U.S National Institutes fo Health | cancer.gov NCI Logo

Cancer Intervention and Surveillance Modeling Network

Modeling to guide public health research and priorities

Comparative Analyses

Cost effectiveness of computed tomographic (CT) colonography

The National CT-colonography Trial-NCTC (American College of Radiology Imaging Network [ACRIN 6664]) was a large multi-site study to assess the accuracy of CT colonography for colorectal cancer screening in the general population and in community based practices. Prior studies had been in more specialize settings or with less specialized equipment or training. The results of the main study were published in the New England Journal of Medicine in September 2008. The CISNET groups who model colorectal cancer have collaborated with affiliate member Dr. David Vanness to conduct the cost-effectiveness analysis of the NCTC study using common inputs but independent analysis from the three models.

Colorectal cancer mortality projections Web site

The Colorectal cancer mortality projections Web site provides a modeling tool that projects future trends in colorectal cancer mortality and evaluates how potential increases in prevention, screening, and access to state-of-the-science cancer treatment may affect future mortality trends. It is intended for policy, legislative, and cancer control planning staff at the federal, state, and local levels, as well as advocacy and professional groups. It also features descriptions of and links to the Healthy People 2010 objectives relevant to colorectal cancer. This work was initiated by the Memorial Sloan-Kettering group (MISCAN colorectal model) (Vogelaar 2006); however, the Web site has comparative analyses across two groups: MISCAN and the University of Minnesota group (SimCRC). Users may select to compare intervention scenarios, results across race and gender groups, or results across the two models.

Results show that almost half of all colorectal cancer mortality can be eliminated by 2020 by more fully utilizing cancer control opportunities that we already know are effective. Lower levels of utilization will substantially reduce those gains. While increased usage of state-of-the-art treatment has the most immediate impact on mortality, screening over the longer term has the largest impact. Changing the risk factor profile of the U.S. population to optimistic but still realistic levels will take many years to influence colorectal cancer mortality trends, but the benefits extend  well beyond colorectal cancer.

Adenoma growth rates and the implications for evaluation of colorectal screening

The CISNET colorectal cancer group is investigating the effect of varying assumptions about the adenoma-carcinoma sequence on projected outcomes for different screening strategies.

Despite the fact that all three colorectal CISNET models were all fit to the same observational data on adenoma prevalence and cancer incidence, they differ in the average time between adenoma formation and cancer diagnosis, a measure for which no direct evidence exists. Average estimates range from 10 to 25 years across the three models. For example, in deciding between adding surveillance colonoscopies following adenoma diagnosis or shortening the screening interval from 15 to 10 years, two models show a small advantage for adding surveillance, whereas the third model has a clear advantage for shortening the screening interval. Studies of this type show the extent to which variation in deep parameters of the natural history model (i.e., adenoma growth rates) influence policy choices and target where our knowledge gaps are most critical.

Evaluating test strategies for colorectal cancer screening: A decision analysis for the U.S. Preventive Services Task Force

The U.S. Preventive Services Task Force (USPSTF) requested a decision analysis for colorectal cancer to assist the task force in determining the age to begin screening, age to end screening, and intervals of screening for multiple screening tests. This is the first time that the Task Force will use a decision analysis in combination with a systematic evidence review to inform their decisions. CISNET models provided standardized comparisons of 145 screening strategies using the best available evidence for consideration by the USPSTF. Several of these screening strategies gave similar gains in life-years—provided that there is equally high adherence for all aspects of the screening process. Under these conditions, the best screening strategies were high-sensitivity FOBT (Hemoccult SENSA or fecal immunochemical test) performed annually, flexible sigmoidoscopy performed every 5 years with Hemoccult SENSA performed every 2 to 3 years, or colonoscopy performed every 10 years. Annual FOBT with a lower-sensitivity test (Hemoccult II) and flexible sigmoidoscopy alone resulted in fewer life-years gained relative to other strategies. These analyses confirmed the current recommendation to begin screening at age 50 in the asymptomatic general population and showed that stopping screening at age 75 after consecutive negative screenings since age 50 provides almost the same benefit as stopping at age 85 but with substantially fewer colonoscopy resources and risk of complications (Zauber 2008).

Threshold analyses for novel colorectal cancer screening tests for Centers for Medicare and Medicaid Services evaluation of National Coverage Determinations

New and evolving tests and technologies are being developed for screening for colorectal cancer. The Centers for Medicare and Medicaid Services (CMS) has the responsibility to assess whether a new test should be covered by Medicare, and, if so, at what level of reimbursement. CMS requested that the CISNET colorectal cancer group perform technology assessments as part of National Coverage Determinants to determine the cost-effectiveness analyses of three tests: immunochemical fecal occult blood test (FIT), DNA stool test, and CT colonography. CISNET modelers conducted a cost-effectiveness analysis for FIT in 2003 (van Ballegooijen 2003), for the DNA stool test in 2007 (Zauber 2007), and for CT colonography in 2008. The analyses determined the per-test cost that would provide comparable cost-effectiveness with currently reimbursable tests. The modeled threshold analysis for FIT resulted in a coverage decision for CMS at $22 per test. A coverage decision for the DNA stool test is pending, and the threshold value for CT colonography will be reported in fall 2008.

Potential decrease in costs to Medicare, Medicaid, and private payers due to increased colorectal cancer screening among the U.S. population aged 50-64

Colorectal cancer screening has reduced colorectal cancer incidence and mortality in randomized trials, but acceptance of screening has lagged in the United States. Recently, colorectal cancer screening rates have reached approximately 50% of the eligible population of those 50 and older. The CISNET colorectal cancer modelers are working with CDC, NCI, and CMS to assess whether there would be a potential cost savings to Medicare if there were an increase in colorectal cancer screening in those aged 50-64, and, if so, whether cost shifting to other medical care providers would occur. This analysis will be utilized by CDC and CMS to determine whether increased spending in colorectal cancer screening will result in a cost savings associated with avoiding colorectal cancer and its treatment costs.