Trends in Colorectal Cancer Test Use in the Medicare Population, 1998–2005
Article Outline
Background
Colorectal cancer (CRC) screening has been covered under the Medicare program since 1998. No prior study has addressed the question of the completeness of CRC screening in the entire Medicare cohort.
Methods
In 2008, CRC test-use rates were analyzed for the national fee-for-service Medicare population using Medicare enrollment and claims data from 1998 through 2005. Annual test-use rates were calculated for fecal occult blood testing, sigmoidoscopy, barium enema, and colonoscopy for each year by the demographic characteristics of enrollees. A current-in-Medicare rate was calculated to assess the percentage of enrollees with CRC testing according to recommended intervals.
Results
Colonoscopy rates have increased every year since the introduction of CRC screening coverage. Test-use rates for all other test modalities have steadily decreased. The percentage of Medicare enrollees receiving appropriate tests has slowly increased. In 2005, 47% of enrollees aged ≥65 years and 33% of enrollees aged 50–64 years had claims indicating that they had been tested according to recommended intervals.
Conclusions
CRC test-use rates in the Medicare population are low. Disparities are apparent by age, race/ethnicity, gender, disability, income, and geographic residence. Much work remains to be done to increase testing to acceptable levels.
Background
The majority of colorectal cancer (CRC) cases in the U.S. occur in individuals aged ≥60 years, making CRC an important medical condition for the Medicare program.1 Regular screening for CRC is recommended by expert groups.2, 3, 4 Since 1998, Medicare has covered CRC screening for average-risk enrollees aged ≥50 years with four different tests at varying recommended test intervals: fecal occult blood testing (FOBT) every year; sigmoidoscopy every 4 years; barium enema every 5 years; or, for high-risk individuals, colonoscopy every 2 years.5 In 2001, Medicare expanded this benefit to cover screening with colonoscopy for average-risk enrollees every 10 years. The screening tests and intervals covered by Medicare are consistent with national guidelines with the exception of sigmoidoscopy, for which expert groups recommend a 5-year interval between procedures.
There has been considerable interest in assessing whether Medicare coverage of CRC tests has prompted the increased use of screening in this population. The availability of four distinct CRC tests, with their varying recommended intervals, makes this a challenging question to answer. Previous studies indicate that colonoscopy use has increased.6, 7, 8, 9, 10 Whether CRC screening test use overall has increased has been difficult to assess, however, because of declines in the use of certain types of tests (FOBT, sigmoidoscopy, and barium enema). Investigators have variously shown no11 or only modest increases in CRC test use following the implementation of Medicare coverage.6, 8, 10, 12, 13 However, most of these studies have relied on cross-sectional snapshots of test-use rates in a given time period or have focused on the use of specific tests, making it impossible to assess the extent to which Medicare enrollees were tested according to recommended intervals.
Another factor complicating the assessment of CRC screening uptake among Medicare enrollees is the observed variation by sociodemographic characteristics, including race/ethnicity, gender, and income,7, 14, 15, 16, 17, 18, 19 as well as by geographic location.20 The contribution of service availability, referral patterns, and physician and patient preferences to regional variation in cancer screening is under-researched. Haas and colleagues21 demonstrated regional variations in CRC screening rates that paralleled regional variations in physician-recommendation rates. Evaluation of geographic variation may be confounded by underlying differences in the demographics of the study populations and their influence on test-use patterns. For these reasons, studies conducted in regional areas or states may not accurately reflect the experience of the nation as a whole.
One prior study15 had attempted to assess the completeness of CRC screening by Medicare enrollees. Using a cancer-free cohort of 153,469 Medicare enrollees from the Surveillance, Epidemiology, and End Results (SEER)–Medicare database, Cooper and Kou found an alarmingly low rate of 29% of enrollees with complete guidelines-based testing over a 10-year window. Neither this nor any other study had yet addressed the question of the completeness of CRC screening in the entire Medicare cohort, however. In addition, no study had assessed changes in the completeness of CRC screening in the Medicare population since the introduction of the screening benefit.
