Executive Summary

Breast cancer is a leading cause of cancer death among women in the United States, but breast cancer mortality has been steadily declining since 1990. Early detection via screening mammography, coupled with improved therapy, has been credited with reducing the number of breast cancer deaths in the United States and other countries. Until research determines a way to prevent breast cancer, screening mammography will continue to be the primary tool in efforts to reduce the toll of the disease. Thus, ensuring the quality of mammography is important for women’s health.1

The quality and accuracy of mammography depend on both technical and human factors. When screening mammography was widely adopted for breast cancer detection in the 1980s, facilities across the country varied considerably with regard to image quality and radiation dose. A voluntary accreditation program achieved limited gains, primarily due to low participation rates, so Congress passed the Mammography Quality Standards Act (MQSA) of 1992 with the aim of ensuring uniform high quality through comprehensive and standardized regulation of mammography. The Food and Drug Administration (FDA) was granted authority to implement and regulate the provisions of MQSA.

Studies indicate that the technical quality of mammography has improved significantly since the implementation of MQSA. Questions remain, however, regarding the quality of image interpretation, which depends on human factors and is difficult to measure in practice. The available evidence indicates that mammography interpretation is quite variable in the United States, similar to other areas of medicine where observation and interpretation are subjective. In preparation for reauthorization of MQSA, Congress commissioned a study from the Institute of Medicine (IOM) to determine whether additional steps could be taken to increase the accuracy of mammography interpretation and whether the current regulations should be modified to improve the oversight process. The IOM Committee was also asked to consider the effect of recommendations on access to mammography services and to identify steps that could be taken to ensure the safe and effective use of other screening or diagnostic tools, given that technology is continually changing. The recommendations, summarized in Box ES-1, represent Committee consensus that was developed through review and discussion of published literature as well as novel survey and modeling results. Most of these recommendations could be implemented immediately without waiting for the next MQSA reauthorization, projected in 2007. However, the Committee stresses that the recommendations are interconnected and implementing the entire set will be important to achieve the objective of further improving the effectiveness of breast cancer detection. In particular, implementing supportive elements in conjunction with additional regulatory requirements will be essential to sustain access to breast imaging services.

1  

This paragraph was added to the report after its prepublication release to provide additional introductory context on mammography.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 1
Improving Breast Imaging Quality Standards Executive Summary Breast cancer is a leading cause of cancer death among women in the United States, but breast cancer mortality has been steadily declining since 1990. Early detection via screening mammography, coupled with improved therapy, has been credited with reducing the number of breast cancer deaths in the United States and other countries. Until research determines a way to prevent breast cancer, screening mammography will continue to be the primary tool in efforts to reduce the toll of the disease. Thus, ensuring the quality of mammography is important for women’s health.1 The quality and accuracy of mammography depend on both technical and human factors. When screening mammography was widely adopted for breast cancer detection in the 1980s, facilities across the country varied considerably with regard to image quality and radiation dose. A voluntary accreditation program achieved limited gains, primarily due to low participation rates, so Congress passed the Mammography Quality Standards Act (MQSA) of 1992 with the aim of ensuring uniform high quality through comprehensive and standardized regulation of mammography. The Food and Drug Administration (FDA) was granted authority to implement and regulate the provisions of MQSA. Studies indicate that the technical quality of mammography has improved significantly since the implementation of MQSA. Questions remain, however, regarding the quality of image interpretation, which depends on human factors and is difficult to measure in practice. The available evidence indicates that mammography interpretation is quite variable in the United States, similar to other areas of medicine where observation and interpretation are subjective. In preparation for reauthorization of MQSA, Congress commissioned a study from the Institute of Medicine (IOM) to determine whether additional steps could be taken to increase the accuracy of mammography interpretation and whether the current regulations should be modified to improve the oversight process. The IOM Committee was also asked to consider the effect of recommendations on access to mammography services and to identify steps that could be taken to ensure the safe and effective use of other screening or diagnostic tools, given that technology is continually changing. The recommendations, summarized in Box ES-1, represent Committee consensus that was developed through review and discussion of published literature as well as novel survey and modeling results. Most of these recommendations could be implemented immediately without waiting for the next MQSA reauthorization, projected in 2007. However, the Committee stresses that the recommendations are interconnected and implementing the entire set will be important to achieve the objective of further improving the effectiveness of breast cancer detection. In particular, implementing supportive elements in conjunction with additional regulatory requirements will be essential to sustain access to breast imaging services. 1   This paragraph was added to the report after its prepublication release to provide additional introductory context on mammography.

