.J;Ilrrl~lrv . I!)!)() SCREENING MAMMOGRAPHY Low-Cost Services Do Not Compromise Quality L* 1 United States G/f%!0 General Accounting Office Washington, DE. 20548 , Human Resources Division *January 10, 1990 The Honorable Lloyd Bentsen Chairman, Committee on Finance United States Senate The IIonorable John D. Dingell Chairman Committee on Energy and Commerce I Iouse of Representatives The Iionorable Dan Rostenkowski Chairman, Committee on Ways and Means IIouse of Representatives The Honorable Barbara B. Kennelly Ilouse of Representatives This report responds to a provision of the Medicare Catastrophic Coverage Act of 1988 (P.L. 100-360) that required the Comptroller General to review the quality of screening mammography provided in a variety of settings. The report also responds to a request from Representative Barbara Kennelly to provide additional information on screening mammography, including the services constituting a complete mammography examination and the procedural differences between screening and diagnostic mammography. We are sending copies of this report to certain other House and Senate committees and subcommittees, the Secretary of Health and Human Services, the Director of the Office of Management and Budget, state health departments, and other interested parties, We also will make copies available to others on request. Please call me on (202) 275-6195 if you or your staff have any questions about this report. Other major contributors are listed in appendix VI, Mark V. Nadel Associate Director for National and Public Health Issues E&ecutiveSummary The Medicare Catastrophic Coverage Act of 1988 (P.L. 100-360) intro- Purpose duced screening mammography for symptom-free women as a new bene- fit for Medicare-eligible women, to become effective in 1990. The act / limited the charge for Medicare-funded screening to approximately $50. Some members of Congress were concerned that this limit could compro- mise women’s ability to obtain quality services. To help assure that I quality services would be provided, the act required (1) the Secretary of Health and Human Services (HHS) to establish standards to assure the safety and accuracy of this test and (2) the Comptroller General to report on the quality of screening mammography performed in a variety of settings. Representative Barbara Kennelly also asked GAOto provide additional information on screening mammography, including what con- stitutes a complete screening examination. In November 1989, the Congress repealed most provisions of the Medi- care Catastrophic Coverage Act of 1988, including the mammography benefit. However, legislation to restore the mammography benefit has been introduced. Breast cancer causes over 40,000 deaths per year in the United States. Background The best method for improving a woman’s chance of survival is early detection, and the most effective tool for early detection is mam- mography, an X-ray that can find cancers too small for the woman or her physician to feel. Leading medical organizations recommend that women begin periodic mammographic screening at the age of 40; regular screening can reduce mortality rates by 30 percent. Screening mam- mography, which is performed on women without symptoms to detect unsuspected abnormalities, can be provided more economically than diagnostic mammography. The latter is used to provide more detailed information about abnormalities that have been discovered. (See pp. lo-12-j Despite the recommendations for regular screening mammography, studies indicate that over half of women over 40 have never had the test. A new Medicare benefit could greatly increase the number of women receiving such screening. (See pp. 12-13.) GAOinterviewed mammography experts to identify quality standards that contribute to optimal screening mammography and sent a question- naire to 1,485 facilities providing mammography in four states-cali- fornia, Florida, Idaho, and Michigan. The facilities were asked to Page 2 GAO/HRD-90-32 Quality of Screening Mammography Executive Summary categorize their type of setting and supply information on services pro- vided, equipment, personnel performing and interpreting mammograms, quality assurance activities, reporting and record-keeping, volume, and charges for screening mammography services. In addition to obtaining information on federal oversight of mammography and laws and regula- tory programs in the four states reviewed, GAOvisited 15 screening mammography providers to supplement the survey data. (See pp. 14-16.) Many providers lack adequate quality assurance programs. This may Re:zlts in Brief contribute to the wide range of image quality and patient radiation dose that occurs in current mammography practice. GAOfound no relation- ship between the price charged for screening mammography and adher- ence to quality standards. Providers with higher mammography volume were more likely to comply with many quality standards than were those with lower volume. There is evidence that high volume permits economies of scale and does not compromise quality. Federal and state oversight programs have been limited by the absence of legally binding quality standards. In September 1989, however, 1111s published proposed regulations for Medicare-funded screening mam- mography that parallel professional quality standards. Such regulations would help federal and state regulators in assuring that mammography providers deliver quality services. Because the Medicare Catastrophic Coverage Act of 1988 was repealed, HHSwill withdraw its proposed regulations. GAO’s Analysis Quality Standards Exist, The quality of screening mammography depends on providers comply- bug Image Quality and ing with a wide range of quality standards. The primary goal is a mam- mographic image of good quality, obtained with low radiation dose to Patient Dose Vary the patient and followed by accurate interpretation of the image. Although professional groups have established quality standards to guide facilities, these standards are not uniformly followed, and image quality and dose vary widely in current mammography practice. (See Y pps 23-25.) Page 3 GAO/HRD-90-32 Quality of Screening Mammography Executive Summnry Qublity Assurance To maintain a consistent level of quality, providers should perform qual- Stdndards Met Less Often ity assurance activities, such as periodic inspections of equipment per- formance. Many providers that GAOsurveyed, however, were not Thbn Other Standards complying with standards for quality assurance programs. Inadequate / ! quality assurance programs may be a cause of the variation in image quality and dose. Primary care physicians and multispecialty clinics were the providers reporting the lowest rates of compliance with stan- , dards for quality assurance activities. (See pp. 26-32.) M re High-Volume High-volume providers were more likely to comply with many quality Pr4 viders Comply W‘ith standards than were low-volume providers. However, providers charg- ing higher fees were no more likely to adhere to quality standards than Qu(ality Standards were those charging lower fees. Several providers that GAOvisited reported that they met many quality standards while charging $50 or less for screening mammograms; all were high-volume facilities. (See pp. 32-34). E&nomies Possible With Compared with diagnostic mammography, screening mammography is a Screening Mammography less complex process, which permits certain economies of scale that make it less costly to provide. The radiologist need not be present during Not Always Realized the examination to read the films, but can read a large batch of films at once, resulting in a more efficient use of the radiologist’s time. In addi- tion, increasing patient volume results in more efficient use of expensive equipment. (See pp. 12 and 20-21.) Most providers operate at relatively low volume-fewer than 50 mam- mograms per week. Survey respondents reported charging a wide range of fees for screening mammography, with the average about $100. One reason most charges are higher than $50 may be that providers base these charges on the more complex and expensive diagnostic procedure. (See pp. 21-22.) Limits of Current State oversight of mammography is limited. Three of the four states GAO Regulation Make HHS reviewed have no legally binding standards for image quality or radia- tion dose; nor do they require the use of equipment specifically designed Standards Important for mammography. The federal government currently does not regulate the quality of mammography. The Food and Drug Administration (FDA) * is responsible for ensuring the correct manufacture and installation of X-ray equipment, but not for overseeing its subsequent use. FDAinspects a relatively small number of mammography machines. (See pp. 35-40.) Page 4 GAO/HRD-90-32 Quality of Screening Mammography - Executive Summary The limitations of existing oversight and evidence of problems with image quality and radiation dose underscore the importance of quality standards the Medicare Catastrophic Coverage Act of 1988 required the Secretary of HHSto issue for Medicare-funded screening mammography. Proposed standards published by HHSbefore repeal of the screening mammography benefit paralleled professional quality standards. Such standards would have filled existing regulatory gaps, (See pp. 41 and 43.) - GAOmakes no recommendations in this report. Rethommendations Page 5 GAO/HRpBO-82 Quality of Screening Mammography contents Exbcutive Summary 2 Chbpter 1 10 Inqroduction Screening Mammography Can Reduce Breast Cancer 10 I Mortality Purposes of Screening and Diagnostic Mammography 12 / Differ Efforts Made to Increase Use of Screening Mammography 12 / Medicare Benefit for Screening Mammography Proposed 13 Objectives, Scope, and Methodology 14 Chapter 2 17 V+iation in Screening Hospitals and Radiology Offices Predominate Among 18 Mammography Settings Mtimmography Lack of Distinction Between Screening and Diagnostic 19 Ch.arges May Be Mammography Contributes to Fee Variations Mammography Volume Linked to Charges 21 Litiked to Volume Ctiapter 3 23 High-Volume Compliance With Professional Standards Important to 23 Quality Providers Most Likely Problems in Meeting Image Quality, Radiation Dose 25 to Adhere to Quality Standards Revealed Many Providers Do Not Meet Standards for Quality 26 Standards for Assurance Programs Screening High-Volume Providers More Often Adhere to Quality 32 Standards Compliance With Quality Standards Not Related to 32 Charge Chapter 4 35 Federal and State State Oversight of Mammography FDA Regulates Manufacture, Assembly of Mammography 35 39 Regulation Equipment of Screening IIIIS Proposed Standards for Screening Mammography 41 Mammography Is Limited y Page 6 GAO/HRD-90-32 Quality of Screening Mammography Content43 Ch’apter 5 42 C+clusions Aqpendixes Appendix I: GAO’s Survey Methodology 44 Appendix II: GAO Questionnaire on Mammography 46 Appendix III: Selected Quality Standards for 65 Mammography and Related Survey Results Appendix IV: State Oversight of Screening Mammography 69 Appendix V: Additional Results From GAO’s Screening 71 Mammography Survey Appendix VI: Major Contributors to This Report 77 Tables Table 2.1: Distribution of Survey Respondents and 18 I Mammography Volume, by Setting I Table 2.2: Survey Respondents’ Charges for Screening 19 Mammography Table 2.3: Survey Respondents’ Weekly Mammography 22 Volume Table 3.1: Survey Respondents’ Compliance With Key 27 Quality Standards for Screening Mammography Table 3.2: Relationship Between Volume and Providers’ 34 Compliance With Quality Standards Table 4.1: Michigan Mammography Equipment With 38 Image Quality and/or Radiation Problems, by Facility Type (,Jan.-Nov. 1988) Table 4.2: Florida Mammography Equipment Not in 38 Compliance (Fiscal Year 1988) Table 4.3: FDA Inspections of Newly Installed 40 Mammography Equipment in Four States (1986- 1988) Table 4.4: Noncompliances Identified by FDA in Four 40 States (1982-1989) Table I. 1: Response to GAO Questionnaire, by State 45 Table V. 1: Settings of Respondents Performing Screening 71 Mammography, by State Table V-2: Respondents’ Charges for Screening 72 Mammogram, by Locale and Setting Table V-3: Volume of Mammograms Performed, by Setting 74 Table V-4: Survey Respondents Accredited by ACR 74 Mammography Accreditation Program, by Setting and State (As of April 1989) Page 7 GAO/HRD-90-32 Quality of Screening Mammography Contents Table V,5: Survey Respondents Reporting Annual 75 Inspection by Radiological Physicist, by Setting and State Table V.6: Survey Respondents Reporting Annual 75 Physicist Inspection of Beam Quality (HVL), Average Glandular Dose, and Phantom Image Quality, by Setting and State Table V.7: Survey Respondents Reporting Compliance 76 With Selected Quality Assurance Standards, by Setting Table V.8: Survey Respondents’ Retention of Original 76 Mammographic Images, by Setting Figures Figure 3.1: Annual Inspection Performed by Radiological 28 Physicist, by Setting Figure 3.2: Annual Inspection of Selected Features by 29 Radiological Physicist, by Setting Figure 3.3: Daily Processor Sensitometry, by Setting 30 Figure 3.4: Semiannual Phantom Image Checks, by Setting 31 Figure 3.5: Selected Quality Assurance Activities, by 33 Setting Page 8 GAO/HRD-90-32 Quality of Screening Mammography ---- Contents Abbreviations ACIZ American College of Radiology ACS American Cancer Society AWN1 American Registry of Clinical Radiographic Technologists AHIlT American Registry of Radiologic Technologists HCDDI Breast Cancer Detection Demonstration Project I3SE breast self-examination CC craniocaudal CDRII Center for Devices and Radiological Health CIICPD Conference of Radiation Control Program Directors DHS Department of Health Services DHH Division of Radiological Health FDA Food and Drug Administration HCFA Health Care Financing Administration HHS Department of Health and Human Services IIII' Health Insurance Plan HMO health maintenance organization HVI, half value layer ICS Investigation and Compliance Section MAP Mammography Accreditation Program NC1 National Cancer Institute NCRP National Council on Radiation Protection and Measurements CWA Office of Technology Assessment PPRC Physician Payment Review Commission XMRC X-ray Machine Registration and Control Page 9 GAO/HRD-90-32 Quality of Screening Mammography Chabter / 1 Iritroduction Breast cancer is a leading cause of death and illness in the United States; over 40,000 breast cancer deaths are expected to occur in 1989. No pre- ventive strategies are known, but early detection through periodic screening mammography can lower a woman’s risk of dying from breast cancer by 30 percent. The Medicare Catastrophic Coverage Act of 1988 (P.L. 100-360) created a new benefit for periodic screening mam- mography for Medicare-eligible women that was to go into effect in Jan- uary 1990. However, on November 22, 1989, the Congress repealed most provisions of the act, including the mammography benefit. Legislation that would restore a Medicare screening mammography benefit has been introduced in the House of Representatives. If such a benefit is enacted, millions of women would become eligible for low-cost screening, which could have a significant impact on breast cancer morbidity and mortal- ity in the United States. To help contain costs, the act generally limited the fee providers could charge for a screening mammogram to $50. Some members of Congress expressed concern that the charge limit could compromise the quality of Medicare-funded screening mammography. To help assure that quality services would be provided, the Congress required the Secretary of Health and Human Services (HHS) to establish quality standards for facilities providing screening mammography to Medicare beneficiaries. To provide the Congress with information on the quality of current screening mammography services, the act also required GAOto assess quality of care in a variety of settings. In response to this requirement and a request from Representative Barbara Kennelly, we developed information on which settings currently provide screening mam- mography, what they charge, whether quality of care varies by setting or charge, and government regulation of screening mammography. Breast cancer is the most common cancer and second leading cause of Screening cancer deaths in women in the United States, and its incidence is Mammography Can increasing. One in 10 women will develop breast cancer during her life- Reduce Breast Cancer time; only 4 years ago, the rate was 1 in 13. The American Cancer Soci- ety (ACS)estimates that during 1989, nearly 143,000 women in the Mortality United States will develop breast cancer and 43,000 will die from it. Increasing age is the most important risk factor for developing breast cancer. There is a dramatic increase in risk after age 40, and over one- third of the cases diagnosed occur in women over 65. ” At present, the best method known to reduce breast cancer mortality is early detection, which permits treatment that greatly increases the Page 10 GAO/HRD-90-32 Quality of Screening Mammography - --l----- Chapter 1 Introduction - chance for survival. Detection of breast cancer is accomplished through mammography, clinical breast examination, and monthly breast self- examination. Of these methods, mammography, an X-ray of the breast, is the most effective for detection of early stage breast cancer. The value of mammography for breast cancer screening is that it can detect cancers that are too small to be felt through physical examination (palpation), and these early stage cancers can be 90 to almost 100 per- cent curable. When detected at a later stage, they are much more likely to have spread to the axillary lymph nodes or distant sites, and the 5-year survival rate can drop as low as 18 percent.’ Not only is treat- ment then more debilitating, it also is much less effective. The National Cancer Institute (NCI) and ACS believe that a breast physi- cal examination is also an essential element of breast cancer screening, as a small percentage of cancers is identified by palpation but cannot be seen on a mammogram. However, this examination need not necessarily occur at the same time as the mammogram. Through technological refinement of mammography equipment, the effectiveness of mammography has increased, while the amount of radi- ation exposure has dropped dramatically. Current levels are one-tenth of those produced during the 1960s. They are considered safe enough so that the value of the mammographic examination for women eligible for screening far outweighs any risk from the exposure. Studies indicate a significant decrease in breast cancer mortality for women who have screening mammograms at recommended intervals. Among the studies are those of the Health Insurance Plan (IIIP) of Greater New York Screening Project and the Breast Cancer Detection Demonstration Project (RCDDP),a collaborative effort of NCI and ACS. Leading medical organizations, including NCI, ACS,and the American Col- lege of Radiology (ACR),~ have endorsed the following breast cancer screening guidelines: . An annual or biennial mammogram for women 40-49 and l An annual mammogram for women age 50 and older. ‘Survival rate in white women. “AC12is a professional and educational association of 20,000 board-certified rddiok)gists and radiolog- ical physicists. Page 11 GAO/HRD-90-32 Quality of Screening Mammography Chapter 1 Introduction Mammography is performed for two different purposes, screening and Ptirposes of Screening diagnosis, anfl Diagnostic M&nmo&aphy Differ l Screening mammography is an examination of a woman without breast symptoms to detect breast cancer before a lesion can be felt by her or her physician. It is done simply to detect unknown abnormalities in women who appear to be disease-free. 0 Diagnostic mammography is an examination of a woman who exhibits a ,I symptom, such as a lump, that indicates the possible presence of breast cancer. It is performed to fully characterize lesions, providing as much I information as possible. The process of performing the mammograms is the same for both. A diagnostic procedure, however, may require additional breast views and other tests, such as ultrasound, to provide more information about a suspicious lesion. Because of its more limited purpose, screening mam- mography can take advantage of certain economies not possible during diagnostic mammography. For example, a radiologist need not be on the premises for immediate interpretation of screening mammograms. Instead, the day’s films can be read all at one time, allowing greater effi- ciency in the costly use of a radiologist’s time. Despite the recommendations for regular screening mammography, Efforts Made to studies indicate that around 60 percent of women age 40 and over have Increase Use of never had a screening mammogram. The most common reasons women Screening give for not being screened are that they do not think they need screen- ing mammograms and that their physicians did not recommend them. Mammography Other reasons include fear of excess radiation and high fees. Fear of detecting breast cancer also may act as a deterrent, because until recently, such a diagnosis meant the likelihood of death or at least loss of a breast. In an effort to increase the number of women who are screened, ACShas been supporting numerous local campaigns since 1987 to establish pro- grams that promote and provide low-cost screening mammography. Common features of these programs include low charges for screening ranging from $36 to $50 and criteria for participating facilities to assure provision of high-quality services. Some states also sponsor programs to Y increase the use and awareness of mammography, and some mandate health insurance coverage of screening mammography. Page 12 GAO/HRD-90-32 Quality of Screening Mammography Chapter 1 Introduction Currently, Medicare covers diagnostic mammography only, as it gener- Medicare Benefit for ally pays only for treatment, not preventive services. The only excep- Sc ‘eening tions have been immunizations for pneumococcal pneumonia and M f mmography hepatitis B. During 1987, Medicare spent approximately $75 million on diagnostic mammograms. Prbposed The Medicare Catastrophic Coverage Act of 1988 (P.L. 100-360) intro- duced a new Medicare benefit for screening mammography, effective in 1990. This would have represented the first mass screening procedure supported by Medicare. The population eligible for this benefit includes nearly 18 million women age 65 and over and a portion of the more than 1 million disabled women under age 65. The Health Care Financing Administration (HCFA) estimated that the cost of providing this benefit would be $150 million in fiscal year 1990 and $275 million in fiscal year 1991. Because of the potential cost of the benefit, the law established an indexed limit on the amount providers could have charged. Participating physicians (those who agree to accept Medicare-approved charges as payment in full for services) could have charged a maximum of $50 for screening mammography performed in 1990, including both examina- tion and interpretation, of which Medicare would have reimbursed 80 percent. In years subsequent to 1990, the charge limit would have been the preceding year’s limit increased by the percentage increase in the Medicare Economic Index.” Nonparticipating physicians could have charged 125 percent of the $50 limit ($62.50) in 1990, and their maxi- mum charge would have decreased to 115 percent of the indexed limit by 1992. The Congress specified that the Secretary of HHSwas to establish stan- dards to assure the safety and accuracy of Medicare-funded screening mammography. The law also set limits on the individual’s use of the screening mammography benefit. These frequency limitations generally coincided with NCI’S guidelines for screening mammography, with the following exception. For women over age 64, Medicare would have pro- vided reimbursement for a screening mammogram every 2 years instead of every year. “After 1991, the Secretary of HHS could have reduced the limit to the amount required to assure the availability of convenient screening mammography of good quality. Page 13 GAO/HRD-90-32 Quality of Screening Mammography Chapter 1 Introduction The Medicare Catastrophic Coverage Act of 1988 required the Comp- Okjjectives, Scope, and troller General to report to the House Committees on Ways and Means Methodology and Energy and Commerce and the Senate Committee on Finance on the quality of care of screening mammography provided in a variety of set- tings. Congressional conferees indicated that these settings include hos- pital outpatient departments, clinics, radiology practices, physicians’ offices, and other facilities where Medicare beneficiaries could obtain screening mammography.’ In addition, Representative Barbara Ken- nelly, in a letter of June 28, 1988, asked GAOto provide additional infor- mation on mammography. In response to the statutory requirement and the request from Repre- sentative Kennelly, and as agreed with congressional staff, our objec- tives were to collect and analyze data across settings on the provision of screening mammography, charges for services, and quality assurance mechanisms. We developed information to answer the following questions: l What provider settings currently offer screening mammography pro- grams, and where do Medicare patients currently obtain mammography? . What do different settings that provide screening mammography charge for their services? . What are the factors necessary for screening mammography of accept- able quality, and do quality factors vary by setting or by charge? l What governmental and professional regulations and oversight pro- grams are in place to assure the quality of screening mammography? We conducted our work in four states: California, Florida, Idaho, and Michigan. The criteria used to select these states were availability of information on facilities with mammography equipment, geographic diversity, size of the Medicare population, population density, and inclu- sion of a variety of mammography settings. Our principal source of information on the first three questions was a mail survey WCconducted of 1,485 facilities (757 in California, 313 in Florida, 35 in Idaho, and 380 in Michigan) providing mammography in the four states. These were all the facilities identified by the state radio- logical health departments as having mammography equipment. Our ‘l’hc~ law also required the Physician Payment Review Commission to report on the cost of providing screening mammography in a variety of settings and at different volume levels. Page 14 GAO/HRD-90-32 Quality of Screening Mammography : , Chapter 1 Introduction questionnaire requested information on providers’ screening mam- mography practices, focusing on equipment, personnel, services pro- vided, quality assurance mechanisms, reporting and record-keeping, volume, and charges. (Our survey methodology is described in more detail in app. I, and a copy of the questionnaire is presented in app. II.) The response rate was 82 percent overall, and, for each state, at least 80 percent. Questionnaire items were based on the standards and recommendations developed by the ACR for use in granting accreditation to screening mam- mography programs (see p. 24 for more detailed information on ACR accreditation), the requirements of Public Law 100-360, and factors identified by other experts as associated with quality in screening mam- mography. The questionnaire was reviewed by officials from ACRand NCI. When analyzing the questionnaire responses, we used the following set- ting categories: . Primary care physician: Office of a primary care physician such as a gynecologist, internist, or surgeon; . Radiology private practice: Individual or group radiology practice; . Hospital: Hospital radiology department, outpatient clinic, or ambula- tory care center; . Hospital breast clinic: Breast screening clinic located in a hospital; . Breast clinic: Freestanding facility for screening and/or treatment of breast disease; HMO:Health maintenance organization; Multispecialty clinic: Multispecialty group practice or outpatient clinic; Mobile van: Mobile van fitted with mammography equipment that may or may not be affiliated with a hospital, clinic, or physician’s practice; and . Other: Includes freestanding imaging centers not owned by radiologists, military primary care clinics, and other outpatient facilities. Our survey results are based completely on self-reporting in the ques- tionnaire, pertain only to the four states, and cannot be projected nationwide. The data from the questionnaire are reported in the aggre- gate, but any marked differences between the states are pointed out. To gather information about topics not addressed in the questionnaire, we conducted site visits in each state at some facilities participating in the survey. We visited 15 facilities-5 in California, 4 each in Florida Page 16 GAO/HRD-90-32 Quality of Screening Mammography -&-..---- I Chapter 1 Introduction and Michigan, and 2 in Idaho. The types of setting visited included pri- mary care physician, radiology practice, hospital, breast clinic, multi- specialty clinic, mobile van, and “other.” (The latter was a practice in which portable mammography equipment was transported among sev- eral rural hospitals and physicians’ offices.) In the four states studied, we identified state laws and regulations per- taining to mammography, interviewed officials responsible for oversight of mammography facilities, and analyzed data on state inspections. To analyze the federal role in oversight of screening mammography and promoting quality assurance, we interviewed Food and Drug Adminis- tration (FDA) and HCFAofficials and obtained data on FDAinspections. For information on quality standards and on the role of screening mam- mography in early detection of breast cancer, we reviewed the literature and interviewed many health care practitioners and experts from a wide range of organizations. Our work was performed from September 1988 to July 1989 in accord- ance with generally accepted government auditing standards. Page 16 GAO/HRD-90-32 Quality of Screening Mammography Variation in ScreeningMammography Charges May l3eLinked to Volume The majority of screening and diagnostic mammograms currently are done in hospitals and radiology offices, where most Medicare-funded diagnostic mammography also occurs. The fees charged for screening mammography vary widely, from $50 or less to over $150, our survey showed. More than two-thirds of our respondents reported charges in the $51-125 range, with an average charge of $104. One reason for the wide range of charges for screening mammography is providers’ lack of distinction between screening and diagnostic services. Most of our respondents began providing screening within the past 5 years, and many did not distinguish between screening and diagnostic purposes. Screening programs can operate at a higher patient volume and take advantage of certain economies of scale that allow them to pro- vide services at a lower fee. Mobile vans typically concentrate on screen- ing mammography. Of the mobile vans in our survey, 60 percent charged $50 or less, and none charged more than $125. The Physician Payment Review Commission found that, at facilities that did not differ- entiate between screening and diagnostic mammography, the average charge was $103. However, at providers that did make a distinction, the average charges for screening and diagnostic mammography were $53 and $113 respectively. Most survey respondents perform fewer than 50 mammograms per week, a relatively low volume. Practitioners and studies have indicated a relationship between high volume and the ability to lower the price of screening, and our survey data showed a similar trend. w Page 17 GAO/HRD-9032 Quality of Screening Mammography Chapter 2 Variation in Screening Mammography Charges May Be Linked to Volume The majority of our respondents (65 percent) were hospitals and radiol- HoGpitals and ogy private practices, and the majority of screening and diagnostic Radiology Offices mammograms (57 percent) were performed in those settings, as table 2.1 shows.i (For information on setting distribution by state, see table V. 1.) Prddominate Among M WiwPhY Set y ings 1 I Tablei2.1: Distribution of Survey Respffndents and Mammography Distribution (percent) Volun~e, by Setting Setting Respondentsa Volumeb I Hospital -...-- 40 33 Radiology private___-practice 25 25 Multispecialty clinic 10 8 .--.____--- -. ~--.~ Primary care physician 9 4 -.. -______ I Breast clinic 8 13 HMO 4 IO / Hospital breast clinic -.