Familial risk for anxiety and depression: intergenerational effects and genetic transmission

Abstract

Symptoms of anxiety and depression are common and debilitating emotional problems, experienced by both adults and children. Described collectively as ‘internalising’ problems, they share much of the same genetic aetiology and typically co-occur in individuals. Robust evidence exists for the clustering of internalising problems in families, as symptoms in parents are associated with symptoms in offspring. Many questions remain as to the pathways underpinning such familial risk. Parent-offspring associations could reflect causal pathways, whereby parents’ symptoms directly influence child symptoms, and vice versa. However, associations could be non-causal if the same genes influence symptoms in both parents and offspring. Understanding these processes can help to shed light on the pathways that shape our mental health and ultimately help to refine intervention targets to prevent the development of psychiatric problems.

In this thesis I use genetically informative research designs to approximate and control for genetic effects in intergenerational associations involving parent and offspring internalising symptoms. Results help us to understand whether it is ever reasonable to draw causal inferences about the influence of parent and offspring internalising symptoms on one another. I present five studies to contribute both clinically relevant and methodological lessons. Specifically, I seek to explore the possibility of transactional effects or co-development between parent and offspring symptoms across time; seek ways to improve generalisability of findings by including data from a wider pool of participants; and attend to questions surrounding statistical power to detect genetic effects in families.

In my first study I conduct a meta-analysis to show that concurrent associations between parent anxiety and offspring internalising problems withstand correction for genetic confounding, while associations involving prenatal exposure to maternal anxiety do not. In my second study I use data from a longitudinal adoption cohort to show prospective prediction from child anxiety symptoms to mother anxiety symptoms during middle childhood; and prospective prediction from father anxiety symptoms to child anxiety symptoms. In my third study I show that associations between parental criticism and adolescent internalising problems withstand correction for genetic confounding, using an extended Children-of-Twins design. I present power analyses for the detection of genetic confounding using this design and explore the possible direction of causation between generations. In my fourth study I introduce a novel approach to combining developmental methods with a statistically powerful Multiple-Children-of-Twins/Siblings model. I show that mothers’ internalising symptoms do not co-develop with offspring temperament during early childhood, although baseline stability in mothers’ symptoms is associated with baseline stability in offspring emotionality, via both genetic and non-genetic pathways. In the final chapter I discuss the findings and limitations of these approaches, alongside possible avenues for future research.

Date of Award	1 Jan 2021
Original language	English
Awarding Institution	King's College London
Supervisor	Thomas McAdams (Supervisor) & Thalia Eley (Supervisor)