Interactive graphics for expressing health risks: development and qualitative evaluation

Abstract

Recent findings suggest that interactive game-like graphics might be useful in communicating probabilities. We developed a prototype for a risk communication module, focusing on eliciting users' preferences for different interactive graphics and assessing usability and user interpretations. Feedback from five focus groups was used to design the graphics. The final version displayed a matrix of square buttons; clicking on any button allowed the user to see whether the stick figure underneath was affected by the health outcome. When participants used this interaction to learn about a risk, they expressed more emotional responses, both positive and negative, than when viewing any static graphic or numerical description of a risk. Their responses included relief about small risks and concern about large risks. The groups also commented on static graphics: arranging the figures affected by disease randomly throughout a group of figures made it more difficult to judge the proportion affected but often was described as more realistic. Interactive graphics appear to have potential for expressing risk magnitude as well as the feeling of risk. This affective impact could be useful in increasing perceived threat of high risks, calming fears about low risks, or comparing risks. Quantitative studies are planned to assess the effect on perceived risks and estimated risk magnitudes.

Figure 1A

A random stick figure graphic used in focus groups; the blue figures represent those with the disease and are randomly distributed throughout the group. (The illustrations used in the focus groups showed 1000 stick figures; for clarity, illustrations in this article contain fewer than 200.)

Figure 1B

A sequential stick figure graphic; blue figures representing people with disease are arranged sequentially.

Figure 2

In this portion of the interactive prototype, viewers were shown a grid of green squares with one blue square (left). Using the mouse (depicted here by an arrow) to click on the blue square revealed the stick figure underneath as well as all the other figures. If the central figure was affected by disease, it appeared in blue; if not, it was green. (The software showed 1000 stick figures; for clarity, this illustration contains fewer than 200.)

Figure 3

In this potential modification, the viewer would click many squares, revealing the figures underneath one by one (left). Only when a blue figure is found would the entire grid of squares “flip” to reveal all the figures (right).