The conventional wisdom that surgical volume is associated with breast cancer survival is based largely on unpublished data from a National Cancer Data Base (NCDB) study of more than 173,000 patients treated in 1,238 hospitals between 1985 and 1991. An abstract presented at a clinical oncology meeting reported that five-year survival was increased significantly when surgery was performed in hospitals with at least 25 breast cancer surgeries per year. Harcourt and Hicks challenged this conventional wisdom, stressing that other factors are more significant in breast cancer survival and proposing that concentrating services in a few centers could adversely influence survival in rural patients.

The authors studied cases of breast cancer in areas of northeast Oregon and southeast Washington state that are served by the Blue Mountain Regional Tumor Registry. The 10 participating facilities are a considerable distance from major metropolitan centers. The researchers analyzed data on presentation, treatment, and survival rate of 2,409 patients with breast cancer treated at local facilities between 1980 and 1994.

The average age of patients was 64 years, and the average stage at presentation ranged from 1.46 between 1980 and 1984 to 1.26 between 1990 and 1994. The approach to treatment changed during the study period. In the early years, only 24 percent of women received systemic therapy, but this rate increased to 56 percent during the 1990 to 1994 period. The five-year relative survival rate rose from 79 percent in 1980 to 1984 to 87 percent from 1990 to 1994. The average annual new case volume increased for each of the participating hospitals. In the 1980 to 1984 period, the range was three to 22; by 1990 to 1994, the range increased to 12 to 52 new cases per year. The five-year survival for 1980 to 1984 ranged from 64 to 87 percent, and for 1990 to 1994, the range was 76 to 95 percent.

No significant correlation was found between annual new case volume and survival ratio, either overall or when hospitals with more than 10 cases were compared with those with 10 or fewer cases per year. Survival correlated strongly with the stage of cancer at diagnosis. Centers with higher volumes tended to see younger women and women at an earlier stage, but the dominant variable in survival was the stage at diagnosis.

The authors conclude that case volume is not a major factor in the survival of patients with breast cancer. They argue that while the stage at diagnosis is the strongest determinant of survival, a concentration of services imposes barriers for women in rural areas and women without access to specialist breast surgery services. The authors argue for more dispersed services for breast cancer treatment with a greater emphasis on accessible local education and screening services, as well as treatment facilities, so that all women receive quality care.

editor's note: The authors are careful to avoid an antagonistic rural versus urban interpretation of their findings, and we must not assume that their conclusions can be applied to all cancer surgeries. Breast cancer procedures have been simplified over the years, and the key factor in outcome is prompt, individualized treatment at the earliest possible stage. As services currently are organized, the regional system may be at a disadvantage because of delays and difficulties for many women in beginning care. As family physicians, we must help each woman and her family choose the best course of action for this frightening diagnosis. The second message from this study is that we should use evidence and not assumptions when advising patients. How many of us have taken the time to find out success rates of locally based treatments or those of the specialist centers to which we refer? We must advocate for more research into outcomes of common problems. Bigger may not necessarily be better, and our patients trust us to give them the facts, not impressions.—A.D.W.