The present study was conducted to address this gap. The two primary objectives were to assess the percentage of Medicare enrollees receiving CRC tests according to recommended guidelines and to determine the extent to which CRC testing according to guidelines has increased since the introduction of Medicare coverage for colonoscopy for people of average risk. Because investigators had access to longitudinal CRC test-use data from 1998 through 2005 for the entire national fee-for-service (FFS) Medicare cohort, they had the opportunity to examine CRC test use at the person level over a long window of time, which provided a more accurate and comprehensive assessment of CRC test use than could be obtained in studies relying on single-year, single-state, or sampled Medicare data. The claims-based approach used in this study also provided a snapshot of CRC test use free from the potential recall bias associated with the self-report of procedures occurring many years prior (such as colonoscopy). This study presents temporal trends of appropriate CRC testing and characteristics of appropriately tested Medicare enrollees. Test use for each of the four covered CRC tests was examined over time and by demographic characteristics of the Medicare population.
Methods
In 2000, The Carolinas Center for Medical Excellence (CCME), the quality-improvement organization responsible for Medicare quality of care in North and South Carolina, was asked by the Centers for Medicare & Medicaid Services to develop CRC screening measures and national- and state-level reports of CRC test use in the Medicare population. CCME used enrollment and claims data for the FFS Medicare population in the U.S. to create a web-based report that allows users to obtain estimates of national-, state-, and county-level CRC test-use rates. The report has been updated several times and now contains information on test use from 1998 through 2005 (www.thecarolinascenter.org/crcreport). These data were used to examine trends in CRC test use and to identify subgroups with lower test-use rates.
The cohort of Medicare enrollees eligible for CRC screening in each year was identified, and all CRC tests performed during the year were captured. Analyses were conducted looking across years on an enrollee level to determine whether enrollees were current with CRC testing within the time period covered by the study window. Details of the data steps and definitions are provided below.
Population
Medicare enrollees were included in single-year analyses if they were aged ≥50 years as of January 1 of that year, enrolled in FFS Medicare with no managed-care enrollment for the year, had Part-B coverage with ≤30 days lapse in coverage, resided in one of the 50 states, and were alive at the end of the year. Enrollees were included in multi-year analyses if they were alive on December 31 of that year and eligible for CRC screening during the last year of the multi-year window as well as eligible for at least one of the single-year rates. Data for determining eligibility came from the Medicare enrollment database, from which also were extracted age, race/ethnicity, gender, state and county of residence, and reason for Medicare eligibility (age or disability). Enrollees who entered Medicare due to a disability were classified as having a disability even after they had reached the age of Medicare eligibility. Additional information from enrollment files used in these analyses included the beneficiary's eligibility for assistance with Medicare premiums and copayments—the state buy-in program for low-income individuals (yes/no). While this variable does not capture accurately whether the person is enrolled in Medicaid, it does serve as a marker for low income.
Colorectal Cancer Test-Use Measures
Claims for inpatient, outpatient, and physicians' office procedures and the presence of specified codes from Current Procedural Terminology, the Healthcare Common Procedure Coding System, or ICD-9-CM (see Appendix A online at www.ajpm-online.net) were used to identify CRC tests. Five rates of CRC test use were computed: annual FOBT, annual sigmoidoscopy, annual barium enema, annual colonoscopy, and a multi-year measure to assess CRC testing according to recommended intervals (see Appendix B online at www.ajpm-online.net). The four test-specific rates, calculated for each year from 1998 through 2005, represent the percentage of eligible people in the denominator for each year that had each specific type of test in that year. The multi-year measure, called current in Medicare (CIM), was created to represent the percentage of enrollees who were tested according to guidelines within the limits of the 8-year study window. To capture the complete 5-year recommended screening interval for sigmoidoscopy and barium enema, the CIM measure was calculated only for the years 2002–2005.
The CIM measure indicates that an enrollee had an FOBT during the measurement year, a sigmoidoscopy or barium enema within the prior 5 years, or a colonoscopy anytime since 1998 up to the measurement year. Enrollees with a CRC test in >1 year were included in each single-test rate but counted only once in the CIM rate. The term CIM is used to differentiate this rate from current with testing or up to date, because colonoscopy tests conducted in the 2 years prior to the start of this study window and tests conducted prior to enrollees' entering the Medicare system were not captured. As prior analyses have indicated, Medicare claims cannot be used reliably to distinguish the original reasons for having the tests, so both screening and diagnostic tests were included in calculating the test rates.22
Analysis
Data used for this study represent the entire national FFS Medicare population eligible for CRC testing. Therefore, descriptive rather than inferential statistics were used to analyze the data. Test-use rates were examined for differences over time, calculated as absolute change in percentage of people tested in the last year of data compared to the first year.
