Medical economics of whole-body FDG PET in patients suspected of having non-small cell lung carcinoma--reassessment based on the revised Japanese national insurance reimbursement system.

Abstract

Focusing on the savings expected from the revised Japanese national insurance reimbursement system in the management of patients suspected of having non-small cell lung carcinoma (NSCLC), cost-effectiveness was assessed using decision tree sensitivity analysis on the basis of the 2 competing strategies of whole-body FDG PET (WB-PET) and conventional imaging (CI). A WB-PET strategy that models dependence upon chest FDG PET scan, WB-PET scan, and brain MR imaging with contrast was designed. The cost of a FDG PET examination was updated and determined to be US dollar 625.00. The CI strategy involves a combination of conventional examinations, such as abdominal CT with contrast, brain MR imaging with contrast, and a whole-body bone scan. A simulation of 1,000 patients suspected of having NSCLC (Stages I to IV) was created for each strategy using a decision tree and baselines of other relevant variables cited from published data. By using the WB-PET strategy in place of the CI strategy for the management of patients suspected of having NSCLC in hospitals with an NSCLC prevalence of 75%, the cost saving (CS) for each patient would be US dollar 697.69 for an M1 prevalence of 20% and US dollar 683.52 for an M1 prevalence of 40%, but the CS gradually decreases as the NSCLC prevalence increases. The break-even point requires less than an 80% prevalence in order for the WB-PET strategy to gain life expectancy (LE) per patient. By using the WB-PET strategy in place of the CI strategy for the management of patients suspected of having NSCLC in hospitals with an NSCLC prevalence of 75%, the gain in LE for each patient would be 0.04 years (11.06 vs. 11.02 years) for an M1 prevalence of 20% and 0.10 years (10.13 vs. 10.03 years) for an M1 prevalence of 40%. The maximum cost of a PET study without losing LE would be US dollar 1322.68 per patient for prevalences of 75% NSCLC and 20% M1 disease. The present study quantitatively showed WB-PET, employed in place of CI for managing NSCLC patients, to be cost-effective in the Japanese revised insurance reimbursement system. However, the present cost is very low from the industrial viewpoint.