OBJECTIVES: To assess the impact of insulin pen needle (PN) length on persistence of prescribed insulin therapy among patients recently diagnosed with type 2 diabetes mellitus (T2DM). METHODS: Patients 18 years and older diagnosed with T2DM who had at least one indexed claim from May 2009 through September 2010 were assessed from the Truven Health MarketScan Research Database. Inclusion required 1 year continuous enrollment pre and post-index insulin claim with ≤ 1 prescription for an oral anti-diabetic drug during the pre-index period. Patients diagnosed with T1DM or using specialty self-injectables during the pre-index period were excluded. Patients were grouped into three PN length cohorts: short (4 & 5mm), intermediate (6mm) and long (8 & 12.7mm) needle initiators. Persistence was evaluated at 6 and 12 months by the absence of 90 day gaps between insulin prescriptions. Differences in persistence between cohorts were compared for statistical significance using a Fisher’s exact test (95% CI, two-tail)