Background

Colorectal Cancer as a Public Health Concern

• Colorectal cancer is the second leading cause of cancer deaths in the United States (1).
• Estimates show that more than 52,000 Americans will die from the disease in 2007 (1).
• Early detection and treatment are the best defenses against colorectal cancer. The five-year survival rate for cancers caught early is more than 90% (2).

The Importance of Screening

Most health groups, including the American Cancer Society (3), the National Colorectal Cancer Roundtable, U.S. Preventive Services Task Force (4) and the GI consortium, support routine screening for average-risk persons age 50 and older with any ONE of the following options:

• yearly fecal occult blood test (FOBT)
• flexible sigmoidoscopy every 5 years
• combination of annual fecal occult blood tests and flexible sigmoidoscopy every 5 years
• colonoscopy every 10 years
• a double contrast barium enema every 5-10 years as an alternative to sigmoidoscopy or colonoscopy

Medicare Coverage of Colorectal Cancer Tests

Age is a major risk factor for colorectal cancer (1). Thus, the disease imposes a significant burden on the Medicare population (5). Medicare covers colorectal cancer screening for enrollees ages 50 and older, covering four screening tests (FOBT, sigmoidoscopy, colonoscopy and barium enema), each with a different coverage schedule (Appendix 1). Prior to implementation of this benefit (1998), Medicare paid for colorectal cancer tests only if they were billed as diagnostic tests.

The CMS Colorectal Cancer Screening Projects

In October 2000, the Centers for Medicare & Medicaid Services awarded The Carolinas Center for Medical Excellence (CCME) a colorectal cancer screening project that had two components:

• Design and evaluation of interventions to increase colorectal cancer screening. The National Cancer Institute collaborated with CCME through an inter-agency agreement. A description of the interventions and evaluation results can be found at: www2.thecarolinascenter.org
• Calculating national and state rates on the use of colorectal cancer tests. CCME responded by producing a web-based report summarizing the national, state and colorectal cancer tests.
• Since the first CMS colorectal cancer screening project, CCME has been awarded additional contracts to update the Medicare colorectal cancer testing rates:
• This report is the result of our fourth data update and contains test use information from 1998 – 2005.

Methods

Population

Medicare enrollees were classified as “eligible” for colorectal cancer tests and included in rates if they met the following criteria for the year of measurement:

• Age 50 to 114
• Continuous enrollment in Medicare Part B for at least one year with no more than a 30-day lapse in coverage
• Non-HMO Medicare enrollees
• Alive at the end of the year

Data Sources

The data for these analyses come from Medicare enrollment and claims files. We examined paid claims for four screening and diagnostic colorectal cancer tests: FOBT, sigmoidoscopy, colonoscopy and barium enema, conducted in hospitals (both inpatient and outpatient) or physicians’ offices between January 1, 1998 and December 31, 2005. Tests were identified using ICD-9, CPT and HCPCS codes (Appendix 2).

Test Rate Calculations

We calculated the percent of eligible Medicare enrollees who had been tested for each type of test, each year and across the eight-year window of available data. The following rates are included in the report:

• Current in Medicare – the percent of eligible Medicare enrollees who were current with colorectal cancer testing based on Medicare claims between 1998 and the measurement year. Enrollees are reported as current in Medicare if they had at least one of the following: an FOBT in the year of measurement; a sigmoidoscopy in the previous five years; or a colonoscopy claim in 1998 forward. This measure is only available for 2002-2005.
• Any CRC test – the percent of eligible Medicare enrollees who had any one of the four recommended colorectal cancer tests during the report interval: one colonoscopy OR one sigmoidoscopy OR one barium enema OR one FOBT. This rate is presented for each year.
• FOBT, COLO, SIGM and BE test use rates – the percent of eligible Medicare enrollees each year that had the test of interest during the year. Within the test-specific rate, enrollees were counted only once even if they had multiple tests (for example: a person who had two colonoscopy tests would be counted only once in the COLO rate). Enrollees who had more than one type of test (such as an FOBT and a sigmoidoscopy) were included in the rates for each type of test.

The criteria for calculating the rates presented in this report are detailed in Appendix 3.

