Investigating causal links between chronic physical illness and major depressive disorder and anxiety disorders in youth

Children with chronic physical illnesses, such as asthma, diabetes and migraine, have a higher rate of mental disorders, including depression and anxiety, than their physically healthy counterparts. With chronic illnesses estimated to affect around 25% of children, it is important to consider how these conditions contribute to early-onset mental disorders. Whilst previous studies have indicated associations between chronic illness and mental ill-health, causality has not been determined and cannot be concluded due to the potential of reverse causality or confounding factors (e.g. poverty, preterm birth). Therefore, we aim to use genetic approaches to investigate causal links between childhood chronic physical illness and major depressive and anxiety disorders in youth. Using data from the Norwegian Mother, Father and Child cohort study (MoBa) and the Norwegian National Patient Registry, we first investigated epidemiological associations between maternal reports of child chronic illnesses by 14 years (arthritis, asthma, cerebral palsy, chronic fatigue syndrome (CF), coeliac disease, diabetes, epilepsy, migraine and hearing loss) with a depression diagnosis (ICD-10: F32/F33) and anxiety disorder diagnosis (ICD-10: F40/41) prior to 18 years, including birth year and sex as covariates where appropriate. Two-sample Mendelian randomisation (MR) analysis was conducted using SNPs associated with common childhood chronic physical illnesses from existing GWAS. Initial analyses using outcome data from the largest depression and anxiety GWASs were undertaken. We will extend this analysis using outcome data from a GWAS of child/adolescent-onset depression ( Childhood arthritis (OR = 3.1 [1.1-7.1], p=0.02), cerebral palsy (OR = 3.1 [1.1-6.9], p = 0.02), CF (OR = 2.7 [0.9-6.3], p=0.03), epilepsy (OR = 2.9 [1.6-5.0], p=3 × 10-4) and migraine (OR = 1.4 [1.0-1.8], p=0.02) were associated with a diagnosis of anxiety disorder by 18 years. CF was the only illness also associated with a depression diagnosis (OR = 6.2 [2.5-13.0], p=1 × 10-5). Initial MR results did not indicate causal associations between juvenile arthritis, childhood-onset asthma, type 1 diabetes and eczema and major depressive disorder (MDD). However, migraine genetic liability was associated with MDD (b=2.0 ±0.99, p=0.04). For anxiety disorders, genetic liability to type 1 diabetes had a weak protective effect (b=-0.04 ±0.02, p=0.03), but there was no causal association for juvenile arthritis, childhood-onset asthma, eczema, nor migraine. Having a chronic physical health condition is associated with mental ill-health in young people, however, initial MR analysis indicates the majority of these conditions are not causally associated with depression nor anxiety disorders. However current GWAS do not focus on early-onset depression so MR findings using these data will also be presented.