OCR for page 1
Improving Breast Imaging Quality Standards BOX ES-1 Summary of Recommendations to Improve Breast Imaging Quality Improve mammography interpretation Revise and standardize the required medical audit component of MQSA. Facilitate a voluntary advanced medical audit with feedback. Designate specialized Breast Imaging Centers of Excellence and undertake demonstration projects and evaluations within them. Further study the effects of CME, reader volume, double reading, and CAD. Revise MQSA regulations, inspections, and enforcement Modify regulations to clarify their intent and address current technology. Streamline inspections and strengthen enforcement for patient protection. Ensure an adequate workforce for breast cancer screening and diagnosis Collect and analyze data on the mammography workforce and service capacity. Devise strategies to recruit and retain highly skilled breast imaging professionals. Make more effective use of breast imaging specialists. Improve breast imaging quality beyond mammography Mandate accreditation for nonmammography breast imaging methods that are routinely used for breast cancer detection and diagnosis, such as ultrasound and magnetic resonance imaging (MRI). IMPROVING IMAGE INTERPRETATION The effectiveness of mammography greatly depends on the quality of image interpretation, but reading mammograms and assessing interpretive performance are both challenging. MQSA regulations include requirements for interpreting physicians regarding initial training, Continuing Medical Education (CME), continuing experience, and medical audits. The Committee addresses three primary questions relevant to interpretive performance: (1) whether the current audit procedures are likely to ensure or improve the quality of interpretive performance, (2) whether additional audit procedures could allow for more meaningful estimates of performance, and (3) whether the current CME and continuing experience requirements enhance performance. MQSA requires that mammography facilities track their performance by obtaining outcome data on women recommended for breast biopsy based on an abnormal mammogram. However, the Committee concludes that current MQSA medical audit requirements are inadequate for measuring or improving the quality of image interpretation. Ideally, medical audits should be designed to link practice patterns to patient outcomes in a way that can influence interpretive performance. Interpreting physicians need to know and understand their current level of performance before they can determine whether and how it could be improved. The first two recommendations aim to improve the quality of mammography interpretation through enhanced and standardized medical audits, one mandatory and one voluntary. The third recommendation, for demonstration projects

OCR for page 1
Improving Breast Imaging Quality Standards within Breast Imaging Centers of Excellence, attempts to combine many approaches—including double reading, high interpretive volumes, and systematic feedback—because in concert, these approaches appear to have contributed to successful screening programs in other countries. However, there is insufficient evidence to evaluate the individual components of these programs; thus the Committee cannot recommend incorporating them into MQSA regulations. Furthermore, the pluralistic organization of health care delivery in the United States precludes the adoption of some components such as systematic feedback. Therefore, the Committee believes that incentives to participate in the voluntary programs put forth in Recommendations 2 and 3 will encourage facilities to strive for a higher level of performance and that experience with these programs will eventually demonstrate their feasibility to achieve higher standards for all facilities. The fourth recommendation suggests studies that are needed to develop a stronger evidence base for specific program components, including continuing education and experience requirements, which could potentially improve interpretive performance. Recommendation 1: The FDA should revise and standardize the medical audit component of MQSA to make it more meaningful and useful. The required basic medical audit of mammography interpretation should be enhanced to include the calculation of three core measures for internal review at the mammography facility, as follows: Positive predictive value 2 (PPV2; proportion of women recommended for biopsy [BI-RADS2 4, 5] who are subsequently diagnosed with breast cancer) Cancer detection rate per 1,000 women Abnormal interpretation rate (women whose mammogram interpretations lead to additional imaging or biopsy) The group of women that facilities are required to track should include not only women with BI-RADS 4 and 5 assessments, but also all women for whom additional imaging is recommended (defined in BI-RADS as Category 0; incomplete—needs additional imaging)3 to facilitate resolution of all cases so that women for whom biopsy is recommended at final assessment will be included in the calculation of PPV2. All performance measures (PPV2, cancer detection, and abnormal interpretation rate) should be stratified by screening and diagnostic mammography. Facilities should have the option of combining medical audit measures for physicians interpreting at multiple facilities to allow for more meaningful data. 2   The Breast Imaging Reporting and Data System (BI-RADS) provide a standardized method of categorizing mammography results. Biopsy is recommended for women assigned BI-RADS Category 4 or 5. 3   FDA regulations and BI-RADS also permit the assignment of assessment Category 0 when awaiting comparison with prior films. All of such Category 0 assessments should be given final assessments within 30 days.

