A B S T R A C TBackground: Depression is highly prevalent among people with MS, and determinants thereof would be useful. Objectives: We examined the relationship of demographic and clinical factors with positive depression-screen and change in depression over 2.5 years in people with MS. Methods: Positive depression-screen assessed by Patient Health Questionnaire (PHQ)-2 and PHQ-9. Associations of demographic and clinical factors with depression-screen and change thereof assessed using multivariable regression models, adjusted for age, sex, disability, fatigue, antidepressant use, and baseline PHQ-2, as appropriate. Results: Overweight/obese BMI, comorbidity number, fatigue, and disability were associated with positive depression-screen, while married/partnered state, being employed, higher perceived socioeconomic status, and greater education were inversely associated with depression-screen. After adjustment, only marital status, socioeconomic status, antidepressant medication use, and fatigue were associated with risk of newly positive depression-screen. MS type, relapse number and immunomodulatory medication use were not associated with depression-screen after controlling for disability and fatigue. Conclusion: In a large prospective cohort study of depression in people with MS, we substantiated several potential determinants of a positive depression-screen and depression trajectory, particularly fatigue. Given that fatigue is the most common and most significant clinical symptom for people with MS, efforts to reduce fatigue may have follow-on benefits for reducing depression.T the frequency of positive depression-screen increased by 7.3% per year of follow-up over 2.3 years, strongly predicted by disability (Wood et al., 2012). The BEYOND clinical trial (n = 891 with depression data) examined depression using the Beck Depression Inventory-II, finding an average frequency of positive depression-screen of 25-30%, not greatly varying over three years of follow-up (Schippling et al., 2016). Koch and colleagues followed a subgroup of 94/228 patients with baseline depression data for ten years, finding the proportion with clinically significant depression as measured by the Center for Epidemiologic Studies Depression Scale increased from 47.9% to 52.1% (Koch et al., 2008).While there are many cross-sectional studies examining mood and depression in MS, there is a comparative lack of prospective studies of change in depression in large representative samples. We undertook a prospective cohort study among an international sample of people living with MS followed over 2.5 years, examining the demographic and clinical characteristics of positive depression-screen at 2.5-year review, and of change in depression-screen during follow-up. Of note, the lifestyle determinants of depression-screen during this same period are described in another article . S. Simpson, et al.