Skip to main content
Free access
Breast Cancer
March 20, 2007

Surveillance Testing Among Survivors of Early-Stage Breast Cancer

Publication: Journal of Clinical Oncology

Abstract

Purpose

Guidelines recommend against routine surveillance testing for women who have had breast cancer. We described follow-up care for breast cancer survivors, examined how surveillance testing varies by the types of physicians seen, and assessed changes in testing rates over time.

Methods

Using Surveillance, Epidemiology, and End Results–Medicare data, we studied a population-based cohort of 44,511 women age ≥ 65 years diagnosed with stage I/II breast cancer during 1992 to 1999 and observed through 2001. We measured bone scans, tumor antigen tests, chest x-rays, and other chest/abdominal imaging during 3 consecutive surveillance years. We described physicians seen in follow-up and used repeated-measures logistic regression to assess the relationship with testing and to assess testing trends over time.

Results

Nearly half of breast cancer survivors saw a medical oncologist in surveillance year 1, but only 27% saw a medical oncologist annually for 3 years. In adjusted analyses, women seeing medical oncologists had more bone scans, tumor antigen testing, chest x-rays, and chest/abdominal imaging than other women (all P < .001). Nevertheless, rates of testing decreased over time (all P < .001). Rates of tumor antigen testing and chest x-rays decreased faster and chest/abdominal imaging increased slower among women seeing medical oncologists than among other women (all P < .05).

Conclusion

Nonrecommended testing for early-stage breast cancer patients has decreased over time. Although most breast cancer survivors did not see oncologists annually, those who did had substantially higher rates of testing than others; whether such testing in this low-risk population was due to more symptoms or excessive surveillance is an important question for additional study.

Introduction

With more than 200,000 women diagnosed with invasive breast cancer yearly,1 more than 2 million breast cancer survivors are currently alive in the United States.2 National guidelines recommend that women with a history of breast cancer undergo annual mammography but advise against other routine surveillance testing (at regular predefined intervals), including complete blood counts, chemistry studies, chest x-rays, bone scans, liver ultrasounds, abdominal computed tomography scans, and tumor markers.3-5 These recommendations are based on randomized studies demonstrating that routine testing for distant metastatic disease provides no benefit in survival or health-related quality of life.6-8 Moreover, intensive surveillance costs from $260 million to $630 million more annually than an approach that includes only history, physical examination, mammography, and additional testing when indicated.9-11
Despite the lack of evidence supporting routine surveillance testing other than mammography for breast cancer survivors and the recommendations against its use, such testing is performed by many physicians,12-16 under the assumption that detecting and treating recurrences early results in better outcomes. Because many physicians, particularly medical oncologists,17 may favor intensive surveillance, the types of physicians seen by patients after breast cancer diagnosis and treatment may influence the surveillance care received. Moreover, recent reports emphasize the importance of coordination of care for cancer survivors,5,18 and better coordination may limit unnecessary testing. Few data are available about the mix of physicians caring for breast cancer survivors or the associated patterns of surveillance testing. In a prior study, we identified underuse of surveillance mammography among breast cancer survivors, with higher rates of mammography among women who visited cancer specialists in the years after diagnosis.19
In this study, we assessed bone scans, tumor antigen testing, chest x-rays, and other chest/abdominal imaging over a 3-year period in a population-based cohort of early-stage breast cancer survivors at low risk for recurrence. We described patterns of follow-up care and examined how testing varied by the types of physicians seen after primary treatment. In addition, we examined trends in testing over time to assess whether practice changed after studies and guidelines related to surveillance testing were published.3,4,6,7

Methods

Data and Patients

The study methods have been described in detail previously.19 Briefly, we used the Surveillance, Epidemiology, and End Results–Medicare data,20 which combine uniformly reported data from 11 population-based cancer registries covering approximately 14% of the United States population21 with Medicare claims data. The study cohort included 44,511 women with a first diagnosis of stage I/II breast cancer during 1992 to 1999, who were ≥ 65 years old and enrolled in Parts A and B of fee-for-service Medicare when diagnosed, and who underwent primary surgery.19 We censored patients at the end of 2001 or sooner (if they died), discontinued enrollment from Parts A and B fee-for-service Medicare, developed a new cancer, or had evidence of a possible recurrence, using a relatively conservative definition of recurrence to ensure we had disease-free cohorts for assessment of surveillance testing.19
Because mammography peaks in 1-year intervals after diagnosis19,22 and because we did not want to capture testing associated with primary treatment, we defined 3 surveillance years: months 7 to 18, 19 to 30, and 31 to 42 after diagnosis. For inclusion in analyses of a surveillance year, women had to be alive and not censored through the end of that surveillance year. Of the 44,511 early-stage breast cancer survivors, 37,967 were observed through surveillance year 1, 30,406 were observed through surveillance year 2, and 23,016 were observed through surveillance year 3.19

Surveillance Testing

We identified bone scans, tumor antigen testing, chest x-rays, and chest/abdominal imaging (computed tomography, magnetic resonance imaging, and ultrasound) based on inpatient, outpatient, and physician supplier files (Appendix Table A1, online only). For each 12-month surveillance period, we assessed whether at least one of each type of test was performed, recognizing that this would underestimate the total extent of testing. We originally examined each subcategory of tumor markers and chest/abdominal imaging separately; however, because the numbers were small for some subcategories and patterns of testing were similar, we combined them into the larger categories (Appendix Table A1). For analyses examining testing over time, we also assessed mammography (recommended annually).

