Visualizing Personalized Health Risk Information to Enhance Understanding: A Mental Models and User-Centered Design Approach to the Design and Testing of Visualizations for Breast Cancer Screening and Treatment

The increasing availability of personalized health risk information through the rise of AI and prognostic models raises the question of how to present this information in an understandable way to citizens and patients. This thesis addresses this question regarding personalized risk information in two contexts: breast cancer treatment (Part A) and breast cancer screening (Part B). Personalized information in the context of screening involves risk-based screening, which means that models are used to predict an individual’s breast cancer risk so that screening intervals and methods can be tailored accordingly, improving the harm-benefit balance. This thesis addressed several factors assumed to influence people’s understanding of personalized health-related risk information (see Figure 2 in the General Introduction, Chapter 1). Specifically, the extent to which visualizations of both quantitative and qualitative risk information enhance comprehension was investigated. Potential differences in understanding among individuals with varying levels of skills, such as health literacy and numeracy, were also examined. To identify patients’ and citizens’ mental models and information needs and to include them in the development of the information materials and visualizations, involving the target group throughout the design process was central to the approach used. The aim of this thesis is twofold: 1. To investigate how personalized risk information in the context of breast cancer treatment and breast cancer screening could be designed and visualized using an approach combining mental models research and participatory design. 2. To investigate the effect of the designed information materials, including the visualizations, on comprehension among individuals with varying skill levels.