Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
PART III The IOM Clinica Workshops Condition
The lOM Condition Workshops: Introduction Kenneth I. Shine The individuals who participated in the IOM clinical condition work- shops on breast cancer, hip fracture, and acute myocardial infarction believe that the Health Care Financing Administration (HCFA) has taken a very positive step in seeking to make its vast administrative data bases available for effectiveness research. They recognize, however, that HCFA has taken only a first step in what will be a very long journey. Many participants- and all of them on this panel- come from academic medicine and from a tradition of rigorous science and analysis. Not surprisingly, they caution prudence in the use of such data because of a realistic awareness of pitfalls to be avoided and obstacles to be surmounted. The following three papers reflect this prudent caution. In breast cancer, for example, treatment often involves in-hospital surgery and subsequent radiation therapy or chemotherapy administered in an outpatient setting. However, existing Medicare data are primarily hospital-based, and HCFA does not yet have good ambulatory data. Evaluating mammography for its effectiveness in breast cancer screening and diagnostic uses, to take another example, will require long-term follow-up data, not simply data on acute care encounters between the patient and a provider. On the other hand, Medicare data can help immeasurably in focusing attention on the similari- ties and differences between Medicare-age women and younger women. Valerie D. Jackson has a special research interest in breast imaging. She brings that expertise to bear on diagnosing and treating breast cancer in the elderly. Hip fracture was approached as a relatively straightforward clinical problem. A single bone is involved, diagnosis is clear and consistent across different practitioners, and surgery is the recommended intervention, followed by rehabilitation. Yet as the committee delved into the issues, we realized how 51
52 EFFECTIVENESS A]VD OUTCOMES IN HEALTH CARE complicated it was to assess the effectiveness of prevention, of different surgical and medical interventions, and of rehabilitation programs. Preven- tion, for instance, requires that effective efforts begin long before an individual reaches the age of Medicare eligibility. Research along these lines, therefore, must link Medicare data bases to Medicaid and private insurance data bases for a younger population, linkages that span federal and state as well as public and private boundaries. In addition, different sites of care are required for treating hip fracture, and unexplained geographic differences exist in rates of fracture. David G. Murray, an orthopedic surgeon, examines these and other effectiveness issues related to treating hip fracture. One of the major issues confronted in the clinical workshops was how to make the best use of administrative data bases for effectiveness and outcomes research. Barbara J. McNeil, a radiologist and investigator with experience in using large data bases, addresses the opportunities and the limits of using claims data in acute myocardial infarction and other conditions.
9 Breast Cancer Valerie P. Jackson Breast cancer, the second leading cause of cancer death in American women, is a major health problem for women in the Medicare age group because its incidence increases with advancing age. Currently, the Ameri- can Cancer Society estimates that 1 in 10 American women will be affected by this devastating and highly emotional disease during her lifetime. Sev- eral studies have shown that screening mammography can detect breast cancer at a more favorable stage, resulting in improved prognosis for screened women found to have breast cancer (1-10~. THE PROBLEM OF POOR COMPLIANCE The American Cancer Society and the American College of Radiology have recently been joined by most of the medical groups in this country in recommending the following guidelines for screening mammography: . Baseline mammogram by age 40 · Mammogram every one to two years between ages 40 and 49 . Yearly mammograms after age 50 The overall mortality statistics for breast cancer in the United States have changed little in several decades, however, largely because of poor compli- ance with screening mammography guidelines. Thus, screening mammography has proven efficacy, but its effectiveness is diminished because it is underutilized. Although we can detect tumors as small as 3 to 5 millimeters with mammography (Figure 1), we continue to see too many large, clinically obvious carcino- mas that carry a poor prognosis (Figure 2~. Noncompliance is a problem for all groups of women, but it is particularly prevalent in our elderly and indigent populations. 53
54 EFFECTIVENESS AND OUTCOMES IN HEALTH CARE FIGURE 1 Oblique left mammogram of an asymptomatic 60-year-old woman. There is a 4 millimeter slightly irregular mass in the upper portion of the breast (arrow), which was found to be a small invasive ductal carcinoma at surgery. She has had no evidence of spread to the axillary lymph nodes or elsewhere in her body in one year of follow-up, and her prognosis is excellent. As shown in Table 1, there are a number of potential reasons for poor compliance. Cost is a major factor, particularly for elderly women on fixed incomes. Many mammographers are working to decrease the cost of screening mammography to approximately $50. If Medicare paid for screening (as opposed to just diagnostic) mammograms, cost might no longer be a deter- rent and compliance would improve.