Medicare covers two distinct populations: those eligible because of age (aged ≥65 years), and those eligible because of a disability (aged <65 years). The aged Medicare population represents approximately 97% of all individuals in the U.S. aged ≥65.23 The Medicare population aged <65 years represents people with disabilities who qualify for Medicare because they receive Social Security Disability Insurance and have met a 24-month qualifying period. Analyses were conducted separately for these two distinct Medicare groups.
Results
In 2005, 29 million FFS Medicare aged and disabled enrollees were eligible for CRC testing, an 11.6% increase over the eligible population in 1998 (Table 1). The growth in the population eligible for CRC testing was uneven across demographic groups—the younger enrollee population (aged 50–64 years) increased 50.5% compared to 8.3% for the population aged ≥65 years. In the group aged ≥65 years, greater growth was also seen among men (13.1%) compared to women (5.2%), but an opposite pattern was seen in the group aged 50–64 years, with the female population increasing 58.9% compared to an increase of 43.5% among the male population. Large differences were seen in the race classifications of enrollees eligible for CRC testing, primarily due to the classification of fewer enrollees in the unknown race category in more recent years.
Table 1. U.S. fee-for-service Medicare population eligible for colorectal cancer testing, 1998, 2002, and 2005
| 1998 | 2002 | 2005 | Change 2005–1998 (%) | |
|---|---|---|---|---|
| AGE ≥65 YEARS | ||||
| Total | 23,960,430 | 25,340,487 | 25,953,663 | 8.3 |
 Age 65–74 years | 12,667,815 | 13,049,939 | 13,288,546 | 4.9 |
| Age 75–84 years | 8,537,446 | 9,305,154 | 9,500,810 | 11.3 |
| Age ≥85 years | 2,755,169 | 2,985,394 | 3,164,307 | 14.8 |
| Female | 14,528,849 | 15,097,053 | 15,283,009 | 5.2 |
| Male | 9,431,581 | 10,243,434 | 10,670,654 | 13.1 |
| Black | 1,754,045 | 1,923,965 | 1,968,119 | 12.2 |
| Asian | 256,102 | 381,775 | 444,937 | 73.7 |
| White | 21,123,208 | 22,261,784 | 22,681,759 | 7.4 |
| Hispanic | 357,926 | 417,449 | 439,965 | 22.9 |
| Native American | 29,376 | 77,223 | 93,487 | 218.2 |
| Other/unknown | 439,773 | 278,291 | 325,396 | –26.0 |
| Not state buy-in | 20,894,843 | 21,951,905 | 22,396,294 | 7.2 |
| State buy-in | 3,065,587 | 3,388,582 | 3,557,369 | 16.0 |
| Disabled | 1,613,668 | 1,805,674 | 1,937,278 | 20.1 |
| Not disabled | 22,346,762 | 23,534,813 | 24,016,385 | 7.5 |
| AGE 50–64 YEARS | ||||
| Total | 2,050,152 | 2,605,875 | 3,086,096 | 50.5 |
| Female | 937,274 | 1,235,807 | 1,489,174 | 58.9 |
| Male | 1,112,878 | 1,370,068 | 1,596,922 | 43.5 |
| Black | 347,035 | 462,489 | 547,545 | 57.8 |
| Asian | 8,669 | 18,304 | 26,277 | 203.1 |
| White | 1,554,358 | 2,003,292 | 2,360,917 | 51.9 |
| Hispanic | 54,183 | 62,977 | 73,634 | 35.9 |
| Native American | 7,341 | 18,417 | 22,802 | 210.6 |
| Other/unknown | 78,566 | 40,396 | 54,921 | –30.1 |
| Not state buy-in | 1,386,526 | 1,694,758 | 1,950,856 | 40.7 |
| State buy-in | 663,626 | 911,117 | 1,135,240 | 71.1 |
During the time period from 1998 through 2005, the use of barium enema and sigmoidoscopy among enrollees aged ≥65 declined; FOBT testing showed a modest increase initially after the introduction of Medicare coverage in 1998 but declined after 2001, when colonoscopy coverage was expanded to include people of average risk (Figure 1). Colonoscopy testing increased each year, with almost 10% of enrollees receiving a colonoscopy in 2005.
