Early cancer detection is critical for improving survival rates. However, current population-level cancer screening programmes often adopt a one-size-fits-all approach, leading to potential inefficiencies. Two key sources of inefficiency are high false-positive rates and substantial unnecessary follow-up examinations, such as biopsies, which can cause patient anxiety, overtreatment, and additional healthcare costs. While more frequent screening may improve early detection, it can also increase these burdens. Public healthcare organizations continue to debate the optimal age to begin and end screening and how frequently to screen to balance early detection benefits against potential harms.
Although previous research has advanced personalized screening, screening decisions and post-screening diagnostic decisions are often studied separately. However, these decisions are inherently interconnected: screening frequency influences the occurrence of false positives and the need for follow-up diagnostics, while diagnostic outcomes can inform the timing and intensity of future screening. Studying these decisions independently may lead to suboptimal care pathways.
This research addresses these challenges by integrating screening and diagnostic decisions within a unified framework to develop adaptive, personalized policies that guide patients throughout the care pathway. The proposed framework will be applied to breast cancer screening as a key application area.