Quality of breast cancer care in NCCN centers as assessed by the ASCO/NCCN quality measures: Overall performance and reasons for nonconcordance

Abstract

6506 Background: To assess the quality of breast cancer care in the National Comprehensive Cancer Network (NCCN), we compared practice against the ASCO/NCCN quality measures (QM). Methods: Using the NCCN Outcomes Database, we studied the care of women with newly diagnosed stage I-III breast cancer treated at 8 NCCN centers in 2003–6 to determine the proportion whose care was consistent with the 3 QMs (tamoxifen or anastrozole within 1 year of diagnosis for HR+ >1 cm tumors [HT]; post- lumpectomy radiation within 1 year of diagnosis for women <70 yo [RT]; and chemotherapy within 120 days of diagnosis of HR- >1 cm tumors for women <70 yo). Based on chart review, reasons for non-concordant care were classified as: 1) treatment recommended, but declined; 2) MD recommended against treatment; 3) non-recommended treatment administered; or 4) system problem (referral to the relevant specialist not documented or specialist not seen, delayed treatment initiation, or no relevant documentation in the chart). Results: 5,175 women were analyzable on at least one QM. Overall, treatment was consistent with the QM for 6,628/7,265 (91%) of the unique patient-recommendation pairs. Non-concordant care was received by 11% of patients eligible for the HT QM, 5% for the RT QM, and 13% for the chemotherapy QM. The reasons for non-concordance are shown in the table. Conclusions: Overall, breast cancer care in the NCCN is highly concordant with the ASCO/NCCN quality measures. System problems rather than patient refusal or physician recommendations account for the majority of care that deviates from the QM recommendations. This suggests that quality improvement targeting systems rather than physicians would be especially efficient and effective. [Table: see text] No significant financial relationships to disclose.