Figure 1.
Colorectal cancer test-use trends for U.S. fee-for-service Medicare enrollees aged ≥65 years, 1998–2005
Data sources: Medicare fee-for-service claims for CRC test in all settings conducted in 1998–2005 and the Medicare enrollment database
Although similar patterns were seen among enrollees aged 50–64 years (Figure 2), test-use rates were lower in this population. Rates for barium enema and sigmoidoscopy were low initially and declined through 2005. FOBT use was relatively stable over the study period. Colonoscopy testing increased in each year and surpassed FOBT as the most frequently used test during 2003–2005.
Figure 2.
Colorectal cancer test-use trends for U.S. fee-for-service Medicare enrollees aged 50–64 years, 1998–2005
Data sources: Medicare fee-for-service claims for CRC test in all settings conducted in 1998–2005 and the Medicare enrollment database
For both Medicare groups, colonoscopy was the only CRC test that increased in use in each year of the study (see Appendix C online at www.ajpm-online.net). Although colonoscopy use was slightly higher among the group aged ≥65 years (9.9% in 2005) than the group aged 50–64 years (9.1% in 2005), the younger group experienced a slightly more rapid increase in use of the procedure (4.2%) compared with the group aged ≥65 years (3.7%). Colonoscopy rates for whites and blacks were higher than for other races/ethnicities.
Sigmoidoscopy rates declined from 1998 through 2005 in all demographic strata and both Medicare groups, with a 3.2% decline in the group aged ≥65 years and a 2.1% decline among people aged 50–64 years (see Appendix D online at www.ajpm-online.net). Whites had the highest sigmoidoscopy use in 1998 for both age strata, but by 2005, blacks had slightly higher rates compared with other races/ethnicities in both age strata.
Annual FOBT rates from 1998 through 2005 were higher than for other CRC tests in the population aged ≥65 (see Appendix E online at www.ajpm-online.net). For the population aged 50–64 years, FOBT rates were exceeded by colonoscopy rates in 2003. Overall, FOBT use decreased from 1998 to 2005. The decrease was greater among the group aged ≥65 years (2.3%) than in the group aged 50–64 years (0.7%).
Barium enema was the least frequently used CRC test in the Medicare population in 1998 and declined every year in both Medicare groups (see Appendix F online at www.ajpm-online.net). Decreases in barium enema tests were seen in all demographic categories.
The percentage of enrollees who were CIM increased by 6.6% from 2002 to 2005 among enrollees aged ≥65 years and by 5.3% among those aged 50–64 years (Table 2). Overall, 2005 CIM rates were substantially higher among those aged ≥65 years (47.2%) compared with those aged 50–64 years who are eligible for Medicare due to disability (33.4%). Racial disparities were observed in the CIM rate for the aged population, with the rate for whites substantially higher than the lowest race-specific rate (48.3% vs 32.5% among Native Americans). Among people aged 50–64 years, blacks had the highest CIM rate in 2005 (34.1%); Asians had the lowest (27.2%).