___-- 2 4 Mobile van 1 2 Other 1 1 Totals 100 100 aA total of 1,026 respondents reported doing screening mammography and reported where it was done. “Based on reported weekly volume of screening and diagnostic mammography. The distribution of settings our respondents reported is similar to the distribution of settings where Medicare beneficiaries currently receive diagnostic mammography services. Most mammograms reimbursed by Medicare in 1987 took place in an outpatient hospital setting or a radiol- ogist’s office, according to data provided by HCFA. About 83 percent of respondents that did screening mammography reported an urban or suburban location and 17 percent a rural location. Of the four states we focused on, only Idaho reported that a majority of its facilities were in rural areas (66 percent). ‘Respondents were asked to name the principal setting from the list we provided (see ch. 1 and app. II). Page 18 GAO/HRD-90-32 Quality of Screening Mammography Chapter 2 Variation in Screening Mammography Charges May Be Linked to Volume Our survey respondents reported charging a wide range of fees for Labk of Distinction screening mammography. Since many providers do not differentiate Bebween Screening between screening and diagnostic mammography, they may not take an@Diagnostic advantage of the economies of scale possible with screening mam- mography. This may explain why most respondents currently charge M+nwpwhy more than the $60 Medicare generally would have allowed under the Co//tributes to Fee screening benefit. Vdriations ScqeeningCharges Vary Questionnaire recipients’ reported charges for a screening mammogram, Widely including both performance and interpretation of the mammogram, ranged from $50 or less2to $275, with an average of $104. The median charge was $100. About 7 percent of the charges were in the lowest ($1-50) of five charge ranges that we established (see table 2.2). This range coincides with the reimbursement limit set for screening mammograms that would have been funded by Medicare. Overall, about 48 percent of respondents reported charging less than $100, and 52 percent charged $100 or more. Table 2.2: Survey Respondents’ Charges for Screening Mammography No. of Percent of Charge range respondents respondent@ $1 - 50 ___.I__ ---__ 71 __- 7 ---._~ ~~ __. -__ .--- $51. -~.- 99_._. - ~~_.-.-_-- ___.-__ --- 395 41 --. $100 - 125 ______ .___264 28 $126- 150 -_.-..-.- __-- -. 145 _---..~----~ 15 ~ Over$150 81 9 Totals 956 100 Qespondents reporting a charge of $0 are excluded In comparing charges by state, we found that almost all Idaho facilities (97 percent) charged less than $100. This was a substantially higher percentage than in California (44 percent), Florida (54 percent), and Michigan (47 percent). ‘About 2 percent of respondents, most of them HMOs, reported charging either nothing or only a nominal amount for screening mammography, presumably because it is a covered service for which there is no charge or a small copayment. We exclude no-charge cases from the remainder of this analysis. Page 19 GAO/HRD90-32 Quality of Screening Mammography Chapter 2 Variation in Screening Mammography Charges May Be Linked to Volume Screening mammography charges were lower in mobile vans than in other settings. Sixty percent of the mobile vans reported charging $50 or less and none over $125. Typically, mobile vans concentrate on screen- ing services, which allow them to take advantage of such procedural economies as batch reading of films (see p. 21). In contrast, no primary care physicians reported charging $50 or less, and a large percentage of hospitals and hospital breast clinics (31 and 37 percent respectively) charged more than $125. For additional information on charges by set- ting and by state, see table V.2. In regard to location, rural facilities tended to charge somewhat less than facilities in urban/suburban locations, which may contribute to the lower charges in Idaho. Forty-seven percent of the urban/suburban facilities and 56 percent of the rural facilities charged less than $100. Conversely, 25 percent of the urban/suburban and 15 percent of the rural facilities charged over $125. Limited Use of Economies The medical community is at an early stage in distinguishing between of Scale in Screening screening mammography, a relatively new service, and diagnostic mam- mography. Almost two-thirds of the respondents to our survey reported Mammography Affects that they began providing screening services after 1984 and almost half Charges after 1985. Over half of the 15 facilities we visited did not distinguish between the two kinds of mammography or differ in the way they provided these services or in their fees. This may account for the wide range of charges reported. In a 1989 report to Congress” on the cost of providing screen- ing mammography, the Physician Payment Review Commission (PPRC) noted a similar relationship. In surveying 125 randomly selected mam- mography providers, PPRC found that at facilities that did not differenti- ate between screening and diagnostic mammography the average charge was $103. However, at providers that did make a distinction, the aver- age charge for screening was $53 and for diagnostic mammography, $113. The higher fees for screening mammography where providers do not distinguish between the two types of service may result from limited use of procedural efficiencies permitted by screening mammography. These include: “PI’RC, The Costs of Providing Screening Mammography, June 30, 1989. Page 20 GAO/HRD-90-32 Quality of Screening Mammography Chapter 2 Variation in Screening Mammography Charges May Be Linked to Volume l Batch reading of films. A major factor driving up the price of diagnostic mammography is the need for a radiologist to read the films while the patient is present. This does not apply to a screening situation. l Maintaining a large volume of patients. This generates additional reve- nue without requiring additional investment in equipment. l Requiring payment at time of service. This eliminates costs associated with billing and bad debt. Mammography experts have indicated that high volume is essential for Mhography providing screening mammography at lower fees. Many of our survey V@rne Linked to respondents reported a relatively low weekly mammography volume, Charges and our analysis showed some association between higher volume and lower charges. Generally Low Volume of The majority of facilities responding to our survey performed a rela- Mammography Reported tively low volume of mammography.4 The weekly volume ranged from 2 to 500 mammograms, but the average was 52 and the median 35, indi- cating a concentration at the lower end of the range. About one-third of respondents providing data on volume performed fewer than 25 mam- mograms per week, while only 21 percent did 75 or more weekly (see table 2.3). Analysis by setting reveals great variation in mammography volume. At the lower end, 70 percent of the primary care physicians reported weekly volume under 25, and none performed more than 100 mam- mograms per week. Seventy-one percent of the hospitals reported a weekly volume under 50, with 7 percent doing over 100. In contrast, over half (58 percent) of the hospital breast clinics do at least 75 mam- mograms per week, with 37 percent doing over 100. (For additional information, see table V.3.) 4The volume data reported represent total mammography volume, including both screening and diag- nostic mammography. Although we asked for separate data, some respondents provided only com- Y bined volume data and others used a definition of screening different from the one we provided. Our definitions of screening mammography and diagnostic mammography (that conducted on asymptom- atic patients versus symptomatic patients) coincided with those of mammography experts, but some respondents evinced difficulty with these concepts. For example, one respondent stated that although many of the facility’s patients were asymptomatic, it did not perform screening mammography. Page 21 GAO/HID-90-32 Quality of Screening Mammography Chapter 2 Variation in Screening Mammography Charges May Be Linked to Volume Table .3: Survey Respondents’ Weekly Mam nr, ography Volume No. of mammograms Percent of No. of performed weekly respondents cases I l-24 34 336 / 25-49 28 269 50-74 17 167 75- --- 100 10 100 Over100 11 104 Totals 100 976 Higher Volume Associated Lower charges tended to be linked with a higher volume of mam- With Lower Charges mography in our survey. The proportion of facilities charging $50 or less was almost three times greater among those performing over 100 mammograms weekly (15 percent) than among those doing fewer than 25 a week (6 percent). Another indication of this relationship is the fact that of the facilities performing more than 100 mammograms per week, over twice as many charged $50 or less (15 percent) as those that charged over $150 (7 percent). The relationship is reversed for facilities performing fewer than 25 mammograms per week. In that group, 6 per- cent charged $50 or less and 10 percent, over $150. Experts on screening mammography emphasize that a key factor that makes screening for a lower fee possible is increased volume of service. To offer inexpensive mammography, one expert practitioner has stated, a provider must take advantage of certain economies of scale that require at least 15 patients daily. This minimizes periods when expen- sive equipment is idle. A break-even price for 16 patients a day of about $64, for 20 patients about $54, and for 25 patients about $46 was reported by a Florida physician who gave us information on the costs of a mobile van providing screening mammography. At a stationary setting with average costs, the cost for providing one screening mammograms ranges from about $34 at a volume level of 50 exams per day to $107 at a volume of 5 exams daily, PPRC said in its report. The unit cost at between 15 and 20 exams per day is around $50. Thus, WHCconcluded, $50 is sufficient payment for screening mam- mography if volume levels are high enough.(; “lncludcs nonphysician cost per exam and physician fee. “I’PHC also recommended increasing the Medicare payment to $60 in rural areas of low population density where it is not possible to generate sufficient volume to support a $50 payment. Page 22 GAO/HRD-90-32 Quality of Screening Mammography Chgpter 3 @gh-VolumeProviders Most Likely to Adhere t$ Quality Standaxis for Screening Professional groups have developed quality standards for providers of screening mammography to follow. They include using dedicated equip- ment (i.e., equipment designed specifically for mammography instead of general purpose X-ray equipment), employing staff with proper creden- tials to perform and interpret mammograms, reporting and retaining records of mammographic results, and having an adequate quality assurance program. The primary goal of these standards is to produce a mammogram with good image quality, while limiting the patient’s radia- tion dose, and to provide an accurate interpretation of the image. How- ever, these standards are not uniformly followed, and there is great variation in image quality and radiation dose in current mammography practice. Compliance with many of the professional quality standards, such as using equipment specifically designed for mammography, was wide- spread among the providers we surveyed. However, many did not com- ply with standards for quality assurance activities, such as annual inspection by a radiological physicist. Primary care physicians and multispecialty clinics reported the lowest rates of compliance with stan- dards for quality assurance programs; in general, hospital breast clinics, HMOS,and mobile vans reported the highest. Providers reporting higher levels of mammography volume showed a greater degree of compliance with quality standards than those per- forming fewer mammograms. Higher charges, however, did not necessa- rily buy higher quality. We found no consistent relationship between charge and adherence to quality standards. Mammography is a complex process that requires providers’ adherence Compliance With to numerous quality standards to produce good results. Professional Professional groups have established widely recognized quality standards to guide Standards Important providers. Because of the complexity of the process, failure to comply with any of the standards can compromise the quality of the results. tb Quality Some of the standards, such as using dedicated equipment and employ- ing qualified personnel, relate directly to provision of mammography services. Others, such as daily inspection of the film processor, relate to the facility’s quality assurance program. Adhering to quality assurance standards allows facilities to test their systems to ensure that they are Y providing high-quality mammograms and interpretations. Most of the features considered necessary for quality screening mam- mography contribute to the goal of obtaining good image quality with Page 23 GAO/HRD-9052 Quality of Screening Mammography Chapter 3 High-Volume Providers Most Likely to Adhere to Quality Standards for Screening minimal risk to the patient. Because a mammogram is among the radio- graphic images most difficult to read, it must have optimal clarity. If image quality is poor or the interpretation faulty, the interpreter may miss cancerous lesions. This could delay treatment and result in an avoidable death or mastectomy. Problems with images or interpretation also can lead to unnecessary testing and biopsies if normal tissue is mis- read as abnormal. The American College of Radiology has made a comprehensive effort to establish quality standards for screening mammography through its vol- untary Mammography Accreditation Program (MAP). Started in 1987, it offers peer review and evaluation of a facility’s equipment, staff qualifi- cations, examination procedures, reporting practices, recordkeeping, and quality control and assurance programs. ACR also obtains informa- tion on image quality and radiation dose, which are evaluated through the USCof a breast phantom’ and dosimeter. A set of the facility’s own mammographic images is also evaluated for image quality.’ A relatively small number of facilities have applied for and received accreditation. As of April 1989,8 percent of facilities that responded to our questionnaire were accredited, according to data ACR supplied.:’ The setting with the highest proportion of accredited facilities was breast clinics (13 percent), followed by radiology practices (11 percent). Table V.4 provides additional information on respondents’ ACRaccreditation by setting and state. Other groups also have issued mammography guidelines. They include the National Council on Radiation Protection and Measurements (NCRP),~ some of whose standards ACRincorporated ’ Phantoms arc objects designed to simulate breast tissue when exposed. The ACR phantom is a block with a wax insert cont,aining fibers, specks, and masses that simulate growths that could be cancer- ous. Tht: facility exposes the phantom with its equipment, and the visibility and clarity of the objects imbcddcd in the phantom are evaluated. ‘When facilities submit clinical images to ACR as part of its MAP, the features that are assessed include proper patient positioning, acceptable image contrast, compression, and adequate visualiza- tion of structures within the breast. “No Idaho facility was accredited, but 3 percent of the California, 6 percent of the Florida, and 19 percent of the Michigan respondents were. An ACR official said that participation in the accreditation program has been high in states where ACS chapters have required accreditation for participation in KS screening programs. Michigan ACS and state officials have encouraged participation. ‘NCRI’, Mammography-A User’s Guide, Report No. 85, Mar 1, 1986. The NCRP is a nonprofit corpo- ration chartered by the Congress to make recommendations on radiation protection and related matters. Page 24 GAO/HRD-90-32 Quality of Screening Mammography . Chapter 3 High-Volume Providers Most Likely to Adhere to Quality Standard6 for Screening into its accreditation program, and the Conference of Radiation Control Program Directors (CRCPD).