| To Register SMDM Homepage | |
|
|
|
||||
Many clinical problems involve sequential decisions and multiple strategies. Current analytic software enables decision analysts to determine the optimal first decision, but it is difficult to identify recommended downstream decisions and to perform sensitivity analyses. The diagnosis and management of solitary pulmonary nodules (SPN) is one such clinical area in which multiple combinations of sequential diagnostic tests exist. We sought to develop a software tool which would facilitate the evaluation of such clinical problems.
Using Decision Maker, we developed a Markov model to identify optimal strategies for patients with SPN, comparing 40 clinically plausible sequences of 5 diagnostic interventions for SPN management (CT, PET, biopsy, surgery, and watchful waiting). We developed ALCHEMIST-Frontier, a web-based tool which uploads Decision Maker models and evaluates both costs and effectiveness of all possible strategies. ALCHEMIST-Frontier identifies strategies that are eliminated by strict or extended dominance, plots the remaining strategies on an efficiency frontier, and calculates incremental cost-effectiveness ratios by comparing strategies with the next most effective, non-dominated alternative. ALCHEMIST-Frontier also enables the user to reduce the number of potential strategies by either requiring a particular test be performed, or by eliminating any strategies that include a specified test.
ALCHEMIST-Frontier evaluated the SPN model and displayed both graphically and in a table the 40 different strategies and their lifetime costs and life expectancy. In patients with low pre-test probability of malignancy, 35 strategies were eliminated either through strict or extended dominance. Assuming a cost-effectiveness threshold of $50,000/QALY, the optimal strategy began with CT and used PET imaging selectively when CT results were indeterminate, surgery when PET results were positive, and needle biopsy when CT results were benign or when PET results were negative. When the analyst indicated that PET was unavailable, ALCHEMIST-Frontier eliminated 33 strategies that included PET and 3 of the remaining strategies by strict dominance. The most effective strategy that cost less than $50,000/QALY was to perform CT-guided needle biopsy in all patients, to choose surgery when biopsy results were malignant, and watchful waiting when biopsy results were benign or non-diagnostic.
ALCHEMIST-Frontier facilitates analysis of decision models with sequential decisions and multiple strategies. While we have demonstrated here its utility in the management of SPN, it can be readily used to analyze decision models in other clinical domains.
See more of Poster Session - Utility Theory; Health Economics; Patient & Physician Preferences; Simulation; Technology Assessment
See more of The 26th Annual Meeting of the Society for Medical Decision Making (October 17-20, 2004)