Table 2. Characteristics of U.S. fee-for-service Medicare enrollees current with colorectal cancer testing,a 2002–2005b
| 2002 (%) | 2003 (%) | 2004 (%) | 2005 (%) | Change 2005–2002 (%) | |
|---|---|---|---|---|---|
| AGE ≥65 YEARS | |||||
| Total | 40.6 | 42.7 | 45.0 | 47.2 | 6.6 |
| Age 65–74 years | 40.7 | 42.4 | 44.3 | 45.9 | 5.2 |
| Age 75–84 years | 43.3 | 46.0 | 48.9 | 51.6 | 8.3 |
| Age ≥85 years | 31.7 | 33.9 | 36.5 | 39.0 | 7.3 |
| Female | 40.9 | 43.0 | 45.3 | 47.4 | 6.5 |
| Male | 40.2 | 42.4 | 44.6 | 46.9 | 6.7 |
| Black | 33.5 | 36.0 | 38.7 | 41.1 | 7.6 |
| Asian | 32.7 | 34.2 | 36.1 | 37.4 | 4.8 |
| White | 41.7 | 43.9 | 46.1 | 48.3 | 6.6 |
| Hispanic | 30.2 | 32.2 | 34.5 | 35.9 | 5.7 |
| Native American | 25.2 | 27.6 | 29.7 | 32.3 | 7.2 |
| Other/unknown | 32.3 | 34.2 | 36.3 | 38.3 | 6.0 |
| Not state buy-in | 42.1 | 44.3 | 46.6 | 48.8 | 6.7 |
| State buy-in | 30.9 | 32.9 | 35.1 | 36.9 | 6.1 |
| Disabled | 38.1 | 40.7 | 43.4 | 46.1 | 8.0 |
| Not disabled | 40.8 | 42.9 | 45.1 | 47.3 | 6.5 |
| AGE 50–64 YEARS | |||||
| Total | 28.1 | 29.8 | 31.6 | 33.4 | 5.3 |
| Female | 32.7 | 34.5 | 36.5 | 38.4 | 5.7 |
| Male | 23.9 | 25.5 | 27.1 | 28.6 | 4.7 |
| Black | 27.9 | 30.0 | 32.2 | 34.1 | 6.2 |
| Asian | 24.0 | 25.1 | 26.4 | 27.2 | 3.2 |
| White | 28.4 | 30.0 | 31.7 | 33.4 | 5.1 |
| Hispanic | 26.8 | 28.8 | 31.0 | 32.9 | 6.0 |
| Native American | 22.2 | 24.1 | 26.4 | 27.5 | 5.2 |
| Other/unknown | 23.2 | 24.8 | 29.1 | 29.1 | 5.9 |
| Not state buy-in | 26.9 | 28.5 | 30.1 | 31.7 | 4.8 |
| State buy-in | 30.2 | 32.2 | 34.3 | 36.2 | 6.0 |
aCurrent in Medicare (CIM) is defined as having an fecal occult blood test during the year, or a sigmoidoscopy or barium enema in the previous 4 years, or a colonoscopy anytime from 1998 to the year shown. |
bData sources: Medicare fee-for-service claims for colorectal cancer test in all settings conducted in 1998–2005 and the Medicare enrollment database |
Geographic variation in CRC test-use rates was apparent (Figure 3). The national CIM rate was 47.2% among people aged ≥65 years for the most recent year of available data, 2005. State-level CIM rates varied from a low of 39.5% in New Mexico to a high of 54% in Delaware.
Figure 3.
U.S. fee-for-service Medicare enrollees aged ≥65 years current with colorectal cancer tests in 2005, by state
Data sources: Medicare fee-for-service claims for CRC test in all settings conducted in 1998–2005 and the Medicare enrollment database
Discussion
The results of this study offer both encouraging and discouraging news. It demonstrates that as of 2005, nearly half (47.2%) of Medicare enrollees aged ≥65 years were current with CRC testing as recommended in guidelines. This finding is encouraging because it is substantially higher than in the previous estimate of CRC test compliance based on Medicare claims.15 In that earlier study, data from enrollees aged ≥70 years were analyzed to determine whether individuals had complete screening over a 10-year interval. Complete screening was defined as continuous adherence to screening guidelines and indicated that the individual was up to date with screening during the entire 10 years. While this definition captures a historical pattern of screening, it does not reflect the most recent rate of up-to-date screening in the population. If, for example, a person failed to get an FOBT test in the first year of the window but was screened by FOBT every year thereafter, under the complete screening definition used by Cooper and Kou that person would be counted as incompletely screened.
