respectively, had a 6.2 and 6.6 times increased risk of a non-postpartum AD in the years following first childbirth and a 27 and 46 times higher recurrence rate of postpartum AD following a second birth.