BACKGROUND Breast‐conserving surgery (BCS) with radiation (BCSR) requires a multidisciplinary care approach between surgeons and radiation oncologists. METHODS This retrospective cohort study examined the use of preoperative radiation oncology consultation and whether use of or distance to this care was associated with treatment choice among 1188 women age ≥ 65 years who were diagnosed with local or regional breast carcinoma in Washington State in 1994 and 1995. Study outcomes included rates of BCSR; BCS alone; and mastectomy; and radiation therapy among women who underwent BCS. RESULTS Only 29% of patients in the current study consulted with a radiation oncologist preoperatively, and less than half of the patients (46.6%) consulted with either a medical oncologist or a radiation oncologist. Among women who underwent either BCSR or mastectomy, the odds of undergoing BCSR among women who had a preoperative radiation oncology consultation were 6.7 times the odds of women who did not have the consultation (P ≤ 0.001). Similarly, the odds of receiving radiation therapy among women who underwent BCS and had a preoperative radiation oncology consultation were 5 times the odds of women who did not have the consultation (P < 0.001). The 3.4% of women who lived > 50 miles from the radiation therapy center had the lowest BCSR rate (15.8%) and had the lowest radiation therapy rate among women who underwent BCS (54.5%), although these findings were not statistically significant in adjusted analyses. CONCLUSIONS A preoperative visit with a radiation oncologist was associated strongly with BCSR use. More should be done to evaluate the role of multidisciplinary consultation in the decision to use BCSR. Cancer 2004;100:701–9. © 2004 American Cancer Society.

BACKGROUND It is not well understood whether breast density is a marker of cumulative exposure to estrogen or a marker of recent exposure to estrogen. The authors examined the relationship between bone mineral density (BMD; a marker of lifetime estrogen exposure) and breast density. METHODS The authors conducted a cross‐sectional analysis among 1800 postmenopausal women ≥ 54 years. BMD data were taken from two population‐based studies conducted in 1992–1993 (n = 1055) and in 1998–1999 (n = 753). The authors linked BMD data with breast density information collected as part of a mammography screening program. They used linear regression to evaluate the density relationship, adjusted for age, hormone therapy use, body mass index (BMI), and reproductive covariates. RESULTS There was a small but significant negative association between BMD and breast density. The negative correlation between density measures was not explained by hormone therapy or age, and BMI was the only covariate that notably influenced the relationship. Stratification by BMI only revealed the negative correlation between bone and breast densities in women with normal BMI. There was no relationship in overweight or obese women. The same relationship was seen for all women who had never used hormone therapy, but it was not significant once stratified by BMI. CONCLUSIONS BMD and breast density were not positively associated although both are independently associated with estrogen exposure. It is likely that unique organ responses obscure the relationship between the two as indicators of cumulative estrogen exposure. Cancer 2004. © 2004 American Cancer Society.

T his remarkable collection of articles draws to close a generation of research regarding the promotion of cancer screening. Beginning in the mid-1980s with breast cancer screening using mammography, there has been an unprecedented growth in our understanding of how to encourage the use of screening tests, and a transformation in how people conceptualize the issues. 1 This supplement highlights the lessons learned from that generation of research and provides the foundation for the next. Looking through this volume offers the rare opportunity to reflect on an entire body of research, representing hundreds of grants and articles published by investigators who worked during what I believe was a golden age for screening.These last 20 years were a prime time to develop our understanding of screening; we had relatively simple tests for things people could see. Cervical cancer screening had been around since the 1940s and the regular Papanicolaou (Pap) smear was part of the annual examination. 2 People could understand breast cancer and knew of it occurring in their friends and family. The work of the 1980s focused on improving followup to the familiar Pap smear and promoting mammography. Investigators initially examined how patient characteristics were associated with screening and then considered the characteristics of those physicians who did not appear to be ordering the appropriate tests. 3 The belief was if we could just get women to have the right attitude and the physicians to do their job, screening and follow-up would happen.However, screening promotion became a more complex task as the 1990s evolved. First, it was not so clear that mammography benefited all women. 4 There was some backlash with regard to what may have been an overselling of screening. 5 It also emerged that mammography was not equally available to all people, and especially not the underinsured. 6 The context of screening mattered with regard to whether screening was available and ordered. 7 It was not just a question of the right attitude and the right physician; it also became an issue of the right environment.Additional tests then began to appear such as prostate-specific antigen, sigmoidoscopy, fecal occult blood testing, genetic screens for mutations in specific genes, and tests for the human papillomavirus. Simultaneous with this growth in tests was the growth in "evidencebased" medicine. People began to be more critical concerning the basis for recommending medical interventions such as screening, and pressure grew to go beyond just ordering the test but rather to discuss what was known. Rimer et al. 8 describe this growing area of work and 1105