OCR for page 1
Improving Breast Imaging Quality Standards Audit data collection and analysis should be verified at FDA inspection, but not collected by FDA. Reimbursement rates for mammography should be increased to account for the additional costs of these new audit procedures. The principles of RBRVS (resource-based relative value scale, which compares relative physician work for different types of services) should apply to both the physician and facility. Rationale: The current medical outcomes audit mandated by MQSA does not require mammography facilities to calculate any specific performance statistics. In practice, most facilities use one type of PPV as a performance measure. However, no single measure accurately reflects interpretive performance, and the method of PPV calculation varies widely. Given the challenges and limitations of the various possible performance measurements, the committee recommends use of PPV2 because it provides more useful information than PPV3,4 and because it is easier to calculate than PPV1.5 Calculating the rates of cancer detection and abnormal interpretation would facilitate appropriate interpretation of PPV2, which is influenced by the prevalence of cancer in the screening population. Universal adoption of these three measures, which are standard practice in successful screening programs of many other countries such as those in the United Kingdom, Sweden, and British Columbia, would be a significant and achievable, albeit imperfect, improvement over the current audit requirement, under which facilities need only collect and review data on the outcomes of women for whom a biopsy is recommended. This combination of measures would provide facilities with consistent and meaningful feedback, thus making it more feasible for audit interpreting physicians to identify the need for performance improvement plans. Measuring sensitivity and specificity would be more useful, but calculating these measures in community practices that do not have linkage with a tumor registry to determine the final diagnosis is not feasible. Because most screened women who are referred for biopsy begin with a recommendation for additional imaging, the tracking requirement under current MQSA regulations produces a biased assessment of performance by focusing on only examinations initially assigned BI-RADS 4 and 5 assessments. Combining medical audit data for screening with diagnostic examinations, as is permitted under current MQSA regulations, confounds the meaning of the results, making it difficult to interpret and compare performance with current literature or established databases. Medical audit data are more meaningful when larger numbers of examinations are analyzed. Individual interpreting physicians at a particular facility may generate 4   PPV3 is the proportion of all women biopsied due to the interpreting physician’s recommendation who are diagnosed with cancer at the time of biopsy. 5   PPV1 is the proportion of all women with positive examinations (Category 0, 4, or 5) who are diagnosed with breast cancer.

OCR for page 1
Improving Breast Imaging Quality Standards insubstantial data; combining data from the multiple facilities where they practice would be more meaningful. The current medical outcomes audit data are reviewed during inspection to ensure that the facility is complying with the regulations, but FDA does not collect the data. Each facility must designate an “audit interpreting physician” to review the audit data, notify other interpreting physicians of the result, and document the nature of any follow-up actions. No change in this procedure is warranted because regulators would not be able to verify the accuracy of audit data. The workload and costs associated with meeting MQSA requirements are significant, and the new audit procedures proposed here will add to the workload and expense of adhering to MQSA requirements. Historically these costs have not been factored into reimbursement, placing a considerable financial burden on facilities. However, the Centers for Medicare and Medicaid Services (CMS) and other health care payers should account for both the technical and professional costs of adhering to federally mandated audit procedures when establishing reimbursement rates for mammography. Recommendation 2: Facilities should be encouraged to participate in a voluntary advanced medical audit with feedback. This should be facilitated by incentives for participation and the formation of a data and statistical coordinating center. In addition to all tracking, measurements, and assessments in the enhanced basic required audit described in Recommendation 1, the voluntary advanced audit should include the collection of patient characteristics and tumor staging information from pathology reports. A central data and statistical coordinating center, independent of a regulatory authority, should be established and maintained to collect and analyze the advanced audit data to provide feedback to interpreting physicians for quality assurance and interpretive improvement. Sufficient funding should be provided for (1) data collection, analysis, and feedback; (2) appropriate hardware and software for data management; and (3) appropriate information technology support personnel for data maintenance. The coordinating center should: Electronically collect, analyze, and report advanced-level audit data and provide regular feedback to interpreting physicians. Help develop, implement, and evaluate self-improvement plans for interpreting physicians who do not achieve performance benchmarks. Routinely release aggregate summary data on interpretive performance, including recall rates, PPV2, and cancer detection.