~ I Despite the existence of professional standards intended to result in Pqoblems in Meeting optimal levels of image quality and radiation dose, ACRand state regula- In;lage Quality, tory agencies have identified significant problems in current mam- Rbdiation Dose mography practice. An analysis” of data collected from the ACR accreditation process revealed wide ranges of image quality and dose. Sdandards Revealed The author of the analysis concluded that the underlying reason for these variations is the lack of universal compliance with quality assur- ante standards by mammography facilities. The analysis was based on data collected from 647 providers that had completed the ACKaccreditation process as of February 1, 1989. Twenty- nine percent of the applicants did not meet ACR’S criteria and were not granted accreditation. Of these, about 36 percent failed because of poor clinical images, 38 percent because of poor phantom image quality, and 15 percent for both reasons. An additional 3 percent failed because of excessive radiation dose, and 8 percent failed to meet both dose and clinical image criteria. These findings were echoed in Michigan’s inspec- tion program, which also found significant problems when it evaluated equipment using ACRstandards. (See pp. 37-38 for more detailed information.) For the states we reviewed, the following applicants who had completed the ACR process as of May 1989 had received accreditation, according to ACH: l Michigan and Florida, about two-thirds of the 109 and 32 applicants, respectively; l California, 82 percent of the 34 applicants; and l Idaho, neither of 2 applicants. “CHCPD, Mammography Screening Guide, Conference Publication 87-4, Feb. 1987 (prepared in coop- eration with the Ccntcr for Devices and Radiological Health (CDRH) of the Food and Drug Adminis- tration; it also uses NCRP standards). The CRCPD is an association of state radiation control officials that has worked to improve the quality of mammography. “H. Edward Hendrick, Ph.D., “Quality Control in Mammography: The American College of Radiology’s Mammography Screening Accreditation Program,” Current Opinion in Radiology, Vol. I, 1989, p. 203. Page 25 GAO/HRD-90-32 Quality of Screening Mammography Chapter 3 High-Volume Providers Most Likely to Adhere to Quality Standards for Screening We found widespread compliance with many quality standards, particu- MaGy Providers Do larly those directly related to providing mammographic services, such as Not! Meet Standards employing certified or licensed technologists to perform the mam- for Quality Assurance mograms. But many facilities did not comply with standards for quality assurance programs, our survey showed. A comprehensive quality Pro@ams assurance program is essential to evaluate both equipment and staff performance, as problems in image quality and radiation dose have been attributed to the lack of such programs. Primary care physicians and multispecialty clinics reported the lowest levels of compliance with quality assurance standards; hospitals, hospital breast clinics, HMOS, and mobile vans generally reported the highest levels. Almost all of our respondents reported adhering to a number of key quality standards (see table 3.1). Additional information about the importance of these practices and the results of our survey appears in appendix II. However, smaller percentages of providers reported com- plying with professional standards for quality assurance activities, as we discuss in more detail below. -’ An&al Inspection by ACRpolicy states that a mammography system should be inspected by a Radiological Physicist radiological physicist at least once a year. About two-thirds (69 per- cent)’ of the facilities responding to our survey reported that either a Varies by Setting staff or consultant radiological physicist conducts this annual inspec- tion. Compliance with this standard varies greatly by setting, ranging from 43 percent of primary care physicians to 91 percent of mobile vans (see fig. 3.1). For a more detailed analysis by setting and state, see table v.5. 71’ercentagesby state were: California, 73; Florida, 62; Idaho, 62; and Michigan, 68. Page 26 GAO/HRD-90-32 Quality of Screening Mammography Chanter 3 l&&-Volume Providers Most Likely to Adhere to Quality Standards for Screening ib(e 3.1: Survey Respondents’ pliance With Key Quality Standards Percent of respondents creening Mammography Quality standard reporting compliance Service delivery standards: ___- ._--- -_- Takina medical history 100 Using ..-...__.-dedicated mammography ---. __--equipment 97 Taking 2 __.-_..----. or more breast views 99 Certified or licensed _..-- .-..- -~ .._.__ mammography operator ---~.- .~- 97 Interpretation of mammograms by radiologist ____---.--. 99 .-. _- -~~ Reporting results of abnormal mammograms to patient and/ or physician: When patient has designated physician 99 When patient has not designated physician - 99 Retention of original mammographic images: 5 years or more 98 Over 10 years 49 Instructing patients -_ -- on breast self-examination .._______. ____-.. 90 Quality assurance standards: ..~___ ___~___-. _.I-_--..---...~ -~~.- Annual inspection .-~by-__.__-. radiological physicist 69 Annual physicist inspection of beam quality, average glandular dose, and phantom image ._________ quality ___.- 55 Daily processor sensitometry 35 Semiannual phantom image check by facility 46 Monitoring repeat mammograms and doing one other quality assurance activitya 44 Performing second reading of mammograms within facility -____. 29 _-.----...--.-.-----__- Following up on patient biopsies 76 aOther activities include second reading of mammogram within facility, peer review of readings, and follow-up on patient biopsies. -Fewer Pmvi (lers _.._- --_.- Have Three items that a radiological physicist should inspect annually are: (1) Annual Inspection of beam quality,H (2) average glandular dose,” and (3) phantom image qual- ity. (Beam quality is related to both image quality and dose.1o) About 55 Selected Features percent of our respondents reported annual inspection of these three ‘Ream quality is measured in terms of half value layer (HVL), which is the amount of filtration necessary to reduce the intensity of the beam to half of the original value. “The measurement of radiation absorbed by the breast that best characterizes radiation risk from mammography, according to the NCRP. “‘If the energy of the beam is too low, the radiation dose will be excessive; if too high, the contrast will be too low, resulting in poor image quality. Page 27 GAO/HRD-90-32 Quality of Screening Mammography Chapter a IUgh-Volume Providers Most Likely to Adhere to Quality Standarda for Screening Figure 3.1: Annual Inspection Performed byRa?iological Physicist, by Setting / loo Pomni 00 00 70 80 50 40 a0 20 10 0 Note: Total 1,026 respondents. factors by a radiological physicist. As with overall inspection by a radio- logical physicist, primary care physicians and multispecialty clinics reported the lowest rates of compliance. Hospital breast clinics showed the highest rate of compliance (see fig. 3.2). There was considerable variation across states.ll Table V.6 presents more detailed information. Daily Processor Some quality control procedures should be performed regularly by the screening mammography staff. There is a growing consensus that facili- Ser@itometry Advised ties doing screen-film mammography12 need to do daily sensitometry of the film processor, a procedure that checks whether the processor that develops the film is operating properly. Of respondents using the screen- * I ‘Percentages by state were: California, 69; Florida, 46; Idaho, 21; and Michigan, 59. “The two principal techniques for performing mammography are the screen-film and xeroradi- ography methods. Of the facilities responding to our survey, 84 percent used only the screen-film method, 11 percent used xeroradiography, and 6 percent used both. Page 28 GAO/HRD-90-32 Quality of Screening Mammography Chapter 3 High-Volume Providers Most Likely to Adhere to Quality Standards for Screening 1 Fig re 3.2: Annual lnrpection of Sel cted Features by Radiological 100 Phyt iclst, by Setting 90 80 70 80 60 40 30 20 10 Note: These selected features include beam quality (HVL), average glandular dose, and phantom image quality. Total 1,026 respondents. film method, 35 percent’:’ reported doing daily processor sensitometry; an additional 17 percent did it at least weekly. By setting, there was wide variation in compliance with the daily sensitometry standard, ranging from 10 percent of primary care physicians to 59 percent of IIMOs, as shown in figure 3.3. Semiannual Phantom Although evaluating a phantom image is one component of the physi- cist’s inspection, this procedure also may be done by trained personnel Image Checks Done by at the facility. As discussed on page 24, exposing a phantom shows the Half of Respondents quality of image the system is producing and can indicate the existence of specific problems. Recommendations for frequency of this practice vary. Overall, 46 percent of the facilities responding to our survey reported doing a phantom image check at least every 6 months. By set- Y ting, the percentages ranged from 28 at primary care physicians to 70 at “‘Percentages by state were: California, 40; Florida, 33; Idaho, 37; and Michigan, 26. Page 29 GAO/HRD-SO-32 Quality of Screening Mammography Chapter 3 High-Volume Provldem Most Likely to Adhere to Quality Standards for Screening Senaifometry, by Setting loo Poment I Note: Only repondents who perform screen-film mammography have been included in this analysis. Total 910 respondents. hospital breast clinics and 73 at mobile vans, as shown in figure 3.4. We also analyzed the extent to which facilities that did not report annual inspection by a radiological physicist performed semiannual phantom image tests, Although the need for such tests is probably greater at those facilities, the proportion performing them (30 percent) was smaller than for all facilities (46 percent). SelectedQuality Significant findings on compliance with several other quality assurance Asbrance Activities activities emerged from our survey: Analyzed . Monitoring repeat mammograms. Monitoring the number of repeat mam- mograms and analyzing the reasons the original images had to be dis- carded provides information about possible problems with either the * equipment or the work of the technologist, such as improper positioning. Page 30 GAO/HRD-90-32 Quality of Screening Mammography Chapter 3 High-Volume Providers Most Likely to Adhere to Quality Standards for Screening 8 3.4: Semiannual Phantom Image ks, by Setting Note: Total 1,026 respondents. Just over half of our respondents (62 percent) reported doing such mon- itoring, ranging from 44 percent at primary care physicians to 67 per- cent at HMOS. l Performing a second reading of mammograms within the facility, This provides a check on the radiologist’s interpretation. About 29 percent of respondents reported doing second readings; responses varied by state.14 . Submitting mammograms to peer review panels for second readings. This practice provides feedback on the entire mammography process, both production and interpretation of the image. Eleven percent of respondents said they submit images for peer review. l Following up on patient biopsies. This also provides feedback on the entire mammography process and was the most common quality assur- ance activity reported by respondents. Three-fourths indicated adhering to this standard. (See table V.7 for additional data on selected quality assurance activities by setting.) 14Percentagesby state were: California, 34; Florida, 32; Idaho, 10; and Michigan, 21. Page 31 GAO/HRD-90-32 Quality of Screening Mammography . Chapter 3 High-Volume Providers Meet LLkely to Adhere to Quality Standards for Screening ---I T---.-^------ We analyzed the proportion’ of facilities that monitored repeat mam- mograms and did at least one of the other quality assurance activities discussed above. About 44 percent of respondents did this, ranging from 35 percent of primary care physicians to 64 percent of HMOS, as shown in figure 3.5. A strong relationship existed between the volume of mammography per- Hig$-Volume formed and the rate of compliance with many quality standards. Exam- iders More Often ples of quality practices where the facilities with the lowest rate of compliance were in the lowest volume range and those with the highest rate of compliance were in the highest volume range appear in table 3.2. This association between high volume and adherence to quality stan- dards is significant, because, as discussed on page 22, high volume is a critical factor in reducing the price of screening mammography. A recent report in the Journal of the Florida Medical AssociationI” noted that not only is high volume necessary to lower costs, but it also contrib- utes to quality by giving radiologists sufficient work to increase the pro- ficiency of their interpretations. For several quality standards, we found no direct relationship between Compliance With the charge for screening mammography and the degree of compliance Quality Standards Not with professional standards. For other standards, there was a relation- Related to Charge ship, but its direction was not consistent. That is, in some cases provid- ers charging the lowest fees had the highest rate of compliance with a quality standard, while in other cases those with the highest fees had the highest rate of compliance. For example, 44 percent of respondents charging over $150 reported doing semiannual phantom image checks, while 56 percent of those charging $50 or less said they did so. Similarly, 27 percent of the facili- ties charging $50 or less reported following the practice of both sending positive mammogram reports to the patient and/or physician and then reminding the patient to contact her physician. This proportion was at least twice as high as any other charge category. However, the charge category with the highest proportion of facilities inspected annually by a radiological physicist was those charging over $150 (79 percent). For “Hobcrt A Clark, M.D., et al., “Screening Mammography in the Tampa Bay Area: Current Status and Implications for the Next Decade,” Journal of the Florida Medical Association, May 19S9, pp. 449-453. Page 32 GAO/HRD-90-32 Quality of Screening Mammography Chapter 3 High-Volume Providers Most Likely to Adhere to Quality Standards for Screening Figbre 3.5: Selected Quality Assurance Acthlties, by Setting 100 80 a0 70 60 50 40 I I n \ 30 20 10 0 Note: Activities are monitoring rate of repeat mammograms plus at least one of the following: (1) second reading of mammograms within facility, (2) peer review of interpretation, and (3) follow-up of patient biopsies. Total 1,026 respondents. daily processor sensitometry, the highest rates were in the lowest (45 percent) and highest (44 percent) charge ranges. Page 33 GAO/HRD-90-32 Quality of Screening Mammography Chapter 3 High-Volume Providera Most Likely to Adhere to Quality Standards for Screening Table 3:2: Relationship Between Volume and Pr+Aders’ Compliance With Quality Percent of respondents reporting StandaIds compliance Low-volume High-volume providers providers Quality standarda.-.~-- __.-.-__ (<25/week) (>lOO/week) Annual inspection by radiological physicist ---.-____ 58 87 -.-..~~ Annual physicist inspection of beam quality, average glandular dose, and phantom image 44 74 ..-.._---quality ___- ____.__-.- Daily processor sensitometry -- ------.-- 24 -_____ 50 -.-...-.-_______ Semiannual phantom image check by facility 33 73 ------_____ Monitoring repeat mammograms and doing one other quality assurance activityb 40 __- 60 Performing second reading of mammogram within facility -- 25 45 . .-..--- _.--- Following -..