Seeing Is Believing: Graphical Depictions of Low-Probability Outcomes for Eliciting Treatment-Related, Benefit-Risk Tradeoff Preferences

• Activity Number: 440
• Type: Invited
• Date/Time: Wednesday, August 3, 2011 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Reed Johnson*+ and Lauren Donnalley and Juan Marcos Gonzalez
• Companies: RTI International and RTI International and RTI International
• Keywords: Numeracy ; Benefit-Risk Analysis ; Health Risk Communication ; Visual Aids

Abstract

Individuals with low numeracy skills have been shown to have different perceptions of presented risk then those with high numeracy skills. Effective risk communication is essential and challenging when asking subjects to evaluate treatment benefits and risks. Researchers have proposed visual aids for communicating quantitative risks; however, empirical evidence is sparse about which format is most effective for specific judgment tasks. This study evaluated graphical depictions of medical-outcome probabilities from 19 otherwise similar discrete-choice experiments conducted in 9 different countries (n=8,111) to evaluate how graphic format affected understanding of numerical probabilistic risks. Three risk-grid variations were tested with either 100 or 1000 units. Random-effects panel regression analysis of correct interpretation of the graphic parameters yields negative and significant parameters for age and not being a US subject, and positive and significant parameters for education and humanoid icons compared to square or dot icons. Better comprehension of quantitative risks also is shown to improve the statistical precision of preference-parameter estimates.