Cost-effectiveness on age- and gender-specific CRC screening strategies

Carolina Vivas-Valencia, Purdue University


Colorectal Cancer (CRC) is the third most commonly diagnosed cancer and third leading cause of cancer death in the US. CRC is known as a silent disease. This is because no symptoms that lead to diagnostic screening are easily detected on earlier CRC stages and most patients are not aware of CRC presence until it advances to a later stage where the survivability is low. However; effective screening allows preventing CRC by enabling the detection and removal of polyps before it progress to cancer. Many organizations recommend a screening colonoscopy program to all Americans at age 50, this recommendation is based on economical prediction models that simulate the natural history of CRC; however current prediction models not consider any age and gender difference during simulation. Recent clinical evidence has suggested that men tend to face noticeable risk of CRC incidence at lower ages than women. In this thesis, I extended the Vanderbilt-NC State model to account for age gender-specific population subgroups and performed economic studies that allowed us to make recommendations for personalized screening strategies. To extend the Vanderbilt-NC State model, I developed a calibration methodology based on Design of Experiments, which allows the estimation of model parameters that have a higher impact in the cancer response. After calibration, the extended model was validated with respect to the CRC incidence. Finally I conducted a cost-effectiveness analysis considering age, race and gender subgroups, and four screening strategies that were compared against the natural history of CRC. As a result of this work, I was able to propose a new and more efficient calibration procedure based on model-based parameter estimation for personalized CRC screening strategy that takes into consideration the natural history of the disease. This procedure can be further applied to other simulation-based economic studies on individualized medicine. Finally, simulated screenings strategies with the updated Vanderbilt-NC State model presented in this thesis, suggest that men should start a screening program at age 40, and sigmoidoscopy is the strategy that saves both cost and lives.




Kong, Purdue University.

Subject Area

Biomedical engineering|Health care management

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server