The goal of the collaborative pre-workshop exercise was to shed light on the degree to which difference across models could be reduced by standardizing the values of key input parameters, or assumptions, across models. Any residual variation in model outcomes would be the result of differences either in parameters that remained unstandardized or in the structure of the models themselves. Secondary objectives were to demonstrate the benefit of collaborative interactions among modelers and to ascertain the research resources (time and money) required to mount such exercises.
Five research teams with published CEAs of colorectal cancer screening agreed to participate in a comparative modeling exercise to further explore the reasons for disparate cost-effectiveness findings. Each of the models can track (via computer) a hypothetical cohort of average-risk Americans, beginning at age 50, over their remaining lifetimes and can estimate the number of years of life lived and the medical costs incurred by the members of that cohort.1 The participating research teams were:
The Harvard Model (Frazier et al., 2000), led by Karen Kuntz, Ph.D.;
The Ladabaum Model (Ladabaum et al., 2004a; Song et al., 2004), led by Uri Ladabaum, M.D.;
The Miscan Model (Loeve et al., 1999, 2000), led by Marjolein van Ballegooijen, M.D.;
The Vanderbilt Model (Ness et al., 2000) led by Reid Ness, M.D.; and
The Vijan Model (Vijan et al., 2001), led by Sandeep Vijan, M.D.
At the workshop, each team leader described essential features of the model’s structure and assumptions. (See the appendixes with speakers’ presentations.) The teams further agreed to provide cost-effectiveness results for a set of five specific screening strategies across 10 different combinations of assumptions, starting with the assumptions in their original models.
All the strategies included in the pre-workshop exercise envisioned periodic screening of all average-risk Americans beginning at age 50 and ending at age 80. The five selected strategies were:
F/S: Annual fecal occult blood testing in combination with a flexible sigmoidoscopy every five years;
S: Sigmoidoscopy every five years;
R: A prototype radiology procedure every five years, with specific test characteristics and costs;
C: Colonoscopy every 10 years; and
F: Annual fecal occult blood testing.
These strategies were selected not for any posited superiority over other CRC screening approaches, but for the frequency with which they are advocated by practitioners today. Some of them represent strategies that have been recommended by professional groups (Smith et al., 2004; U.S. Preventive Services Task Force, 2002; Winawer et al., 2003). They also represent a wide range of procedure cost and test accuracy.