IOM CLINICAL CONDITION WORKSHOPS 55 FIGURE 2 Oblique left mammogram of a 60-year-old woman with bloody nipple discharge and a large, hard, palpable mass behind the nipple. The mammogram demonstrates a 6 centimeter mass, found to be invasive ductal carcinoma with posi tive axillary lymph nodes at surgery. She died 6 months after mastectomy. Vigorous compression is necessary in order to minimize radiation dose and maximize image quality. Many women worry about pain from this compression, and pain has been cited as a possible deterrent to mammography. However, compression is adequately tolerated by the majority of women (11) and is unlikely to be a significant factor in compliance. Many women have difficulty finding the time to get a mammogram. There are probably many psychological reasons for this, but we need to
56 EFFECTIVENESS AND OUTCOMES IN HEALTH CARE TABLE 1 Noncompliance with Screening Mammography Guidelines Potential Factor Potential Solution 1. Cost of mammograms 2. Pain from compression 3. Time 4. Radiation exposure 5. Interpretation errors 6. Fear of finding cancer Medicare payment for screening Lower cost of mammograms Education about advantages of compression Compassionate technologist improved access to screening mammography facilities Less time necessary for examination State-of-the-art equipment, appropriately used, monitored, and maintained Patient arid physician education Consistently high-quality mammograms Greater mammographer experience and education Improved diagnostic criteria Adjunctive use of ultrasound in selected cases Patient education make mammography facilities more accessible for patients and minimize the time necessary for the examination. Exposure to radiation was a serious consideration in previous years (12~; however, mammography equipment and film systems have improved mark- edly, and the radiation dose from properly performed mammography is so small that radiation exposure is no longer a problem (131. It has been estimated that, for women over age 50, the risk of having yearly mammograms is one-tenth the risk of early death caused by failure to diagnose breast cancer by screening mammography (14~. Both women and their physicians are worried about interpretation errors (that is, false negative and false positive studies). Unfortunately, there is overlap in the appearances of benign and malignant processes, necessitating biopsy or mammographic follow-up for differentiation. In spite of the extensive experience with mammography in this country, there are some cancers that are missed, either by negligence or because they are not mammographically visible, even with a good quality film. Although education and careful attention to technique and quality control will minimize these unfortunate occurrences, it is unlikely that we can completely eliminate false negative mammograms in the foreseeable future. Breast cancer is a very emotional disease, and many women are afraid of having breast cancer discovered. This fear may paralyze a woman to the point where she will not undergo screening or diagnostic evaluations. Increased
IOM CLINICAL CONDITION WORKSlIOPS 57 patient education regarding the benefits of early detection and treatment should be targeted at specific geographic or socioeconomic groups, or both. CONTROVERSIES SCREENING A number of controversies surround breast cancer screening. First, who should be screened and at what intervals? The American Cancer Society guidelines represent our "best guess" at appropriate intervals for women over age 35. However, these are likely to be modified as our knowledge of the biology of breast cancer increases. Currently, the majority of controversy surrounds screening for women age 40 to 49 (15,16~. Although the efficacy of screening mammography for women over age 50 is well established, we do not know at what age screening should stop. Obviously, this will depend on the patient's physiological status and whether she would benefit by having a small, potentially curable, cancer detected, in light of her other disease processes. Is it effective to screen all women over the age of 35 or only those at high risk? Unfortunately, most women with breast cancer do not have identifiable risk factors, other than the fact that they are women who are getting older. Thus, targeting specific "high-risk" groups for screening is of limited value. Why is screening mammography underutilized, and how can compliance be improved? These are major issues in determining the effectiveness of breast cancer screening in this country. Current investigations are studying the reasons women do not go for mammograms and the reasons their physicians do not order them. In the future, we must define and test interventions that will improve compliance and improve our overall breast cancer mortality statistics. TREATMENT A number of controversies surround treatment of breast cancer as well. Debates rage over appropriate surgical approaches to various types of breast cancer. In the past, the standard surgical treatment was a mastectomy (generally, a modified radical mastectomy), which obviously left the woman without a breast. In recent years, many surgeons have offered some women a less mutilating approach: segmental resection with an axillary node dissection, usually followed by radiation therapy. Large randomized controlled trials have shown that these two treatments result in equal prognoses for most women (17,18~. Unfortunately, many women, particularly elderly women, may not be given any choice of surgical therapy (191.
