iCase: Examining causes and consequences of postpartum depression through large scale biobank data

MRC’s iCASE awards provide students with experience of collaborative research with a non academic partner, enabling the student to spend a period of time with the non-academic partner (usually no less than three months over the lifetime of the PhD).

Students who are successfully awarded an iCASE studentship are entitled to an additional £2,500 per year as a supplement to their stipend and an annual cash contribution of at least £1,400 towards the cost of the project.  The iCase project additional funding is only secured once contracts between the industrial partner and University of Edinburgh are signed. 

Background

Postpartum depression (PPD) is a common mental health condition, affecting around 10–20% of mothers worldwide (1). PPD is associated with substantial long-term consequences, including impaired maternal health, adverse offspring development, and worse family functioning (2). Despite the prevalence and impact, the true causes and long-term consequences of PPD are still poorly understood. Current treatment for PPD remains ineffective in many cases (3) and it is currently unclear whether PPD represents a distinct clinical entity to major depressive disorder (MDD) (4). 

The purpose of this PhD proposal is to address these challenges using large scale biobank data, causal inference methods and statistical genetics. This project will leverage an ongoing collaboration with Biogen to improve understanding of the true causes and consequences of PPD, identify promising targets for intervention/prevention for the right person at the right time and further uncover the genetic epidemiology of PPD and its distinction from MDD. This will be achieved through a series of inter-related aims. 

Aim 1: The first aim will bring together large population/biobank/registry datasets to examine observational associations between PDD, and its specific antecedents and outcomes. These datasets include the Avon Longitudinal Study of Parents and Children (ALSPAC; n=14k), The Norwegian Mother, Father and Child Cohort Study (MoBa, n=90k), UK Biobank (n=250k) and Nordic Clinical Registries (>1 million). Using a combination of these cross-sectional and longitudinal datasets, this first part aims to bring together different datasets to identify the most important causes and consequences of PPD, which will be carried forward throughout the PhD. This will also explore heterogeneity within PPD (i.e., persistent or remising patterns of PPD). The potential causes will include: genetic risk of depression, previous mental health history and adverse life events, along with protective factors such as access to social support and access/adherence to treatments. Outcomes will include later mental/physical maternal health, partner mental/physical health and interpersonal relationships and offspring mental/physical health and social functioning. 

Aim 2: The second aim will probe which of the above variables are indeed causal for PDD or a direct consequence of PDD using a range of causal inference methods. The student will use methods such as counterfactual designs (i.e., propensity score analysis) and Mendelian Randomization (MR) to probe these causal questions using the datasets listed above, including additional genetic data generated by Biogen. These methods provide opportunities to test the direct impact of PPD on self, partner, and offspring outcomes and will leverage new methods such as intergenerational Mendelian Randomization. 

Aim 3: Using newly collected data in collaboration with Biogen, this final part will aim to uncover genetic differences between PPD and MDD, potentially enabling the identification of key drug targets and biological mechanisms. In addition, any genetic differences between PPD and MDD can be explored with other health/social/behavioural traits through genetic correlation, polygenic score analysis and Mendelian Randomization, further paving the way for precision psychiatry. 

Training outcomes

This proposal draws on expertise in psychiatry, longitudinal epidemiology, causal inference, and statistical genetics, building on an already established collaborations across institutions including the University of Edinburgh, Karolinska Institute, and Biogen. The student will receive advanced training in: causal inference methods (both within and external to Edinburgh), statistical genetics and bioinformatics (GWAS, polygenic scoring, Mendelian randomization, gene prioritization), and data science skills in R/Python. Training will be complemented by lab exchanges and student placements in both academic and industry settings, providing exposure to novel resources not normally available to PhD students (i.e., novel genetic data held by Biogen). The project places strong emphasis on impact and the student will be uniquely placed to help translate findings into real world targets for intervention and prevention with collaborators in Biogen. This project will equip the student with cutting-edge, interdisciplinary skills in psychiatric epidemiology, statistical genetics and public health, with applications extending across perinatal and mental health research more broadly.

References

1. World Health Organization (2015). https://www.who.int/teams/mental-health-and-substance-use/promotion-prevention/maternal-mental-health

2. Netsi E, Pearson RM, Murray L, Cooper P, Craske MG, Stein A. Association of Persistent and Severe Postnatal Depression With Child Outcomes. JAMA Psychiatry. 2018;75(3):247–253. doi:10.1001/jamapsychiatry.2017.4363

3. Fitelson, E., Kim, S., Baker, A. S., & Leight, K. (2010). Treatment of postpartum depression: clinical, psychological and pharmacological options. International journal of women's health, 3, 1–14. https://doi.org/10.2147/IJWH.S6938

4. Nguyen, T. D., Harder, A., Xiong, Y., Kowalec, K., Hägg, S., Cai, N., Kuja-Halkola, R., Dalman, C., Sullivan, P. F., & Lu, Y. (2022). Genetic heterogeneity and subtypes of major depression. Molecular psychiatry, 27(3), 1667–1675. https://doi.org/10.1038/s41380-021-01413-6