Abstract Body (Do not enter title and authors here): Introduction
Metabolic dysregulation is posited to underlie observed links between chronic distress and higher cardiometabolic disease risk. We recently identified a plasma metabolite-based distress score (MDS) related to prevalent depression and associated with increased risk of incident diabetes and cardiovascular disease.

This study leverages repeated metabolomics measures in an ongoing cohort of women to assess the longitudinal relationship between depression and MDS.

Hypotheses
We hypothesized that (1) depression at baseline is associated with greater increase in MDS over follow-up and (2) more severe distress is associated longitudinally with higher MDS.

Methods
This is a secondary analysis with participants drawn from a harmonized data source of nested case-control studies within the Nurses’ Health Study (N=1152). White women with metabolomics data, depression measured by the Mental Health Inventory (MHI-5), and covariate data at baseline (1990) and follow-up (2000) were included. Women who developed major health conditions within two years of blood draws were excluded. To test if baseline depression status (MHI-5 ≤ 52) was associated with changes in the MDS, we performed multiple linear regression analyses (H1). To assess if depression severity (continuous MHI-5 score) was associated with MDS levels longitudinally, we fitted linear mixed models with random individual intercepts (H2). Both analyses adjusted for potential confounders incrementally in nested models: (1) baseline MDS and matching factors, (2) medical covariates, and (3) biobehavioral covariates.

Results
At baseline, the average age was 56 (sd = 7), 53 women (4.6%) had severe depression, and mean MDS (scaled to sd = 1) was -0.064, increasing to -0.026 at follow-up.

After adjusting for all relevant covariates, severe depression at baseline was associated with larger increases in MDS (β=0.29, 95% CI [0.05, 0.53]). A one-point increase in MHI-5 (i.e., less severe depressive symptoms) was linked to lower MDS levels (-0.004, [-0.007, -0.0004]).

Conclusions
We establish novel longitudinal explorations revealing that depression severity is associated with increased MDS. Depression earlier in life could potentially lead to adverse longitudinal changes of metabolic profiles linked to cardiometabolic health over the course of ten years, which may have important implications for screening and managing cardiometabolic health in individuals with psychological distress.