With minor differences, most national colorectal cancer (CRC) screening programs in Europe consist of one-size-fits-all age-based strategies. In a joint effort with the UNIRIOJA team, researchers from the Mathematical Sciences Institute (ICMAT) in Spain have developed a decision model for screening allocation that considers CRC risk factors beyond age, including body mass index, smoking, alcohol consumption, diabetes, hypertension and some others.
Represented as an influence diagram, the decision model can identify potential screening strategies, their impacts, and how to evaluate them. The model considers attributes influencing the decision, such as the implementation costs of the screening method, the associated comfort level, potential complications that may occur during the process of screening, and the amount of information each method can provide. A general “parametrizable utility function” then combines all these criteria while accounting for the decision maker’s risk aversion. As such, a screening protocol is devised for an individual based on a “maximum expected utility” algorithm.
Several use cases are considered, including examples underscoring the need for personalised screening strategies that adapt to CRC risk; the assessment of a national strategy showcasing how its design may be improved for better identification of CRC positive cases; and a methodology to assess and benchmark new screening tools and methods, such as those soon to be delivered in ONCOSCREEN, and determine which of them are more closely related to people’s risk patterns.
The details of this publication are as follows:
Corrales, D., Insua, D. R., & González, M. J. (2025). A decision support model for colorectal cancer screening. Computers in Biology and Medicine, 196, 110755. https://doi.org/10.1016/j.compbiomed.2025.110755.