- --__~_--__ up on patient biopsies 70 89 instructing patients on breast self- examination 84 97 aWe did not include in this table the standards shown in table 3.1 with which almost all respondents reported complying. bOther activities include second reading of mammogram within facility, peer review of readings, and follow-up on patient biopsies. Our site visits also tended to dispel the concern that quality would be compromised at facilities charging lower fees for screening mam- mography. We visited three facilities that reported complying with many quality standards and that charged $50 or less for screening mam- mograms. All reported volume levels of at least 200 mammograms per week. One used a significant amount of volunteer labor to lower operat- ing costs, but the other two did not. The quality standards and recom- mended practices present at these facilities include . trained, experienced radiologists and certified radiologic technologists; l inspection by a radiological physicist at least annually; . daily processor sensitometry; . phantom image checks at least semiannually; l monitoring of repeat mammograms and following patients with abnor- mal findings; . proper record retention and reporting practices; and . extensive instruction on breast self-examination. Page 34 GAO/HRINO-32 Quality of Screening Mammography Chapter 4 Federal and State Regulationof Screening Qbmmography Is Limited Currently, both the Food and Drug Administration and the states have responsibility for regulating mammography equipment and services. FDA'Srole is to ensure the proper manufacture and installation of equip- ment, so it has no standards for mammographic image quality or patient radiation dose. Of the states we reviewed, only Michigan has a law requiring the use of dedicated mammography equipment and the setting of image quality and radiation dose standards. The lack of such stan- dards in the other three states limits their ability to regulate screening mammography services. Even with limited oversight of mammography, both FDAand state inspections have found noncompliance with perform- ance standards. The limited scope of FDA and state regulation and the problems found by inspectors underscore the importance of the quality standards that Con- gress required the Secretary of HHSto issue for Medicare-funded screen- ing mammography. On September 1, 1989, HHSpublished proposed regulations that closely parallel the professional quality standards we discussed in chapter 3. However, HHSplans to withdraw the regulations due to the repeal of the Medicare Catastrophic Coverage Act of 1988. State regulation of mammography equipment in the four states we vis- State Oversight of ited is limited. Only Michigan requires use of dedicated equipment and Mammography has established standards for image quality and radiation dose. Idaho does not regulate operators of mammography equipment, and the other three states we visited vary in the qualifications required for persons interpreting mammograms. State inspections have found image quality problems at mammography providers, but narrow legal authority often limits state enforcement efforts. 1leeulation of --v-- --- - - In three of the four states we reviewed, state regulation applies to use of Mammography Equipment all X-ray equipment, and emphasizes protecting the equipment operator and bystanders. These states have no separate standards for mam- and Personnel Limited mography services. In June 1989, Michigan enacted legislation adopting ACRstandards for mammographic image quality and radiation dose, becoming the only one of the states we reviewed to require the use of equipment specifically designed for mammography. Each of the states we reviewed has an office responsible for the over- sight of all types of X-ray equipment. The states require registration of X-ray equipment at the time of installation, as well as periodic reregistration. Page 38 GAO/HRD-90.32 Quality of Screening Mammography . Chapter 4 Federal and State Regulation of Screening Mammography Is Limited , The competence of the person operating the mammography equipment is critical to producing a useful image. Both California and Florida have requirements for and license operators of X-ray equipment, including mammography equipment. Idaho neither licenses the people who take mammograms nor sets minimum qualifications they must meet. The leg- islation enacted by Michigan in June 1989 requires state regulators to set standards for mammography operators. (For additional information on licensing requirements in these states, see app. IV.) The four states also differ in their regulation of persons who interpret mammograms, another critical component of the screening process. Michigan has no law governing who may interpret mammograms. According to state officials, Florida and Idaho permit any licensed phy- sician to interpret mammograms. California requires interpreters to have a state license in the healing arts. Frequency and Content of Each state we visited has an inspection program for all X-ray equip- ment, including mammography machines, The programs vary with Inspections Vary regard to frequency of inspection, staffing, and contents of the inspec- tions, although some equipment features are examined in all four states. The criteria of the Conference of Radiation Control Program Directors call for state inspections of new facilities within the first year of opera- tion In Florida and Idaho, newly registered equipment is inspected within 1 year, and Michigan will begin inspecting new equipment within 60 days. Subsequent periodic inspections occur annually in Florida and every 3 years in California. Michigan’s new law will increase subsequent inspections there from every 3 years to at least annually. Idaho bases inspection priorities on the setting in which the X-ray equip- ment is used. After the initial inspection during the first year, the sched- ule calls for inspection of mammography equipment in hospitals once every year and in physicians’ offices once every 2 to 3 years. Hospitals are given priority, state officials told us, because most X-ray examina- tions are performed in that setting. Because of a staffing shortage, how- ever, Idaho is not completely adhering to this inspection schedule. Of Idaho’s three state inspector positions, two have been vacant for over 1 year. Consequently, equipment in hospitals is being checked about once every 2 years and in physicians’ offices, about once every 3 years. The inspection procedures of the four states include a number of com- mon elements. For example, all measure the beam quality (WI,) and Page 36 GAO/HRD-90-32 Quality of Screening Mammography ChnDter 4 Federal and State Regulation of screen& Mammv@aphy Ia Limited average glandular dose. California, Idaho, and Michigan use a phantom (see footnote 1, p. 24) to inspect image quality, and Florida plans to begin this procedure in early 1990. Both California and Michigan use the phantom ACRuses in its accreditation program. Except for Michigan, the states lack legally binding standards regarding use of dedicated mam- mography equipment, limitations on the radiation dose received by patients, and minimum image quality. Although California uses the same phantom to evaluate image quality that ACRuses in its accredita- tion program, it does not have image quality standards. Florida and Michigan focus on whether the facility is using equipment specifically designed for mammography. Until recently, neither state prohibited the use of nondedicated equipment, although both encouraged facilities to use only dedicated equipment. As a result of Michigan’s efforts, the number of facilities using general purpose machines for screen-film mammography dropped from 25 to 3 between May 1987 and November 1988. Now, as indicated earlier, Michigan law requires use of dedicated equipment. (For additional information on inspection personnel, see app. IV.) State Inspections Find Of the states we visited, only Michigan had analyzed inspection data on Problems mammographic image quality and radiation exposure, basing its evalua- tions on ACRstandards. Between January and November 1988, Michigan inspected 95 mammography machines, over 20 percent of those in the state. As shown in table 4.1,35 percent of the 95 machines produced poor image quality and 11 percent registered excessive radiation dose. To determine if there were differences in adherence to certain quality standards in different settings, the Michigan inspection program catego- rized the inspection data by facility type. Mammography machines located in medical offices were more than twice as likely to produce poor image quality than machines located in hospitals and had more problems than those located in radiology offices. Y Page 37 GAO/HRD-90-32 Quality of Screening Mammography Chapter 4 Federal and State Regulation of Screening Mammography Is Limited , Table i4.1: Michigan Mammography Equipbent With Image Quality and/or Percent experiencing problem ion Problems, by Facility Type By facility0 ov. 1988) All equipment Rad’,“;g Medical Equipment problem inspected Hospital office Poor image quality 35 23 36 57 High radiation exposure 11 9 10 14 aMichigan’s principal facility categories are hospital, radiology office, and medical office. A clinic could be included as either a radiology or medical office. In compiling data on compliance with state law and regulations concern- ing mammography (see table 4.2), Florida officials did not distinguish between minor and serious violations, even though some of the types of noncompliance cited could affect patient dose. California and Idaho offi- cials were unable to provide summary data from their inspections. Table~4.2: Florida Mammography Equipjment Not in Compliance (Fiscal Year No. of Percent of 1988) Facility inspections noncompliance _------- --..--___ Hospital 173 9 Medical doctora 216 ~- 11 Osteopath 3 0 Mobile lab 15 0 aThis category includes both physicians and clinics Absence of Legal Officials from all four states told us that when they find violations of Authority Hinders state standards, facilities usually respond readily to efforts to bring about compliance. Steps such as court action are rarely necessary. How- Enforcement Efforts ever, because there are few state regulations pertaining to mam- mography, state officials have limited authority to require providers to correct problems. When there is no legal requirement in effect, such as for use of dedicated equipment or limiting radiation dose to the patient, state officials must depend on persuasion to correct the problem. Offi- cials told us they are often successful in these compliance efforts, even without the force of law. However, if a provider chooses not to comply with a recommendation, state radiation control officials have no author- ity to apply sanctions. The problems this situation can create are illustrated by our site visit to a California mobile van. A June 1988 inspection report by Los Angeles County noted several deficiencies, including unacceptable temperature of the developing solution in the film processor and poor phantom image Page 38 GAO/HRD-90-32 Quality of Screening Mammography Chapter 4 Federal and State Regulation of Screening Mammography Is Limited quality, No follow-up actions occurred, and since there is no legal stand- ard for image quality, the state could not require the facility to upgrade its technique. The facility has to repeat 10 percent of its mammograms, a facility official told us; this rate is five times that considered accepta- ble by the Conference of Radiation Control Program Directors. (See pp. 30-31 for a discussion of the quality assurance purpose of monitor- ing repeat mammograms.) 1 FDAregulates the manufacture and initial assembly of mammography IDA Regulates equipment. Its standards, which govern all types of X-ray machines, not Manufacture, just mammography equipment, apply to manufacture and installation. Pissembly of ’ Although FDAinspects only a small percentage of mammography equip- ment, it has found noncompliance with its standards. @--nmography $quipment FDA’s Performance FDA'Srole is to ensure that diagnostic X-ray equipment (including mam- SZandardsApply to mography equipment) is correctly manufactured and installed. Conse- quently, its standards apply only to the manufacturer and assembler of Manufacturer, Not User the equipment, not to the user, such as the mammography provider. FDAperformance standards cover such factors as equipment alignment and measurement of radiation leakage. There is no standard for the radiation dose received by the patient, as FDAconsiders that to be a practice of medicine issue not within its purview. FDAevaluates compli- ance with the standards during inspections of newly installed radio- graphic equipment. Relatively Few Inspections Partially through contracts with 34 state radiological health agencies, Performed including those of California and Florida, FDAarranges for field testing of newly installed diagnostic X-ray equipment. In states without con- tracts, FDAstaff do the inspections. Inspectors test to determine whether the equipment complies with FDA'Sperformance standards, focusing on equipment installed within the past 12 months. Only within this time frame does FDAtake regulatory action, and few inspections occur beyond the first year. * States decide which X-ray facilities to inspect, with some guidance from FDA as to the percentages of each type of equipment to check. The FDA Page 39 GAO/HRD-90-32 Quality of Screening Mammography . , Chapter 4 Federal and State Regulation of Screening Mammography Is Limited goal is to inspect up to 30 percent of new X-ray equipment within 1 year of installation, but the proportion of mammography machines inspected 1 in the four states we visited was much lower (see table 4.3). This is because FDAgives priority to general purpose equipment, the type most I, frequently used. -- Table 4.3: FDA Inspections of Newly Installed Mammography Equipment in Four States (1986-l 988) / / 1986 1987 1988 Installed Inspected installed Inspected Installed Inspected State ’ .. ...!“O.) No. Percent (no.) No. Percent (no.) No. Percent Califorr 1la 207 11 5 193 16 8 79 35 44 Florida~ 100 12 12 113 4 4 41 8 20 Idaho / 5 -_--.-- 0 0 ___- 7 0 0 6 0 0 Michi& 84 5 6 88 0 0 33 0 -6 --.L_ Subkantial Noncompliance FDA'Sinspections in the four states we visited have found a substantial With Standards Found amount of noncompliance with its performance standards (see table 4.4). If noncompliance is found within 1 year of equipment installation, FDA assumes it is due to improper installation. The assembler is responsi- ble for correcting the problem unless obvious misuse is apparent. FDA sends a notice of noncompliance to the assembler instructing repair within 30 days. The assembler must then notify the FDAdistrict office when the problem has been corrected. Table 4.4: Noncompliances Identified by FDA in Four States (1982-I 989) No. of No. of Percent of State .-__...---_-..--- _..-...--.. II- inspections” noncompliances noncompliances California 86 9 16 Florida 35 8 23 Idaho ..-_ .--.--- .._--_-.. 