509Healthy-lifestyle-scores associated with lower subsequent fatigue risk in multiple sclerosis using inverse probability treatment weighting

Abstract

Abstract Background Several lifestyle factors have been associated with the onset and progression of multiple sclerosis (MS). Combining these lifestyle factors into scoring indices is an efficient way to assess their collective relationship with clinical outcomes. We examined the association of two lifestyle scores with clinically significant fatigue and change thereof over two years’ follow-up. Methods Data on sociodemographic, lifestyle and clinical characteristics surveyed from the international HOLISM cohort of people with MS at baseline and 2.5-year follow-up. Fatigue was defined by Fatigue Severity Scale (FSS), and healthy lifestyle by the Healthy Lifestyle Index Score (HLIS), and SNAP (Smoking, Nutrition, Alcohol, Physical Activity) score. Analyses by standard logistic and inverse probability treatment weighting (IPTW) models adjusted for age, sex, MS type, disability, comorbidity number, immunomodulatory medication use, prescription antifatigue medication use, and ongoing relapse symptoms; change in fatigue models also adjusted for baseline fatigue. Results 1,160 participants completed the FSS questionnaire at both timepoints, and roughly 62% had fatigue at each timepoint. By logistic regression, baseline HLIS and SNAP were each associated with lower risk of being fatigued at follow-up, persisting on adjustment. Using doubly-robust IPTW these associations were attenuated but high HLIS (OR = 0.89, 95% CI = 0.81-0.97) and SNAP (OR = 0.82, 95% CI = 0.73-0.90) were each associated with lower risk of fatigue at follow-up. Evaluating change in fatigue, however, there was no consistent association of either HLIS or SNAP score by either statistical method. Conclusions In this sample of people with MS, healthy lifestyle scores were associated with less fatigue 2.5 years later, though not with change in fatigue over this interval.