Genetic liability to major psychiatric disorders contributes to multi-faceted quality of life outcomes in children and adults
Yingjie Shi,Barbara Franke,Nina Roth Mota,Emma Sprooten
DOI: https://doi.org/10.1101/2023.01.17.23284645
2023-01-18
MedRxiv
Abstract:Importance: Psychiatric disorders can have an immense impact on socioeconomic, physical, and social-psychological facets of life. Psychiatric disorders are also highly heritable. Under a liability threshold model, an important question arises as to what extent genetic liability for psychiatric disorders relates to, and possibly impacts on, different aspects of quality of life in the general population. Objective: To characterize the link between psychiatric genetic liability and diverse aspects of quality of life in childhood and adulthood. Design setting and participants: We used data from two multi-site, population-based cohorts, i.e. preadolescent children in the USA enrolled at age 9-10 years from the Adolescent Brain Cognitive Development (ABCD) study (N=4,645) and white British adults between age 40-69 years from the UK Biobank (UKB) study (N=377,664). Due to the current limitations of our genetic methods, only data from unrelated individuals of European descent could be included. Main outcomes and measures: To derive robust measures capturing multiple domains of quality of life in each of the cohorts, we integrated an array of measurements of academic, economic, and physical status, as well as social well-being, in a second-level three-factor confirmatory factor analysis. The genetic liabilities to seven major psychiatric disorders were quantified by a set of polygenic scores (PGSs) derived from the largest genome-wide association studies to date, independent of the target cohorts, of major depressive disorder (MDD, N=142k-173k), anxiety disorders (ANX, N=22k-144k), attention-deficit/hyperactivity disorder (ADHD, N=226k), autism spectrum disorder (ASD, N=55k), schizophrenia (SCZ, N=130k), bipolar disorder (BIP, N=353k-414k), and cannabis use disorder (CUD, N=384k). Using general linear models we assessed associations between PGSs and the estimated latent factors, controlling for age, sex, site, genotyping batch, plate, and genetic ancestry. Results: In each cohort, three latent factors indexing distinct but correlated quality of life domains, (1) educational performance and cognition (Edu, in ABCD) / social economic status (SES, in UKB), (2) physical health (Hea), (3) adverse social experience (Adv, in ABCD) / social well-being (Soc, in UKB), were estimated with excellent model fit indices. In addition, a general factor was derived that captured the covariances between the three latent factors (QoL). In the ABCD cohort, ADHD-PGS was significantly associated with Edu (β = -0.13, t = -8.29, p = 1.53e-16), Adv (β = -0.09, t = -5.79, p = 7.81e-09), and general QoL (β = -0.14, t = -8.74, p = 3.37e-18) factors. In the UKB cohort, all examined disorder PGSs were significantly associated with the general QoL latent factor and at least one first-order subdomain, with ADHD-PGS (β = -0.06 ~ -0.10, t = -29.1 ~ -52.5, p < 5.91e-186) and MDD-PGS (β = -0.04 ~ -0.07, t = -23.8 ~ -36.3, p < 3.63e-125) showing the largest effects. Conclusions and relevance: The present study reveals an inverse relationship between psychiatric genetic liabilities and multiple quality of life metrics, with ADHD-associated genetic risk being the main contributor in both children and adults, and MDD additionally showing effects in adults. All effect sizes observed were small, as expected. Understanding potential real-world outcomes of quantitative measures of disorder-related genetic risks in the general population can provide a scientific foundation for societal intervention and policy-making processes, with profound implications for promoting a flourishing society.