OBJECTIVES: To describe real-world utilisation of three echinocandins available in England for the treatment of invasive fungal infections: anidulafungin, caspofungin, and micafungin. METHODS: A longitudinal, retrospective cohort study was conducted using the Hospital Treatment Insights database, which links pseudonymised patient data with hospital pharmacy dispensing in England. The Jan 2011 to Apr 2013 analysis period contained 2.7m unique patients. Included patients were adults with ≥2 dispenses of the same echinocandin. Reporting is obscured where <6 hospitalisations observed a particular dose dispensed. RESULTS: Of the anidulafungin hospitalisations (n=53), slightly more exceeded the licensed Day 1 dose than matched it (200mg, 34.0%; >200mg, 39.6%); 75.5% of Day ≥2 doses were 100mg as per licence; 15.1% were 200mg. For caspofungin (n=1079) the most frequently dispensed dose was 70mg as per licence (Day 1: 42.5% [<70mg, 24.7%; >70mg, 32.6%]; Day ≥2: 62.8%). When including permitted adjustments for weight/hepatic impairment, 84.5% doses were as per licence on Day ≥2; 15.5% were not. For micafungin (n=177) Day 1 doses were: 100mg, 35.0% (as per licence); 100mg-200mg, 48.5% (permitted only for inadequate response), >200mg, 16.4% (not permitted). Most Day ≥2 doses were as per licence (100mg, 68.4%; >100mg 21.5%; >200mg, 10.2%). A subsequent antifungal was required in a greater proportion of micafungin hospitalisations (30.5%), compared to caspofungin (22.9%) and anidulafungin (20.8%). CONCLUSIONS: While observed usage of anidulafungin and caspofungin was largely consistent with product licences, most Day 1 micafungin doses exceeded 100mg, suggesting inadequate treatment response or clinical cautiousness. Exceeding licensed doses may lead to higher drug costs than might be assumed based on acquisition cost and product licence. The greater requirement for subsequent antifungal treatment following micafungin may also suggest increased costs.