Visits With Physicians

We identified all outpatient visits with physicians (Appendix Table A1) and obtained each physician's specialty after linking with the American Medical Association Physician Masterfile.23 We categorized each physician's self-reported primary specialty as primary care physician (family practice, internal medicine, general practice, geriatrics), medical oncologist, general surgeon (including a small number of surgical oncologists), radiation oncologist, medical subspecialist (eg, cardiologist, endocrinologist), surgical subspecialist (eg, vascular surgeon, thoracic surgeon), or other specialist (eg, obstetrician/gynecologist [rarely seen by these elderly cancer survivors], dermatologist, ophthalmologist). We also calculated the total number of office visits during each surveillance year.

Control Variables

The Surveillance, Epidemiology, and End Results registries document each patient's age at diagnosis, race, Hispanic ethnicity, marital status, history of other cancer, residence in a metropolitan county, and tumor characteristics. We characterized education and income of the census tract of residence using 1990 Census data, classified into quartiles within the registry before combining across registries to avoid misclassification due to regional variations. We measured comorbid illness using Diagnostic Cost Group (DCG) scores,24 a risk-adjustment tool used by the Centers for Medicare & Medicaid Services to predict disease burden and future costs for Medicare beneficiaries using diagnostic information from inpatient and ambulatory claims. DCGs capture 184 conditions (v 18 for the Charlson score25), predict mortality in cancer patients,26 and may also capture patients’ propensity to seek health care services. We calculated summary DCGs scores (excluding breast cancer codes) based on the 12-month period beginning at the month of diagnosis to characterize illness at a time when decisions about testing would be made. Finally, we used registry data and claims19,27-30 to classify primary treatment as mastectomy, breast-conserving surgery with radiation, or breast-conserving surgery without radiation. Variables are categorized as in Table 1.

Analysis

We first described bone scans, tumor markers, chest x-rays, and abdominal/chest imaging, and the physicians seen during follow-up. We used χ2 tests to examine bivariate associations between types of providers seen during surveillance year 1 and testing during that year. Similarly, we examined bivariate associations between providers seen during the 3 surveillance years (annual, at least one but not annual, or no visits to each provider type) and annual testing during years 1 to 3.
For each category of test, we used repeated-measures logistic regression with generalized estimating equations31 to assess the odds of testing associated with seeing a certain provider type, adjusting for the patient characteristics described, and visits (categorized in quintiles for each surveillance year). The modeling strategy allowed women to be included in analyses for all surveillance years in which they had complete data and allowed estimates of the effect of seeing certain types of providers in only some surveillance years. For example, if a patient saw a medical oncologist in years 1 and 3, information from both years 1 and 3 is used in estimating the effect of seeing a medical oncologist on testing.
We examined change in mammography rates and other testing over time using indicator variables for year of diagnosis and calculated P values for the overall effect of year of diagnosis on testing. We calculated adjusted predicted rates of testing according to year of diagnosis using regression model coefficients.32 These predicted rates reflect the average rates of testing for the surveillance years for which each patient has complete data. We also tested the interaction of diagnosis year and visits with a medical oncologist. Because analyses suggested linear trends in testing, we modeled year of diagnosis in these models as a continuous effect.
Given that many women saw more than one type of provider in a given year, we repeated the models after categorizing women into four exclusive categories reflecting types of providers seen during each surveillance year: primary care provider but no cancer specialist (medical oncologist, radiation oncologist, surgeon; reference group), cancer specialist but no primary care provider, both cancer specialist and primary care provider, and neither cancer specialist nor primary care provider.
All tests of statistical significance were two sided. We converted adjusted odds ratios to risk ratios to better reflect the true relative risks.33 We conducted analyses using SAS statistical software, version 8.2 (SAS Institute Inc, Cary, NC). The Harvard Medical School Committee on Human Studies approved the study protocol.

