Modeling the Cost-Effectiveness of Learning Health Systems in Diagnostic Radiology and Nuclear Medicine: A Theoretical Framework Using Healthcare Economic Evaluation Techniques

Abstract Diagnostic radiology and nuclear medicine are increasingly scrutinized for their value, demanding systems that improve quality while controlling costs. This paper proposes a theoretical economic evaluation framework to model the cost-effectiveness of implementing a Learning Health System (LHS) in this domain. Utilizing a Markov model structure, the framework compares the long-term costs and outcomes of a Standard Care pathway versus an LHS, focusing on key economic drivers such as the reduction in repeat scan rates, optimization of staff workflow, and avoidance of high-cost adverse radiation events. The analysis adopts a hospital/payer perspective over a 5-year time horizon with a 3% annual discount rate. Illustrative quantification, based on a hypothetical cohort of 30,000 annual exams, demonstrates that the LHS can generate significant annual net savings (€115,000) and achieve a rapid payback period (0.78 years). Crucially, the model suggests that the LHS operates in the dominant quadrant of the cost-effectiveness plane, with an Incremental Cost-Effectiveness Ratio (ICER) of approximately -€3,842 per QALY gained. A Probabilistic Sensitivity Analysis (PSA) confirms the robustness of this conclusion, showing a 100% probability of cost-effectiveness at a willingness-to-pay threshold of €30,000/QALY. This theoretical work provides a necessary foundation for future empirical studies and supports the strategic investment in continuous learning infrastructure within high-volume imaging departments.