1 1 100 Michiaan 8 1 13 “Between Jan. 1982 and Feb. 1989 Problems with image receptor alignment are the most frequent type of noncompliance found. If the beam and image receptor are not properly aligned, part of the body that should not be exposed to radiation is exposed, and part that should be included in the image is not. Page 40 GAO/HRD-90-32 Quality of Screening Mammography Chapter 4 Federal and State Regulation of Screening Mammography Ie Limited The 1988 Medicare Act required the Secretary of HHSto develop regula- H/HS Proposed tions to ensure that screening mammography reimbursed by Medicare Standards for would meet safety and accuracy standards. The law specifically called for requirements on equipment, personnel, and film retention. On Sep- tember 1, 1989, HHSpublished its proposed regulations. These included standards closely paralleling those developed by professional organiza- tions, such as ACR. Such standards, along with an adequate enforcement mechanism, would help to assure the provision of high quality screening mammography. However, since the Medicare Catastrophic Coverage Act of 1988 was repealed, HHSplans to withdraw its proposed regulations. Page 41 GAO/HRD-90-32 Quality of Screening Mammography Cha&!r fi C&wlusions Breast cancer is increasing in American women, and mammography is the best method of detecting breast cancer at its earliest, most curable stage. Symptom-free women who follow recommended guidelines for periodic screening mammography can lower their risk of dying from breast cancer by 30 percent. Despite broad agreement in the medical community on the importance of regular screening, relatively few women participate in screening mammography programs. A Medicare screening benefit such as that provided in the Medicare Cat- astrophic Coverage Act of 1988 would make millions of high-risk women eligible for low-cost screening mammography. However, some members of Congress were concerned that the act’s $50 limit on what providers could charge might reduce the availability of high-quality services. We found facilities that charged $50 for screening mammography and reported complying with quality standards. Providers can better do so if they distinguish between screening and diagnostic mammography and operate high-volume screening practices. This enables them to use such procedures as batch reading of films and make more efficient use of equipment to lower the cost of providing screening services. The Physi- cian Payment Review Commission, which found that providers that make a distinction between screening and diagnostic services charge less for screening, concluded that $50 is an appropriate charge if volume is sufficiently high. High volume also was associated with high quality in our survey. The facilities that reported the highest rates of compliance with many qual- ity standards were those providing the highest volume of mam- mography services. However, we found no consistent relationship between charge and compliance with quality standards. Quality standards are not always adhered to in current mammography practice, our survey and other studies have found. Oversight of mam- mography services by the states and FDAis limited. Of particular con- cern, there is little regulation of mammographic image quality and the radiation dose patients may receive. In conclusion, Medicare coverage for screening mammography would bring a valuable life-saving tool to a large population of women at risk of developing breast cancer. The evidence we found suggests that limit- J ing charges to $50 should not jeopardize the quality of care available to Medicare beneficiaries. To the contrary, the charge limit would Page 42 GAO/IBID-90-32 Quality of Screening Mammography Chapter 6 Conclusions -’ encourage provision of screening mammography in high-volume set- tings, which we found were most likely to comply with quality stan- dards. By reinforcing the trend to high-volume settings, which can best provide high quality at lower prices, such a charge limit also could increase the accessibility of screening mammography for younger women eligible for screening. At the same time, we found a need for strong federal standards to assure the quality of screening mammography. Although there are pro- fessional standards for acceptable image quality and radiation dose, there have been few legal mechanisms to enforce them. The result is a wide range of image quality and patient dose in current mammography practice. The regulations that were proposed by HHSwould have helped to fill this regulatory gap. Page 43 GAO/HRD90-32 Quality of Screening Mammography Apr>e/ndix I GAO’sSurvey Methodology In January 1989, we mailed a questionnaire to all facilities identified by state radiological health departments in California, Florida, Idaho, and Michigan as having mammography equipment. We assumed that these facilities were currently in business in the respective states. This appen- dix contains a technical description of our questionnaire design, pretest procedures, and response rate. The questionnaire was designed to obtain information concerning the practices and procedures of facilities performing screening mam- mography. It was reviewed by officials from the National Cancer Insti- tute and American College of Radiology. / Before distributing the questionnaire, we pretested it in person with Qu&tionnaire officials at nine facilities in the Washington, D.C., metropolitan area and Pretested in Two two facilities in Providence, Rhode Island. These facilities represented Arehs the types of settings likely to be found among the facilities to be sur- veyed. Pretesting the questionnaire assured us that the questions were generally understandable and free of confusion and error. During the pretest, the officials completed the questionnaire as if they had received it in the mail. Our staff noted the time it took to complete each question and any difficulties the respondents experienced. Once the question- naire was completed, we used a standardized approach to elicit descrip- tions of difficulties and issues encountered with each item. Using the pretest results, we revised the questionnaires to ensure that (1) the potential respondents could and would provide the information requested and (2) all questions were fair, relevant, easy to answer, and relatively free of design flaws that could introduce bias or error into the study results. In addition, we tested the questionnaire to ensure that the task of completing it would not place too great a burden on the respondent. Of a total of 1,485 questionnaires mailed to facilities in four states, IuJ 3 .’v.r3bculbesponse 1,242 were returned. We adjusted our universe to 1,369 to exclude ques- 1 *? Percent tionnaires that (1) were mailed to facilities no longer in operation, (2) were duplicates,’ or (3) were returned as nondeliverable.” This resulted ‘The list the state provided indicated a different address or contact person, but the recipients informed us the facilities were identical. “After making several unsuccessful attempts to contact these facilities by telephone and mail, we assumed they were not currently in business. Page 44 GAO/HRD90-32 Quality of Screening Mammography Appendix I GAO’e Survey Methodology in an overall response rate of 82 percent. The initial and adjusted uni- verse and the number of responses are shown by state in table 1.1. ladle 1.1: Response to CIAO Questionnaire, by State No longer in Initial operation, No. of universe of undeliverable Duplicate Adjusted usable State facilities questionnaires facilities universe __--.- responses - ~-~. Cal I fornia 757 10 70 677 553 Flo Ida 313 35.________-- 06 91 298 34 _-..__. .-~~~- 239 o Idaho . . __--.--- __.-... ---_-----__ -.-. -.~ MICc lgan 380 ___-- 3 17 360 304 Totbls 1.485 19 97 1.389 1.126 Page 45 GAO/HRD-90-32 Quality of Screening Mammography /ppc&dix II k(l.0 Questionnaireon Mammography - United States General Accounting Office Screening Mammography Survey ‘Ihe United States General Accounting Office (GAO), an agency of the U.S. Congress, is conducting a survey of facilities identified by state offices of radiologic health or radiation control as having equipment used to perform mammgraphy. The purpose of this questionnaire is to obtain information about mammography conducted on asymptcmatic patients. In this questionnaire, we refer to such mannnography as screening mamnography. This questionnaire should take about 20 minutes to ccmplete. Your responses will be kept confidential. We will report your responses only in sum~ry with those of other facilities that respond to this questionnaire. Your responses will not be made knm to anyone outside of the GAO. This questionnaire should bs axnpleted by the person(s) most familiar with your facility’s screening mammography practices. Before you begin, because of the variety of information requested, you may want to briefly review the questionnaire to determine the necessary sources of information you will need and whan you may want to consult. In the event you receive more than one questionnaire, please ccanplete only one questionnaire for each facility where mamncgraphy is performed, mark the addTiona1 questionnaire(s) “duplicate”, and return all questionnaires in the same lusiness reply envelope provided. It is possible that your state records both individual avnership as well as facility location. Iherefore, since we sent a questionnaire to each address rovidd, we may have included in our survey the owner of the equipnent as well as the Pacility. In order for us to report accurate information to the Congress, it is very imbrtant that all questionnaires be returned. LABEL Please provide the name, title, and telephone number of the primary person we may contact if additional information is required concerning your responses. N~MZ of Primary Contact Person: Official Title: Telephone Nuder: ( ) Please return the questionnaire in the enclosed business reply envelcpe, or if the envelope is misplaced, send it to the address shown on the back of the questionnaire. If you have any questions, please call Helene Toiv at (202) 426-0842 (or 426-0800). Y Page 46 GAO/HRD-90-32 Quality of Screening Mammography , yl!$ Appendix II GAO Questionnaire on Mammography 1. Is mazmgraphy performed at your facility? (CHECKcm.) a. ( ) Yes b.( 1 NO ->(Stqpl PLEASE HEIUHN‘IHIS QUESTIONNAIRE. IT IS IMKXIANl’ ‘IHAT YOU HETURNTHIS QUESTIONNAIRE.1 2. Does your facility provide manmgraphy to -- (a) as to ss screening PPanmography), (b) sy;ptan$ic pat6%%%$%% :ms:o mmmgra&), or (c) both asymptanat c a symptanatic patients? (CHECKCNE.) a. ( ) Screening mnmgraphy only ->(SKIP lO -CN 5.1 b. ( ) Diagnostic mammgraphy only c. ( ) Both diagnostic and screening manmmgraphy ->(SKIP Xl Quesnrcrs 5.) 3. Using the categories provided helm, please indicate the ofm kind of setting where the majority of your diagnostic mammgrams are performed, (CHECKCNE.) a, ( ) Primary care physician’s office (e&, gynecologist, inteY?IiStr SUrgeOn) b. ( ) Radiology private practice c. ( ) Hospital -- Radiology Department d. ( ) Hospital -- Cutpatient clinic/ Ambulatory Care center e. ( ) Breast Center/ Clinic (i.e., freestanding facility for screening and/or treatment of breast disease) f. ( ) Health Maintenance Organization (HMO) g. ( ) Mu;;l.;Eyialty clinic (e.g., multispecialty group practice, outpatient h. ( ) Mobile van -- Affiliated with a hospital, clinic, or physician’s practice i. ( ) Mobile van -- Not affiliated with a hospital, clinic, or physician’s practice j. ( ) Other (PLEASE SPECIFY) Page 47 GAO/HID-DO-32 Quality of Screening Mammography Appendix II GAO Queatiouuaire on Mammography 4. How would you describe the physical location of your facility? (CHECKONE.) a. ( Urban b. ( Suburban C. ( Rural 5. Using the categories provided below, please indicate the am kind of setting where the majority of your screening tnamnogrm are performed. (CHECKCNE.) a. ( Primary care physician's office (ea, gynecologist, internist, surgeon) b. ( Radiology private practice C. ( Hospital -- Radiology Department 4. ( Hospital -- Outpatient clinic/ Ambulatory Care Center e. ( Hospital -- Breast Screening clinic (i.e., screening clinic located in hospital, but for outpatients only) - f. ( Breast Center/ Clinic (i.e., freestanding facility for screening and/or treatment of breast disease) 4. ( Health Maintenance Organization (HMO) h. ( Multispecialty clinic (ea, multispecialty group practice, outpatient clinic) i. ( Mobile van -- Affiliated with a hospital, clinic, or physician's practice j. ( Mobile van -- Not affiliated with a hospital, clinic, or physician's practice k. ( Other (PLEASE SPECIFY w Page 48 GAO/HRD-90-32 Quality of Screening Mammography Appendix II GAO Questionnaire on Mammography 6. Ho.+would you describe the physical location of your facility? (CHECKONE.) a. ( 1 Urban b. ( ) Suburban c. ( ) Rural 7. Approximately during what month and year did your facility screening manznography services? begin provid ing 8. During a normal week for each day listed below, indicate the total number of hours your facility provides screening manuqraphy. Total Nmber of Ham3 Screenirq Memmgraphy Is Prwided a. Monday Hours b. Tuesday Hours C. Wednesday Hours d. Thursday Hours e. Friday Hours 3. IXlring a normal week, does your facility provide screening marsqraphy after 6:00 p.m.? (CHECKONE.) a. ( 1 Yes b.( ) No ->(SKIP ‘R) q.JETrlm 11.1 10. Approximately how many evening IXIULX (after 6:00 p.m.) during a normal week does your facility provide screening mamnography? Total evening hours screening mammographyprovided Page 49 GAO/HRD-90-32 Quality of Screening Mammography Appendix II GAO Questionnaire on Mammography << Answer only with mmect to the cm setting ylou identified in question 5 >> 1. Check below when, if at all, your facility is open to provide screening mammographyduring the weekend. (Saturday and/or Sunday) (CHECKONE.) a. ( ) Every weekend b. ( ) 3 weekends a month c. ( ) 2 weekends a month d. ( ) 1 weekend a month e. ( ) Not open during the weekend !. During a normal week, haw many diagnostic and/or screening -rams does your facility perform? a. Number of diagnostic mananograms b. Number of acmeming mammograms I. Listed Bela are various pieces of information which might be collected fran a patient as part of the screening manmqraphy process. Indicate the information you usually collect. (CHECKALL 7xxr APPLY.) a. ( ) Demographic data (e2, age, marital status, ethnic background) b. ( ) Current breast symptans (e.g., breast tenderness, pain, lump, or nipple discharge) C. ( ) Previous mammography information (e&, date, where performed) d. ( ) Surgical history -- breast surgery e. ( ) Family history of breast cancer f. ( ) Current medication history (e.g., hormone) g* ( ) Other (PLEASE SPECIFY) h. ( ) Do not collect information Page SO GAO/HRD-90-32 Quality of Screening Mammography Append& II GAO Questionnaire. on Mammography << - -a with nwpect to the Qy) wttinn you identified in qweticm !j >> 4. At the time a screening mamnogram is performed at your facility, is a breast physical examination (palpation) routinely conducted? (CHECKONE.1 a. ( ) Yes b. ( ) No->(sKIPmQuEsTIm17.) 5. When a breast physical examination is performed at your facility, who usually does the examination? (CHECKONE.) a. ( 1 Radiologist (who interprets the films) b. ( 1 Radiologist (not necessarily interpreter) C. ( 1 Other physician d. ( ) Technologist who performs the mmmgram 8. ( ) Nurse f. ( ) Other (PLEASE SPECIFY1 6. Does your facility charge a separate fee for the breast physical examination? (CHECKONE.) a. ( 1 Yes -> Please indicate fee charged $ b. ( ) No Page 61 GAO/HRD90-32 Quality of Screening Mammography Appendlv XI GAO Questionnuke on Mammography << m c&y with Lwspect to the one settitlg you idmtified in qwstim 5 >> 17. Indicate what mechanism(s) is used by your facility to inform the patient about breast self-examination. (CHECKALL ‘IMAT.APPLY.) a. ( ) Video b. ( ) Pamphlet C. ( ) Staff person instructs the patient d. ( ) Other (PLEASE SPECIFY) e. ( ) No information provided 18. For screening manmwgraphy, indicate if your facility perfO?ZiW Screening -rams using (a) dedicated mammographyequipment (i.e., equipment manufactured for the sole purpose of nwmwgraphy M general radio&$& equipnent that is modified for mamwgraphy only and cannot be used for general radiographic *purposes), (b) general purpose radiographic equipment, or (c) both dedicated manwqraphy and general purpose radiographic eguipnt? (CHECKONE.) a. ( ) Dedicated rwmwgraphy equiprsent only ( i.e. , equipment manufactured or rrcdified for manwqraphy only) - b. ( ) General plrpoee radiographic equipment C. ( ) Both dedicated wmmwgraphy and general purpose radiographic eguipnent 19. Indicate the type of m manuwqraphy your facility uses. (CHECKONE.) a. ( ) Screen-film mwtwgraphy b. ( ) Xeranarranography C. ( ) Both screen-film and xermraphy Page 62 GAO/HRD-90-32 Quality of Screening Mammography . Appendix II GAO Questionnaire on Mammography << Aninrer akly with m3pect to the one 8ettirx.g yau identified in question 5 >> 20. Which views does your facility usually do for a screening bilateral manmrgram? (CHECKALL ‘JXAT APPLY.) a. ( ) Cranio-caudal or Cephalo-caudal (1 view per breast) b. ( ) True Lateral (1 view per breast) c. ( ) Oblique--Mediolateral (1 view per breast) d. ( ) Other (PLEASESPECIFY) 21. Listed belaw are various categories representing individuals who might screening mamnography. Indicate the ane category which represenfs the %%%a1 at your facility who usually performs the screening marsnograms. (CHECKONE.) a. ( ) AFW registered radiologic technologist b. ( ) State licensed radiologic technologist CO ( ) Padiologic technician d. ( ) Nurse e. 1 ) Radiologist f. 1 ) Technician trained to perform marsnography 4* ( ) Other (PLEASE SPECIFY) 22. Does any other individual(s) perform screening, marsnogramsat your facility? (CHECKONE.) a. ( 1 Yes b. ( ) No ->(!xIP ‘1[10-CN 24.) Page 63 GAO/HRD-90-32 Quality of Screening Mammography ,, - Appendix II GAO Questionnaire on Mammography <~krewercml~withre3mect to theone sf3ttingyar identified inque~tim 5 >> 23. Indicate each category