OCR for page 1
Improving Breast Imaging Quality Standards Test different methods of delivering audit results to improve interpretive performance. Undertake studies of randomly selected facilities using required basic audit procedures to ascertain the impact of these new measures on interpretive quality. Protect from discoverability the data collected for purposes of quality assurance. Incentives must be provided to encourage mammography facilities to strive for a high standard of interpretive performance through participation in the advanced audit program. Facilities participating in the voluntary advanced audit should be exempt from FDA inspection of medical audit data (documentation of participation in the advanced audit process would be sufficient). CMS and other health care payers should pay for performance by providing a higher reimbursement rate in exchange for meeting performance criteria. Specific performance criteria should be determined (and periodically updated) by an informed group of experts and patient advocates. Rationale: Information from pathology reports can be difficult for facilities to obtain, but audit data will be more useful if the sizes of invasive cancers are recorded because interpretation of PPV2 is informed by tumor size. Other staging indices, such as axillary lymph node status, are important to assess whether cancers are being detected early. In addition, performance data on individual interpreting physicians may be misleading without adequate consideration of patient characteristics such as age, family history, breast density, presence of prior films, and time since last mammogram. Data collection is the catalyst for change, but data are inadequate without resources for accurate and uniform analysis and feedback to improve quality. A statistics and analysis group is crucial for quality assurance in successful organized screening programs of other countries. An analogous research effort was initiated by the original MQSA authorization. Uniform feedback should help participating U.S. facilities attain a higher level of quality assurance than is feasible now. However, data are lacking on how feedback can best be used to improve performance, so studies are needed to optimize the impact of this approach. The statistical coordinating center could also aid the basic required audits by developing national benchmarks that facilities and interpreting physicians could use to assess their performance, and by testing the impact of the basic audit procedures. Public release of aggregate summary data and benchmarks will benefit all mammography participants. Protecting audit data from discoverability is important to ensure accurate reporting and widespread participation. Regulatory oversight is unnecessary for a voluntary program, and there is potential for conflict if a regulatory body also provides analysis and feedback for quality improvement.

OCR for page 1
Improving Breast Imaging Quality Standards The Breast Cancer Surveillance Consortium (BCSC) has already developed effective procedures and guidelines for mammography data collection, and has demonstrated the feasibility of such an undertaking through a cooperative agreement with the National Cancer Institute (NCI). Given the established expertise and success of the BCSC, it is a model for this new endeavor. The Agency for Healthcare Research and Quality (AHRQ) also has many characteristics that make it a viable option. CMS is developing a pay-for-performance policy, and private health insurers often follow the lead of CMS. The extensive quality assurance procedures proposed for the voluntary advanced audit justify the use of such an approach for mammography. Participation in the voluntary advanced medical audit will likely lead to a higher quality of performance and care, but it also will entail a considerable increase in cost and workload. Supportive elements such as feedback and protection of quality improvement data to improve interpretive performance may not be implemented as easily as regulations, but in their absence, the advanced audit would be viewed primarily as an added burden by mammography facilities and personnel, thereby limiting the number of facilities that participate. Recommendation 3: Establish a demonstration and evaluation project to designate and monitor the performance of specialized Breast Imaging Centers of Excellence that attempt to incorporate aspects of successful organized breast cancer screening programs. This undertaking should be rapidly initiated with high priority and will likely require cooperative efforts of several organizations, perhaps including NCI, CMS, and AHRQ. Patient advocates should be involved in the design and oversight of the Centers. Centers would participate in the basic and advanced medical audits described in Recommendations 1 and 2, but would also test additional approaches to improving the quality and effectiveness of mammography. Centers of Excellence should incorporate and test the effects of the following attributes: high volume, double reading, demonstration of proficiency via comprehensive medical audit as defined in the BI-RADS atlas (which goes beyond even the voluntary advanced audit in Recommendation 2), a documented quality assurance process, and patient reminder systems. In addition, pilot projects should be established within selected centers to further develop and evaluate interpretive skills assessment exams. Centers should also incorporate expertise in accepted nonmammographic imaging modalities for breast cancer diagnosis (for example, ultrasound and magnetic resonance imaging, or MRI), with accreditation where applicable. Incentives to seek this designation should be similar to those described in Recommendation 2 (e.g., higher reimbursement rates). Centers could also use the designation to recruit patients and referrals. In addition, high-performing interpreting physicians who work in Centers of Excellence should have the opportunity to participate in a program to determine the feasibility of a no-fault medical liability system.

