[A population-based study on variations in the use of adjuvant radiotherapy in breast cancer patients]

The assessment of the compliance with consensus recommendations for adjuvant radiation therapy among women with breast cancer. The study is based on data obtained in a population-based cohort-study, which was performed to evaluate the quality of health care for patients with breast cancer.

About one million inhabitants live in the study region Eastern Thuringia, 2,031 cases with invasive breast cancer without distant metastasis (MO) or inflammatory spread were registered from 1995 to 2000. Out of these 1,700 with complete documentation of covariates were included in multivariate analysis. To examine the simultaneous influence of all clinical factors and "caseload" on the likelihood to receive adjuvant radiation therapy a logistic regression model was fitted for radiation therapy after mastectomy. In order to describe the impact of each individual clinic on treatment decision as "caseload" was replaced by the clinics with more than 30 primary treatments.

Following breast conserving therapy (BCT) 90.6% of the patients received adjuvant radiation therapy. In the univariate analysis older age was negatively associated with the use of radiation therapy among women with BCT (Table 1). Furthermore, comorbid conditions were negatively associated with the use of radiation therapy. For all other cofactors no associations were found. Subsequent to mastectomy 33.0% of the women underwent radiation therapy (Table 2). Associations between the use of radiation therapy and age, tumor category, number of positive lymph nodes, multiple tumors, histologic differentiation grade, residual tumor as well as hormone receptor status were found. In the multivariate analysis only older age (>/= 70 years) was identified as negative indicator for the utilization of radiation therapy. Among patients with mastectomy increasing tumor size was a positive predictor on radiation therapy (Table 3). In addition more than three positive lymph nodes, multiplicity, poor histologic differentiation grade (G3/4), medial localization, and younger age (35-49 years) were positive predictors for the application of radiation therapy. For R-classification, hormone receptor status and comorbidity no influence on the use of radiation therapy was observed. Among clinics adjusted for case-mix variations in the compliance of treatment recommendation both following BCT and mastectomy were found (Figure 1).

Following BCT compliance with treatment recommendations regarding radiation therapy was high. Women in older age or with comorbidities received less often radiation therapy. Subsequent to mastectomy indicators for the use of radiation therapy were identified. Among clinics differences in the adherence to treatment recommendations were observed.