I conducted several cohort studies to quantitatively assess the risks and benefits of using antidepressants. Overall, my findings suggest that continuing antidepressant treatment reduces the risk of relapse and has limited or no adverse effects on children. Here are the key findings:
1. Individuals on antidepressant treatment for less than 4, 4-6, 7-9, and 10-12 months had cumulative incidences of restarting antidepressants within one year at 37.4%, 35.1%, 35.0%, and 32.8%, respectively. To minimize the risk of relapse after discontinuation, a minimum of 10-12 months of treatment seems preferable.
2. Discontinuing antidepressants during pregnancy increased the risk of psychiatric emergencies compared to continuing treatment (5.0% for discontinuation versus 3.7% for continuation). Approximately for every 76 pregnant women who discontinued antidepressants, there was an additional psychiatric emergency. Continuing antidepressant treatment throughout pregnancy may reduce the risk of psychiatric emergencies.
3. In collaboration with a Norwegian group, we found four similar treatment trajectories. Discontinuing antidepressants early in pregnancy reduced the risk of worsening depressive symptoms postpartum compared to continuers, while discontinuation in late pregnancy increased the probability of increased depressive symptoms. These results were consistent between the two countries.
4. I also examined whether the reduced relapse risk observed among women using antidepressants during pregnancy extended to those with mental disorders other than affective disorders, such as eating disorders and obsessive-compulsive disorders. The beneficial effect was also seen among women who continued antidepressants during pregnancy and postpartum to manage these disorders.
5. The timing of antidepressant discontinuation played a crucial role in relapse risk. Therefore, I conducted a study to examine the determinants of the timing of discontinuing antidepressants. I identified four treatment trajectories: early discontinuers (23%), late discontinuers (24%), interrupters (15%), and continuers(38%). Women who continued or discontinued antidepressants late in pregnancy exhibited more severe mood disorders. However, these differences were not reflected in their polygenic risk scores.
6. Psychiatric disorders that occurred at different timing may indicate distinct severity. I investigated this by employing polygenic scores and found genetic liability to bipolar disorder, but major depression or schizophrenia was associated with a higher risk of postpartum psychiatric disorders that occurred within 30 days after childbirth.
7. To provide a balanced overview of antidepressant treatment efficacy and safety, I also investigated various health outcomes in offspring. I found a slightly increased risk of preterm birth, low birth weight, postnatal adaptation syndrome, neonatal admission, and neonatal persistent pulmonary hypertension among children exposed to antidepressants during pregnancy. However, the relative risks were modest, and the absolute risks were low. Additionally, I found no evidence supporting that maternal genetic liability explains the associations.
I published 12 peer-reviewed papers and have three papers under review. Furthermore, I have disseminated our findings through various channels to reach a wide audience:
1. Conference presentations: I have presented our findings at several important conferences, including the Danish Epidemiological Society Annual Meeting, the NordicEpi, iPSYCH Research Forum Meetings, and the 13th Annual Meeting of the Danish Society for Pharmacology.
2. Press releases and social media engagement: I have issued press releases highlighting our research's key insights and updates. Additionally, I have effectively utilized social media platforms such as Twitter to disseminate our findings.
3. Invited keynote speech: I was invited to deliver a keynote speech at the Aarhus Institute of Advanced Studies (AIAS) conference on June 14, 2022.
4. Presentations at prominent institutes: I presented my research at five renowned institutes, including the Institute for Molecular Bioscience and the Child Health Research Centre at The University of Queensland.