OCR for page 1
Improving Breast Imaging Quality Standards Designated centers should be encouraged to serve as training centers for breast imaging and as regional readers of mammograms to alleviate access problems in underserved areas. Ideally these Breast Imaging Centers of Excellence should be linked with facilities that provide comprehensive and multidisciplinary breast care. Rationale: Several countries have successfully developed integrated and centralized breast cancer screening programs that incorporate these attributes, but in the United States, screening is decentralized and offered in diverse practice settings. Although adapting quality assurance programs of countries with national health care systems to the fragmented delivery of U.S. health care may not be fully feasible, the challenge is not insurmountable—it has occurred within some integrated health plans and through the NCI Breast Cancer Surveillance System. It is urgent to further test the concept by designating specialized Centers of Excellence in Breast Imaging that would strive to achieve a higher level of integration, performance, and quality assurance in breast cancer detection and diagnosis by adopting many components of successful organized programs. This multifaceted approach should contribute to the optimal performance of mammography, but mammography is not the only imaging tool used for breast cancer diagnosis. The Centers of Excellence could provide multidisciplinary training and work environments for diagnosis. Centers could increase job satisfaction, retention of practitioners, and the productivity and quality of all members of the breast care team—high-quality facilities would attract high-quality personnel at all professional levels. As noted in Recommendation 2, supportive elements and incentives are critical to encouraging facilities and personnel to strive for higher quality. In their absence, meeting the requirements for designation as a Center of Excellence would likely be viewed as an unnecessary burden, thereby limiting participation. Nonetheless, incentive programs such as the no-fault insurance system must be tested as well. As noted previously in the IOM report Fostering Rapid Advances in Health Care: Learning from System Demonstrations (2003), a no-fault system linked to high quality assurance could potentially benefit both patients and providers. Such a system has worked effectively for many years in other countries such as Sweden. In contrast, medical malpractice lawsuits generally have not been found to have a positive influence on quality of care, and compensation for misdiagnosis under the current system is inconsistent. The goal would be to provide patients with access to higher quality care, while offering consistent and fair compensation in the event of a misdiagnosis and eliminating the need for long, difficult lawsuits with uncertain outcome. Highly skilled physicians meeting stringent qualification criteria would practice in a safe harbor, without fear of lawsuits and the label of fault when cancers are missed by mammography. The threat of lawsuits and the cost of malpractice insurance are major concerns for interpreting physicians and mammography facilities, but even the most skilled and experienced interpreting physicians will miss some cancers because of the inherent limitations in mammography tech-

OCR for page 1
Improving Breast Imaging Quality Standards nology and interpretation. Approximately 30 percent of breast cancers can, only in retrospect, be seen on previous mammograms interpreted as normal. Centers would have the expertise to develop and host training programs in breast imaging. Also, interpretation at centralized facilities could help alleviate access problems in low-volume areas. Implementation of centralized reading at designated high-volume/high-quality centers could be facilitated by either high-tech approaches (softcopy telemammography) or low-tech approaches (shipping films). In the United States, there is a lack of continuity among screening, diagnosis, treatment, and follow-up care. Breast imaging centers that incorporate all aspects of screening, diagnosis, and imaging-based patient management, and are linked with facilities providing comprehensive and multidisciplinary nonimaging breast care, can facilitate seamless transitions for patients as needed. Recommendation 4: Although some evidence suggests that additional approaches could also improve the quality of mammography interpretation, the data currently available are insufficient to justify regulatory changes. Thus, studies are urgently needed to: Demonstrate the value of CME specifically dedicated to improving interpretive skills. Determine the effects of reader volume on interpretive accuracy, controlling for other factors that improve interpretive performance. Improve and measure the impact of double reading and computer-aided detection (CAD) on interpretive performance over time, in different practice settings, and at different levels of experience. The funding for these studies should be provided through NCI, with the expectation that study results will motivate additional improvements to future reauthorizations of MQSA. Rationale: Requiring that a portion of MQSA-mandated CME be dedicated to mammography interpretive skills assessment could enable interpreting physicians to identify weaknesses and take steps toward improving interpretive performance. However, there is an urgent need to study specific educational approaches and determine their effectiveness. Continued development of innovative teaching interventions to improve mammography interpretive skills is also important. Although the current minimum requirement for continuing experience (960 mammograms every 2 years) is quite low given the prevalence/incidence of cancer in the screening population (generally 4 to 6 cancers detected per 1,000 women), there is insufficient evidence at this time to recommend an increase in minimum interpretive volume. Published reports provide conflicting evidence regarding the relationship between