Which test use rate should you use? The answer depends on your purposes. If you want to know:

• The percent of Medicare enrollees tested according to recommended guidelines
  • Unfortunately, this rate is not available. The rate which most closely captures this information is the Current In Medicare (CIM) rate with the following caveats:
    • The repeat interval for colonoscopy is 10 years and this report only includes 8 years of claims
    • Medicare has only covered screening colonoscopy for average risk since 2001
    • Some fecal occult blood tests will be paid for out-of pocket or completed without submitting Medicare claims
    • Tests occurring prior to enrollment in Medicare are not included
• The percent of Medicare enrollees getting CRC tests in any year
  • Use the Any CRC test rate for the year of interest.
• The percent of Medicare enrollees using a specific type of test
  • Use the FOBT, SIGM, COLO and BE Rates, whichever is of interest, for the year and location of interest

Analyses were conducted separately for each state, with national analyses representing the sum of all states, Puerto Rico, Washington DC, and the Virgin Islands. Within each state, county level test use rates are presented. Rates for multi-state regions used by CMS for the quality improvement program are also available.

The following Medicare enrollee characteristics were used in stratified analyses to examine variations in test use rates:

• Age in four groups: 50-64, 65-74, 75-84, and 85 and older
• Ethnicity in six groups: African American, Caucasian, Asian, Hispanic, North American Native and Other
• Gender
• Medicaid eligibility (Y/N) based on any one month of eligibility during a calendar year
• Eligible for Medicare due to a disability (Y/N) based on original or current reason for eligibility

For the rate that spans multiple years (CIM) enrollees were categorized based upon the characteristics associated with their last year of eligibility.

Results

Demographics of the Population

In 2005 there were 29.3 million Medicare enrollees eligible for colorectal cancer screening. A few highlights of the eligible population:

• Persons ages 65-74 comprised the largest eligible group (46%)
• There were more females (58%) than males
• Whites were the largest racial group (86%)
• Seventeen percent (17%) of the population were eligible for both Medicare and Medicaid
• Persons eligible for Medicare because of a disability made up 17% of the population

National Test Rates

The percent of enrollees who have Medicare claims indicating they are current with testing has risen over the years from 39.3% in 2002 to 45% in 2005.

A few highlights of test use variation from 2005 across demographic groups:

• Current in Medicare test rates were higher among females (46%) than males (44%)
• Current in Medicare test rates were highest among Caucasians (46%) followed by African Americans (39%), Asians (37%), persons of Hispanic descent (35%) and Native Americans (31%)
• There was considerable disparity between the test rates for those eligible for only Medicare (47%) and persons eligible for both Medicare and Medicaid (37%)

In 2005, FOBT was the most commonly used test with a rate of 10.3%, followed closely by colonoscopy (9.8%). The test use rates for sigmoidoscopy and barium enema were both under 1%.

State Rates

There was considerable variability in the testing rates across the states and territories Delaware and Florida had the highest Current in Medicare Test rates (53% and 51% respectively). Among the 50 US states, New Mexico had the lowest Current in Medicare Test rate with 38%. The Virgin Islands had the lowest Current in Medicare rate measured: 27%.

Trends

The proportion of Medicare enrollees receiving a CRC test in any given year has changed very little since 1998 when the CRC screening benefit was first introduced. In 1998, 19.5% of the eligible population had a CRC test. In 2005, 18.8% of enrollees were tested.

However, the proportion of enrollees receiving colonoscopy has increased: from 6% in 1998 (when only high risk enrollees were eligible), to 9.2% in 2002 (when coverage for average risk enrollees was added), to 9.8 in 2005

Because colonoscopy needs to be repeated only every 10 years, the shift from other types of tests to colonoscopy has resulted in an increasing Current in Medicare rate: from 39.3% in 2002 to 45% in 2005

Discussion

Observations

By any measure shown in this report, the Medicare covered colorectal cancer screening benefit is underused. Although no part of the Medicare population has high rates of test use, some groups are clearly at risk of not being tested. Disparities in test rates can be seen by age, with lower rates for those aged 50-64 and those aged 85 and older, by race, with African Americans and other minority races having lower rates than Caucasians, and by eligibility status, with those eligible for Medicaid, and persons eligible for Medicare because of a disability having lower test use.

Next Steps

• Physicians and other health care providers can use the data from this report to assess their community colorectal cancer screening levels and to identify groups of patients that are at risk of not being tested.
• Local, state and national groups interested colorectal cancer screening can use the data from this report to monitor progress with interventions, or develop additional targets for programs designed to increase colorectal cancer test use.
• For a brief summary of promising interventions for increasing colorectal testing rates in the Medicare population, see the summary of the CCME pilot project at www2.thecarolinascenter.org or visit some of the sites included in the links section.