|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2007.12.8058 on April 21 2008 © 2008 American Society of Clinical Oncology. Mammography Before Diagnosis Among Women Age 80 Years and Older With Breast Cancer
From the Departments of Surgical Oncology and Breast Medical Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, TX Corresponding author: Gildy Babiera, MD, Department of Surgical Oncology, Unit 444, The University of Texas M.D. Anderson Cancer Center, 1400 Holcombe Blvd, PO Box 301402, Houston, TX 77030-1402; e-mail: gvbabiera{at}mdanderson.org
Purpose Screening mammography guidelines for patients age 80 years and older are variable. We determined the effect of mammography use on stage at breast cancer diagnosis and survival among women of this age range.
Patients and Methods We used the linked Surveillance, Epidemiology, and End Results–Medicare database to evaluate 12,358 women Results Percentages of women with nonuse, irregular use, and regular use of mammography during the 5 years preceding diagnosis were 49%, 29%, and 22%, respectively. On multivariate analysis, patients were 0.37 times less likely to present with late-stage cancer for each mammogram obtained (odds ratio, 0.63; 95% CI, 0.63 to 0.67). Breast cancer–specific 5-year survival among nonusers was 82%, that among irregular users was 88%, and that among regular users was 94%. However, survival from causes other than breast cancer was also associated with mammography use, suggesting a bias for healthier patients to undergo mammography.
Conclusion Regular mammography among women
The benefit from use of screening mammography for patients age 50 to 69 years is commonly accepted. That mammography can save lives is clear from studies documenting earlier detection and improved survival.1 However, the eight prospective, randomized trials conducted to date providing this evidence excluded patients older than 74 years from participation. In the one trial that included patients age 70 to 74 years, the benefit for screening mammography lacked statistical significance on subgroup analysis.2 The exclusion of older patients from these studies does not fit the changing demographics of the United States. Beginning in 2011, the baby boomer generation (those born from 1946 through 1964) will reach age 65 years, sharply increasing this segment of the population.3 Among those older than 65 years, the proportions of people older than 80 years have already increased, increasing by 26% for ages 80 to 84 years and by 38% for ages 85 years and older for the years 1990 to 2000.3 Octogenarians are living longer and healthier lives, with a mean life expectancy of approximately 7 years from age 85 years.4
Physicians attempting to follow health maintenance recommendations for this expanding population have several guidelines from which to choose. The American Cancer Society recommends annual mammography with no upper age limit for women in good health.5 The United States Preventive Services Task Force recommends annual to biennial mammography if life expectancy is not compromised by comorbid disease.1 The American Geriatrics Society guidelines for those older than 75 years are biennial or triennial mammography if life expectancy is more than 4 years.6 The reported percentages of older women undergoing screening mammography vary widely. Although self-reported data systems have traditionally displayed high levels of screening (70% to 80%), population-based studies suggest that approximately one half of women
For patients older than 80 years, exclusion from large prospective randomized trials and variability among consensus recommendations have led to confusion regarding the optimal screening strategy for mammography. Therefore, the purpose of this study was to determine mammography use among women
Study Population The study population consisted of women in the Surveillance, Epidemiology, and End Results (SEER) –Medicare linked database who were 80 years and older at diagnosis of incident breast cancer during 1996 to 2002. Because we wanted to measure mammography use during the 5 years before diagnosis, we limited the study to women who were enrolled in Medicare Part B on the month of their diagnosis and for the 5 previous years and were not members of a health maintenance organization during that time. According to the SEER-Medicare database, 148,023 women were diagnosed with breast cancer from January 1996 to December 2002. The following women were excluded from our final study population: 2,936 women with a previous cancer diagnosis, 2,562 women without histologic confirmation of breast cancer, 328 women with unknown month of diagnosis, 9,801 women with a second cancer diagnosis, 107,359 women who were younger than 80 years at diagnosis, 6,134 women for whom breast cancer stage information was missing or stage 0, 249 women who were not covered by Medicare B for 5 years before the diagnosis, and 5,392 members of health maintenance organizations. The remaining 13,262 women were included in this analysis. To identify chemotherapy within 6 months of diagnosis, 904 observations with incomplete Medicare coverage were further excluded, and 12,358 patients were included in the final model.