Results

The mean age of the 44,511 breast cancer survivors was 75.3 years at diagnosis and 11% were nonwhite (Table 1). Of 37,967 patients observed throughout surveillance year 1, 13.3% had at least one bone scan, 29.2% had a tumor antigen test, 10.9% had chest/abdominal imaging, and 58.8% had a chest x-ray during surveillance year 1. Of 23,016 patients observed through the end of surveillance year 3, 2.1% had yearly bone scans, 16.0% had yearly tumor antigen testing, 1.0% had yearly chest/abdominal imaging, and 29.2% had yearly chest x-rays.
Most breast cancer survivors visited a primary care provider during surveillance year 1 (Table 2), although only 51% did so annually during 3 years. Forty-two percent of patients saw a medical oncologist in year 1, 46% saw a general surgeon, and 24% saw a radiation oncologist, but few patients saw these types of physicians annually. For example, only 27% saw a medical oncologist annually; those who did were younger (73.4 v 75.3 years; P < .001),were more often diagnosed with stage 2 cancers (47.8% v 31.1%; P < .001), and had lower comorbidity scores (0.57 v 0.62; P < .001), but did not differ by race (4.6% v 4.9% were black; P = .27).
Women who had visits with medical oncologists had the highest rates of bone scans and tumor antigen testing and high rates of chest/abdominal imaging (Table 3). Women seeing medical oncologists annually had high rates of annual testing (eg, 37% received a tumor antigen test annually v 6.7% of the women with no visits to a medical oncologist; P < .001).
In multivariable models, adjusted rates of mammography and chest/abdominal imaging increased by year of diagnosis and adjusted rates of other testing decreased over time (all P < .001; Fig 1). These trends followed linear patterns, without abrupt changes after presentation or publication of major clinical trials (1994) or guidelines (1997 and 1999).
In adjusted analyses, cancer survivors who visited medical oncologists or radiation oncologists were more likely than women not seeing physicians of these types to have bone scans, tumor antigen tests, and abdominal/chest imaging, with risk ratios for each of these tests highest among patients seeing medical oncologists (Fig 2). Women seeing primary care providers were less likely than other women to have bone scans and tumor antigen tests, and women seeing surgeons did not differ from other women in use of bone scans and tumor antigen tests. Cancer survivors seeing medical subspecialists, surgeons, and radiation oncologists were more likely than other women to undergo chest/abdominal imaging. Women seeing primary care physicians, medical oncologists, medical subspecialists, and other specialists were more likely than other women to have chest x-rays.
When testing the interaction of diagnosis year and whether patients had seen a medical oncologist in a given surveillance year, we found that adjusted mammography rates increased faster for women seeing medical oncologists (adjusted rate of 75.8% for women diagnosed in 1992 to an adjusted rate of 83.0% for women diagnosed in 1999) than for other women (67.2% to 73.4%, respectively; P = .04). Tumor antigen testing rates decreased faster among women seeing medical oncologists (43.2% to 36.3%) versus other women (17.6% to 15.4%; P = .04), as did rates of chest x-rays (71.8% to 53.0% if seeing medical oncologists v 57.3% to 45.1%; P < .001). Adjusted rates of chest/abdominal imaging increased more slowly among women seeing medical oncologists (10.4% to 13.4%) than other women (6.8% to 11.9%; P < .001). These findings suggest oncologists may have been more aware of research results and recommendations for surveillance testing than other providers. The rate of decrease in bone scans did not differ by whether patients were observed by a medical oncologist.
When we categorized women based on combinations of physicians seen in any particular year, we found that compared with women who saw a primary care provider but no cancer specialists, women who saw other combinations of providers had higher rates of bone scans, tumor antigen testing, and abdominal/chest imaging (Table 4).

