OBJECTIVES: To compare healthcare resource utilization (HRU) and associated costs for patients with neuromyelitis optica spectrum disorder (NMOSD) with or without a relapse.

METHODS: This retrospective cohort study evaluated administrative claims data from the Magellan database between January 1, 2016, and January 31, 2020, to identify adults (≥18) with NMOSD. Patients were grouped into 2 cohorts based on relapse status: active (≥1 relapse) or less active (no relapse) in the follow-up period (first claim for NMOSD [index] to study period end). Patient characteristics, HRU, and costs were assessed.

RESULTS: Of 132 adults included in the study, 60 had less active relapse status and 72 had active relapse status in the follow-up period (mean: 15.2 months, standard deviation [SD]: 1.9 months). Baseline characteristics were similar between cohorts. Each cohort included patients who experienced ≥1 relapse during the 1-year, pre-index baseline period (less active: 21.7%; active: 51.4%). A greater proportion of active relapse status patients had ≥1 emergency department visit (69.4% vs 45.0%, P=0.0046), inpatient hospital encounter (63.9% vs 6.7%, P<0.0001), outpatient hospital encounter (97.2% vs 80.0%, P=0.0014), or pharmacy claim (98.6% vs 88.3%, P=0.0137) versus less active relapse status during follow-up. Those with active relapse status had ~double annualized outpatient hospital encounters (mean [SD]: 9.1 [8.1] vs 5.0 [5.6]; P=0.0017) and more than twice the mean annualized costs per patient versus less active relapse status ($80,228 vs $33,059; P=0.0023). Mean (SD) length of stay was ~4 days greater for those who relapsed during follow-up (4.1 [7.4] days) versus those who did not (0.2 [0.0] days).

CONCLUSIONS: Patients with NMOSD who had a relapse incurred more outpatient hospital encounters, longer lengths of stay, and greater mean annualized costs compared with patients who did not relapse. Because of the significant impact relapses have on HRU, therapeutic options focusing on reducing relapses should be considered.