Measures
Demographic information includes year of diagnosis (1996 to 2002), age at diagnosis (80 to 84, 85 to 89, and Disease stage at diagnosis was assigned according to the 2002 staging system of the American Joint Committee on Cancer14 and categorized as early stage (I to IIa) or late stage (IIb to IV) for multivariate analysis. All-cause mortality, breast cancer–specific mortality (with cause of death identified from ICD-8 or –9 codes as 1749 or ICD-10 codes as C509), and mortality from other diseases were determined in the Patient Entitlement and Diagnosis Summary File.
Statistical Analyses
For women with stage I to III breast cancer, we used a multivariate analysis of variance to examine the effect of mammography use on tumor size at diagnosis. The covariates were the same as in the logistic model. Least squares mean of tumor size were estimated for each cohort by using a general linear model procedure.16,17 Cox models were built to determine the effect of mammography screening on survival probability among age
The characteristics of the study sample (Table 1) were stratified according to use of mammography in the 5 years preceding diagnosis. Regular mammography use (  three mammograms in 5 years) increased with time, whereas nonuse (no mammograms in 5 years) declined. The proportion of patients with irregular use (one to two mammograms over 5 years) remained relatively stable throughout the study period. Percentages of women who had regular mammograms were 27% for those age 80 to 84 years, 17% for those age 85 to 89 years, and 8% for women older than 90 years. Stratified according to race/ethnicity, regular mammography use was found in 23% of the white population, 12% of the black population, and 16% among those remaining ("other"). Regular mammography use was also associated with higher educational status, being married, and having a higher income level.
Tumor characteristics and treatments are also listed in Table 1. Regular users had fewer positive lymph nodes at surgery, whereas nonuse was associated with larger numbers of positive nodes. Tumor grade and estrogen receptor status also demonstrate an association with mammography use. Receipt of breast conservation surgery and radiotherapy were both associated with more frequent mammograms. Finally, nonusers had higher comorbidity scores, indicating poorer baseline health than regular users. Figure 1 illustrates mean tumor size in relation to use of mammography for women with stage I to III breast cancer. Mean tumor size for nonusers was 5.22 cm (95% CI, 4.57 to 5.86 cm); that for irregular users was 3.40 cm (95% CI, 2.70 to 4.10 cm), and that for regular users was 2.90 cm (95% CI, 2.15 to 3.65 cm). After adjusting for diagnosis year, age, race/ethnicity, SEER region, marital status, income, education level, and comorbidity score, mammography use remained a significant factor in association with mean tumor size at presentation. Figure 2 displays disease stage according to use of mammography during the 5 years before diagnosis.14 Regular users of mammography had larger percentages of stage I breast cancer than did either irregular users or nonusers. The percentages of stage II, III, and IV breast cancer were largest in the nonuse and irregular-use groups.
We classified early-stage disease as stage I to IIa and late-stage disease as stage IIb to IV, similar to other investigators.10,18 Results of multivariate logistic regression for the probability of late-stage disease are listed in Table 2, with mammography use displayed as a continuous variable. Each mammogram obtained resulted in a 37% decrease in risk of late-stage cancer (odds ratio, 0.63; 95% CI, 0.63 to 0.67). Stage of presentation was adjusted for SEER region. Late-stage disease was more common in the older age groups and among the black patient group. Geographic areas with the highest percentages of people with less than 12 years of education demonstrated the greatest amounts of late-stage disease. Finally, a higher Charlson comorbidity score was correlated with a higher probability of advanced (late-stage) cancer.
