OBJECTIVES: This research uses a comparative analysis framework between national health care systems and continues the theoretical development of the 3P theory.  It demonstrates that sets of cost reduction strategies and by physicians, Pharmacists and Patients as well as different meanings of cost (cost the system, cost to the physician, and cost to the patient) choices per group of physicians lead to very different decision points in each system and variations in sets of clinical choices for similar patients. A random utility model is proposed. METHODS: Data are extracted from the endep/biomed database for 600 physicians, transcripts from qualitative focus groups and estimates from the centralized database of 6 patients’ surveys on cost of medicines (www.endeplux.org). The thinking about cost is classified in Cost S (cost to the System), Cost Ph (cost to the Physician) Cost Pa (cost to the Patient).The conceptual framework has been mainly developed from pair of country comparisons, especially from the French, German and Italian physicians analysis (Huttin, Andral; 2003; Atella et als; 2003; Brenner et als, 2002). RESULTS: A comparative intercountry framework is used to weight differently combinations of (Cost S, Cost Ph, Cost Pa) in the system. A generalization will be proposed with a list of different possible combinations aCostS+bCostPh+CostPa for each physician ij among N physicians in J Health financing systems. CONCLUSIONS: This research step aims to propose a link between a research line on transaction cost politics and several statistical developments for a stated revealed preference disease econometric model. It will help to identify the type of random utility models that would clearly model how variations of preferences from Physicians, Pharmacist , Patients that could help to manage variations between different national health care financing systems.