Women with breast cancer are cared for by a variety of providers over the course of their diagnosis and treatment and during the years following treatment. In this study, we examined the providers seen by older women at risk for and once diagnosed with breast cancer and assessed the extent to which care from various providers affects stage at breast cancer diagnosis, primary treatment, surveillance, and use of hospice. We observed that:

• rates of diagnosis with advanced-stage cancer were highest among women with no visits (36.2%) or with visits to physicians other than primary care physicians or medical specialists (15.3%), compared with women who visited either a primary care physician (8.6%) or medical specialist (9.4%) or both (7.8%). The rate of advanced cancer also decreased with increasing number of visits. Thus, even within this insured population, a substantial minority of elderly women had limited or no outpatient care in the two years before breast cancer diagnosis, and these women had a markedly increased risk of advanced-stage diagnosis. These women, many of whom were unmarried and living in poor and urban areas, may benefit from targeted outreach or coverage for preventive care visits.
  
  
• about one-fifth of older women consulted with a medical oncologist before surgery, and those who did were more likely to undergo care recommended by guidelines. Thus, elderly women for whom less than definitive surgery is being considered may benefit from preoperative visits with a medical oncologist. Further research is needed to better understand the quality and content of discussions that women, particularly elderly women, have with various providers about breast-conservation versus mastectomy.
  
  
• many older breast cancer survivors do not undergo annual surveillance mammography, particularly women who are older, black, and unmarried, and this underuse was not explained by access to follow-up care, including visits with cancer specialists. These findings suggest that new strategies are needed to increase use of surveillance mammography and decrease variations based on non-clinical factors that are likely unrelated to appropriateness of medical care.
  
  
• many older breast cancer survivors receive nonrecommended surveillance testing. Rates of testing were significantly higher among women seeing medical oncologists; whether these high rates among low-risk women are due to more symptoms or excessive surveillance is an important question for further study.
  
  
• among older women diagnosed with advanced breast cancer, rates of hospice enrollment were higher for hospitalized patients than for those not hospitalized, and higher among patients seeing a cancer specialist and primary care provider (PCP) and those seeing a cancer specialist and no PCP than among those seeing a PCP and no cancer specialist or neither type of physician in the six months before death.