To Örebro University

IMPORTANCE: Clinical diagnoses of psychiatric disorders are associated with cardiometabolic diseases (CMDs) such as type 2 diabetes and ischemic heart diseases. Studying how genetic liability for psychiatric disorders relate to CMD risk will offer novel insight into the relationship between psychiatric disorders and CMDs. OBJECTIVE: To evaluate the associations between psychiatric polygenic risk scores (PRSs) and clinically diagnosed CMDs while accounting for cross-disorder pleiotropy.

DESIGN SETTING AND PARTICIPANTS: This study computed PRSs for attention deficit-hyperactivity disorder (ADHD), major depressive disorder (MDD), anxiety disorder, post-traumatic stress disorder (PTSD), bipolar disorder, and schizophrenia. The analysis was conducted in three population-based Northern European cohorts: the Swedish Twin Registry (STR, N=17,378 genotyped samples), the Estonian Biobank (EstBB, N=208,383), and the Norwegian Mother, Father and Child Cohort Study (MoBa, N=129,398). Associations between psychiatric PRSs and clinical diagnoses of 10 major CMDs (including metabolic diseases such as hyperlipidemia, obesity, and type 2 diabetes, and cardiovascular diseases such as hypertensive disease, arteriosclerosis, ischemic heart disease, heart failure, thromboembolic disease, cerebrovascular disease, and arrhythmias) were estimated using models that mutually adjusted for all psychiatric PRSs. Supplementary analyses were performed by additionally controlling for self-reported body mass index (BMI). A discordant twin-pair analysis was conducted in the STR (N=70,619) to assess the association between self-reported lifetime MDD and subsequent CMD risk while adjusting for familial factors shared between monozygotic and dizygotic co-twins.

MAIN OUTCOMES AND MEASURES: Psychiatric PRSs were constructed based on both all available genetic risk variants and genome-wide significant risk variants from large-scale GWASs. Clinical diagnoses of psychiatric disorders and CMDs were ascertained through electronic health records (with primary care records used exclusively in the EstBB). Lifetime self-reported MDD in the STR was assessed via the Composite International Diagnostic Interview Short Form.

RESULTS: PRSs for ADHD and MDD were associated with increased risk of all CMDs. The ADHD PRS showed stronger associations with metabolic disease, whereas the MDD PRS showed stronger associations with cardiovascular diseases. PRSs for anxiety disorder, PTSD, and bipolar disorder showed only limited associations with CMDs, while increased levels of schizophrenia PRSs were associated with decreased risk of CMDs. These associations remained after adjustment for BMI. Finally, twins endorsing lifetime MDD were found to have an increased risk of subsequent CMD diagnoses compared to their unexposed co-twins.

CONCLUSIONS AND RELEVANCE: PRSs for ADHD and MDD showed robust associations with risk of CMDs and self-reported MDD was associated with subsequent CMD risk even after adjusting for familial factors shared between co-twins. These findings provide robust evidence for genetic overlap between ADHD and MDD with CMDs.