Improvements in breast cancer pathology practices among medicare patients undergoing unilateral extended simple mastectomy.

The information contained in pathology reports of breast cancer specimens is of critical importance to treating physicians for selection of local regional treatment, adjuvant therapy, evaluation of therapy, estimation of prognosis, and analysis of outcomes. This information is also of great importance to patients and their families. In 2000, a Breast Cancer Pathology Advisory Group was formed to advise on the design of a project to assess the quality of pathology reports on unilateral extended simple mastectomy (ICD-9-CM procedure code 85.43) specimens from Medicare patients in New York State. This group comprised clinical pathologists, breast surgeons, medical oncologists, clinical breast cancer specialists, and a radiation oncologist. The group suggested that the reports be examined for several elements (quality indicators) that are relevant to patient care and prognosis. Baseline random sample data assessing these elements were established from a random sample of all cases for the calendar year 1999. A random sample of 748 cases (43.5%) of unilateral extended simple mastectomy was chosen from among 1718 cases for the calendar year 1999. Of these, 555 (74.2%) were suitable for review. The remaining 193 (25.8%) cases did not satisfy the inclusion criteria. Aggregate performance on 7 quality indicators (presence of carcinoma, laterality of specimen, number of lymph nodes present, number of positive nodes, documentation of lymph nodes, histologic type, and largest dimension of the tumor) was 83.7% or better, whereas performance was 69.4% or less on 10 others (resection margin status, verification of tumor size, gross observation of the lesion, histologic grade, angiolymphatic invasion, nuclear grade, location of the tumor, mitotic rate, extent of tubule formation, and perineural invasion). The last, perineural invasion, was used as a control element and was not considered an evaluative quality indicator. Performance levels for New York State were significantly lower for histologic grade, resection margin status, and angiolymphatic invasion than in similar studies elsewhere. In addition, there were significant interhospital disparities in the performance levels for these quality indicators. Whereas some hospitals always recorded certain indicators, others never did. This in part reflects differing degrees of adoption of recommended specialty society protocols. The second phase of the project consisted of an educational feedback program involving the directors of pathology laboratories in New York State. The aggregate findings of the baseline study were shared with all the pathologists. In addition, each hospital that performed unilateral extended simple mastectomies during the study period received its own specific data so that it could compare its performance with the aggregate performance. The results of the baseline study also were shared with the New York Pathological Society and the New York State Society of Pathologists. The latter described the results in its newsletter. A postintervention review of the medical charts of a sample of 297 Medicare patients discharged from New York State acute care hospitals with an ICD-9-CM procedure code of 85.43 (unilateral extended simple mastectomy) was conducted for the 6-month period from December 1, 2001, through May 31, 2002. The 8 quality indicators, performance for which was below 84% in the baseline, were chosen for this remeasurement. Statistically significant improvements (P < .0001) occurred in all the 8 quality indicators, ranging from 12.6% to 19.9%. The results of this study indicate that the issues identified by breast cancer pathology reports are amenable to improvement. Such improvement can serve both the patients and the treating physicians better in making adjuvant treatment decisions, estimating prognosis, and evaluating outcomes. It also will be of help to patients and their families in making other life decisions.