BACKGROUND: Between 1986 and 1996, the overall mean overnight length of stay for all diagnoses in Sweden decreased from 20.8 to 7.1 days. OBJECTIVES: The study describes changes in surgical technique, from mastectomy to breast-conserving surgery, in treatment of female breast cancer and the parallel change in average length of hospital stay, and discusses the possible link between the trends. RESEARCH DESIGN: The study was performed as a descriptive register study on hospital admission data from the Swedish Hospital Discharge Register over a 16-year period (1980-95). RESULTS: During the study period, the mean length of stay for surgical curative breast cancer treatment in Sweden decreased by 56%. In 1980, the proportion of women receiving conservative surgery was 7%. At the end of the period, this share had increased to 51%. Breast-conserving surgery had an approximately 30% shorter mean length of stay compared with mastectomy. The gap was remarkably stable during the study period. The shift from mastectomy to breast-conserving surgery had a limited effect on the share of patients that went through lymph node dissection. Neither age nor the number of operations per woman could, to any significant extent, explain the decrease in mean length of stay. Approximately 14% of the overall decline can be attributed to the changes in technique. CONCLUSIONS: Clinical practice style, in this case the surgical technique, has had an effect on length of stay, but the surgical technique can only to some extent explain the trend.