Table 3 lists 5-year overall, breast-cancer specific, and non–breast cancer–related survival according to mammography use. Statistically significant improvements in overall and breast-cancer specific survival were associated with increasing mammography use. However, similar improvements were also noted for non–breast cancer–related survival, reflecting the potential bias for healthier patients to obtain screening mammography. Figure A1 (online only) displays the survival curves stratified for mammography use and further demonstrates the effect of mammography on survival, which is not limited to breast cancer–specific survival. In an attempt to account for the confounding effect of comorbidity in this analysis, we used a multivariate analysis of effect of mammography use on survival (Table 4). The effect of mammography on all-cause mortality adjusted for comorbidity (hazard ratio = 0.87; 95% CI, 0.85 to 0.90) was similar to that not adjusted for comorbidity (hazard ratio = 0.88; 95% CI, 0.85 to 0.90). The effect for breast cancer–specific mortality and breast cancer–unrelated mortality were also unchanged when adjusted for comorbidity. To determine whether certain comorbidities negated the survival effect from mammography, we analyzed the effect of mammography stratified for specific comorbid conditions. The effect of mammography on breast cancer–specific mortality was decreased for patients with a Charlson comorbidity score of 2 or more. However, for many of the specific conditions, including myocardial infarction, congestive heart failure, peripheral vascular disease, chronic obstructive pulmonary disease, diabetes, peptic ulcer disease, and rheumatologic disease, the effect of mammography on overall and breast cancer–specific survival did not change (data not shown).
In this study, we found that use of mammography by women 80 years or older was associated with earlier disease stage at presentation of breast cancer. The most benefit was seen for regular users of mammography, thus supporting an interval of annual to biennial screening. On multivariate analysis, race/ethnicity, age, education level, income level, comorbidity score, and regular mammography use were associated with early diagnosis of breast cancer. Improvement was shown for breast cancer–specific survival but also for breast cancer–unrelated survival, an effect that we were unable to separate based on comorbidity.
The major strength of this study is that it aids in the understanding of recommendations for screening mammography for women
Our findings are consistent with those of other reports demonstrating improvements in stage presentation for older women undergoing routine mammography, although none have focused specifically on individuals Our study demonstrates some of the effects of racial disparities in health care. Black race and the SEER race category "other" show a decrease in mammography use and an increase in advanced stage of disease. The effect of race on stage of presentation persists even when adjusted for mammography use. This is not a novel finding and has been well documented by other investigators. A recent SEER-Medicare analysis found mammography use accounted for 12% to 30% of the excess late-stage disease among black women.19 Another important aspect of screening for older women is cost effectiveness. Although not addressed in our analysis, studies addressing cost effectiveness suggest continuing screening for women without significant comorbid conditions.20,21 This similarity between the cost effectiveness recommendations and clinical recommendations allow for easy integration but again stress the importance of incorporating comorbidity into the decision-making process. Another factor for screening in older patients involves multiple previous attempts to determine whether breast cancer in older women is more indolent compared with that of younger patients.22-24 Our study demonstrates a significant percentage of estrogen receptor–positive tumors and well-differentiated to moderately differentiated tumors. In addition, our mean tumor size, even among regular users, was larger than might be expected. However, we have no data to support that a less virulent histology is present in our older population. We have attempted to examine cancer mortality as a consequence of screening mammography with caution because of the clear relationship between comorbidity/mortality and use of mammography. Specifically, this selection bias reflects the tendency for healthier patients, who have longer life expectancies, to undergo screening mammography. In addition, comparisons of SEER areas with the total United States population suggest that caution is needed in extrapolating cancer mortality across non-SEER areas. An overview of SEER-Medicare data found the United States cancer mortality rate among older persons to be 9% higher than that among older persons who live in SEER areas.25 We have also included breast cancer stage and tumor size because of the well described association with survival, although these end points may be subject to the same bias as SEER cancer mortality.26,27 Several limitations are worth noting regarding the use of Medicare claims data to assess mammographic screening. The first is the difficulty in distinguishing screening from diagnostic mammography, in that screening may be converted to diagnostic if abnormalities are detected on the screening mammogram.28 This underestimation of screening may contribute to the low rates of screening found in our population. To address this concern, we used an algorithm similar to that of previous reports that excludes mammograms obtained within the 2 months before diagnosis and also repeat mammograms within 11 months.10 Second, not accounting for patients who pay for screening with sources other than Medicare or attend free programs would also underestimate mammography use. This may not represent a significant limitation, however, as Medicare is the primary health insurer for the vast majority of the older adult population, and we excluded patients who belonged to health maintenance organizations. Third, SEER-Medicare data for the older patient population tend to include more urban and more affluent individuals and fewer white individuals compared with the total United States population.25 Despite these limitations, the SEER-Medicare database provides data on a large and geographically diverse population that can answer important questions pertaining to screening in the older adult population, especially in lieu of the exclusion of older adult populations from large controlled trials.29,30 Our findings add to the accumulating evidence that the use of regular mammography may be beneficial for older women. This population-based analysis demonstrated that mammography use was associated with improved (earlier) breast cancer stage at diagnosis in older women, with no upper age limit to indicate when to stop recommending mammography. However, the true efficacy of screening with respect to survival is still unknown and will remain so until a clinical trial of screening mammography is performed in this age group, which is unlikely. Health care providers should consider discussing the potential benefits of screening mammography with their older patients, particularly for those without significant comorbidity.
