Article Text

Download PDFPDF
278 Assessing the impact of (interactive) visualizations of risk information on comprehension of risk-based breast cancer screening results
  1. Inge SVan Strien-Knippenberg1,
  2. Danielle RM Timmermans1,
  3. Maaike Weber1,
  4. Yasmina Okan2,
  5. Carla HVan Gils3,
  6. Mireille JM Broeders4,
  7. Olga C Damman1
  1. 1Department of Public and Occupational Health, Amsterdam University Medical Center, Vrije Universiteit Amsterdam, Amsterdam Public Health Research Institute, Amsterdam, The Netherlands
  2. 2Department of Communication, Pompeu Fabra University, Barcelona, Spain
  3. 3Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, the Netherlands
  4. 4Department for Health Evidence, Radboud University Medical Center and Dutch Expert Centre for Screening, Nijmegen, The Netherlands


Introduction We aimed to assess whether, in the context of communicating results of risk-based breast cancer screening, the use of visualizations results in better comprehension of: (a) numerical probability information and (b) qualitative information about risk concepts involved (e.g., risk factors).

Methods An online experiment using a 3 (information format: text – text with static visualizations – interactive) x 4 (risk category: slightly decreased, average, slightly increased, increased risk) between- subjects design. Dutch women aged 40–74 years participated, recruited by a panel. Women were assigned to a risk category and asked to imagine receiving results from risk-based breast cancer screening. Primary outcomes were verbatim and gist comprehension of numerical probability information and verbatim comprehension of qualitative risk information. Secondary outcomes included risk perception and evaluation of the information materials. Ordinal logistic regressions were used to assess the effects of information format and risk category.

Results In total 1,047 women participated, of which 413 (39.4%) had a low educational level. No significant differences were found in comprehension between the ‘text with static visualizations’ and the ‘text’ format (comprehension of numerical probability information verbatim OR .871 (95% CI, .580 – 1.051); gist OR 1.037 (95% CI, .775 – 1.388); comprehension of qualitative risk information OR .769 (95% CI, .590 – 1.002)). Verbatim comprehension of numerical probability information and qualitative risk information was worse in the ‘interactive’ format compared to the ‘text’ format (OR .479 (95% CI, .351 – .654); OR .507 (95% CI, .388 – .663)). Those assigned to a risk category other than average exhibited lower comprehension.

Discussion The lack of differences in comprehension between the risk visualizations and text format may be due to the quite optimal text format, the amount and difficulty of the information, or the way comprehension was assessed.

Conclusion(s) Visualizations may not enhance comprehension of risk-based breast cancer screening results beyond well-designed text formats.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.