There are several technologies available to screen for colorectal cancer (CRC). While prior research suggests that screening with available technologies is cost-effective, patient adherence with available screening technologies is poor. Major efforts are being made to develop new screening techniques. Among the most promising of these is the use of genetic markers. Two major types of genetic screening have been proposed and evaluated in early studies. The first is the use of stool DNA markers of neoplasia to identify subjects with large adenomatous polyps and/or CRC. The second is the use of serum markers of the risk of developing polyps as a risk-stratification marker for screening. Specifically, the project aims are: 1) To estimate the cost-effectiveness of screening for CRC with stool DNA markers; 2)To estimate the cost-effectiveness of risk-stratifying subjects based upon serum markers of polyp and CRC risk; 3)To explore the impact of variation in the accuracy of risk markers, costs, and adherence on the costs and effectiveness of screening programs; and 4)To explore the degree to which genetic screening can improve the efficiency of screening programs by reducing demand for the limited resource of colonoscopy. Methods: Study Design: Simulation modeling. Subjects: A modeled cohort of patients 50 to 80 years of age at average-risk for CRC. Setting: the US health care system. Methods: The investigators will use a Markov simulation model that they have previously published as the basis of the analyses. The model will be adapted and updated in order to examine the costs and effectiveness genetic-based screening technologies. For stool DNA markers, we will use test characteristics from published and ongoing trials to estimate cost effectiveness compared to current screening methods. However, we will also explore in detail several novel factors that may impact cost-effectiveness: the effect of test accuracy on screening frequencies, and based on incremental cost-effectiveness analysis, the optimal screening frequency; the possible impact of tradeoffs between frequency and adherence; the impact of using stool DNA for surveillance for those with polyps; and the impact of a more efficient mode of screening on the demand for follow-up colonoscopy. For serum markers, more extensive model revisions will be required. In particular, the effectiveness of a serum-screening test lies in its ability to stratify subjects into high and low-risk subgroups; thus the natural history model will be altered to fit a stratified distribution of risk. This will allow detailed exploration of the possibility of individualizing screening recommendations based upon level of risk. For both types of analyses, we will conduct detailed multi-way sensitivity analyses using Monte Carlo modeling of population risk-distribution, test cost and accuracy, and adherence. Outcome measures: The risks of developing or dying from CRC, the costs of screening, and the incremental cost-effectiveness of various screening regimens.
| Sussman, Jeremy B; Vijan, Sandeep; Choi, HwaJung et al. (2011) Individual and population benefits of daily aspirin therapy: a proposal for personalizing national guidelines. Circ Cardiovasc Qual Outcomes 4:268-75 |
| Vijan, Sandeep; Hwang, Inku; Inadomi, John et al. (2007) The cost-effectiveness of CT colonography in screening for colorectal neoplasia. Am J Gastroenterol 102:380-90 |
