A decision support model for colorectal cancer screening

With minor differences, most national colorectal cancer (CRC) screening programs in Europe consist of one-size-fits-all aged-based strategies. This paper provides a decision analysis-based approach to personalized CRC screening in a general population setting, supporting decisions concerning whether and which screening method to consider and/or whether a colonoscopy should be administered.

We use an influence diagram model characterizing CRC risk with respect to different variables of interest, and including comfort, costs, complications, and information as decision criteria, the last one assessed through information-theoretic measures. The criteria are integrated with a multi-attribute utility model. Optimal screening policies are then computed by maximizing expected utility.

The proposed model is used to support personalized individual screening based on relevant features beyond age. It serves to assess existing national age-based screening programs as well as design new risk-based ones. In particular, it suggests replacing current age-based strategies prevalent in many European countries by more personalized strategies based on risk dependent on individual features. Additionally, the model facilitates benchmarking of novel screening devices. Software to implement the model and reproduce the results is included.

This work develops a framework for personalized CRC screening that improves upon current age-based screening strategies and highlights how CRC screening strategies could be redesigned and optimized.