Survival data for non-small cell lung cancer is typically reported from clinical trials that include patients fit enough to meet treatment criteria. The denominator of all patients from which the gefitinib-treated population is derived has rarely been reported and the impact of gefitinib on population-based outcomes is difficult to measure. We have retrospectively reviewed data of 626 patients who received gefitinib in Ibaraki Prefecture (with a population of 3 million) in Japan from July 2002 until September 2007. Overall response rate was found to 30.8%, and the median survival time was 8.0 months (95% confidence interval: 7.0–9.0 months). Female gender, good PS, and adenocarcinoma were significantly associated with prolonged survival. Adverse events were generally mild and were mostly skin reactions and diarrhea. Our population-based study has generated similar results to those previously reported in published clinical trials, which had restrictive criteria for eligible patients.