The formation of multidisciplinary breast teams across the UK is intended to concentrate the assessment and treatment of breast cancer into the hands of high volume specialists. We undertook a retrospective population-based study in order to determine the trends in surgeon breast cancer workload in Yorkshire, UK, and to investigate whether patients treated by low-workload surgeons had poorer survival. Of 11 329 female breast cancer patients diagnosed in 1989-1994 in Yorkshire, 6% were managed by surgeons with a mean annual workload of less than 10 new patients, while surgeons with workloads of 10-29, 30-49 and >50 treated 21, 21 and 52%, respectively. Over the study period, increasing number of patients were managed by surgeons with higher workloads. Patients treated by low-workload surgeons had poorer survival. Five-year survival was 60% in the lowest workload category compared to 68% in the highest category. The relative risk of death was increased by 15% (RR=1.15, 95% CI 1.03-1.28) and by 10% (RR=1.10, 95% CI 1.02-1.18) for patients managed by surgeons with workloads <10 and 10-29 cases per annum in comparison to patients managed by surgeons with workloads of >50. The results of this study suggest increasing site specialisation in breast cancer among general surgeons. It also provides further evidence that the management of patients by surgeons with low workloads decreases overall survival.