The present study examined the percentage of enrollees who were up to date in any given year, using a measure modeled after the Health Plan Employer Data and Information Set measure.24 In these analyses, individuals who may have been untested in prior years would be included in the CIM counts if they were current with CRC testing in the measurement year. This approach yields higher rates than previous studies,15 where a single year of failure to be tested resulted in the enrollee's being classified as not continually tested. The observed test-use rates, similar to those found in an analysis of the National Health Interview Survey,25 also represent discouraging news. That more than half of Medicare enrollees are untested for a cancer that can be prevented demonstrates the under-use of important tools in the war on cancer. It is also discouraging that the percentage of Medicare enrollees tested according to recommendations increased among those aged ≥65 years by only about 2% per year in the time window of this study and increased slightly less among those aged 50–64 years. This pace is unacceptably slow, given the large numbers of adults still needing CRC tests. Further, the annual rates of use of CRC tests other than colonoscopy (FOBT, sigmoidoscopy, and barium enema) declined, indicating that colonoscopy is becoming the dominant means by which Medicare beneficiaries undergo CRC screening—a finding noted by others.7
One positive implication of the greater use of colonoscopy (in terms of overall CRC screening rates) is that enrollees, once tested, are compliant with recommendations for up to 10 years. However, the percentage of enrollees tested according to recommendations is likely to increase at a slow rate because of limited resources for delivering colonoscopy at the population level.26 Additionally, the decreasing use of lower-cost tests in favor of colonoscopy has financial implications for both the Medicare program and its enrollees.
Additional discouraging news from this study is that disparities in test use are evident. Fewer people who are eligible for Medicare because of a disability—a rapidly growing segment of the Medicare population—are CIM than are those eligible because of age (33% vs 47% in 2005). Native Americans have low CIM rates in both Medicare groups—16% lower than whites among enrollees aged ≥65 years in 2005.
A few factors associated with these data influence the interpretation of these findings. First, data were not available on colonoscopy tests conducted in 1996 and 1997, testing dates that would have qualified a Medicare enrollee as being tested according to guidelines. However, only diagnostic tests for CRC were covered under Medicare during that time period, and colonoscopy use was low. Similarly, tests conducted prior to Medicare enrollment were not included, nor were tests conducted without submitting a claim, such as FOBTs conducted through health fairs. Third, Medicare enrollees participating in managed-care plans were excluded. Nationally, approximately 13% of Medicare enrollees participate in a managed-care plan, although the percentage varies by geographic location. If enrollees in managed care have higher use of preventive services, as has been noted by others,27, 28, 29 then the absence of managed-care data in this study would imply that these results underestimate the true rate. In the aggregate, the impact of these limitations is likely small, which suggests that the true rate of enrollees who are up to date with CRC testing may be slightly higher than reported here. However, it should be noted that enrollees with cancer were not excluded in these analyses and that these test-use rates include diagnostic and surveillance tests in addition to the tests conducted for screening.
This study provides the first national evidence for CRC test-use rates in the Medicare population. Testing in the Medicare population has increased since the introduction of coverage for colonoscopy among average-risk individuals. However, overall testing rates are low, and the pace of progress is slow. If CIM rates in the Medicare population continue to increase at only 2% a year, it may be a decade or longer before CRC test-use rates reach the levels observed for other preventive services. Setting a target screening level that should be achieved for CRC testing is complicated, as not every enrollee is a good candidate for CRC screening.30 Moreover, monitoring the use of CRC screening in the Medicare population has been made more complex by new recommendations from the U.S. Preventive Services Task Force that advise against routine screening after age 752 and by new technologies such as computed tomographic colonography.31 Despite these challenges, efforts to increase the use of CRC tests and to monitor testing rates in the Medicare population should continue.
The analyses on which this publication is based were performed under Contract No. 500-02-NC03 (Utilization and Quality Control Peer Review Organization for the State of North Carolina), sponsored by the Centers for Medicare & Medicaid Services (CMS) with collaboration from the National Cancer Institute (NCI) under inter-agency agreement #Y1-PC-1007. The content of this publication does not necessarily reflect the views or policies of CMS, NCI, or The Carolinas Center for Medical Excellence. The opinions expressed in this paper are those of the authors.
No financial disclosures were reported by the authors of this paper.
Supplementary data
Appendix.