The author(s) indicated no potential conflicts of interest.
Conception and design: Brian D. Badgwell, Sharon H. Giordano, Gildy Babiera Financial support: Sharon H. Giordano Administrative support: Sharon H. Giordano Provision of study materials or patients: Sharon H. Giordano Collection and assembly of data: Brian D. Badgwell, Sharon H. Giordano, Zhigang Z. Duan, Shenying Fang Data analysis and interpretation: Brian D. Badgwell, Sharon H. Giordano, Zhigang Z. Duan, Shenying Fang, Gildy Babiera Manuscript writing: Brian D. Badgwell, Sharon H. Giordano, Henry M. Kuerer, Gabriel N. Hortobagyi, Gildy Babiera Final approval of manuscript: Brian D. Badgwell, Sharon H. Giordano, Isabelle Bedrosian, Henry M. Kuerer, S. Eva Singletary, Kelly K. Hunt, Gabriel N. Hortobagyi, Gildy Babiera
published online ahead of print at www.jco.org on June 1-5, 2007. Presented in part at the 43rd Annual Meeting of the American Society of Clinical Oncology, June 1-5, 2007, Chicago, IL. Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Humphrey LL, Helfand M, Chan BK, et al: Breast cancer screening: A summary of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med 137:347-360, 2002 2. Tabár L, Vitak B, Chen HH, et al: The Swedish Two-County Trial twenty years later: Updated mortality results and new insights from long-term follow-up. Radiol Clin North Am 38:625-651, 2000[CrossRef][Medline] 3. Hetzel L, Smith A: The 65 Years and Over Population: 2000 Washington, DC, Census 2000 Brief, C2KBR/01-10, United States Census Bureau. www.census.gov/prod/2001pubs/c2kbr01-10.pdf, accessed 2/15/07 4. Federal Interagency Forum on Aging-Related Statistics: Older Americans Update 2006: Key Indicators of Well-Being. Washington, DC, United States Government Printing Office, 2006. www.agingstats.gov/update2006/Health_Status.pdf, accessed 2/15/07 5. Smith RA, Saslow D, Sawyer KA, et al: American Cancer Society guidelines for breast cancer screening: Update 2003. CA Cancer J Clin 53:141-169, 2003 6. Breast cancer screening in older women: American Geriatrics Society Clinical Practice Committee. J Am Geriatr Soc 48:842-844, 2000[Medline] 7. Kagay CR, Quale C, Smith-Bindman R: Screening mammography in the American elderly. Am J Prev Med 31:142-149, 2006[CrossRef][Medline] 8. Breen N, Wagener DK, Brown ML, et al: Progress in cancer screening over a decade: Results of cancer screening from the 1987, 1992, and 1998 National Health Interview Surveys. J Natl Cancer Inst 93:1704-1713, 2001 9. Blackman DK, Bennett EM, Miller DS: Trends in self-reported use of mammograms (1989-1997) and Papanicolaou tests (1991-1997): Behavioral Risk Factor Surveillance System. MMWR CDC Surveill Summ 48:1-22, 1999[Medline] 10. Randolph WM, Goodwin JS, Mahnken JD, et al: Regular mammography use is associated with elimination of age-related disparities in size and stage of breast cancer at diagnosis. Ann Intern Med 137:783-790, 2002 11. Charlson ME, Pompei P, Ales KL, et al: A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis 40:373-383, 1987[CrossRef][Medline] 12. Romano PS, Roos LL, Jollis JG: Adapting a clinical comorbidity index for use with ICD-9-CM administrative data: Differing perspectives. J Clin Epidemiol 46:1075-1079, 1993[CrossRef][Medline] 13. National Cancer Institute: SEER-Medicare: Calculation of Comorbidity