A recent study by Guadagnoli et al. (1998b) using data collected from medical records and a patient survey reported much higher rates of breast conserving surgery but still substantial regional variation when comparing women treated in 18 hospitals in 1993-1995 in Massachusetts and 30 hospitals in Minnesota in 1993. Although the hospitals participating in Massachusetts were selected randomly (20 were originally selected but two refused to participate), those from Minnesota were part of a consortium formed by the Healthcare Education and Research Foundation and included about 60 percent of patients hospitalized in Minnesota. Overall, 64 percent of women in Massachusetts and 38 percent of women in Minnesota with Stage I and II breast cancer received breast conserving surgery. However, after excluding patients with contraindications to breast conserving surgery (e.g., prior BCS, tumor multifocal, tumor centrally located), the proportions of eligible women receiving breast conserving surgery increased to 74 and 48 percent, for Massachusetts and Minnesota, respectively. Importantly, 27 percent of women in Minnesota and 15 percent in Massachusetts who underwent mastectomy, even though they were eligible for breast conserving surgery, reported that their surgeon had not discussed BCS with them. This suggests that a significant proportion of the variation in rates of breast conserving surgery reflects the fact that women have not been given an informed choice of procedure.
Data compiled by Wennberg et al. (1996) in the Dartmouth Atlas of Health Care using 1992-1993 Medicare claims show that widespread variation in the proportion of women offered breast conserving surgery remains. Without stratifying on stage in this predominantly 65 and older population, the proportion of inpatient cancer surgery that was a breast conserving procedure ranged from 1.4 to 48.0 percent by Medicare hospital referral region. In light of the data from Guadagnoli et al. (1998b) that up to one-quarter of women who undergo mastectomy have not been provided information about BCS, the persistent widespread regional variation in the performance of breast conserving surgery would appear to indicate that many women are not being offered a choice.
Other studies also have found significant variation in the use of breast conserving surgery according to hospital characteristics (Nattinger et al., 1992, 1996). A study by Nattinger et al. (1996) found that 55 percent of breast conserving surgeries performed on Medicare patients occurred in only 10 percent of the hospitals submitting claims in 1986 to 1990. Increased use of BCS was associated with larger hospital size, the presence of a radiation facility, the presence of a cancer program, being a teaching hospital, not-for-profit status, and the volume of breast cancer surgeries performed at the hospital. The study by Guadagnoli et al. (1998b) also found a positive association between breast conserving surgery and the teaching status of the hospital. The odds of breast conserving surgery were 2.4 times higher for patients treated at teaching hospitals in Massachusetts, and 1.5 times in Minnesota, compared with patients treated at nonteaching facilities (Guadagnoli et al., 1998b). Interestingly, in contrast to the Nattinger et al. (1996) study, Guadagnoli et al. found no relation between BCS and hospital size or presence of a radiation facility, perhaps indicating that these factors are no longer significant once higher rates of breast conserving surgery have been achieved overall in the community.
These studies suggest that many women are not offered a choice in the type of breast surgery and that steps have to be taken to increase the implementation of recommendations to promote informed decision making. Further research is needed to determine what proportion of patients are aware that breast conserving surgery is available, whether they are given a fully informed choice of surgery, and which procedure they ultimately receive. To evaluate this aspect