OBJECTIVES: To be effective, osteoporosis (OP) therapy must be taken consistently and as prescribed. Persistence is critical for successful outcomes, including fracture risk reduction. Few studies compare the persistence of oral bisphosphonate (BPs), IV BPs and denosumab beyond 1 year. This retrospective database analysis evaluated 2-year persistence to oral BPs, IV BPs and denosumab following treatment initiation and the risk of treatment discontinuation. METHODS: From the German IMS®LRx database, we included women aged ≥45 years who initiated an OP treatment after 1-July-2010 (index date = treatment initiation) with ≥2 years of follow-up until 31-Dec-2014. Persistence (prescription refill gap ≤ 60 days and no drug switch) was measured for 2 years from index date, and a Cox proportional hazard regression model was used to estimate the risk of treatment discontinuation (i.e. non-persistence). RESULTS: Data from 159,993 women were included in the analysis. Two years after treatment initiation, 39.8 % of those receiving denosumab (n=21,154), 24.8 % receiving IV ibandronate (n=20,472), 21.2 % receiving IV zoledronic acid (n=3,966), and 16.7 %„Ÿ17.5 % of those receiving oral BPs were persistent. Compared with those receiving denosumab, women receiving IV ibandronate or IV zoledronic acid were at higher risk of treatment discontinuation (HR [95% CI]: 1.65 [1.61-1.69] and 1.28 [1.23-