Prospective iterative data visualization (DV) study to enhance health literacy in prostate cancer (PC).

Daniel Eidelberg Spratt University Hospitals Seidman Cancer Center, Case Western Reserve University School of Medicine, Cleveland, OH info_outline Daniel Eidelberg Spratt, Patrick Bingham, John Randall Eckardt, Shelby Moneer, Matthew Pagano, Kirsten York

Authors Daniel Eidelberg Spratt University Hospitals Seidman Cancer Center, Case Western Reserve University School of Medicine, Cleveland, OH info_outline Daniel Eidelberg Spratt, Patrick Bingham, John Randall Eckardt, Shelby Moneer, Matthew Pagano, Kirsten York Organizations University Hospitals Seidman Cancer Center, Case Western Reserve University School of Medicine, Cleveland, OH, ZERO Prostate Cancer, Alexandria, VA, Janssen Global Medical Affairs, Raritan, NJ, buzzback, New York, NY, ENTRADA, New York, NY Abstract Disclosures Research Funding Janssen Background: Methods used to visualize and present comparative effectiveness research (CER) impact patient (pt) and caregiver (CG) understanding and ultimately treatment decisions. Evidence-based approaches to optimize DV and comprehension in PC are lacking. We studied the presentation and DV of CER with key stakeholders aiming to overcome health disparity/inequity barriers related to understanding of CER DV. Methods: A panel representing pt and advocate, industry, and health care provider viewpoints provided qualitative feedback on multiple CER DV methods and generated a survey for US PC pts and CGs. Respondents were identified by a market research agency (buzzback, n=192) or a pt advocacy group (ZERO Prostate Cancer, n=58). Respondents first reported on comfort with and understanding of scientific data on PC using a Likert scale, then evaluated a randomly assigned graphic (1 of 2 versions each depicting the same information) for each of 4 topics. For efficacy and prostate-specific antigen (PSA) data, figures showed superiority of treatment over placebo; for safety, figures positively or negatively represented outcomes. For quality of life (QoL), overall QoL and specific QoL domains were illustrated. For each graphic, respondents first suggested their own key takeaway, then selected the best from 4 options, and reason for selection. Graphics were then assessed side by side with better key takeaway selected to create an overall story. Results: Respondents (pts, n=199; CGs, n=51) enrolled between Aug and Sep 2023; 54% were non-White; 52% had an annual income <$75,000/yr; 45% had no college degree. Some respondents reported being uncomfortable/neutral toward PC scientific data (26%) or found data difficult to understand (35%) or overwhelming (29%). >70% identified correct takeaways for Efficacy and PSA. Figures with numerical data showing a difference between treatment groups (Efficacy, PSA) more effectively communicated key takeaways than those without (Safety, QoL). Figures representing pts without toxicity (positive reporting) versus pts with toxicity (negative reporting) were better understood (62% vs 37%) and communicated (67% vs 33%). Respondents (~55-60%) chose the correct key takeaway for QoL figures, not containing quantitative data, and were divided as to which version best communicated the key takeaway. There were no consistent, robust patterns of differences in accuracy of interpretation based on respondent characteristics, especially for top performing Efficacy and PSA graphics. Conclusions: This is one of the first prospective multistakeholder studies in PC to explore comprehension and preferences from a diverse group of pts and CGs. Figures showing a difference between treatments were more accurately interpreted than those conveying parity. We provide a benchmark for best practices for CER DV for pts and CGs to assist in their education and optimize shared decision-making.