Weights. Http://healthservices.cancer.gov/seermedicare/program/comorbidity.html, accessed 11/16/06 14. Greene FL, Page DL, Fleming ID, et al: AJCC Cancer Staging Manual (ed 6). New York, NY, Spring-Verlag, 2002 15. Stokes M, Davis C, Koch G: Categorical Data Using the SAS System. Cary, NC, SAS Institute Inc, 2000 16. Draper NR, Smith H: Applied Regression Analysis (ed 3). New York, NY, J Wiley, 1998 17. SAS Institute: SAS/STAT User's Guide, Version 8. Cary, NC, SAS Institute, 1999 18. McCarthy EP, Burns RB, Freund KM, et al: Mammography use, breast cancer stage at diagnosis, and survival among older women. J Am Geriatr Soc 48:1226-1233, 2000[Medline] 19. McCarthy EP, Burns RB, Coughlin SS, et al: Mammography use helps to explain differences in breast cancer stage at diagnosis between older black and white women. Ann Intern Med 128:729-736, 1998 20. Mandelblatt JS, Schechter CB, Yabroff KR, et al: Toward optimal screening strategies for older women: Costs, benefits, and harms of breast cancer screening by age, biology, and health status. J Gen Intern Med 20:487-496, 2005[CrossRef][Medline] 21. Mandelblatt J, Saha S, Teutsch S, et al: The cost-effectiveness of screening mammography beyond age 65 years: A systematic review for the U.S. Preventive Services Task Force. Ann Intern Med 139:835-842, 2003 22. Diab SG, Elledge RM, Clark GM: Tumor characteristics and clinical outcome of elderly women with breast cancer. J Natl Cancer Inst 92:550-556, 2000 23. Singh R, Hellman S, Heimann R: The natural history of breast carcinoma in the elderly: Implications for screening and treatment. Cancer 100:1807-1813, 2004[CrossRef][Medline] 24. Evron E, Goldberg H, Kuzmin A, et al: Breast cancer in octogenarians. Cancer 106:1664-1668, 2006[CrossRef][Medline] 25. Warren JL, Klabunde CN, Schrag D, et al: Overview of the SEER-Medicare data: Content, research applications, and generalizability to the United States elderly population. Med Care 40:IV-3-18, 2002 (suppl) 26. Gamel JW, Meyer JS, Feuer E, et al: The impact of stage and histology on the long-term clinical course of 163,808 patients with breast carcinoma. Cancer 77:1459-1464, 1996[CrossRef][Medline] 27. Carter CL, Allen C, Henson DE: Relation of tumor size, lymph node status, and survival in 24,740 breast cancer cases. Cancer 63:181-187, 1989[CrossRef][Medline] 28. Freeman JL, Klabunde CN, Schussler N, et al: Measuring breast, colorectal, and prostate cancer screening with medicare claims data. Med Care 40:IV-36-42, 2002 29. Smith-Bindman R, Quale C, Chu PW, et al: Can Medicare billing claims data be used to assess mammography utilization among women ages 65 and older? Med Care 44:463-470, 2006[CrossRef][Medline] 30. Mouchawar J, Byers T, Warren M, et al: The sensitivity of Medicare billing claims data for monitoring mammography use by elderly women. Med Care Res Rev 61:116-127, 2004 Submitted May 29, 2007; accepted February 6, 2008.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2008 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 tests. Logistic regression was used to test for the effect of mammography use on disease stage (early v late), controlling for year of diagnosis, age, census tract income, education level, marital status, SEER region, race/ethnicity, and comorbid conditions. The fit of this model was tested by using the Hosmer-Lemeshow goodness-of-fit statistic.
