OBJECTIVES: Fatigue is recognized by patients and physicians as one of the most burdensome symptoms in autoimmune (AI) diseases. The objective of this study was to evaluate if fatigue increases healthcare resource utilization and related costs in selected AI diseases. METHODS: Cohorts of patients with Multiple Sclerosis, Psoriatic arthritis, Systemic Lupus erythematosus, Ankylosing Spondylitis and Sjögren syndrome were identified using ICD-9/10 diagnosis codes in Marketscan, a large US claims database. In each cohort, patients were considered to suffer from fatigue if they did not have fatigue ICD-9/10 code one year prior to the disease cohort entry and had at least one fatigue code after. Risk, total number and costs of inpatient, outpatient and emergency visits were used to compare healthcare resource utilization in presence or absence of fatigue during the first year post cohort-entry date. Confounding due to age, gender, region, comorbidity burden and disease stage (estimated through use of specific medications) was addressed via propensity score matching using the k-nearest neighbors matching method. RESULTS: The results show that presence of fatigue in patients with AI diseases significantly increases the number outpatient visits per year and significantly increases a risk of inpatient and emergency visits. Presence of fatigue was also shown to lead to a significant and important overall medical cost increase. CONCLUSIONS: This study provides real-world evidence on the incremental economic burden of fatigue for patients with AI diseases. Given that the presence of fatigue is underreported in Marketscan, these results may underestimate the real impact of fatigue.