one-half (52 percent) of patients had at least one unjustified dose reduction. The total dose intensity for women whose doses were reduced was not reported. Older women and treatment in a ''clinic'' versus academic or private practice were associated with nonadherence to the treatment schedule.
Other studies using breast cancer process measures suggest that the proportion of women receiving standard treatment decreases with age, although the results of the published literature are not concordant. Several studies using cancer registry data to assess care for Medicare patients have shown that elderly women do not receive the recommended treatments for breast cancer as often as younger women, even when controlling for comorbid illness. Using New Mexico Cancer Registry data from 1984 to 1986, Goodwin et al. (1993) found that while only 43 percent of women age 85 and older, and 84 percent of women age 75 to 84, received definitive treatment for localized breast cancer compared with 92 percent of women age 65 to 74 (Goodwin et al., 1993). Definitive breast cancer treatment was defined as lumpectomy or excisional biopsy followed by radiation therapy or mastectomy. Age remained significant even when controlling for women's access to transportation, physical activity levels, income, social support, ability to perform activities of daily living, mental status, and the presence of other medical illnesses. In a comparable population of women in Virginia in 1985-1989, also using cancer registry data, Hillner et al. (1996) found that although the reported number of women age 65 to 69 receiving radiation therapy after breast conserving surgery was inappropriately low at 66 percent, only 7 percent of women age 85 and older had received radiation therapy (Hillner et al., 1996). In addition, although adjuvant therapy is recommended for all patients with node-positive disease, only 44 percent of patients with positive lymph nodes received any adjuvant therapy and only 33 percent received hormone therapy. Several studies, using data on treatment collected by the SEER cancer registries, have also noted that the use of radiation therapy in women who have undergone breast conserving surgery is lower than expected and declines with age (Ballard-Barbash et al., 1996; Farrow et al., 1992; Lazovich et al., 1991). Ballard-Barbash et al. (1996) found that whereas 76 percent of women age 65-69 received radiation therapy after breast conserving surgery for Stage I or II cancer, 68 percent of 70 to 74-year-olds, 56 percent of 75-to 79-year-olds, and 24 percent of women 80 years or older were given radiation treatment (Ballard-Barbash et al., 1996). They found that although controlling for differences in comorbidity was associated with a decrease in the frequency of radiation therapy after breast conserving surgery across all age groups, the decline with age persisted.
An earlier study, using data obtained directly from the medical record in seven southern California hospitals in 1980 through 1982, also found that rates of appropriate breast cancer treatment declined with age (Greenfield et al., 1987). Greenfield et al. (1987) reviewed patients' medical records to determine whether they had received diagnostic testing, staging evaluation, and treatment that was consistent with stage-specific consensus recommendations at the time (e.g., radiation therapy after breast conserving surgery; adjuvant chemotherapy for premenopausal women with node-positive breast cancer; adjuvant hormone therapy for postmenopausal women with node-positive estrogen receptor-positive tumors). Although the proportion of women receiving the recommended diagnostic and staging evaluations did not vary with age, the proportion receiving the recommended treatment did. According to the authors, 83 percent of women age 50 to 69 received the recommended breast cancer treatment compared with 67 percent of women age 70 and older. This difference remained significant when controlling for comorbidity, stage of breast cancer, and hospital at which treatment occurred.
In contrast, in a study of postmenopausal women age 50 and older with Stage I and II breast cancer treated in 1993 at 30 hospitals in Minnesota, using data collected through patient