Do icon arrays help reduce denominator neglect?

Med Decis Making. Nov-Dec 2010;30(6):672-84. doi: 10.1177/0272989X10369000. Epub 2010 May 18.

Abstract

Background and objective: Denominator neglect is the focus on the number of times a target event has happened (e.g., the number of treated and nontreated patients who die) without considering the overall number of opportunities for it to happen (e.g., the overall number of treated and nontreated patients). In 2 studies, we addressed the effect of denominator neglect in problems involving treatment risk reduction where samples of treated and non-treated patients and the relative risk reduction were of different sizes. We also tested whether using icon arrays helps people take these different sample sizes into account. We especially focused on older adults, who are often more disadvantaged when making decisions about their health.

Design: . Study 1 was conducted on a laboratory sample using a within-subjects design; study 2 was conducted on a nonstudent sample interviewed through the Web using a between-subjects design.

Outcome measures: Accuracy of understanding risk reduction.

Results: Participants often paid too much attention to numerators and insufficient attention to denominators when numerical information about treatment risk reduction was provided. Adding icon arrays to the numerical information, however, drew participants' attention to the denominators and helped them make more accurate assessments of treatment risk reduction. Icon arrays were equally helpful to younger and older adults.

Conclusions: Building on previous research showing that problems with understanding numerical information often do not reside in the mind but in the representation of the problem, the results show that icon arrays are an effective method of eliminating denominator neglect.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Analysis of Variance
  • Bias
  • Communication
  • Decision Support Techniques*
  • Female
  • Health Education / methods*
  • Health Literacy
  • Humans
  • Male
  • Mathematics*
  • Middle Aged
  • Risk Reduction Behavior*
  • Risk*
  • Statistics as Topic
  • Visual Perception*
  • Young Adult