58 EFFECTIVENESS AND OUTCOMES IN HEALTlI CARE Chemotherapy and hormonal therapy are relatively new interventions for women with breast cancer and have improved the prognosis for selected patients. Until recently, chemotherapy was generally reserved for women with positive axillary lymph nodes at the time of surgery or for very young women with a generally poor prognosis. However, recent studies (20-24) have led the National Cancer Institute (NCI) to recommend that all breast cancer patients, even those with node-negative tumors, have chemotherapy or hormonal therapy, depending upon their age and a number of other tumor factors. This has provoked considerable controversy among oncologists and may be a deterrent to compliance with screening mammography. In the past, we were able to tell women that if they had early cancer detected by screening, they probably would not require chemotherapy, with its unpleasant side effects. If the NCI recommendations are followed, women may feel there is no advantage to early detection. We need to find new tests to identify subgroups of women with node-negative breast cancer who are most likely to benefit from chemotherapy. THE ROLE OF EFFECTIVENESS STUDIES IN BREAST CANCER There are many long-term considerations for women with breast cancer. For example, what are the appropriate methods and intervals for follow-up of women with cancer? What are the psychological needs of these patients, particularly of elderly women? What are their reconstructive surgery op- tions? Most important, what therapies improve quality of life as well as mortality statistics? Surprisingly, these areas have received relatively little attention in the past. Effectiveness studies may provide answers to these questions. Initial breast cancer effectiveness studies should involve mammography. For optimal studies, we need much more data than are currently available from the Medicare data bases. Because Medicare currently reimburses only for diagnostic mammograms (meaning that the patient has some sort of a problem, such as a palpable lump), we do not have data on screening mammograms done on asymptomatic women. If Medicare pays for screen- ing mammography, it will be a golden opportunity to study utilization and effectiveness of breast cancer screening; however, we must accurately de- termine and record the reason for the mammography (that is, screening vs. diagnostic examination). We need to track the interpretations and outcomes for all women who have mammograms. Some mammograms are going to be interpreted as "negative" for cancer; others are going to be called "posi- tive." Of those women with negative mammograms, we must find out how many subsequently go on to have a breast cancer that was missed on th mammogram (false negative mammogram). Tracking this information may require years. Of women who have a positive mammogram (where a lesion
IOM CLINICAL CONDITION WORKSlIOPS 59 is called suspicious), we must find out how many women actually have cancer (true positive rate), how many have a benign lesion (false positive rate), and how many women do not undergo further evaluation. Occasion- ally a mammogram is interpreted as abnormal, but the patient and her phy- sician are never notified or they choose not to do anything about it. It is also crucial that we identify the temporal relationship between mammography and biopsy. For example, if a woman has a mammogram in January 1989 and has a biopsy positive for cancer in December 1989, the study and the surgery may have very little relationship. We must know if the mammogram prompted the biopsy and the length of time between the mammogram and the surgery. Ideally, we should record inpatient and out- patient diagnosis, treatment, and follow-up data on all of these women. It is also very important that we have standardized terminology and recording of data on tumor stage, cell type, and hormone receptor status. All of these factors must be analyzed in order to obtain accurate data about the effec- tiveness of breast cancer screening in the United States. REFERENCES 1. Shapiro, S. Evidence on Screening for Breast Cancer from a Randomized Trial. Cancer 39:2772-2782, 1977. 2. Shapiro, S., Venet, W., Strax, P., et al. Ten- to Fourteen-Year Effect of Screening on Breast Cancer Mortality. Journal of the National Cancer Institute 69:349-355, 1982. 