Discussion

We examined testing for older, low-risk, early-stage breast cancer survivors and found relatively high rates of testing. Although some breast cancer survivors require testing for evaluation of symptoms, the frequent annual testing for some tests suggests that they may be ordered routinely for certain women, despite evidence-based recommendations against this practice.3-5 We also found that most women in this population-based sample did not see a cancer specialist annually, although those who did had substantially higher rates of testing than other patients. Because the evaluation of suspected recurrence may be more often performed by cancer specialists than by other physicians, even in a low-risk population, understanding whether these high rates of testing are due to more symptoms or excessive surveillance is an important question for additional study.
Despite relatively high rates of testing overall, we observed that rates of nonrecommended testing generally decreased over time, whereas mammography increased over time. These changes were more pronounced for patients seeing medical oncologists, suggesting that medical oncologists were most influenced by results of clinical trials and guidelines related to surveillance testing published during the 1990s.3,4,6,7
As the proportion of women diagnosed with early-stage breast cancer increases, so too does the population of breast cancer survivors, emphasizing the importance of follow-up care for these women. Randomized controlled trials have found that routine testing for distant metastatic disease provides no benefit in survival or health-related quality of life,6-8 and an intensive approach to surveillance is costly.9-11 Moreover, although many physicians17 and patients34 may favor intensive surveillance, patients overestimate the value of laboratory and imaging studies and may incorrectly perceive the significance of a normal test.34
It is unclear which type of provider is most suitable to care for breast cancer survivors. According to guidelines, physicians should be experienced in the surveillance of cancer patients and in the examination of irradiated and normal contralateral breasts, and care should be coordinated and not duplicated by multiple specialists.3,5 Data suggest that follow-up care need not be performed by a cancer specialist. British and Canadian trials comparing primary care–centered and specialist-centered follow-up of breast cancer patients found no difference in time to diagnosing recurrence, anxiety, or quality of life35,36; in addition, patients observed in primary care settings were more satisfied.37
Nevertheless, British and Canadian data suggest that although generalist physicians are comfortable or even prefer taking primary responsibility for routine follow-up of their breast cancer patients,38,39 most breast cancer specialists prefer to continue observing their patients and do not believe that general practitioners have the necessary skills to provide adequate follow-up care.38 Although data suggest that breast cancer surveillance care is being provided primarily by specialists,39,40 these and other14-16 studies of surveillance testing included relatively small samples of patients treated at cancer centers, large hospitals, or large group practices. Given that breast cancer survivors remain at risk of recurrence for 20 years or more,3 it may be impractical for patients to continue to visit cancer specialists for extended periods of time, especially as they age and develop conditions requiring the expertise of other physicians.
Our finding that most older survivors of early-stage breast cancer in the United States do not visit a medical oncologist or other cancer specialist annually heightens the need to ensure that generalist physicians have the skills to care for cancer survivors, are aware of current recommendations, and can communicate with cancer specialists when necessary. This may be especially important because although elderly patients who do not follow up with cancer specialists receive fewer nonrecommended tests than other women, they also receive fewer surveillance mammograms.19,41 Patients and their physicians would benefit from greater education about surveillance care recommendations.5 In addition, a clear plan regarding which doctor is primarily responsible for patients’ routine follow-up care, including a schedule for visits and testing that patients should expect,18 would likely increase mammography rates and decrease rates of nonrecommended testing, and could also alert patients and physicians about symptoms that might prompt additional testing. Such a plan could be developed by a patient's oncologist after primary treatment is completed.18 Finally, better communication among providers sharing re-sponsibility for a patient's care, perhaps using innovative electronic applications, could help to decrease duplicate testing and expedite evaluation of symptoms when necessary.
Our study has several limitations. First, our data do not allow us to ascertain whether tests were ordered for surveillance or evaluation of symptoms. The patterns of use for some tests (eg, more abdominal/chest imaging among patients who did not see a primary care physician or cancer specialist) suggest that not all tests were performed for breast cancer surveillance. Such noncancer specialists may have a low threshold for ruling out cancer recurrence when patients present with nonspecific symptoms or they may be more closely monitoring other conditions. However, the high rates of annual testing for some tests in our low-risk population suggest that they are being ordered routinely. Similarly, our data are not well suited to assess whether more testing led to identification of recurrences; this issue has been addressed more effectively in randomized clinical trials.6-8
Second, we used a claims-based algorithm for identifying potential recurrences to exclude women whose testing was unlikely to be routine, and may have misclassified some women. However, our conservative algorithm was more likely to exclude women without recurrences than to include women who had experienced recurrence, which would underestimate testing among low-risk patients. Third, our method for identifying provider specialty may have misclassified some physicians, particularly those with more than one specialty or practice site. Even so, the American Medical Association data may be the most reliable source of physician specialty.23 Finally, we ascertained receipt of at least one test in each surveillance year, not every test, so our study underestimates the total amount of testing.
In conclusion, our findings demonstrate a high rate of nonrecommended testing associated with surveillance after treatment for early-stage breast cancer, suggesting that these tests may be overused. Testing patterns for our low-risk cohort of patients may underestimate the extent of testing among breast cancer survivors as a whole. Although most breast cancer survivors did not see a cancer specialist annually, those who did had substantially more testing. Given the high cost of breast cancer surveillance testing, eliminating unnecessary testing could provide substantial savings.

Authors’ Disclosures of Potential Conflicts of Interest

Although all authors completed the disclosure declaration, the following author or immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Employment: N/A Leadership: N/A Consultant: John Z. Ayanian, DxCG Inc Stock: N/A Honoraria: N/A Research Funds: N/A Testimony: N/A Other: N/A

Author Contributions

Conception and design: Nancy L. Keating, Mary Beth Landrum, Edward Guadagnoli, Eric P. Winer, John Z. Ayanian
Financial support: Nancy L. Keating
Collection and assembly of data: Nancy L. Keating
Data analysis and interpretation: Nancy L. Keating, Mary Beth Landrum, Edward Guadagnoli, Eric P. Winer, John Z. Ayanian
Manuscript writing: Nancy L. Keating, Mary Beth Landrum, Eric P. Winer, John Z. Ayanian
Final approval of manuscript: Nancy L. Keating, Mary Beth Landrum, Edward Guadagnoli, Eric P. Winer, John Z. Ayanian