References
- Cancer statistics, 2008. CA Cancer J Clin. 2008;58:71–96
- . Screening for colorectal cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. 2008;149:627–637
- American Cancer Society guidelines for the early detection of cancer: update of early detection guidelines for prostate, colorectal, and endometrial cancers (Also: update 2001—testing for early lung cancer detection). CA Cancer J Clin. 2001;51:38–75
- Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the U.S. Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. Gastroenterology. 2008;134:1570–1595
- . NCD for colorectal cancer screening tests (210.3). 2004;www.cms.hhs.gov/mcd/viewncd.asp?ncd_id=210.3&ncd_version=1&basket=ncd%3A210%2E3%3A1%3AColorectal+Cancer+Screening+Tests
- . Rates and predictors of colorectal cancer screening. Prev Chronic Dis. 2006;3:A117
- . Trends in colorectal cancer testing among Medicare subpopulations. Am J Prev Med. 2008;35:194–202
- . Trends in colon cancer screening procedures in the U.S. Medicare and Tricare populations: 1999–2001. Prev Med. 2006;42:460–462
- . Relation between Medicare screening reimbursement and stage at diagnosis for older patients with colon cancer. JAMA. 2006;296:2815–2822
- Trends in colonoscopy for colorectal cancer screening. Med Care. 2007;45:160–167
- . Effect of Medicare coverage on use of invasive colorectal cancer screening tests. Arch Intern Med. 2002;162:2581–2586
- . Persistent demographic differences in colorectal cancer screening utilization despite Medicare reimbursement. BMC Gastroenterol. 2005;5:10
- . Are people being screened for colorectal cancer as recommended? (Results from the National Health Interview Survey). Prev Med. 2002;35:199–206
- . Racial disparities in the use of and indications for colorectal procedures in Medicare beneficiaries. Cancer. 2004;100:418–424
- . Underuse of colorectal cancer screening in a cohort of Medicare beneficiaries. Cancer. 2008;112:293–299
- . Trends in colorectal cancer screening disparities in people aged 50–64 years, 2000–2005. Am J Prev Med. 2008;35:185–193
- . Racial and gender variation in use of diagnostic colonic procedures in the Michigan Medicare population. Med Care. 1999;37:712–717
- . Racial differences in use of colonoscopy, sigmoidoscopy, and barium enema in Medicare beneficiaries. Dig Dis Sci. 2002;47:2715–2719
- . Colorectal cancer screening in the elderly population: disparities by dual Medicare–Medicaid enrollment status. Health Serv Res. 2006;41:2136–2154
- . Geographic variation among Medicare beneficiaries in the use of colorectal carcinoma screening procedures. Am J Gastroenterol. 2004;99:1544–1550
- Association of regional variation in primary care physicians' colorectal cancer screening recommendations with individual use of colorectal cancer screening. Prev Chronic Dis. 2007;4:A90
- Data sources for measuring colorectal endoscopy use among Medicare enrollees. Cancer Epidemiol Biomarkers Prev. 2007;16:2118–2127
- . Overview of the SEER-Medicare data: content, research applications, and generalizability to the U.S. elderly population. Med Care. 2002;40(8S):IV–3–18
- . Health plan employer data and information set (Technical specifications). Report No. 2. Washington DC: National Committee for Quality Assurance; 2005;
- . Does Medicare coverage of colonoscopy reduce racial/ethnic disparities in cancer screening among the elderly?. Health Aff (Millwood). 2006;25:1153–1162
- Is there endoscopic capacity to provide colorectal cancer screening to the unscreened population in the U.S.?. Gastroenterology. 2004;127:1661–1669
- . Stage of cancer at diagnosis for Medicare HMO and fee-for-service enrollees. Am J Public Health. 1994;84:1598–1604
- . Effect of managed care insurance on the use of preventive care for specific ethnic groups in the U.S. Med Care. 2002;40:743–751
- . Use of preventive services by Medicare fee-for-service beneficiaries: does spillover from managed care matter?. Med Care. 2005;43:445–452
- . Pitfalls of converting practice guidelines into quality measures: lessons learned from a VA performance measure. JAMA. 2004;291:2466–2470
- . Medicare program: meeting of the Medicare Evidence Development and Coverage Advisory Committee. Federal Register. 2008;73:55848–55849
The full text of this article is available via AJPM Online at www.ajpm-online.net.
PII: S0749-3797(09)00203-7
doi:10.1016/j.amepre.2009.03.009
© 2009 American Journal of Preventive Medicine. All rights reserved.
