Health Economics in Clinical Practice

Abstract

Health economics -- the discipline that evaluates the costs, consequences, and value of healthcare interventions -- has traditionally operated at a distance from clinical practice, influencing reimbursement and guideline decisions at the policy level but rarely informing individual patient care. This disconnect persists despite the fact that every clinical decision is implicitly an economic decision: choosing between two equally effective treatments that differ in cost by EUR 50,000 per year is as much an economic question as a clinical one. We evaluated five frameworks for integrating health economic evidence into clinical decision-making across 186,424 treatment decisions at four European academic health systems (Vienna, Berlin, Stockholm, and Paris) from 2017 to 2022, covering oncology, cardiology, rheumatology, and diabetes management. We compared: traditional cost-effectiveness analysis (CEA) informing guidelines, real-time cost transparency tools at the point of prescribing, value-based treatment algorithms, patient-level micro-costing with individual ICER estimation, and ML-predicted cost-effectiveness using patient-specific treatment response prediction. Our Clinical Health Economics Score (CHES) integrates value identification accuracy, cost awareness impact, prescribing behaviour change, patient outcome maintenance, and clinician acceptability. ML-predicted individual cost-effectiveness achieved the highest CHES (0.924) by identifying patients for whom expensive therapies are cost-effective (predicted responders) and those for whom cheaper alternatives are equally beneficial, reducing per-patient therapeutic costs by 18.4% without measurable outcome compromise.