Appendix

Fig 1. Adjusted rates of testing by year of diagnosis reflect testing performed during up to 3 surveillance years after diagnosis; therefore, rates reflect testing performed during 1993 through 2002. Women are included for each surveillance year for which they have complete data. P values reflect the overall effect of diagnosis year on testing from repeated measures models. CT, computed tomography; MRI, magnetic resonance imaging; CXR, chest x-ray; RCT, randomized controlled trial; ASCO, American Society of Clinical Oncology.
Fig 2. Adjusted risk ratios and 95% CIs for seeing a provider of each type (v not seeing a provider of that type) during a surveillance year, using repeated-measures logistic regression with generalized estimating equations to adjust for patient and tumor characteristics. Repeated-measures models allowed estimates of the effect of seeing certain types of providers in only some surveillance years. PCP, primary care physician.
Table 1. Characteristics of Early-Stage Breast Cancer Survivors (N = 44,511)
CharacteristicNo.% of Sample
Age, years  
    65-6910,47923.5
    70-7411,73726.4
    75-7910,29723.1
    80-847,04315.8
    85 and older4,95511.1
Race  
    White39,77089.4
    Black2,4515.5
    Other2,1494.8
    Unknown1410.3
Hispanic ethnicity  
    No42,90996.4
    Yes1,3873.1
    Unknown2150.5
Marital status  
    Unmarried24,26454.5
    Currently married19,23743.2
    Unknown1,0102.3
Residence  
    Nonmetropolitan county7,37816.6
    Metropolitan county37,13383.4
SEER region  
    San Francisco, CA3,6278.2
    Connecticut6,32714.2
    Michigan6,96115.6
    Hawaii9852.2
    Iowa6,55814.7
    New Mexico1,7333.9
    Seattle4,77910.7
    Utah2,1074.7
    Georgia2,5345.7
    San Jose, CA2,0504.6
    Los Angeles, CA6,85015.4
Median household income in census tract of residence, quartile  
    1 (lowest)10,57123.8
    211,00824.8
    310,99324.6
    4 (highest)11,10425.1
    Unknown8351.9
% who are high school graduates in census tract of residence, quartile  
    1 (lowest)10,53323.7
    211,04024.8
    310,93124.6
    4 (highest)11,17225.1
    Unknown8351.9
Prior history of non–breast cancer  
    No41,84094.0
    Yes2,6716.0
Stage  
    127,16861.0
    217,34339.0
Tumor grade  
    Well differentiated7,75617.4
    Moderately differentiated16,70837.5
    Poorly differentiated10,69424.0
    Undifferentiated8651.9
    Unknown8,48819.1
Tumor size, mm  
    ≤ 1013,37030.0
    11-1510,49423.6
    15-207,59217.1
    21-308,08318.2
    > 304,59710.3
    Unknown3750.8
Primary treatment  
    Mastectomy23,78553.5
    Breast-conserving surgery with     radiation15,76935.4
    Breast-conserving surgery without     radiation4,94711.1
Year of diagnosis  
    19925,92813.3
    19935,58212.5
    19945,45312.3
    19955,60612.6
    19965,42912.2
    19975,53412.4
    19985,47112.3
    19995,50812.4
Comorbidity, quartile  
    1 (lowest)14,73733.1
    27,15716.1
    311,42625.7
    4 (highest)11,19125.1
Abbreviation: SEER, Surveillance, Epidemiology, and End Results.
Table 2. Visits to Providers During Surveillance Year 1 and Surveillance Years 1-3
Visits During Year 1 (n = 37,967)*% of PatientsVisits During Years 1-3 (n = 23,016)% of Patients
PCP   
    No visits28No visits19
    At least 1 visit72At least 1 visit30
  Visits each year51
Medical oncologist   
    No visits58No visits56
    At least 1 visit42At least 1 visit17
  Visits each year27
General surgeon/surgical oncologist   
    No visits54No visits47
    At least 1 visit46At least 1 visit36
  Visits each year17
Radiation oncologist   
    No visits76No visits78
    At least 1 visit24At least 1 visit15
  Visits each year6
Medical subspecialist   
    No visits55No visits43
    At least 1 visit45At least 1 visit32
  Visits each year25
Surgical subspecialist   
    No visits49No visits31
    At least 1 visit51At least 1 visit47
  Visits each year22
Other specialist   
    No visits49No visits30
    At least 1 visit51At least 1 visit49
  Visits each year20
Abbreviation: PCP, primary care provider.
*
Patients observed through the end of surveillance year 1.
Patients observed thorugh the end of surveillance year 3.
Table 3. Surveillance Testing by Types of Providers Seen
Provider TypeBone Scan Tumor Antigen Test Abdominal/Chest Imaging Chest X-Ray 