OCR for page 1
Improving Breast Imaging Quality Standards reader volume and interpretive accuracy, and there is no basis for specifying a higher level of reader volume. A variety of approaches may improve reader performance, but it is unclear to what extent interpretive volume and experience alone contribute to that improvement. Volume alone cannot contribute to improvement unless it is accompanied by accurate feedback about performance. Given the uncertainty regarding the isolated effect of reader volume on interpretive performance, and given that increasing the minimum reader volume could reduce the supply of mammography services in some areas, maintaining access should be of primary concern. Intuitively, a second look by another reader or a computer program should increase sensitivity, but the benefit has not been unequivocally characterized and quantified in prospective clinical trials. The effects on specificity are also not fully understood. Some studies indicate that any potential increase in sensitivity could be accompanied by a decrease in specificity. In addition, various CAD programs are still being refined, so the effect and use of CAD may change over time. Studies are needed to evaluate the impact of physician experience and lesion characteristics on the effectiveness of CAD. These studies should be done with a standard set of cases that were not used to develop the CAD systems. The findings could help physicians use CAD more effectively and could help improve the underlying algorithms. There are many different ways to approach double reading, each requiring different amounts of personnel and time, with consequent differences in cost and effects on access to mammography services. Consensus double reading (one of the most costly and personnel-intensive approaches) may be the most effective, but formal studies are needed to confirm this. CHANGES TO MQSA REGULATIONS, INSPECTIONS, AND ENFORCEMENT Final MQSA regulations were promulgated in 1999, and a review is warranted at this time to identify areas in need of revision. Advancements in technology, particularly digital mammography, require expansion of the current regulations to ensure quality mammography. At the same time, revising the current FDA regulations, including some deletions, and streamlining the inspection process could reduce the burden on facilities and allow for increased focus on improving interpretation. FDA should be provided with adequate resources to revise, disseminate, implement, and enforce these changes. Recommendation 5: FDA should modify MQSA regulations to clarify the intent and to recognize current technology and applications. For example: FDA should remove the exemption on interventional mammographic procedures such as stereotactic breast biopsy and preoperative mammographic needle localization, and develop specific regulations for these X-ray procedures and equipment.

OCR for page 1
Improving Breast Imaging Quality Standards Regulations specifically addressing the use of digital mammography need to be developed. Regulations regarding mammography reports should be updated to reflect the most recent version of BI-RADS, including a new category for “known biopsyproven malignancy.” FDA should establish luminance standards for viewing mammograms. The requirement for modality-specific CME should be eliminated for interpreting physicians to allow them a broader choice of educational opportunities, including those that focus on interpretation skills. A complete list of recommended changes to the regulations and the rationale supporting such changes can be found in Table 3–1 of Chapter 3. Recommendation 6: FDA should modify MQSA inspections to streamline the process, reduce redundancy, and address current technology and applications. Enforcement needs to be strengthened for further patient protection. For example: Several onsite inspection tests are unnecessary and should be eliminated. FDA should have the authority to require that facilities cease performing mammography after two consecutive unsuccessful attempts at accreditation, even if their MQSA certificate is still valid. FDA should require a facility that closes for any reason or has its certification revoked to notify patients and their referring physicians. In addition, regulations for film retention should apply to facilities that close. Rationale: Several onsite inspection tests are redundant and have exceptionally low rates of failure; quality would not be adversely affected by their elimination because these parameters are monitored regularly. Until 2003, FDA required facilities to cease mammography in such cases. In 2003, FDA informed accrediting bodies that they cannot require facilities to cease mammography if their MQSA certificate is still valid. FDA has received a number of complaints from patients who were not informed when their facility closed and as a result were unable or unsure of how to access their mammography records. If facilities are incapable of notification, FDA should assume responsibility for notification.

