Effects of Information Visualization on Older Adults' Decision-Making Performance in a Medicare Plan Selection Task: A Comparative Usability Study

Abstract

Background: Technology gains have improved tools for evaluating complex tasks by providing environmental supports (ES) that increase ease of use and improve performance outcomes through the use of information visualizations (info-vis). Complex info-vis emphasize the need to understand individual differences in abilities of target users, the key cognitive abilities needed to execute a decision task, and the graphical elements that can serve as the most effective ES. Older adults may be one such target user group that would benefit from increased ES to mitigate specific declines in cognitive abilities. For example, choosing a prescription drug plan is a necessary and complex task that can impact quality of life if the wrong choice is made. The decision to enroll in one plan over another can involve comparing over 15 plans across many categories. Within this context, the large amount of complex information and reduced working memory capacity puts older adults' decision making at a disadvantage. An intentionally designed ES, such as an info-vis that reduces working memory demand, may assist older adults in making the most effective decision among many options.

Objective: The objective of this study is to examine whether the use of an info-vis can lower working memory demands and positively affect complex decision-making performance of older adults in the context of choosing a Medicare prescription drug plan.

Methods: Participants performed a computerized decision-making task in the context of finding the best health care plan. Data included quantitative decision-making performance indicators and surveys examining previous history with purchasing insurance. Participants used a colored info-vis ES or a table (no ES) to perform the decision task. Task difficulty was manipulated by increasing the number of selection criteria used to make an accurate decision. A repeated measures analysis was performed to examine differences between the two table designs.

Results: Twenty-three older adults between the ages of 66 and 80 completed the study. There was a main effect for accuracy such that older adults made more accurate decisions in the color info-vis condition than the table condition. In the low difficulty condition, participants were more successful at choosing the correct answer when the question was about the gap coverage attribute in the info-vis condition. Participants also made significantly faster decisions in the info-vis condition than in the table condition.

Conclusions: Reducing the working memory demand of the task through the use of an ES can improve decision accuracy, especially when selection criteria is only focused on a single attribute of the insurance plan.

Keywords: Information visualization; aging; health-related websites; working memory.

Figure 1

Example layout of a low difficulty decision task in the table condition. Fifteen plan options are shown with four plan attributes (gap coverage, monthly premium, annual deductible, and satisfaction rating).

Figure 2

Color information visualization (color info-vis).

Figure 3

Percent accuracy on low difficulty tasks by plan attribute and decision aid. Error bars represent standard error of the mean.

Figure 4

Decision task time by decision aid for low and high difficulty tasks. Error bars represent standard error of the mean.

Figure 5

Mean decision time (in seconds) by plan attribute and decision aid for the low difficulty condition. Error bars represent standard error of the mean.