OBJECTIVES: The UK National Institute for Health and Clinical Excellence (NICE) recommended gefitinib as cost-effective for use first line in locally advanced or metastatic, EGFR mutation positive NSCLC, when supplied via the SPA scheme, based on the mean duration of treatment observed in the IPASS study (8.8 months). The single fixed payment under the scheme is triggered at the order of the third 30-day pack and covers a patient for their total supply of gefitinib treatment. An initial analysis of the first 18 months of the SPA scheme indicated that the length of gefitinib therapy was at least that anticipated by NICE. This study aims to validate the length of therapy in a larger cohort with up to 3 years follow-up. METHODS: The SPA administrative database contained information on packs (30-days therapy/pack) dispensed to patients from September 2009 to December 2012. This retrospective study included patients registered on the database prior to the end of 2012, fulfilling NICE eligibility criteria and receiving at least 3 packs for which the NHS was invoiced. Patients were considered censored if they had ≥1 pack recorded in the last 3 months of the database. Median time to treatment cessation was estimated from a Kaplan-Meier curve of packs supplied to patients and the mean number of packs dispensed per patients from a parametric failure time model. RESULTS: 883/1160 registered patients met the study eligibility criteria (460/883 censored). These 883 patients, for whom the NHS was invoiced the single fixed payment, received a median of 13 packs 95%CI[12,14], equivalent to 12.8 months to treatment cessation. A mean of 19.4 95%CI[18.0, 21.0] packs were dispensed per patient. CONCLUSIONS: The results of this observational study confirm the average length of gefitinib therapy and mean number of packs dispensed in the SPA scheme exceed that assumed by NICE.