OCR for page 1
Improving Breast Imaging Quality Standards ENSURING AN ADEQUATE WORKFORCE FOR BREAST CANCER SCREENING AND DIAGNOSIS Results from surveys conducted by the American College of Radiology (ACR), the Society of Breast Imaging, and the American Society of Radiologic Technologists, as well as many anecdotes, suggest that the supply of mammography services may be decreasing in some areas. Furthermore, FDA data suggest that the number of interpreting physicians is decreasing. However, no systematic data collection and analysis has been undertaken, making it difficult to assess the current capacity for breast cancer screening and diagnostic services, and even more difficult to plan for the future. Early detection of occult breast cancer is a key element for reducing breast cancer mortality; it is therefore crucial to accurately monitor the capacity of mammography services, as indicated in Recommendation 7, and to ensure adequate access for women. Demand is likely to increase in the future as population demographics change and the number of women eligible for screening mammograms increases. Increased use of other imaging technologies for breast cancer detection and diagnosis could also raise demand for breast imaging. However, in recent years, the majority of fellowship training slots in breast imaging have gone unfilled, and surveys indicate that many radiologists are uninterested in pursuing breast imaging as a specially. Three basic strategies, as noted in Recommendations 8 and 9, could be used to ensure an adequate breast imaging workforce: increasing the number of new entrants to the field, retaining the current workforce, and increasing productivity of new and existing practitioners. Recommendation 7: Data describing the national mammography workforce, volume of services, and capacity should be immediately and routinely collected by FDA and made available to appropriate state and local agencies for tracking and monitoring. For example: Volume information should be added to the data that FDA currently collects in the annual facility inspection, and the Health Resources and Service Administration (HRSA) should produce routine reports on the volume of mammography services by region, state, and type of service. Measures should include the number of facilities, number of certified mammography units per 10,000 women, and number of full-time equivalent (FTE) physicians reading mammograms per 10,000 women,6 and should be stratified by type of service (e.g., screening versus diagnostic) where appropriate. These data should be collected using unique identifiers for all certified physicians, technologists, and medical physicists to facilitate tabulation of volume of services by individual. 6   For purposes of this report, a full-time equivalent physician is one who is employed full-time (37.5+ hours per week of paid employment for 40 or more weeks per year) and spends at least 80 percent of his or her time performing clinical radiology imaging interpretation. Note that an FTE radiologist is not necessarily an FTE breast imager. For example, a full-time radiologist who devotes 40 percent of practice time to breast imaging could be equivalent to a radiologist who works 40 percent of the time and devotes all of his or her practice to breast imaging.

OCR for page 1
Improving Breast Imaging Quality Standards FDA should also collect data by facility on waiting times for screening and diagnostic appointments. These data should be tabulated by region and state and be published routinely. In addition to the funding needs for collection, analysis, hardware, software, and IT support, sufficient funding should also be provided to HRSA to model future workforce supply and demand on a regular basis. Rationale: There is a paucity of accurate “real-time” data available to monitor and track capacity on a national and regional basis. Without accurate tracking, it is impossible to determine the current status of the workforce and it is extremely difficult to plan actions for the future. In the absence of national data, there is also no consensus on appropriate waiting times for screening and diagnostic mammograms. There is an urgent need to begin data collection immediately because identifying trends will take several years. Tracking mammography capacity will also be important in monitoring the impact of new regulations and voluntary programs. If the fragile stability of the breast imaging workforce moves toward crisis, data will be needed to react swiftly and effectively. Collection of workforce and capacity data would also simplify documentation of staff qualifications and continuing education and experience requirements. Recommendation 8: Devise strategies to retain highly skilled breast imagers and increase the number of new entrants into the breast imaging field, particularly in underserved areas. For example: Encourage federal and state agencies and health care payers to develop incentives, both financial and otherwise, to recruit and retain skilled breast imagers. One such approach would be to encourage provision of pro-rated malpractice insurance to interpreting physicians skilled in mammography who wish to work part-time. Appropriately qualified physicians who spend at least 50 percent of their professional time in breast imaging working in underserved areas should be eligible for loan repayment awards through the National Health Service Corps (NHSC) and for J-1 visa waivers authorized by federal and state agencies. In order to target breast imagers to the highest need areas, HRSA should establish a process to identify and designate shortage areas for breast imaging. Rationale: The existing supply of physicians who read mammograms at a high level of interpretive performance is a valuable resource. It is unproductive to invest in efforts to increase the number of entrants into the specialty without also addressing factors that lead to early departures from the existing workforce. Efforts directed at retaining already highly skilled practitioners, even for part-time work, could be a cost-effective way to maintain access to high-quality breast imaging

