Pharmacological treatment during pregnancy has been a huge challenge since the establishment of thalidomide's teratogenicity in the early sixties. Analyses of possible risks associated with drug intake during pregnancy are not possible by performing randomized trials, and interspecies extrapolation is challenging. The best available method is through epidemiological studies. During the past decade use of antidepressants during pregnancy has been associated with negative birth outcomes, such as congenital malformations. In spite of a considerable number of studies on the subject, the data are still conflicting. The main challenge is how to discern between the effects of the drug and the effect of the depression itself. We approached this dire problem conducting a nation-wide register based study analyzing the relation between use of antidepressants during pregnancy and the risk of congenital malformations and perinatal mortality. We performed our analysis with focus on women pausing treatment before pregnancy to account for special characteristics associated with women redeeming a prescription for an antidepressant. Furthermore, we reported prevalences of antidepressant use, in Denmark, in relation to pregnancy and over time, between 1997 and 2010. We found that use of antidepressants during pregnancy has increased from 0.2% in 1997 to 3.2% in 2010. This considerable increase is mostly due to exposure to selective serotonin reuptake inhibitors (SSRIs). In addition, at the time of pregnancy recognition we saw a halving in prevalence of antidepressant exposure and a steep increase after delivery. Our analyses showed an association between being in treatment with an SSRI and congenital malformations. However, this increased risk was also found for women pausing treatment before pregnancy. We conclude that the apparent risk associated with use of SSRIs during pregnancy is not related to the drug exposure, but to unknown characteristics associated with mothers redeeming a prescription for an antidepressant. We found no increased risk of stillbirths or neonatal mortality among off-spring exposed in utero to an antidepressant in any of the three trimesters. The overall conclusion is that antidepressants are not associated with increased risks of congenital malformations and perinatal mortality. However, we cannot rule out a possible causal relation, and treatment must therefore be based on an individual assessment of each woman analysing possible risks versus possible benefit.