 % With Test in Year 1P*% With Test in Year 1P*% With Test in Year 1P*% With Test in Year 1P*
Overall rate of testing during year 1 (n = 37,967)13.3 28.2 10.9 58.8 
Visit during year 1        
    PCP        
        Yes13.7.00227.3< .00111.6< .00161.5< .001
        No12.5 30.6 8.9 51.9 
    Medical oncologist        
        Yes19.0< .00145.6< .00114.9< .00164.3< .001
        No9.2 15.4 7.9 54.7 
    General surgeon/surgical     oncologist        
        Yes13.9.00327.3< .00112.3< .00161.4< .001
        No12.9 29.1 9.6 56.5 
    Radiation oncologist        
        Yes15.2< .00132.3< .00113.7< .00157.6.006
        No12.8 26.9 10.0 59.2 
    Medical subspecialist        
        Yes16.3< .00131.5< .00115.8< .00167.5< .001
        No10.9 25.5 6.9 51.8 
    Surgical subspecialist        
        Yes15.1< .00129.5< .00112.9< .00163.1< .001
        No11.4 26.8 8.8 54.2 
    Other specialist        
        Yes14.9< .00130.0< .00113.5< .00164.8< .001
        No11.7 26.4 8.2 52.6 
Abbreviation: PCP, primary care provider.
*
Using the χ2 test.
37,967 patients were observed through the end of surveillance year 1 and were eligible for analyses of testing in surveillance year 1.
Table 4. Adjusted Risk Ratio of Surveillance Testing by Providers Seen*
Type of Provider SeenBone Scan Tumor Antigen Test Abdominal/Chest Imaging Chest X-Ray 
 Adjusted Risk Ratio*95% CIAdjusted Risk Ratio*95% CIAdjusted Risk Ratio*95% CIAdjusted Risk Ratio*95% CI
PCP but no cancer specialistReferenceReferenceReferenceReference
Cancer specialist but no PCP1.431.34 to 1.531.771.72 to 1.831.361.26 to 1.461.000.98 to 1.02
PCP and cancer specialist1.331.26 to 1.411.711.66 to 1.761.341.26 to 1.431.051.03 to 1.07
Neither PCP nor cancer specialist1.411.30 to 1.521.121.07 to 1.171.531.40 to 1.671.010.98 to 1.03
Abbreviations: PCP, primary care provider; SEER, Surveillance, Epidemiology, and End Results.
*
Using repeated-measures logistic regression with generalized estimating equations to simultaneously adjust estimates for patients’ age, race, ethnicity, marital status, residence in a metropolitan county, SEER region, quartile of area level income and education, stage, history of cancer other than breast cancer, tumor size, tumor grade, comorbid illness, year of diagnosis, surveillance year, type of primary therapy, and quintile of number of visits during each surveillance year. The repeated-measures models allowed women to be included in analyses for all surveillance years in which they had complete data and allowed estimates of the effect of seeing certain types of providers in only some surveillance years, and thus the models include data on testing done during each surveillance year.
In surveillance year 1, 16% of patients saw a PCP and no cancer specialist, 19% saw a cancer specialist but no PCP, 55% saw a PCP and cancer specialist, and 9% saw neither a PCP nor a cancer specialist. In surveillance year 2, 26% saw a PCP and no cancer specialist, 17% saw a cancer specialist and no PCP, 48% saw both a PCP and a cancer specialist, and 11% saw neither a PCP nor a cancer specialist. In surveillance year 3, 31% saw a PCP and no cancer specialist, 19% saw a cancer specialist and no PCP, 30% saw both a PCP and a cancer specialist, and 20% saw neither a PCP nor cancer specialist.
Table A1. Codes Used to Identify Testing
TestCPT CodeICD-9 Procedure CodesICD-9 Diagnosis Codes
Mammography76090-76092V76.11, V76.1287.37
Bone scan78300-78320, 7839992.14 
Tumor antigen tests   
    CEA82378, 86149, 86151  
    CA 15-386300  
    CA 19-986301  
    Shared86316, 86304  
    Other86294  
Chest x-ray71010-7103587.44, 87.49 
Chest and abdominal imaging   
    Abdominal ultrasound76700, 7670588.76 
    Abdominal CT74150-74175, 7563588.01 
    Abdominal MRI74181-7418588.97 
    Chest CT71250, 71260, 71270, 7127587.41, 87.42 
    Chest MRI71550, 71551, 71552, 7155588.92 
Outpatient visits99201-99205, 99211-99215, 99241-99245, 99381-99387, 99391-99397, 99401-99404, 99411-99412, 99420-99429, 99354-99360  
Abbreviations: CPT, Current Procedural Terminology; ICD-9, International Classification of Diseases, 9th revision; CEA, carcinoembryonic antigen; CT, computed tomography; MRI, magnetic resonance imaging.