which represents the other individual(s) who performs screening -rams at your facility. (CHmL THAT APPLY,) a. ( ) APKI registered radiologic technologist b. ( ) State licensed radiologic technologist c. ( ) Radiologic technician d. ( 1 Nurse e. ( ) Radiologist f. ( ) Technician trained to perform manm-ography cl* ( ) Other (PLEASESPECIFY) 24. Listed below are categories representing individuals who might interpret screenirq -rams. Indicate the category which represents the individual at Your facility who usually does the final interpretation of the screening manmxjrams. (CHECKONE.) a. ( ) Radiologist b. ( ) other physician c. ( ) Nurse d. ( ) Technologist e. ( ) Other (PLEASE SPECIFY) 25. Is the final interpretation of the screening mammogramdone by any other individual(s)? (CHECKONE.) a. ( 1 Yes b. ( ) No ->(SKIP ‘X0 UJEJ!KN 27.) Page 64 GAO/HRD-90-32 Quality of Screening Mammography Appendix II GAO Questionnaire on Mammography -+- << e to the ahe aettim yuu identified in qwstion 5 >> 26. Indicate each category which represents the individual(s) who inferprete screening mamnograms at your facility. (CHECKALL THAT APPLY.) a. ( ) Radiologist b. ( ) Other physician C. ( ) Nurse d. ( ) Technologist 8. ( ) Other (PLEASE SPECIFY) 21. Does your facility accept self-referred patients? (CHECKONE.) a. ( 1 No ->(sKIP 10 QxlEsTlm 31.) b. ( ) Yes, but the patient must provide the name of her personal physician of select a hysician fran a list provided by the 15acility ->(sxcn To tJlEsTIm 31.) C. ( ) Yes, even if the patient does not designate a personal physician 28. If a patient has a mtim Dram snd has not designated a personal physician, what is usually done with the mamnogramreport? (CHECKONE.) a. ( ) Report sent to the patient b. ( ) &port sent to the patient and patient telephoned about the results C. ( ) Report not sent, but patient telephoned about the results d. ( ) Raport not sent, but filed at the facility e. ( ) Other (PLEASE SPECIFY) Page 65 GAO/BRD-9052 Quality of Screening Mammography Appendix II GAO Questionnaire on Mammography ~<Mmwercnlywithm8pect to theme settirqyar identifiedinauesticm 5 77 29. If a patient has a positive mammcgramand has not designated a personal physician, mt is usually done with the mamnogramreport? (CHECKONE.) a. ( ) Report sent to the patient b. ( ) Report sent to the patient and patient telephoned about the results C. ( ) Report not sent, but patient telephoned about the results d. ( ) Report not sent, but filed at the facility 8. ( ) Other (PLEASE SPECIFY) 30, List in the space bslm, any other actions your facility takes when a patient has a pcdtive -ram and has not designated a personal physician. 31. For a patient who has a personal phvsician and has a negative nmmmgrm, what is usually done with the mammgrzanreport? (CHECKONE.) a. ( ) Report sent to the patient b. ( ) Report sent to patient's physician C. ( ) Report sent to the patient -and patient's physician d. ( 1 Report not sent, but patient telephoned about the results e. ( ) Report not sent, but patient's physician telephoned about results f. ( ) Report not sent, but patient and patient's physician telephoned about the results 9. ( ) Report not sent, but filed at the facility h. ( ) Other (PLEASE SPECIFY) w Page 56 GAO/HRD-90-32 Quality of Screening Mammography -c Appendix II GAO Questionnaire on Mammography <:( Ansver Cnlv With LBSP3Ctto the One setting you identified ~J-Iquestion 5 77 32. For a patient who has a personal physician and has a positive mammogram,what is usually done with the -ram report? (CHECKONE.) a. ( ) Report sent to the patient b. ( ) Report sent to patient's personal physician C. ( ) Report sent to the patient -and patient's physician d. ( ) Report not sent, but patient telephoned about the results e. ( ) Report not sent, but patient's personal physician telephoned about results f. ( ) Report not sent, but patient and patient's physician telephoned about the results cl. ( ) Report not sent, but filed at the facility h. ( ) Other (PLEASE SPECIFY) 33. What actions , if any, does your facility take in addition to sending or filing the mamrcqram report when a patient has a positive marreqram and has a physician? (CHECKALL THAT APPLY.) a. ( ) No other actions taken b. ( ) Telephone patient's designated or personal physician to discuss marsnogram results C. ( ) Contact patient to remind her to contact her physician d. ( ) Other (PLEASE SPECIFY) Page67 GAO/HRD-9032 Quality of Screening Mammography . Appends II GAO Questionnaire on Mammography -..--:-.--..-- <C~~ailyWithrBSpect tothare settifqyw identified inqwsticm 5 77 34. For each item listed below, check in: Colum 1: Whether your facility keeps the item as part of the patient's record. cdum2: If yes, hew long the item is kept. (CHECKCNE FOR EACH ITEM.) cduml calm 2 Item kept? How long is the item kept? Item No Yes a. Mamxgram report If Yea -7 3 . Original -ram images If Yes-7 c. Patient infonnatlon provided at the time If Yea -7 ofths mamwgram d. Other (Specify) If Yea -7 35. How much does your facility generally charge for a screening bilateral manmqram? The mamnogramcharge should include charges for the mamnoqramand interpretation of the -ram. $ Charge for Screening Bilateral Manwgram Y Page 58 GAO/HRD-90-32 Quality of Screening Mammography Appendix II GAO Queetiouuaire on Mammography << mr only with m to the one settinq yau iclentified in question 5 >> 16. Does your facility periodically have saneone inspect all or part of your mawcgraphy system? (CHECKONE.1 a. ( ) Yes b.( ) No ->(sxIP TD (JwScIm 42.) ‘7. Listed b&w are various individuals who might inspect a mamnqraphy system. Indicate in: columl: Whether your marrmwgraphy system is inspected by each individual. colum 2: If yes, how frequently each individual inspects your system. (CHECKONE.) colum 2 ( How frequently mammography system inspected. (CHECKONE.) Individual a. Consultant Radiolcg ical Physicist If Yes-> (i.e., physicist noXt facility) b. Federal/State/ Local Radiation Control If Yes-> Inspector w Page 69 GAO/HRD-9052 Quality of Screening Mammography Appendix II GAO Questionnaire on Mammography << Anam anlY with mmect to then on8 mttim mu identified in qusrtion 5 >> 38. For each of the item listed below, check in: Colmm It Whether the consultant radiological physicist or radiation control inspector you identified in questian 37 inspects the item. &alum 2r If yes, which individual(s) inspects the item. (CHECKALL 'MAT APPLY.) columl colum2 1 Inspect ( Who inspects item? item? (CHECKALL THAT APPLY) I Fed/State I l-T- Consultant Local Radiological Control Physicist Inspector IfYes-> IfYes-> If Yes--> If Yes-> If Yes-> If Yes-> IfYes-> Page 60 GAO/HRD-90-32 Quality of Screening Mammography . * - , Appendix11 GAOQueetionuaire on Mammography << mr a& with resoect to the one 8etti.q you identified in question 5 >> 9. Is there a radiological physicist on staff at your facility? (CHECKONE.) a. ( 1 Yes b. ( 1 No ->(sKIP TO qJEsMm 41.) 0. Listed belaw are various items a staff physicist might inspect as part of his/her overall quality assurance duties. Check in: columl: Whether the staff physicist at your facility inspects each item. colum 2: If yes, how frequently he/she performs the inspection. (CHECKALL THAT APPLY.) cbluml cblum 2 I Inspect I How trequently item inspected? item? (CHECKALL THAT APPLY) I-T- Item I a I YesI a. Beam quality (Half value layer) b. Focal spot size If Yes-> c. Average glandular dose If Yes-> d . Phototimer If Yes-> e. Consistency of mA station If Yes-> f. KVP If Yes--> g. Phantan image quality If Yes-> h. Other (SPECIFY) Page 61 GAO/HRD-90-32 Quality of Screening Mammography . Appendix II GAO Questionnaire on Mammography << Anmmr anly with m3pect to the am aettim ya identified in Quegtion 5 >> 41. Listed below are additional items that might be inspected as part of your facility's quality assurance program. Check in: colum1: Whether each item is inspected as part of your quality assurance program. C&mm 2: If yes, how frequently the item is inspected. (CHECKCNE.) columl oolum 2 1 Inspect How frequently is the item inspected? Item? (CHECKONE.) I i-r Item No Yes a. Grids If Yes-:b’ . soreens IfYes-: >’ C. Processor Sensitanetry IfYea-: > d. Phantan image quality IfYes-: > ==--?I- I t I 1 LfYes-: 1 /_ I I I I I Page 02 GAO/HRD-90-32 Quality of Screening Mammography Appendix11 GAOQuestionnaireonMammography -c << Anarrer cmly with respect t0 the one settins you identified in question 5 >> 42. A facility might include sane of the follcwing procedures as part of its quality assurance program. Which procedures, if any, does your facility usually perform. (CHECKw THp;r APPLY. 1 a. ( ) Review film quality b. ( 1 Perform a second reading of mammogramswithin the facility C. ( 1 Submit mammogramsfor second reading by peer review panel outside the facility d. ( ) Follow up on patient biopsies e. ( 1 Monitor number of repeat -rams due to equipnent, patient, and/or technologist problems f. ( ) Other (PLEASE SPECIFY) 9. ( 1 NO procedures performed 13. In question 5, did you identify mobile van, either affiliated or not affiliated with a hospital, clinic, or physician’s practice , as the one kind of setting where the majority of your screening -rams are performed? (CHECKCNE.1 a. Yes --#ZIP 'ID @JEXTMON46.) b. No 4. Ooes your facility provide any marranography services using a mobile van? (CHECK ONE.) a. Yes b. No ->WIP 'ID UJBI'IcrJ 46.) 5. Wring a normal week, how many diagnostic and/or screening m%nmzqrams are performed in the mobile van? a. Number of diagnoetic mammograms b. Number of screening mammograms Page63 GAO/HRD-9052Qualityoi'ScreeuingMammography Appendix II GAO Questionuaire on Mammography -. 46. If you have any comments regarding this questionnaire or issues relating to manxwgraphy, please write them in the space provided below. **** +IrfmK ml FOR YaJR COONJCON **** Page 64 GAO/HRD-90-32 Quality of Screening Mammography ,’ Aeocndix III SelectedQuality Standardsfor Mammography and RelatedSurvey Results This appendix contains more detailed information about the quality standards discussed in chapter 3, as well as additional survey results. I To obtain the best mammographic image with the smallest dose of radia- Us ’ of Dedicated tion, it is essential to use dedicated mammography equipment-that Eql1 ipment specifically designed for mammography. Its features enable the operator to obtain high-quality images with much lower radiation exposure than I is possible with general X-ray equipment. Use of dedicated equipment is one of the standards the Congress had mandated for screening mam- I mography reimbursed by Medicare. State-of-the-art screening mammography practice in this country is to Ta$ing Two Breast take two views of each breast. About 82 percent of responding facilities Vie&w indicated they take two views, 17 percent take three views, and only 1 percent do one view. Although almost all respondents reported taking at least two views, we did find one problem. The NCRPrecommends that screen-film mammography consist of the craniocaudal (cc)-oblique com- bination, and specifically advises using an oblique view instead of a true lateral view.! However, about 11 percent of the respondents doing only screen-film mammography reported using the cc-true lateral combina- tion instead of cc-oblique. The settings in which this was more likely to occur were primary care physicians’ offices and hospitals; no mobile vans reported this practice. The person taking the mammogram plays an essential role in providing Performance and quality mammography. Proper positioning of the patient and adjustment Interpretation of of the equipment are vital to producing a good image. ACR’S accreditation Mammograms by standards require, and the CRCPD recommends, that radiologic technolo- gists operating mammography systems be registered with the American Certified or Licensed Registry of Radiologic Technologists (ARRT)”and/or state-licensed and Personnel have specialized training in mammography. Of the 97 percent of facilities reporting that the person who usually per- forms their screening mammograms is ARRT-certified, state-licensed, or both, about 9 percent indicated that another type of staff member (see app. II) sometimes performs screening mammography. The settings with ‘In xeroradiography, the true lateral view is appropriate. ‘CRCPD recommends registration with either ARRT or the American Registry of Clinical Radio- graphic Technologists (ARCRT). Page 65 GAO/HRD-90-32 Quality of Screening Mammography Appendix III Selected Quality Standards for Mammography and Related Survey Results the largest proportions reporting this were primary care physicians (17 percent), radiology practices (11 percent), and hospital breast clinics (10 percent). HMOS and multispecialty clinics had the smallest proportions (6 percent). The CRCPD states that American Board of Radiology-certified radiologists with documented training in mammographic image interpretation should interpret mammograms. ACR requires a board-certified radiologist to supervise accredited programs and interpret the mammograms. The radiologist also must have specific training and experience.:’ About 99 percent of providers responding to our survey reported that the person who interprets the mammogram is a radiologist, and 99.9 percent reported that it is a radiologist or other physician. Guidelines for screening mammography programs stress the importance Reporting Positive of a system for reporting test results-particularly for “positive” mam- Mhmogram Results mograms, those with a finding of possible abnormality-to the patient arrd Following Up and/or her physician, It is especially important for the facility or radiol- ogist to ensure that women without personal physicians are referred for appropriate care when their mammograms are positive. When the woman has designated a personal physician, about 93 percent of respondents reported that they send positive mammogram reports to the physician. An additional 5 percent send the report to both the patient and her physician.4 When the patient has not designated a physi- cian, 99 percent of respondents take action to inform her of the results and/or arrange for her to receive medical care. Experts also recommend telephoning the patient’s physician to ensure that the written report is not accidentally overlooked. Two-thirds of our respondents reported following this practice. Radiology private prac- tices (81 percent), hospital breast clinics (79 percent), and breast clinics (78 percent) were among the settings most likely to take this extra step, and hospitals (61 percent) among the least likely. “The radiologist supervising the program and/or interpreting mammograms must have (1) completed a residency program after 1982, when mammography was added to the radiology board examination; or (2) 3 years’ experience reading at least 10 cases per week; or (3) 40 hours of mammographic education in the past 2 years. ACR is now phasing in additional criteria regarding training and experience. IDue to rounding, table 3.1 shows a total of 99 percent of respondents reporting the results to the patient and/or physician. Page 96 GAO/HRD-SO-32 Quality of Screening Mammography Appendix III Selected Quality Standards for Mammography and Related Survey Results We also asked survey participants whether they contacted patients with positive mammograms who have designated physicians to remind them to contact their physicians. Eleven percent of all respondents reported that they both send the radiologist’s report to the patient and/or physi- cian and contact the patient with a reminder. The settings that most often reported following this practice were mobile vans (36 percent), primary care physicians (29 percent), and hospital breast clinics (26 percent). When a patient with abnormal findings has not designated a physician, many respondents said they take additional steps beyond reporting the mammogram results to her. These include referring the patient to a sur- geon or other physician and sometimes making the appointment for the patient, calling the physician to confirm that the patient was seen, and contacting the patient to ensure that she has sought medical care. Retention of the original mammographic image is extremely important Record Retention for quality screening. Comparison with a previous film can indicate whether a suspected