3. Shapiro, S., Venet, W., Strax, P., et al. Selection, Follow-up, and Analysis in the Health Insurance Plan Study: A Randomized Trial with Breast Cancer Screen- ing. NCI Monograph 67:65-74, 1985. 4. Baker, L.H. Breast Cancer Detection Demonstration Project: Five-Year Sum- mary Report. Cancer 32:194-225, 1982. 5. Seidman, H., Gelb, S.K., Silverberg, E., et al. Survival Experience in the Breast Cancer Detection Demonstration Project. Cancer 37:258-290, 1987. 6. Tabar, L., Fagerberg, C.J., Gad, A., et al. Reduction in Mortality from Breast Cancer After Mass Screening with Mammography: Randomised Trial from the Breast Cancer Screening Working Group of the Swedish National Board of Health and Welfare. Lancet 1 :829-832, 1985. 7. Verbeek, A.L., Hendriks, J.H., and Holland, R. Reduction of Breast Cancer Mortality Through Mass Screening with Modern Mammography: First Results of the Nijmegen Project, 1975-1981. Lancet 1 :1222-1224, 1984. 8. Collette, H.J.A., Day, N.E., and Rombach, J.J. Evaluation of Screening for Breast Cancer in a Non-Randomised Study (the DOM Project) by Means of a Case- Control Study. Lancet 1:1224-1226, 1984. 9. Palli, D., DelTurco, M.R., Buiatti, E., et al. A Case-Control Study of the Efficacy of a Non-Randomized Breast Cancer Screening Program in Florence (Italy). International Journal of Cancer 38:501-504, 1986. 10. Andersson, I., Aspegren, K., Janzon, I., et al. Mammographic Screening and
60 EFFECTIVENESS AND OUTCOMES IN HEALTH CARE Mortality from Breast Cancer: The Malmo Mammographic Screening Trial. British Medical Journal 297:943-948, 1988. 11. Jackson, V.P., Lex, A.M., and Smith, D.J. Patient Discomfort During Screen- Film Mammography. Radiology 168:421-423, 1988. 12. Bailar, J.C. Mammography: A Contrary View. Annals of Internal Medicine 84:77-84, 1976. 13. Feig, S.A. Radiation Risk from Mammography. Is it Clinically Significant? American Journal of Roentgenology 143:469-475, 1984. 14. Ritenour, E.R. and Hendee, W.R. Screening Mammography. A Risk Versus Risk Decision. Investigative Radiology 24:17- 19, 1989. 15. Eddy, D.M., Hasselblad, V., McGivney, W., et al. The Value of Mammography Screening in Women Under Age 50 Years. Journal of the American Medical Asso- ciation 259:1512-1519, 1988. 16. Moskowitz, M. Breast Cancer Screening: All's Well That Ends Well, or Much Ado About Nothing? American Journal of Roentgenology 151: 659-665, 1988. 17. Fisher, B., Bauer, M., Margolese, R., et al. Five-Year Results of a Randomized Clinical Trial Comparing Total Mastectomy and Segmental Mastectomy With or Without Radiation in the Treatment of Breast Cancer. New England Journal of Medicine 312:665 -673, 1985. 18. Veronesi, U., Saccozzi, R., del Vecchio, M., et al. Comparing Radical Mastectomy with Quadrantectomy, Axillary Dissection, and Radiotherapy in Pa- tients with Small Cancers of the Breast. New England Journal of Medicine 305:6-11, 1981. 19. Greenfield, S., Bianco, D.M., Elashoff, R.M., et al. Patterns of Care Related to Age of Breast Cancer Patients. Journal of the American Medical Association 257:2766-2770, 1987. 20. The Ludwig Breast Cancer Study Group. Prolonged Disease-Free Survival After One Course of Perioperative Adjuvant Chemotherapy for Node-Negative Breast Cancer. New England Journal of Medicine 320:491-496, 1989. 21. Mansour, E.G., Gray, R., Shatila, A.H., et al. Efficacy of Adjuv ant Chemo~erapy in High-Risk Node-Negative Breast Cancer. New England Journal of Medicine 320:485-490, 1989. 22. Fisher, B., Costantino, J., Redmond, C., et al. A Randomized Clinical Trial Evaluating Tamoxifen in the Treatment of Patients with Node-Negative Breast Cancer who have Estrogen-Receptor-Positive Tumors. New England Journal of Medicine 320:479-484, 1989. 23. Fisher, B., Redmond, C., Dimitrov, N.V., et al. A Rar~domized Clinical Trial Evaluating Sequential Methotrexate and Fluorouracil in the Treatment of Patients with Node-Negative Breast Cancer who have Estrogen-Receptor-Negative Tumors. New England Journal of Medicine 320:473-478, 1989. 24. Early Breast Cancer Trialists' Collaborative Group. Effects of Adjuvant Tamoxifen and of Cytotoxic Therapy on Mortality in Early Breast Cancer: An Overview of 61 Randomized Trials Among 28,896 Women. New England Journal of Medicine 319:1681-1692, 1988.