Breast Cancer--Screening Data for Assessing Quality of Services --- New York, 2000--2003
The New York State Breast and Cervical Cancer Screening Program (NYSBCCSP) is funded in part through the National Breast and Cervical Cancer Early Detection Program (NBCCEDP) (1,2). A total of 292 mammography facilities participate in NYSBCCSP. In 1998, NYSBCCSP began reviewing the clinical outcomes of clients screened through the program to identify facilities that might need more intensive assessment of quality assurance and corrections of deficiencies. This report describes the quality assurance process of NYSBCCSP, which provides methods for monitoring and assessing the quality of clinical care by program providers. These outcome-based methods can be useful to other programs and providers of breast cancer screening.
NYSBCCSP routinely compares the proportion of abnormal clinical breast examinations (CBE), abnormal screening mammography outcomes, and positive predictive values of breast cancer biopsies (ppvB) for individual program facilities with overall figures generated by the state and national programs (NYSBCCSP and NBCCEDP, unpublished data, 1995--2003). Facilities with proportions of abnormal mammography*, abnormal CBEs, or ppvB§ significantly¶ different from the lower or upper bounds of expected ranges (i.e., outliers) are identified for further evaluation. The expected ranges for abnormal CBEs are based on the NBCCEDP Data Quality Indicator Guide, a program management tool; the ranges for abnormal mammography and ppvB are commonly accepted parameters in mammography (4).
NYSBCCSP compiles additional patient data for the facilities identified with outlier results, including the age, race/ethnicity, and previous screening history of clients served; the proportion of abnormal screenings or ppvB over time; and the diagnostic procedures performed after abnormal screening results. The cancer detection rate also is calculated and compared with rates in the state and national programs. In addition, cross tabulations of screening mammography results by CBE results are reviewed.
The facility and local screening project responsible for coordinating screening and related services in a community are contacted to determine potential explanations for the outliers. A random sample of medical records is reviewed to identify data-reporting errors. If data-reporting errors cannot explain the outlier results, the clinical interpretive skills of providers at the facility are assessed. A board-certified radiologist performs the mammography assessment by evaluating the technical and interpretive quality of the mammograms. A clinician performs the CBE assessment by reviewing the position of patients during examinations, areas of breast tissue examined, search pattern used, and quality of documentation and communication skills of providers.
NYSBCCSP collaborates with the facility and local screening project to develop, implement, and evaluate an action plan to correct any deficiencies identified. The following two examples describe this quality assessment process.
Case 1. During 1996--1999, facility A, a medium-sized hospital (300--400 inpatient beds), provided mammography screening to 544 NYSBCCSP-enrolled women. A total of 73 (13.4%) of the 544 mammograms performed at facility A were classified as Breast Imaging Reporting and Data System (BI-RADS®) category 4 (i.e., suspicious abnormalities that indicating biopsy should be considered), whereas 3,750 (1.5%) of an estimated 250,000 mammograms performed at other facilities in NYSBCCSP during the same period were classified as category 4. Five (3.0%) of 166 women at facility A with abnormal mammograms underwent additional imaging, compared with 13,500 (45.0%) of approximately 30,000 women with abnormal mammograms screened at other NYSBCCSP facilities. In addition, the 166 women at facility A with abnormal mammograms were more than three times as likely to receive a breast biopsy as women at other facilities (52.0% versus 15.0%, respectively). The ppvB for detecting breast cancer from imaging at facility A was 6.9%, compared with 23.9% at other facilities.
On the basis of review of medical records and independent review of mammograms at facility A, two staff radiologists had higher biopsy recommendation rates and use of stereotactic biopsy than those of three other readers at the facility. Facility A agreed to the following corrective plan of action: 1) conduct in-service training for all five radiologists on the use of the BI-RADS® reporting system; 2) obtain a second reading by a different radiologist of all mammograms reported as a BI-RADS® category 4 for 6 months, with final resolutions by the chief of radiology; and 3) obtain facility accreditation for its stereotactic biopsy program.
During January 2001--December 2002 (approximately 4--28 months after completion of the action plan), the proportion of mammograms reported at facility A classified as BI-RADS® category 4 decreased from 13.4% to 4.3%, the proportion of women with abnormal mammograms receiving additional imaging increased from 3.0% to 7.0%, the proportion of women undergoing breast biopsies decreased from 52.0% to 28.0%, and the ppvB increased from 6.9% to 19.0%.
