Objective. To describe clinical characteristics associated with immunotherapy in patients with new-onset refractory status epilepticus (NORSE) and assess its timing and effect on outcomes at hospital discharge after six and 12 months of follow-up. Our secondary aim was to apply the cryptogenic NORSE (C-NORSE) score to subjects in order to evaluate its utility in identifying C-NORSE in our cohort. Methods. This was a retrospective single university hospital cohort study (2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018)(2019)(2020)(2021) of adults and children with NORSE. First-line immunotherapy was defined as corticosteroids, intravenous immunoglobulin (IVIg), and plasmapheresis (PLEX). Early immunotherapy was defined as administration of a first-line agent within seven days of presentation.Results. Twenty-one subjects with NORSE were identified between 2004 and 2021, which was cryptogenic in 18 and immune-mediated in three. All patients received immunotherapy. Seventeen patients received early immunotherapy (81%). There was no significant difference between early versus late immunotherapy regarding "good or favorable" outcomes (mRS 0-2) at hospital discharge or during follow-up. For cryptogenic NORSE patients, 7/11 (64%) achieved good outcomes at six months, 9/11 (82%) at 12 months, and 8/10 (80%) at the last follow-up visit at >13 months. For immune-mediated NORSE patients, 3/3 (100%) achieved good outcomes at six months and 2/2 (100%) at the last follow-up visit at >13 months. In our cohort, a C-NORSE score of 5 was obtained in 12/18 (67%) of cryptogenic cases and a score <5 in all three immunemediated cases. Significance. There is a paucity of published data on the timing of immunotherapy for NORSE. Although at our institution early administration of immunotherapy is feasible, more research is needed to determine which patients may benefit from immunotherapy and if the timing of immunotherapy affects short and long-term outcomes. Among the patients who survived hospitalization, long-term follow-up of our NORSE cohort demonstrated that a subset achieved good mRS (0-2) scores.