Abstract
Inflammation is associated with a range of neuropsychiatric symptoms; however, the nature of the causal relationship is unclear. We used complementary non-genetic, genetic risk score (GRS), and Mendelian randomization (MR) analyses to examine whether inflammatory markers are associated with affect, depressive and anxiety disorders, and cognition. We tested in ≈ 55,098 (59% female) individuals from the Dutch Lifelines cohort the concurrent/prospective associations of C-reactive protein (CRP) with: depressive and anxiety disorders; positive/negative affect; and attention, psychomotor speed, episodic memory, and executive functioning. Additionally, we examined the association between inflammatory GRSs (CRP, interleukin-6 [IL-6], IL-6 receptor [IL-6R and soluble IL-6R (sIL-6R)], glycoprotein acetyls [GlycA]) on these same outcomes (Nmax=57,946), followed by MR analysis examining evidence of causality of CRP on outcomes (Nmax=23,268). In non-genetic analyses, higher CRP was associated with a depressive disorder, lower positive/higher negative affect, and worse executive function, attention, and psychomotor speed after adjusting for potential confounders. In genetic analyses, CRPGRS was associated with any anxiety disorder (β = 0.002, p = 0.037) whereas GlycAGRS was associated with major depressive disorder (β = 0.001, p = 0.036). Both CRPGRS (β = 0.006, p = 0.035) and GlycAGRS (β = 0.006, p = 0.049) were associated with greater negative affect. Inflammatory GRSs were not associated with cognition, except sIL-6RGRS which was associated with poorer memory (β=-0.009, p = 0.018). There was weak evidence for a CRP-anxiety association using MR (β = 0.12; p = 0.054). Genetic and non-genetic analyses provide consistent evidence for an association between CRP and negative affect. These results suggest that dysregulated immune physiology may impact a broad range of trans-diagnostic affective symptoms.