(70) Years of life saved by taking a drug

Download original image
Download PDF of design specifications, testing methods, and results.

The benefit of treatments to prolong life can be described in two main ways: 1) how many months or years they add to an average patient’s life, or 2) how many patients benefit. Many times, a few people benefit a lot but most don’t benefit at all. This image tries to show how the average benefit relates to how many patients benefit.

What’s special about this image?

This graphic has three distinct messages: The top graph shows the average life expectancy in a horizontal bar graph, and the orange circles represent the full population of patients. The middle graph highlights the large benefit achieved by 20% of the population (indicated by the 2 dots). The bottom graph shows that the remaining 80% of patients (shown by the 8 dots) get no benefit.

Note that the legend has the word “patients” misspelled. The graphic was tested as is, but this should be corrected before use.

In testing, this image performed particularly well at helping people realize that the most likely thing is that there will be no additional time benefit for most people. It also resulted in lower perceptions of the likelihood of adding time to one’s life and lower ratings on how good a choice the treatment is. We recommend this graphic because it performed best on the key dimension of clarifying how rare benefit is in this case. Its mildly lower accuracy rates on average time saved is more than compensated for by its good performance on other outcome measures.
  • My goal
  • Differences in likelihood
  • Details or gist?
  • Gist understanding
  • Data I have
  • Risk over time
  • Benefit estimate
  • Health conditions
  • Treatment choice
  • Graphic type
  • Bar graph
  • Icon array