BACKGROUND: Performing cancer surgery in high-volume centres may lead to improved outcomes. This study explored the relationship between annual workload and outcome following resection for carcinoma of the oesophagus and cardia. METHODS: The study was a retrospective case-note review of 1125 patients who had surgery for cardio-oesophageal cancer in the West Midlands region of England. Outcome measures were 30-day mortality and long-term survival. RESULTS: The overall 30-day mortality rate was 10.0 per cent with a median survival of 14 months and a 5-year survival rate of 17.2 per cent. Increasing age, advanced stage of disease and emergency resection independently affected outcome adversely. Forty-one infrequent operators (fewer than four resections per year) performed 146 resections (13.0 per cent), 18 intermediate operators (between four and 11 resections per year) performed 488 resections (43.4 per cent) and five frequent operators (12 or more resections per year) performed 491 resections (43.6 per cent). The 30-day mortality rate was greatest in the infrequent group (15.1 per cent) compared with both the intermediate group (6.6 per cent; adjusted odds 0.40, P = 0.004) and the frequent group (11.8 per cent; odds 0.73, P = 0.28). There were no differences in survival rates between the groups, and no difference in outcome between high- and low-volume hospitals. CONCLUSION: In this unselected population-based series there was little evidence of a trend of improving 30-day mortality rate with increasing workload, or between workload and long-term survival.