Case 2. During April 1995--March 2000, facility B, a private radiology practice, provided 11,972 CBEs and screening mammograms to 9,094 women through NYSBCCSP. The proportion of abnormal CBEs among clients with normal mammography results classified as BI-RADS® categories 1--3 (i.e., negative, benign finding, and probably benign finding, with initial short interval follow-up suggested) was 0.1%, compared with 2.4% for clients screened by other facilities in the program. Among a random sample of 50 clients who received a CBE at facility B, three (9.7%) of 31 women interviewed by program staff reported having received a CBE. Further data analyses determined that the breast cancer detection rate among women screened at facility B during the same period was 1.3 per 1,000 screenings, compared with 5.2 per 1,000 at other facilities. The cancer detection rate did not change after adjustment for age, race/ethnicity, and client previous-screening history. Because the difference between the cancer detection rate at facility B and the rest of the program was greater than what would be expected on the basis of CBE quality concerns alone, an assessment of the technical quality and interpretation of mammograms was conducted. The review identified inadequate technical quality and interpretation of mammograms.
The New York state health commissioner ordered facility B to cease operations immediately. The American College of Radiology also revoked the facility's mammography accreditation.
A total of 9,094 women evaluated previously at facility B were identified as possibly needing repeat breast cancer screenings. The New York State Department of Health (NYSDOH) contacted women by mail and telephone to determine current screening status and the need for rescreening appointments. Of 3,125 women who were rescreened, 734 (23.5%) had abnormal screening results that warranted additional diagnostic evaluation. Among these women, eight (1.1%) had breast cancer diagnosed. Among 1,160 other women screened initially at facility B and later screened at other facilities, 15 (1.3%) women had confirmed diagnoses of breast cancer.
The detection rate among screenings performed or identified through the follow-up rescreening effort was 5.3 cases per 1,000 screenings, approximately the detection rate in the rest of NYSBCCSP. Overall, NYSDOH identified 25 women with breast cancer, including two women screened previously at facility B outside of NYSBCCSP.
Reported by: BJ Hutton, MPH, EK Bradt, MPH, JH Chen, MD, P Gobrecht, J O'Connell, A Pedulla, MPH, T Signorelli, S Bisner, DP Hoffman, MEd, New York State Dept of Health. HW Lawson, MD, Div of Cancer Prevention and Control, National Center for Chronic Disease Prevention and Health Promotion, CDC.
NBCCEDP was implemented to increase screening among low-income, uninsured women and to reduce racial/ethnic disparities in health-related outcomes. Improvements made after applying the quality assurance process described in this report improved patient care not only for women screened through NBCCEDP, but also for women evaluated at participating facilities. The quality assurance activities undertaken by NYSBCCSP were initiated to improve clinical provider skill and have resulted in positive outcomes for facility A and NYSBCCSP.
Locating women evaluated at facility B and scheduling repeat breast cancer screening and other quality assurance efforts served as a basis for media reports about mammography quality, and stimulated discussions about ensuring adequate skill level among radiologists interpreting mammograms during reauthorization hearings for the Mammography Quality Standards Act. Mammography continues to be a highly effective tool for the early detection of breast cancer (5--7). Although neither the benefits of CBE alone nor the incremental benefits of adding CBE to mammography have been determined, CBE remains a common screening practice and detects certain cancers missed by mammography (8--10).
The quality assurance interventions described in this report are a key component of an organized cancer screening program and raise the standards of breast cancer screening throughout the provider community. Efforts to improve the technical quality and interpretation of mammography and CBE will further facilitate increasing the proportion of women whose breast cancers are identified at an early stage, which can further reduce mortality from breast cancer.
* A screening mammogram with a reported result of 4, 5, or 0 based on the American College of Radiology's Breast Imaging Reporting and Data System (BI-RADS®) classification system (i.e., suspicious abnormality, highly suggestive of malignancy, and assessment is incomplete) (3).
A palpable mass, nipple discharge, unilateral nipple inversion, or skin abnormality or unilateral area of thickening that requires immediate diagnostic evaluation.
§ The proportion of breast biopsies that result in a diagnosis of in situ or invasive breast cancer.
¶ One-sided p<0.002 for abnormal CBE and mammogram and one-sided p<0.05 for ppvB.
Disclaimer All MMWR HTML versions of articles are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the electronic PDF version and/or the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.
**Questions or messages regarding errors in formatting should be addressed to email@example.com.
Page converted: 6/3/2004
This page last reviewed 6/3/2004