FEASIBILITY AND ACCEPTABILITY OF MINIMAL MODELING VALUE OF INFORMATION ANALYSES FOR REAL-TIME PRIORITIZATION DECISIONS WITHIN A LARGE CANCER CLINICAL TRIALS COOPERATIVE GROUP
Author(s)
Bennette CS1, Veenstra DL1, Basu A1, Ramsey S2, Carlson JJ1
1University of Washington, Seattle, WA, USA, 2Fred Hutchinson Cancer Research Center, Seattle, WA, USA
OBJECTIVES: Value of Information (VOI) analyses can help align research investments with areas that could have the greatest impact on patient outcomes, but many questions remain concerning its feasibility and acceptability to inform real-world prioritization decisions. Our objective was to develop a process for calculating VOI in “real time” to inform trial funding decisions within SWOG, a large cancer clinical trials group. METHODS: We adapted a novel and efficient modeling approach - minimal modeling VOI - using a sample of nine phase II/III trial proposals from the Breast, Gastrointestinal, and Genitourinary committees reviewed by SWOG’s leadership between 2008-2013. We created decision models for each trial proposal and devised an efficient process to characterize prior uncertainty in treatment effect by linking evidence-based assumptions in a trial’s sample size calculations with the historical success rates of SWOG trials. Expected clinical and economic VOI was calculated using Bayesian updating methods. We customized the process using iterative stakeholder input. RESULTS: The VOI modeling process was feasible and sufficiently captured key expected differences in comprehensive outcomes and attendant uncertainty for 8 of 9 trial proposals. Model construction and calculations took one researcher <1 week per proposal. We accommodated stakeholder input by: a) deconstructing VOI metrics into expected health benefits and incremental healthcare costs, b) assuming treatment decisions were based on health benefits alone, and c) providing both individual and population level results. Following this customization, SWOG generally accepted the VOI framework and results for the retrospective analyses and felt that VOI analyses would likely be useful in informing future trial proposal evaluations. CONCLUSIONS: We developed an efficient and customized process for calculating the expected VOI of cancer clinical trials that is feasible for use in real-time decision-making and is acceptable to stakeholders. Prospective use and assessment of this approach is currently underway within SWOG.
Conference/Value in Health Info
2015-05, ISPOR 2015, Philadelphia, PA, USA
Value in Health, Vol. 18, No. 3 (May 2015)
Code
PRM63
Topic
Methodological & Statistical Research
Topic Subcategory
Modeling and simulation
Disease
Oncology