OCR for page 1
Improving Breast Imaging Quality Standards services. For example, interpreting physicians who wish to work part-time often find it difficult to interpret mammograms because malpractice costs are not adjusted for less than full-time work. Offering pro-rated malpractice insurance premiums would greatly reduce this barrier. The provision of reduced-rate malpractice tail coverage for part-time workers would also be beneficial. The NHSC program and J-1 waivers have been used to bolster the workforce in other shortage areas in medicine and biomedical research. Together, these approaches would increase the number of interpreting physicians in areas with shortages and also increase the aggregate supply. It would also signify to medical students and physicians in training the need and importance of breast imaging specialists. Recommendation 9: Make more effective use of the existing supply of highly skilled breast imaging professionals. Support radiologist assistant (RA) training programs and explore possible new roles for RAs in breast imaging. Support demonstration projects to evaluate the potential for nonphysician clinicians (e.g., nurse practitioners, physician assistants, radiologist assistants, radiologic technologists, etc.) to double read mammograms accurately. These demonstrations should be undertaken within the Centers of Excellence described in Recommendation 3. Support demonstrations to evaluate the roles of ancillary personnel in mammography, such as administrative and data entry personnel, and to assess the impact of alternative staffing configurations on the efficiency, productivity, and quality of breast imaging services. Rationale: RA programs offer an attractive career option for skilled radiologic technologists and also could be an incentive for new entrants into the field. Maximizing the potential role of RAs in breast imaging could improve quality as well as facility productivity and efficiency. Double reading has the potential to improve the accuracy of mammography interpretation, but widespread adoption of double reading would strain the available interpreting workforce. Permitting nonphysician clinicians with special training to serve as second readers, only under the direct supervision of MQSA-certified interpreting physicians, could provide a cost-effective way to expand the use of double reading. Several small studies already suggest that this approach is feasible in mammography. In addition, Pap tests present an important precedent in women’s health, as they are routinely read by nonphysician cytologists under the supervision of a physician. Productivity will be maximized if radiologic technologists focus their efforts on imaging procedures and if breast imaging physicians focus on image interpretation and performing interventional breast imaging procedures. Ancillary personnel could

OCR for page 1
Improving Breast Imaging Quality Standards make an important contribution by taking on nontechnical responsibilities, including quality control and administration. BEYOND MQSA TO BIQSA Medical technology is constantly evolving. Although mammography is still the only recommended breast cancer screening test for the general population, a number of other breast imaging technologies are clinically available, and more are in development. Some of the available imaging technologies, including breast ultrasound and MRI, are already commonly used in the diagnosis of breast cancer. Furthermore, recent studies have suggested a potential role for specific technologies in screening some portion of the population, such as high-risk women. Of concern is the adoption of some technologies for screening despite the limited evidence of their effectiveness. There is no mandatory quality oversight of these other technologies, and quality is known to be variable. Thus, the goal of MQSA to ensure quality breast cancer screening and diagnosis could go unfulfilled if it continues to focus solely on mammography. Recommendation 10: Mandatory accreditation for breast imaging methods not utilizing ionizing radiation that are routinely used for breast cancer detection and diagnosis should be required under the next MQSA reauthorization. This would entail a name change to the Breast Imaging Quality Standards Act (BIQSA). Initially, breast ultrasound and MRI, along with associated interventional procedures, should be subject to mandatory accreditation, although in the case of MRI, accreditation programs specific to breast imaging must first be developed. In addition, a committee of experts and patient advocates should determine if and when other technologies should be subject to accreditation or certification. Rationale: Accreditation programs already exist for breast ultrasound through the American College of Radiology (ACR) and the American Institute of Ultrasound in Medicine. The ACR also has an accreditation program for general MRI, and has begun a dialogue regarding the development of a breast-specific MRI accreditation program. Compulsory accreditation for these breast imaging methods would lead to standardization and could greatly improve the overall quality of breast cancer detection and diagnosis.