Main Findings
This large prospective study aimed to determine and distinguish the main determinants of early and late PPD. While both forms shared some common factors, several factors were specific to only one form of PPD. Another important finding is that PPD is not only associated with psychiatric risk factors (personal or family history), but also with childhood trauma and stressful life events occurring during pregnancy and delivery. More specifically, early PPD was associated independently with a personal history of depression, emotional neglect during childhood, stressful life events during pregnancy and concomitant chronic physical disease. Late PPD was associated with unemployment, personal and family history of mood disorder, emotional abuse during childhood, stressful life events during pregnancy, emergency consultation during pregnancy, and serious post-partum complications.
Stressful life events during pregnancy were associated with both early and late PPD in our cohort, in accordance with previous studies (2,37). However, the majority of women experience such stressful life events and most do not develop PPD. In the present study, we took into account the woman’s subjective experience for each stressful life event and considered the real impact of each event. We have also revealed a cumulative effect of these life events: each additional event increased the risk of PPD. A cumulative effect of stressful life events in the year before pregnancy has been shown to have an impact on depression during pregnancy (38). This “dose-effect” is important in clinical practice: the risk of PPD increases with each additional stressful event.
The impact of childhood trauma on PPD was controversial: some studies reported this association (44,45), while others did not (46,47). Our study not only confirms that childhood trauma is associated with PPD, but also shows that emotional neglect specifically has an impact on early PPD, while emotional abuse is associated with late PPD. Further research is needed to better understand this point.
As with early adversities, we have shown that recent stressful events are also associated with PPD. Regarding obstetric factors before pregnancy, only a concomitant chronic physical disease was associated with early PPD. Infertility and assisted reproductive technology were not associated with PPD, confirming recent findings (39,40). During pregnancy, we highlighted that consulting in the emergency room is an important factor associated with PPD. An Australian study found consistent results, with more emergency room visits during pregnancy among women requiring hospitalization for PPD (41). Whether this is due to an underlying psychological condition, social stress factors, pregnancy complications, or to the stress associated with the emergency consultation itself is not known and needs further exploration. Finally, perineal trauma had an impact only on early PPD, and this effect disappeared with distance from childbirth; confirming the results of Dunn et al. (42). In our study, episiotomy or cesarean section were not associated with PPD. Data on C-section are conflicting (17,43,44).
Having a personal history of a psychiatric disorder was associated with a higher risk of both early and late PPD, which is consistent with the literature (17). One of the most strongly associated factors was a personal history of depression, in line with published data (1,2,17,48). Having a history of anxiety disorder was also associated with early and late PPD although the final model did not retain it, as shown by previous literature (19,49,50). In addition, a history of attempted suicide increased the risk of early and late PPD in our study. In a U.S. population study, a history of attempted suicide was associated with perinatal depression with an adjusted OR of 3.79 (37).
Women with early and late PPD were both more likely to report a family history of mood disorder than controls (53% and 60% vs. 47% respectively). There was a significant difference between women with late PDD and controls for family history of mood disorder disorders (p<.001), but not between women with early PPD and controls, suggesting a smaller effect of family history on early PPD. A recent study based on Danish population-based registers found that parental psychiatric history of depression, bipolar disorder or schizophrenia was associated with postpartum psychiatric disorders, especially among women with personal psychiatric history (51). For women with a psychiatric history, the postpartum period appears to be a time frame of vulnerability for depression, suggesting that genetic factors may play a role in PPD.