abnormality is a benign structure in the woman’s breast or confirm that a new lesion has appeared. ACRpolicy calls for providers to keep mammograms and positive reports for at least 5 years, and the CRCPDrecommends that images be maintained indefinitely. Our survey found that 98 percent of all respondents keep original images for at least 5 years, and 49 percent for over 10 years. (See table V.8 for information by setting.) Almost all respondents told us they retain the mammogram report, with 99 percent reporting they keep it for at least 5 years. Information collected about the patient’s medical history Taking Patient’s Medical History . helps identify women who are not candidates for screening mam- mography but instead should be considered diagnostic patients, such as women with current breast symptoms, and l can be used to identify patients who are at greater risk for developing breast cancer, which may be a factor in determining how often they should be screened. Y With one exception, every respondent to our survey reported collecting some patient information. Page 67 GAO/HRD-90-32 Quality of Screening Mammography Appendix III Selected Quality Standards for Mammography and Related Survey Results Because breast self-examination (BSE) is considered another critical ele- Instruction on Breast ment of breast cancer screening, there is consensus that a good screen- Self-Examination ing program should include information on how to perform BSE!As the ACSwill provide free instructional pamphlets for distribution to patients, cost should not be a barrier to facilities providing this service. At least one form of BSEinstruction is available at 90 percent of the facilities that gave us information on this subject. Three-fourths of responding facilities (78 percent) give pamphlets to patients, 42 percent show a video tape, and instruction by a staff person occurs at 41 percent. When analyzed by setting, the data show that 100 percent of breast clin- ics, hospital breast clinics, and mobile vans provide information on self- examination, while 83 percent of hospitals and 85 percent of radiology practices do so. However, when the analysis includes only facilities that accept self-referred patients, where the need for such information might be greatest, the overall rate of facilities that provide information increases to 96 percent, and the rates for hospitals and radiology prac- tices rise to 93 and 95 percent, respectively. Page 68 GAO/HRD-90-32 Quality of Screening Mammography . A+ppendix IV St@&? Oversight of ScreeningMammography This appendix contains more detailed information about the state over- sight programs described in chapter 4. / California has 32 inspectors, including those in the Radiological Health StaJffing of State Branch of the Department of Health Services (DHS) and those in certain Prc/grams counties. The latter are counties with which DHScontracts for inspection , and enforcement of radiation control regulations in their jurisdictions. , Inspectors must be certified radiologic technologists with supervisory experience. In addition, they must pass a competitive examination and I complete 1 year of on-the-job training prior to becoming an inspector. In Florida’s Office of Radiation Control, the Department of Health and Rehabilitative Services has 35 inspectors. The X-ray Machine Registra- tion and Control (XMRC) Section is responsible for inspecting the state’s 25,000 X-ray machines (as of July 1988) of which 463 are mam- mography machines. According to the manager of XMHC,all inspectors have the education and training necessary to oversee and regulate mam- mography equipment. They are public health physicists, which requires either 9 a bachelor’s degree with a major in radiologic health or radiologic sci- ence, and 1 year of experience in radiologic health, a physical or natural science, radiation control, X-ray technology, or health physics; or l a bachelor’s degree with a major in engineering, mathematics, or one of the physical or natural sciences, and 2 years of the experience described above. A master’s degree and/or doctorate in one of the educational areas described above can substitute for 1 year of required experience. Also, experience can be substituted for education. About one-fourth of the inspectors have a master’s degree. A background in radiology would allow a new inspector to begin work- ing independently sooner than protocol would dictate. A trainee gener- ally goes through 4 to 6 months of on-the-job training before being allowed to conduct an inspection alone. In Michigan, the Division of Radiological Health (DRH) in the Department of Public IIealth has nine inspectors, who are radiological physicists with a I3.S. degree in physics, According to the chief of the DRII Investi- gation and Compliance Section (KS), in 1988 Michigan had about 20,000 Page 69 GAO/HRD-90-32 Quality of Screening Mammography Appendix IV State Oversight of Screening Mammography X-ray machines located in 8,655 facilities. Of these, 446 were mam- mography machines located in 381 facilities. State inspectors participate in FDAtraining courses, the chief of ICStold us, and have on-the-job train- ing that includes observing an experienced inspector for a few months and then conducting inspections under direct supervision for several additional months. Usually after 6 months, new inspectors may perform inspections without direct supervision. Due to a June 1989 law requiring annual inspection of mammography equipment, the ICSchief estimates he will need three additional inspectors on his staff. In California, radiologic technologists must graduate from a state- P rsons Performing approved school of radiologic technology and pass state-approved exam- M,ammography e inations in diagnostic radiation protection and safety and diagnostic radiologic technology. There is no specific requirement for training in mammography, although an official of the Joint Review Committee on Education and Radiologic Technology said that mammography training is considered a standard part of the curriculum of accredited radiologic technology programs. A California official told us that the California examination has no questions on mammography. The Florida Office of Radiation Control issues six different licenses to technologists; mammographers must have the general radiographer license, which is the most advanced level, Generally, licenses must be renewed every 2 years. General radiographers must graduate from a radiologic technology program that meets the guidelines of the Commit- tee on Allied Health, Education, and Accreditation, an arm of the Ameri- can Medical Association. They also must pass either the national ARRT examination or Florida’s examination. The ARRTexamination covers mammography. Y Page 70 GAO/HRD-90-32 Quality of Screening Mammography Additional ResultsFrom GAO’sScreening Mammography Survey Table _-, q.1: Settings of Respondents Performlng Screening Mammography, by State / Distribution of settings, by state I CA FL ID MI Settin % No.” % NO.~ % N0.O % No.~ Hospit: I 38 __..----_-------.- 195 36 79 55 16 45 1.3 I! RadioIcky practice 28 144 25 55 7 2 - 22 58 I-. 1 I _.. -___--...-~..---.~-------_~--~__ Multqkxlalty clinic PrlrnarJ care Dhvsician .- 11 55 --.--__I-____---~-.--.._-_______-- 7 35 11 7 23 1.5 14 3 41 8 --. _--~ 14 22. 38 _- Breast lmlc 7 35 - 15 .--___________32 7 2 --.-..-~. 6 .-- . -~~. .- 16.~ HMO 6 32 1 1 ___- 0 0 1 3 HospItz I breast clinic .~-. ._-..-- _... ~...2.._.... 9 1 3 ~.~ -- - -- -..-______-______-_ 14 4 --_.-. 2.- . -~~-..-- 4 Mobile Ivan 1 6 1 3_____.__-..---0 0 1 2 Other ~ 1 3 ~~~ ~~.-.-___- 3 6 0 0 2 4 All aetiings 50 514 21 217 3 29 26 266 %espondents that reported performlng screening mammography Page 71 GAO/HRD-90-32 Quality of Screen&g Mammography I ,. .._ ~ ,.. ._~--_- Appendix V Additional Results From GAO’s Screening Mammography Survey Table V.2:.__Respondents’ .._.__...I_.. ..--. Charges for Screening Mammogram, by Locale and Setting No. of Distribution of amounts charged,’ by locale/setting (percent) Loca jejsetting respondents -~~~$1-50 .-__ $51-99 $100-125 $126-150 Over $150 All stbtes tlosp(tal 379 7 38 25 16 __-. ~. 15 .- Radi$oyy practice 255 85 48 30 98 24 45 ~____ ,14 4 g .__..._ ~~-.-~~3 Mult~(pec~elty clmic ..___- Pm ry care physman 85 0 45 31 18 7 Rrea HMC)“i91 clinic 83 17 55 l97 14 21 29 ____2g6 ____~.-~~~-.~14 2 tiosp/tal breast clmIc 19 11 37 16 - 32 5 Mot+ van 10 60 30 IO 0 0 otheri 11 0 55 27 9 .~-~~~_.--9 All stittings 954 7 41 26 15 6 Califbrnia tiosp;ta1 181 8 ..~ --___- 34 20__- ___-.-- 18 ~~- 20 Had~Jloyy practice 142. 3 46 30 16 4 -_________~--~-~- ~-~~ Multispecialty clmlc 52~ -8 15 44 29 4 Pm&y care physman 35 0-~~ ---~.. 34----. 34 17 14 Breast clinic 34 24 50 15 9 3 HMO: 10 20 10 IO ____. __-~.. ~ 40 20 t-lo&al breast clmc 9 11 44 11 33 0 Mobk varl 6 67 17 17 __--.. 0 0 Other 2 0 0-- ----..-__-100 __~ 0 .~~ ~.~ .-- 0 All settings 471 6 36 27 16 11 Florida ..__- Hospital 72 6~. 40 28 -~- __~ 14 13 Hadloloyy practice 55 11 47 25 13 4 MlJltlSpCClalty ChnlC 22 5 27 55 14 0 Prmary care physician 14 0 57 43 0 0 Breast clmc 31 13 58 26 ___-~ 0 3 HMO: 1 100 __... 0 .~ 0 0 0 Hospital beast clmic .~ 3~ 0 33 33 --___ 33 -.-----~ 0 Mobk van 2 50 50 0 0 ..._______--------- 0 Oth& 6 0 83 17 0 0 All settings 206 6 46 30 10 6 (continued) Page 72 GAO/HRD-90-32 Quality of Screening Mammography Appendix V Additional Results From GAO’s Screening Mammography Survey I No. of Distribution of amounts charaed,a by locale/setting (percent) Localei/setting respondents $1-50 $51-99 $100-125 $126-150 ---- Over -~-$150 ...-~ Idaho i ..__.-_...__. __-_-~- .__.__.-. -_____ _____- _ HospIt I 16 13 81 _______-_____-_.-~~-.- 6 0 ~~. .~ ~~~ 0 d Radloloigy practice 2 0 100 -.._. _...~ -... ~~-....--_-.-_- _______- 0 0 .~.- 0 Multlsp e clalty cllnlc 4 0 100 0 _.-._ 0 0 Prlrnard care phywian 1 0 100 0 0 0 Breast klinlc 2 100 0 0 0 0 HMO J’ 0 0 0 ---- 0 0 0 Hosplt I breast clinic 3 33 67 0 0 0 Mobile 0 0 0 0 0 0 Other Ian 0 0 0 0 0 0 All set i ings 26 16 79 4 0 0 Michigbn Hospit& Hadloldgy practice Mult&ecialty clmic Prlmar)r care physlcian 35--__.-..-_ --.-~--_-.. .._.__.- _.___~.--._-_- ___.__ 0 49 23 26 3 Breast lclwxc 16 0 69------- 19 13 0 HMO : . ~~.. -_. ~~-3---~~ .~~~...~..0 .~. -~-.- 100 ----. 0 0 0 tlospltal breast clinic 4 --.-.-0 --. ____~~ 0 25 50 25 MoblIe ‘van ~~2 50 50 -.-~.. -~-------~~ 0 .~~~ -~~~...-0~~ 0 Other 3 0 33 0 33 33 All settings 249 4 43 31 15 7 “For mammogram and interpretation. Facilities charging $0 were excluded from analysis Page 73 GAO/HRD-90-32 Quality of Screening Mammography Appendix V Additional Results From GAO’s Screening Mammography Survey Table V.3: Volume of Mammograms Performed, by Setting Distribution of no. of mammogramsa performed weekly, by setting No. of (percent) Setting respondentsb 1-24 25-49 50-74 75-100 Over 100 Hospital 383 37 34 13 9 7 Radrdlogy practice Multr:/pecialty clfnrc 251 98 31 37 27 29 23 17 , 92-- ..~-. .--~~j 10 ___-..--.. ..-_-.... Prrmary care physician 86 70 1.5 8 .._____. 7 -~~ -~~... ~~~0 Breac’t clrnrc ~~~.....~~_..__ 79 ..._ - ~~~... 15 16 ____- 27 15 27 HMO~ 4 35 6 20 17 _--.--..-~~_-~~~ 14 .-~ ~~ 43 Hospital breast clrnrc 19 11 16 16 21 37 Mob& van IO 20 20 10 20 30 Othe! 12 17 25 42 0 17 All settinos 973 34 28 17 10 11 aScreenrng and diagnostic. ‘Total number of respondents in setting category Table V.4: Survey Respondents Accredited by ACR Mammography Percent of respondents accredited Accreditation Program, by Setting and Setting CA FL ID MI All 4~.states .-~--. .~ ...-- State (As of April 1989) Hospital 2 4 0 21 8 kadiology practice ~~~--_____-. 6 11 0 -.--~~26 11 ~~~~ .~~~~-.- Multispecialty clinic 2 9 0 14 6 Primary care physician 0 ______ 0 0 5 2 Breast clinic ___- 6 ______ 9 0 38 13 HMO 0 0 ----.----- 0 .~ 0 0 Hospital breast clinic 11 0 0 0 5 Mobile van ---.. .~--- 17 0 0 _~_~...~~~ 0 ~~~~.. ~~~~ ~~~ 9 Other 0 .-__ 0 0 0 0 All settings 3 8 0 19 8 Page 74 GAO/HRD-90-32 Quality of Screening Mammography . Append& V Additional Results From GAO’s Screening Mammography Survey Table 4.5: Survey Respondents Reporting Annual Inspection by Radiological Physicist, by Setting and State I Percent and no.’ of respondents CA FL ID Ml All 4 states % No. % No. % No. % No. %__-. No. 88 195 82 79 56 16 82 119 84--...- 409 ~- 60 144 ..-....-...- .._--.-.--..-_. 46 55 0 2 64 58 57 259 --55 55 52 23 25 4 32 22 48- 104 - . .-.....--. . . -..- 43 35 47 15 0 1 42 38 43 89 .---..-.-.---..... 83 35 53 32 100 2 94 16 74 85 HMO ~ 81 32 0 1 b 100 3 81 36 Hospltaj breast clinic 89 9 67 3 75 4 100 4 85 20 MoblIe San 100 6 --- 67 3 b 100 2 91 11 Other i .----67 3 83 6 b 25 4 62 13 All setthxts 73 ~m--51~----i,iy- 217 52 29 89 288 89 1,026 aTotal number of respondents in setting category I “No respondents in category. Table V.6: Survey Respondents Reporting Annual Physicist Inspection of Beam Quality (HVL), Average Glandular Dose, and Phantom Image Quality, by Settina and State Percent and no.Oof respondents CA FL ID Ml All 4 states Setting % No. % No. % No. % No. % No. Hospital 73 195 _ 60 79 25 ___- 16 73 119 69 409 Radiology practice .48 144 33 ~~-~~~~-.--o -____-- .-______.. 2___-._ 53 __...58 46 259 Multlspccialty clinic 38 55 --i4 23 0 4 27 22 _- .-~ “-36.._- 104 Primary care physIcIan .29 35 33 focis.- 0 1 26 _._.__._38 28 89 Breast &nlc 66 35 ~41 32 0 2-_.--~._~----.--. 81 16 58 85 HMO 66 32 0 1 b 100 3 67 36 Hospital breast clinic 89 9 33 3 50 ___-_4 100 ___.. 4 __ 75 20 Mobk Gan 67 6 33 3 b 100 2 64 11 Other 67 3 67 6 b 0 ..~~ 4 46 13 All settings 59 514 ~46 217 21 29 59 268 55 1,026 “Total number of respondents in setting category “No respondents In category. Page 75 GAO/HRD-90-32 Quality of Screening Mammography 0 I - --_- Appendix V Additional Results From GAO’s Screening Mammography Survey Table ,V.7: _........ Survey ..___. .I.-_ Respondents Reporting Compliance With Selected Quality Assurance Standardb, by Setting I Percent and no.’ of respondents Second Daily Semianvual Monitoring of reading of Fora;;;; of processor phan;;;$nage repeat mammo ram in 8en8itometryb mammogram8 faci Bity biopsies settine % No. -__ % No. % No. % No. % No. Hosy,&il 52 361 45 409 54 398 37 398 82 398 led&y practice 16 226 46 259 51 257 22 257 71 257 Multlsdzclalty cllnlc 22 95 42 -___ 104 50 101 21 101 70 101 Fknar\/ cart physIcIan ..~~. -. 10 ~~..-~. -~. .-_ 79 ~. .28 - ..~ 89 44 85 20 85 695 __- Brcas cllnlc 37 76 65 85 62 84 33 84 75 84 tHMO 1 . ..-~~ -~ . ..--.59 - ~~~ 32 44 - --.---- 36 67 36 44 36 75 36 Hospi al breast clmlc 39 18 70 20 55 20 25 20 85 20 : Moh~lo van 46 11 73 11 64 11 27 11 82 11 Other I 33 12 ~~~ 54 ~._~~~~______ 13 62 13 23 13 54 13 All settings 35 91oc 46 1,026c 53d 1,005c 29 1,005c 76 1,005c “Total number of respondents in setting category. ‘Analysis excludes respondents who do xeromammography only ‘Numbers of total cases differ because of different numbers that responded to specific questions dTotal differs by 1% from percentage in report because latter IS based on all cases, including those who did not identify setting. Table V.9: Survey Respondents’ Retention of Original Mammographic Percent of respondents Images, by Setting Images kept at Images kept Setting least 5 years over 10 years No.” Hospital 98 49 ____...- 394 Radioloav practice 96 50- 252 Multispecialty clinic ~~~ -__... 98 34 101 Primary care physician -. - ____ 99 ____~ 56 .---__ 86 Breast clinic 99 ____--..-__ 63 83 HMO -.- __-.- 100 42 _--33 Hospital~breast clinic 100 ____. 55 20 Mobile van 91 55 11 Other 100 27 11 All settings 98 49 991 “Total number of respondents in setting category. Page 76 GAO/HRD-90-32 Quality of Screening Mammography Appendix I VI - M#jor Contributors to This Report Issues, (202) 276-545 1 Mark V. Nadel, Associate Director for National Public Health Wadhington, DC. Issues Edwin P. Stropko, Assistant Director / Rodney E. Ragan, Assignment Manager Helene F. Toiv, Evaluator-in-Charge Susan L. Sullivan, Technical Adviser Cynthia L. Booth, University Fellow Detkoit Regional Offf Sharon L. Fucinari, Computer Analyst Gary W. Ulrich, Regional Assignment Manager San’ Francisco Regional Office ( IOHWll) Page 77 GAO/HRD-90-32 Quality of Screening Mammography 6 c
Screening Mammography: Low-Cost Services Do Not Compromise Quality
Published by the Government Accountability Office on 1990-01-10.
Below is a raw (and likely hideous) rendition of the original report. (PDF)