Acknowledgments

This study used the linked Surveillance, Epidemiology, and End Result (SEER) Medicare database. The interpretation and reporting of these data are the sole responsibility of the authors. The authors acknowledge the efforts of the Applied Research Program, National Cancer Institute; the Office of Research, Development and Information, Centers for Medicare and Medicaid Services, Information Management Services Inc; and the SEER Program tumor registries in the creation of the SEER-Medicare database. We thank Laurie Meneades, MS, for expert programming assistance.
Supported by a Clinical Scientist Development Award from the Doris Duke Charitable Foundation (N.L.K.).
The funding organization had no role in design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript.
Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.

References

1.
Jemal A, Murray T, Ward E, et al: Cancer statistics, 2005. CA Cancer J Clin 55::10,2005-30,
2.
Ries LAG, Eisner MP, Kosary CL, et al: SEER Cancer Statistics Review, 1975-2002. http://seer.cancer.gov/csr/1975_2002/
3.
American Society of Clinical Oncology: Recommended breast cancer surveillance guidelines. J Clin Oncol 15::2149,1997-2156,
4.
Smith TJ, Davidson NE, Schapira DV, et al: American Society of Clinical Oncology 1998 update of recommended breast cancer surveillance guidelines. J Clin Oncol 17::1080,1999-1082,
5.
Khatcheressian JL, Wolff AC, Smith TJ, et al: American Society of Clinical Oncology 2006 update of the breast cancer follow-up and management guidelines in the adjuvant setting. J Clin Oncol 24::5091,2006-5097,
6.
The GIVIO Investigators: Impact of follow-up testing on survival and health-related quality of life in breast cancer patients: A multicenter randomized controlled trial. JAMA 271::1587,1994-1592,
7.
Rosselli Del Turco M, Palli D, Cariddi A, et al: Intensive diagnostic follow-up after treatment of primary breast cancer: A randomized trial—National Research Council Project on Breast Cancer follow-up. JAMA 271::1593,1994-1597,
8.
Rojas M, Telaro E, Russo A, et al: Follow-up strategies for women treated for early breast cancer. Oxford, United Kingdom, Cochrane Library, CD001768, 2, 2005
9.
Schapira DV, Urban N: A minimalist policy for breast cancer surveillance. JAMA 265::380,1991-382,
10.
Schapira DV: Breast cancer surveillance: A cost-effective strategy. Breast Cancer Res Treat 25::107,1993-111,
11.
Edelman MJ, Meyers FJ, Siegel D: The utility of follow-up testing after curative cancer therapy: A critical review and economic analysis. J Gen Intern Med 12::318,1997-331,
12.
Loomer L, Brockschmidt JK, Muss HB, et al: Postoperative follow-up of patients with early breast cancer: Patterns of care among clinical oncologists and a review of the literature. Cancer 67::55,1991-60,
13.
Tomiak E, Piccart M: Routine follow-up of patients after primary therapy for early breast cancer: Changing concepts and challenges for the future. Ann Oncol 4::199,1993-204,
14.
Elston Lafata J, Simpkins J, Schultz L, et al: Routine surveillance care after cancer treatment with curative intent. Med Care 43::592,2005-599,
15.
Lash TL, Silliman RA: Medical surveillance after breast cancer diagnosis. Med Care 39::945,2001-955,
16.
Simon MS, Stano M, Severson RK, et al: Clinical surveillance for early stage breast cancer: An analysis of claims data. Breast Cancer Res Treat 40::119,1996-128,
17.
Simon MS, Hoff M, Hussein M, et al: An evaluation of clinical follow-up in women with early stage breast cancer among physician members of the American Society of Clinical Oncology. Breast Cancer Res Treat 27::211,1993-219,
18.
Institute of Medicine: From Cancer Patient to Cancer Survivor: Lost in Transition. Washington, DC, National Academy Press, 2005
19.
Keating NL, Landrum MB, Guadagnoli E, et al: Factors related to underuse of surveillance mammography after breast cancer diagnosis. J Clin Oncol 24::85,2006-94,
20.
Potosky AL, Riley GF, Lubitz JD, et al: Potential for cancer related health services research using a linked Medicare-tumor registry database. Med Care 31::732,1993-748,
21.
Ries LA, Wingo PA, Miller DS, et al: The annual report to the nation on the status of cancer, 1973-1997, with a special section on colorectal cancer. Cancer 88::2398,2000-2424,
22.
Schapira MM, McAuliffe TL, Nattinger AB: Underutilization of mammography in older breast cancer survivors. Med Care 38:281-289, 2000
23.
Baldwin LM, Adamache W, Klabunde C, et al: Linking physician characteristics and Medicare claims data: Issues in data availability, quality, and measurement. Med Care 40::IV-82,2002-IV-95,
24.
Ellis RP, Pope GC, Iezzoni LI, et al: Diagnosis-based risk adjustment for Medicare capitation payments. Health Care Financing Rev 17::101,1996-128,
25.
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::373,1987-383,
26.
Baldwin LM, Klabunde CN, Green P, et al: In search of the perfect comorbidity measure for use with administrative claims data: Does it exist? Med Care 44::745,2006-753,
27.
Keating NL, Landrum MB, Ayanian JZ, et al: Consultation with a medical oncologist before surgery and type of surgery among elderly women with early-stage breast cancer. J Clin Oncol 21::4532,2003-4539,
28.
Keating NL, Landrum MB, Meara E, et al: Do increases in managed care market share of managed are influence quality of cancer care in the fee-for-service sector? J Natl Cancer Inst 97::257,2005-264,
29.
Cooper GS, Virnig B, Klabunde CN, et al: Use of SEER-Medicare data for measuring cancer surgery. Med Care 40::IV-43,2002-IV-48,
30.
Du X, Freeman JL, Goodwin JS: Information on radiation treatment in patients with breast cancer: The advantages of the linked Medicare and SEER data. J Clin Epidemiol 52::463,1999-470,
31.
Zeger SL, Liang K: Longitudinal data analysis for discrete and continuous outcomes. Biometrics 42::121,1986-130,
32.
Little RJ: Direct standardization: A tool for teaching linear models for unbalanced data. Am Statistician 36::38,1982-43,
33.
Zhang J, Yu KF: What's the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA 280::1690,1998-1691,
34.
Muss HB, Tell GS, Case LD, et al: Perceptions of follow-up care in women with breast cancer. Am J Clin Oncol 14::55,1991-59,
35.
Grunfeld E, Gray A, Mant D, et al: Follow-up of breast cancer in primary care vs specialist care: Results of an economic evaluation. Br J Cancer 79::1227,1999-1233,
36.
Grunfeld E, Levine MN, Julian JA, et al: Randomized trial of long-term follow-up for early-stage breast cancer: A comparison of family physician versus specialist care. J Clin Oncol 24::848,2006-855,
37.
Grunfeld E, Fitzpatrick R, Mant D, et al: Comparison of breast cancer patient satisfaction with follow-up in primary care versus specialist care: Results from a randomized controlled trial. Br J Gen Pract 49::705,1999-710,
38.
Grunfeld E, Mant D, Vessey MP, et al: Specialist and general practice views on routine follow-up of breast cancer patients in general practice. Fam Pract 12::60,1995-65,
39.
Worster A, Bass MJ, Wood ML: Willingness to follow breast cancer: Survey of family physicians. Can Fam Physician 42::263,1996-268,
40.
Worster A, Wood ML, McWhinney IR, et al: Who provides follow-up care for patients with early breast cancer? Can Fam Physician 41::1314,1995-1320,
41.
Earle CC, Burstein HJ, Winer EP, et al: Quality of non-breast cancer health maintenance among elderly breast cancer survivors. J Clin Oncol 21::1447,2003-1451,

