The Interplay Between Metabolic Health, Brain Structure, and Cognition in Severe Mental Illness

Abstract

Across three studies, this thesis explores associations between metabolic dysfunction, brain structural, and cognitive alterations in severe mental illnesses (SMI). In a meta-analysis, diabetes emerged as the strongest predictor of brain structural alterations as captured by a machine-learning predicted brain age, highlighting it as a key biological target for intervention. In a sample of first-episode psychosis patients, central obesity was the strongest predictor of brain age, which partially mediated cognitive deficits. In a heterogeneous BD sample, the metabolic syndrome cluster accounted for the most variance in metabolic health, which mediated both structural and cognitive outcomes. These findings suggest that diabetes may pose the greatest biological risk, while obesity, given its early onset and high prevalence, may be the most practical public health target aimed at preventing structural and cognitive alterations in SMI. Longitudinal studies are needed to establish causality and evaluate the therapeutic potential of metabolic interventions in SMI.