Objective:

To evaluate the association of depression with short-term neurological outcomes in a large longitudinal clinical cohort.

Background:

Depression is common in multiple sclerosis (MS). Psychosocial, inflammatory, and neurodegenerative factors have been hypothesized as etiologies. The relationship of depression to MS disease activity is not clear.

Design/Methods:

Standardized clinical and MRI data from baseline and 12-month routine follow-up clinical visits were obtained from Partners-Advancing-Technology-and-Health-Solutions (MS-PATHS) database, a registry of patients from 10 MS centers. Demographic and clinical data, patient reported outcomes, neuroperformance scores (manual dexterity, walking speed, and processing speed tests), and MRI metrics were analyzed. Propensity score weighting, with baseline depression status (depressed and non-depressed: Quality-of-Life-in-Neurological-Disorders (NeuroQoL) depression T-score >45, ≤45, respectively) as exposure, was performed using all baseline covariates. Covariates balance was assessed and the weighting and matching method with best balance was selected. Odds ratios (OR) and 95% confidence interval (CI) of worsened neuroperformance composite, new relapses, and new MRI activity at month-12 were derived via multivariate logistic regression after propensity adjustment.

Results:

1923 patients were included in the analysis. Propensity score weighting showed adequate overlap between depressed and non-depressed groups on all baseline measures. Average-treatment-effects-on-the-treated weighting demonstrated the best covariate balance. Patients with baseline depression compared to those without had greater odds of ≥20% worsening on ≥1 neuroperformance component (OR 1.41, CI 1.05, 1.89), ≥1 new T2 lesion (OR 1.49, CI 0.68, 3.27), ≥1 relapse (OR 1.29, CI 0.89, 1.87), and ≥1 new contrast-enhancing lesions (OR 1.54, CI 0.53, 4.42) at month-12.

Conclusions:

In this large real-world observational cohort, after adjusting for baseline covariates, MS patients with depression at baseline showed more worsening in neurological function, relapses, and MRI lesion activity over the next year compared to non-depressed patients. Our results suggest that depression is not merely a reactive symptom but indicates increased risk of future MS disease activity.