Information & Authors

Information

Published In

Journal of Clinical Oncology
Pages: 1074 - 1081
PubMed: 17369571

History

Published in print: March 20, 2007
Published online: September 21, 2016

Permissions

Request permissions for this article.

Authors

Affiliations

Nancy L. Keating
From the Division of General Internal Medicine, Department of Medicine, Brigham and Women's Hospital; Department of Health Care Policy, Harvard Medical School; and the Department of Adult Oncology, Dana-Farber Cancer Institute, Boston, MA
Mary Beth Landrum
From the Division of General Internal Medicine, Department of Medicine, Brigham and Women's Hospital; Department of Health Care Policy, Harvard Medical School; and the Department of Adult Oncology, Dana-Farber Cancer Institute, Boston, MA
Edward Guadagnoli
From the Division of General Internal Medicine, Department of Medicine, Brigham and Women's Hospital; Department of Health Care Policy, Harvard Medical School; and the Department of Adult Oncology, Dana-Farber Cancer Institute, Boston, MA
Eric P. Winer
From the Division of General Internal Medicine, Department of Medicine, Brigham and Women's Hospital; Department of Health Care Policy, Harvard Medical School; and the Department of Adult Oncology, Dana-Farber Cancer Institute, Boston, MA
John Z. Ayanian
From the Division of General Internal Medicine, Department of Medicine, Brigham and Women's Hospital; Department of Health Care Policy, Harvard Medical School; and the Department of Adult Oncology, Dana-Farber Cancer Institute, Boston, MA

Notes

Address reprint requests to Nancy L. Keating, MD, MPH, Department of Health Care Policy, Harvard Medical School, 180 Longwood Ave, Boston, MA 02115; e-mail: [email protected]

Metrics & Citations

Metrics

Altmetric

Citations

Article Citation

Download Citation

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

For more information or tips please see 'Downloading to a citation manager' in the Help menu.

Format





Download article citation data for:
Nancy L. Keating, Mary Beth Landrum, Edward Guadagnoli, Eric P. Winer, John Z. Ayanian
Journal of Clinical Oncology 2007 25:9, 1074-1081

View Options

View options

PDF

View PDF

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Personal login Institutional Login

Purchase Options

Purchase this article to get full access to it.

Purchase this Article

Subscribe

Subscribe to this Journal
Renew Your Subscription
Become a Member

Media

Figures

Other

Tables

Share

Share

Share article link

Share