Breast cancer is a disease primarily of older women. The cumulative risk for this disease reaches its maximum well into the ninth decade of life (7). It is also a serious disease in older women. The approximate 10-year risk of disease recurrence for women 70 years of age or older who are lymph node negative with 1- to 5-cm tumors is 20%-30%; the risk for women with one to three positive lymph nodes and tumors of any size is 50%; the risk for women with four or more positive lymph nodes and tumors of any size is 80% (2). These risks are especially clinically relevant because recent gains in life expectancy have occurred at the end of life: The average life expectancy of an 85-year-old woman is nearly 6.5 years (3). During the past decade, several studies (4-9) have documented age-related variations in care among patients with early stage breast cancer. These studies were conducted in a variety of health care settings and geographic regions. They have demonstrated age-related differences in diagnostic and prognostic evaluation as well as in initial treatment patterns. As a result of these studies, there has been heightened interest in understanding, in particular, variations in the use of mastectomy versus breast-conserving surgery with or without radiation therapy among older women and the impact of these variations on patient outcomes. In this issue of the Journal, Ballard-Barbash et al. (70) have added to our understanding of age-related variations in breast cancer care through the use of a unique dataset that links Medicare1 claims records with data from nine tumor registries participating in the Surveillance, Epidemiology, and End Results (SEER) Program.2 Studying older women with newly diagnosed early stage disease, these investigators have documented the independent effects of age and comorbidity on the use of breast-conser ving surgery versus mastectomy and on the use of radiation therapy following breast-conserving surgery. Specifically, they found that women aged 80 years or more, those with two or more comorbid conditions, and those with stage I disease were more likely to receive breast-conserving surgery. In contrast, among those receiving breast-conserving surgery, the oldest old (>80 years of age) were much less likely to receive postoperative radiation therapy. In multivariate modeling, both age and comorbidity were independently associated with the receipt of postoperative radiation therapy. The oldest old and those with two or more comorbid conditions were less likely to receive radiation therapy. Particular strengths of this study include the enrollment of a large cohort of patients (n = 18 704) cared for in nine different geographic settings, the careful attention to statistical control for potentially confounding factors, and the use of a validated measure of comorbidity. Nonetheless, the limitations of the data raise questions about the validity and the interpretation of the findings. First, differential ascertainment of comorbidity may have importantly biased the data. A comorbidity score could not be calculated for 15% of the sample. Moreover, the authors note that only 36% of women who underwent breast-conserving surgery without axillary lymph node dissection had a reference hospitalization identified for the purpose of calculating the comorbidity index. Since it is the oldest old who are least likely to undergo axillary lymph node dissection (8), underascertainment of comorbidity in this group may have magnified the independent effect of age on the receipt of treatment. This possibility is supported by the fact that, for example, although there were 1352 women 80 years old or older who underwent breast-conserving surgery, only 220 of the entire sample undergoing breast-conserving surgery had two or more clinically important comorbid conditions. Not only do 80% of persons 65 years old or older have at least one chronic condition, but also the prevalence of chronic diseases increases dramatically with age, and multiple chronic conditions are especially common among older women (/7).